The biggest news at Computex is official, as Nvidia’s CEO Jensen Huang just announced RTX Spark Super Chip family — the company's first-ever all-in-one laptop silicon that takes on Intel, AMD, and Qualcomm in a big way with a GPU equivalent to an RTX 5070.

Set to release this fall, this is not your ordinary ARM chip. The RTX Spark is packed with CUDA cores, which is sure to make that integrated GPU one helluva gaming monster. Plus, Spark is set to change the way you interact with Windows PCs forever with an AI agentic twist. I will be testing the chip extensively during my time out here at Taipei, and talking to a couple of the engineers behind it.

But for now, I’ve taken a deep dive into the specs tables and found what could be a breakthrough moment similar to when Apple brought the M1 to the world back in 2020.

Nvidia RTX Spark Super Chip specs

Nvidia RTX Super Chip devices

Nvidia is expecting a huge lineup of systems to launch with RTX spark — with plans to release over 30 laptop models and well over 10 desktop variations, targeting a range of price points.

But as of now, there are a whole lot of devices launching this fall with the RTX Spark Super Chip inside. On the laptop side, we have:

• Asus ProArt P14 and P16
• Dell SPX 16
• HP OmniBook
• Lenovo Yoga Pro 9
• Microsoft Surface Ultra
• MSI Prestige N16

These devices will be as thin as 14 millimeters, and as light as three pounds — a serious upgrade over the chunky Nvidia gaming laptops you're used to.

But they're not stopping at laptops. In an announcement that probably woke up Anthony Spadafora (our mini PCs king) from his nap, RTX Spark will be coming to small form-factor desktops that are "small enough to disappear on a desk."

These will be made by Dell, HP, Lenovo, Asus, MSI, Acer, and Gigabyte.

Explain it with pizza

This is partly inspired by the tasty ‘za I found while typing this, but also shoutout to Nvidia for specifically pointing out that I just “love to explain stuff with pizza.” You asked for it, and now you’re getting it!

In the laptop space, you’ve got various ways of serving pizza.

• The standard meal (an x86 SoC): This is like ordering a plain cheese pizza. It works perfectly fine for everyday eating, but if you want heavy flavor (think intense gaming or AI generation), it lacks all the necessary toppings. Don’t get me wrong, Intel and AMD have been adding some new garnish, but they’re still a little hampered by architectural limitations.
• The combo meal (x86 CPU + dedicated GPU): For that premium flavor, you have to order the standard pizza plus a massive side of premium toppings. Passing these ingredients back and forth across the table does the job, but it takes up way more space (think thicker gaming laptops) and requires a whole lot of power.
• The efficiency slice (Snapdragon X Elite): Qualcomm is already dealing in making efficient single-crust pizza using a similar recipe (Arm architecture). It’s great for saving power, but Nvidia's world-famous secret sauces (CUDA and RTX) can’t be matched.

So in this backdrop, I can only describe the RTX Spark Super chip as a “super pizza.” Instead of ordering a meal with lots of separate sides, this is the all-in-one slice baked onto a single, highly efficient crust. Half of the pizza is that custom CPU, and the other half is loaded with incredibly spicy, high-end toppings in that Blackwell GPU.

But the magic is in the cheese (the memory). Instead of dividing the cheese between the crust and toppings, Spark uses up to 128GB of “unified cheese” that the entire pizza shares instantly at blazing speeds (five times faster than standard PCIe Gen 5). And because it’s baked so efficiently, you can take this super pizza anywhere without it getting cold (all-day battery life).

…am I stretching this analogy? Absolutely. Do I regret it? No (hi Nvidia if you’re reading this).

Going agentic

With this increase in RTX Spark's brute force, Nvidia is working with Microsoft to shift personal computing on Windows from a passive tool to a proactive agent.

This is a one petaflop AI superchip... For context, some quick math I did while sitting in the keynote, the AMD Ryzen AI Max+ 395 can do around 60 Teraflops. This thing is built differently. To go back to the pizza analogy, not only can it make a mean slice, but it can probably feed it to you, too.

It can work across apps on your device — stringing together different tools like any user would to get the job done. For example, an artist can provide a sketch and a mood board, then use Photoshop not only to generate the art locally but also to turn it into a 3D image and animate it into a video. And this is all done without ever needing cloud compute.

Or if you’re a gamer, you can use G-assist to tweak critical settings like monitor optimization, or (if you’re streaming) tell it to operate your lights, mute the mic, and turn on your be right back scene in OBS.

Essentially, it can take complete control of your system — even the mouse and keyboard to navigate as a human user would. Keeping this all secure is Nvidia OpenShell, which brings enhanced security to ensure your model use works locally and privately. It can even mask personal information before ever using a cloud-based AI model to do anything.

For the players

But of course, there are big gaming capabilities here too. This is Nvidia, after all. Team Green says this falls into the same performance class as an RTX 5070 laptop GPU but uses significantly less power to reach that level.

That means you could be playing the likes of Cyberpunk, Doom, or Indiana Jones at 100 FPS at 1440p resolution. And given its ARM, the RTX Super Chip is capable of this level of performance both when plugged and unplugged.

Of course, battery life will vary based on how much you throw at it, but when I asked Nvidia, a tweak of maximum frame rate and graphics settings can optimize this nicely for long gameplay sessions on flights.

Now, what about the emulation part of it? General apps are included in this, too, but gaming has been the big test for Windows on Arm, and that Prism emulation layer used to translate apps written for other CPUs to this new architecture. Well, Nvidia’s thought of that too.

The company is working with a large list of developers to either port games natively to Windows on Arm or ensure they run well through the emulator. And with that hefty GPU, anything that’s particularly taxing on that will have almost no negative effect on performance.

Plus, Nvidia has collaborated with all first- and third-party anti-cheat software providers to ensure competitive multiplayer games launch without a hitch.

Nvidia RTX Spark vs Apple M silicon vs Snapdragon X2 Elite

So, since I’m calling this Nvidia’s M1 moment, how does it stack up with Apple laptop silicon? While I’m at it, I'm gonna throw Snapdragon into this comparison too, as they’re all similar-ish in Arm architecture.

But the crucial differences show that it’s not about the tools you have, but rather how you use them. You see, Apple silicon isn’t just about the brute force of the hardware; it’s also about its tight integration with software that extracts that performance. In gaming, for example, the MetalFX layer has enabled MacBook Pros to do some pretty great things, like run Cyberpunk 2077.

And while Snapdragon X2 Elite is level pegging in certain benchmarks, there are still some moments where it can feel like a bit of a blunt instrument against Windows 11 — especially in that GPU department.

But with RTX Spark, this is Nvidia we’re talking about. That integrated GPU supports DLSS 4.5, so you can bet your bottom dollar that the out-of-the-box gaming performance is going to be clearly superior to Apple and Snapdragon.

On top of that, Team Green and Windows are rewriting the rules on local agentic AI, so you’re getting a heap of AI coworkers too, with those CUDA cores for the raw brunt.

Nvidia RTX Spark vs Intel Core Ultra Series 3 vs AMD Ryzen AI 400 series

Why did I put these separate from the face-offs with Snapdragon or M5? Well, it comes down to the different ways these chips compute stuff. Nvidia, Apple and Qualcomm all have Arm architectures, which do things very differently to the x86 chips from Intel and AMD.

That being said, the 30+-year-old x86 architecture does bring some material benefits over ARM — namely, zero app compatibility issues whatsoever. It’s a challenge that Microsoft and Nvidia are overcoming with a separate Arm-specific version of Windows 11, alongside working closely with developers. That could narrow the library you can play!

However, for what it may lack in this area, it absolutely makes up for in a sweeter blend of performance and power efficiency. While x86 is all about multitasking its way to doing things you ask it to (not the best on that battery), ARM tackles one task at a time very quickly. That’s easier on the power while bringing high speeds to the table, too.

Outlook

It’s been a long road since the first time we heard rumblings about this chip. In fact, I even put this in my Computex 2025 predictions! But now it’s here, I think it’s fair to say that RTX Spark is far beyond any expectations I had.

Not only is it a beastly integrated GPU for monstrous performance and impressive power efficiency (especially in gaming), but the software tie-ins fundamentally change how a Windows laptop works with you. This is going to be an interesting few months as we count down to the fall launch.

Now the wait is on for an Nvidia RTX Spark gaming handheld…that would slap so hard!

More from Tom's Guide

• I just tested the MSI Claw 8 EX AI+ — Intel Arc G3 is a breakthrough for handheld gaming, but at what cost?
• Intel Arc G3 is here to shake up handheld gaming, and AMD should be terrified — Acer Predator Atlas 8, a new MSI Claw and more launching soon
• I built a touchscreen MacBook Pro using this snap-on accessory, and I'm baffled why Apple hasn't made its own for years

In all my time testing Panther Lake, it’s been clear that Intel Core Ultra Series 3 chips are huge for gaming — packing a beasty internal GPU with all that XeSS 3 AI trickery to boost frame rates to unseen levels in anything outside of a dedicated Nvidia RTX 50-series gaming laptop.

Team Blue knew they were onto something, and confirmed a different version of this chip with that same GPU will be coming to gaming handhelds. And now at Computex 2026, it’s finally happening. Meet the Intel Arc G3 and Arc G3 Extreme processors — you can expect full testing and a deep dive from me soon, but for now, here’s everything you need to know.

Intel Arc G3: Specs comparison

The deep dive

A lot more details (and hands-on time) will be shared at Computex, and you'll see these handhelds start to roll out in June, with broader availability "throughout the year."

Translation: expect to see Arc G3 and Arc G3 Extreme handhelds launch sometime in August.

These chips take that breakthrough integrated GPU from the Core Ultra X series chips — the 10-core Arc B370 for the G3, and the 12-core Arc B390 for the G3 Extreme — and stuff them into a handheld with a lower wattage.

The result? I predict shockingly good gameplay on the go with power efficiency at its heart (thanks to that improved power management and the new efficiency cores being able to take on more of the load).

But the advantages over AMD keep coming. Intel is able to precompile in-game shaders using cloud servers (no more waiting ages for your cloud shaders).

Oh, and XeSS 3 is the big thing here, with AI-based resolution scaling, multi-frame generation up to 4X, and low-latency tech to ensure those AI frames don't bog down your gameplay responsiveness.

What handhelds will pack Arc G3?

At the moment, we know of three.

Acer Predator Atlas 8

As the name suggests, it’s an 8-inch handheld that sports this new silicon to support ray tracing alongside all that XeSS 3 AI-powered goodness. To keep it cool, there’s the Predator AeroBlace thermal management with the first metal fan in a handheld for up to 10% better airflow.

“Handheld gamers want PC-class performance without being tied to a desktop or charger,” Jim Johnson, SVP and General Manager of Client Computing at Intel, commented.

“With Intel Arc G-Series processors and Acer’s Predator Atlas 8, we deliver smooth, high-fidelity gameplay with exceptional battery life in a form factor you can take anywhere.”

Speaking of not being tied to a charger, the Atlas 8 sports an 80Wh battery (for the G3 Extreme — the G3 gets a 60Wh cell), and games are sure to look pretty on that 1920 x 1200 panel with 120Hz refresh rate and a variable refresh rate.

As for the ergonomics, you've got full-size analog sticks, and (what looks like) well-placed buttons directly alongside them, as well as Hall Effect triggers for all your pressure variations in racing games (plus a micro-switch mode to make them instant clicks for shooters).

The others

However, when Intel made the announcement on stage at CES 2026, other brands were connected to it, and Team Blue has confirmed the MSI Claw 8 EX AI+ and OneXPlayer will be launching too.

So all eyes are on Computex as more are inevitably revealed in Taipei! I’ll update this piece when we know more.

Intel Arc G3 vs AMD Ryzen Z2 Extreme

On paper, if you were to look at the specs table, the numbers on that GPU are bigger on AMD’s silicon. But as I found out testing Panther Lake, it’s not about what you have, it’s about how you use it.

XeSS 3 has already blown my mind in a laptop, to the point that I begged Intel way back in October to bring this tech to the handheld space. The secret sauce is that resolution scaling and 4x multi-frame generation that extracts so much more performance-per-watt out of these machines. I tested it (in a laptop) against Nvidia’s DLSS 4, and my mind was blown.

At 1080p, there’s definitely a gap in frame rates, but nowhere near the kind of gap you’d expect in integrated vs dedicated graphics. To get something akin to this on something as small as a handheld is going to be a truly generational shift.

Throw in the fact that you’re only getting 44ms of latency on Intel vs Nvidia in-game, and you start to see just how earth-shaking this could be.

Now, what about AMD? Well, the Ryzen Z2 Extreme was a minor uplift over the Ryzen Z1 Extreme — it uses an RDNA 3.5 architecture, and at the moment, its own FSR resolution scaling and frame generation doesn’t hold a candle to XeSS.

That being said, Team Red did recently say that its latest AI-fueled FSR upscaling 4.1 will be coming to RDNA 3 GPUs. In July, this will be limited to just the desktop-class RX 9700 card, but given the cores are the same, this is AMD’s big chance to keep up.

Bottom line

You can expect more hands-on testing from me in the very near future as I head to Computex. But for now, it's fair to say that if you’ve been waiting for an actual next-generation handheld (maybe you’ve been holding onto that Steam Deck for a few years now), it’s happening now.

Intel Arc G3 and G3 Extreme are really bringing the heat in terms of a monstrous integrated GPU and XeSS 3 to deliver what will be one of the best handheld gaming experiences you can get.

One thing we don’t know is price, and given the massive price increase of Steam Deck recently, that makes me nervous. But time will tell on this around whether the price-to-performance is still there.

More from Tom's Guide

• Computex 2026 predictions: I'm hoping for Nvidia N1X, Intel Arc G3 handhelds, a new Asus ROG Ally and the industry's fightback against RAMageddon
• I can finally turn my MacBook Pro into a gaming laptop (sort of) — GameHub just fixed my Mac’s biggest weakness in minutes
• I’ve driven over 10,000 miles in Forza Horizon 6, and can confirm it is a dream game for car lovers

For three years, if you wanted a great desktop CPU for gaming, you bought AMD, period. Specifically, anything with “X3D” on the end of the name. And it’s been that way because Intel’s previous shot at desktop chips (the original Core Ultra Series 2 or “Arrow Lake”) dropped the ball.

There were performance and power issues, an abnormally high latency between the chip and memory, and all of this led to gaming performance that fell behind even Intel’s older CPUs. One thing was clear: Team Blue had to go back to the drawing board.

And they’re back with what seems to be a refresh on paper — the Intel Core Ultra 5 250K Plus ($199) and Core Ultra 7 270K Plus ($299). Strong pricing, but I was skeptical about what “plus” actually meant.

So I did what any normal person would do: I put them in an insane mismatch scenario against the AMD Ryzen 7 7800X3D and Ryzen 7 9800X3D. With more L3 Cache memory (AMD brands it as 3D V-Cache), you can always rely on these for strong, smooth game performance.

But Intel’s found a cheat code. Starting with the fundamentals of speeding up the Die-to-Die frequency (the travel time for data inside the chip), and optimizing it with an all-in-one Intel Platform Performance Package (IPPP) that brings so much more potential out of it, these chips come too close for Team Red’s comfort.

Don’t get me wrong, if gaming is your priority, AMD still takes the win. But only by a whisker, which is a concern because Intel’s chips are so much cheaper — the Core Ultra 5 being $250 less than the 7800X3D, and the Ultra 7 $180 less than the 9800X3D. And that’s before even getting into the data that shows the Core Ultras being fantastic all-rounders. But I digress, let’s get into it.

How Tom’s Guide reviews CPUs

We already test pretty much every CPU that’s announced in many different ways — either packed into one of the best laptops or being the mega brain inside the best gaming PCs.

But this is the first proper CPU review Tom’s Guide has done in its many years of existence. So, as Managing Editor of Computing, I’m gonna be upfront and set the ground rules on what to expect from CPU reviews from us. It all starts with answering the three key questions anybody has before buying one:

• How well does it play my games?
• What work can I get done with it?
• Is the price (short-term and long-term) fair for what I get?

That means taking a whole lot of testing, deciphering the advanced architectural and software processes happening here, and boiling them down to what they mean for you: the user. It won’t be complex, and that’s by design because I want to answer these three questions for everyone. Whether you’re an experienced PC builder, someone getting into making your first PC, or you’re buying a pre-built and making sense of the spec list.

If you’re looking for a super-dense level of detail, this is where I point you: my friends (and old family — used to work on their team) at Tom’s Hardware! This review is for the rest of us.

Intel Core Ultra 5 250K Plus & 7 270K Plus: Cheat Sheet

• What are these? These are a new line of Intel desktop processors — a refresh from the last gen that addresses a lot of the previous issues.
• Who are they for? Enthusiasts and creators who want a great all-around balance of productivity and gaming.
• What does it cost? Final retail pricing may vary (thanks, RAMageddon), but MSRP at the moment is $199 for the Ultra 5 and $299 for the Ultra 7.
• What do we like? These chips take a massive multi-threaded lead in productivity over AMD’s X3D chips in productivity and content creation, gaming performance has been upgraded nicely, and everything runs surprisingly cool.
• What don’t we like? Intel's use of a complex software stack like this performance package means you’re not getting peak power out of the box (a couple of extra steps to get it), and 1% frame-rate lows don’t match AMD’s hardware-cache-heavy design.

Intel Core Ultra 5 250K Plus & 7 270K Plus: Specs

Intel Core Ultra 5 250K Plus & 7 270K Plus: The ups

You could see what Intel was trying to go for with the O.G. Arrow Lake — a balance between work and play, which it didn’t deliver on. This is Team Blue’s comeback moment, and they’ve taken full advantage.

I’ve tested all four in their best possible light — enabling XMP for Intel and Resizable BAR for AMD. More on those in an upcoming piece because you could be leaving a lot of potential performance on the table if you don’t turn them on! This will cause some differences on a motherboard-by-motherboard basis (I used the Asus ROG Maximum Z890 Apex), and RAM speeds will give Intel a slight leg up. But as somebody who drops a lot of cash on your PC, these will be the settings you’ll go to to make the most of what you buy, so it’s only fair to do the same here.

Aiming for the king

So Intel calls this its “fastest gaming processor.” Let’s put that to the test, shall we?

What is fascinating here is that these are actually beginning to trade blows with X3D chips, and you can see that when put into real-world gaming performance, too.

Now you may be looking at the frame-rate gap AMD has over Intel, at Intel’s lead in a couple of 3DMark synthetic benchmarks, and thinking “WTF!?” Well, that’s where we’ve got to go a bit geeky:

• Monolithic vs tiled: AMD’s X3D chips use a single monolithic chip with 8 cores. This means all the cores can immediately communicate with the memory and the GPU with very low latency. Intel uses a “tiled” architecture — basically different cores of the chip do different things, and while that Die-to-Die (D2D) frequency is 900MHz higher, that’s still a longer time for cores to talk to each other, which can lead to lower frame rates (even when the synthetic speeds may be higher).
• The Cache gap: AMD’s chips pack 96MB of L3 cache (known as the 3D V-Cache), which acts as a massive waiting room for game data like physics and NPC logic. The Core Ultra 7 270K only has 36MB, so even though your chip cores are technically faster, it still needs to sort out an overflow waiting room on the system RAM, which is going to be slower.

There are situations where this can be felt most (more on them later), but the big thing here is that Intel is back on the right track in terms of gaming performance, with nice gains over previous-gen Arrow Lake chips and closing the gap on AMD.

Beasty productivity performance

Same as the previous chips, these are productivity monsters. Whether acting alone or as a conductor to the GPU, the performance and efficiency core combo makes better sense of both day-to-day tasks and intense workloads to deliver zippier speeds.

So whether it’s pushing local AI to its limits (currently training a version of myself…it’s getting weird), complex 4K renders or just simple multi-tab web browsing, these won’t let you down.

And it’s also being done while they keep their cool much better, too. At peak stress of a multi-thread render, you can see that the Core Ultras sit far prettier at around 154-degrees fahrenheit, while the thermal density of putting 3D V-cache atop AMD CPUs like a blanket of sorts does spice things up a little.

Solid bang for your buck

But going band-for-band misses the bigger point, which is the lower price here. Yes, AMD is still ahead in terms of gaming, but with that gap narrowed and the price of Intel’s options being kept quite lower, the performance-per-dollar swings to Team Blue’s favor.

And Intel’s able to do this with some key technologies that allow the chip to work smarter, not harder. Before the IPPP, you can imagine the CPU as a chaotic kitchen — getting tons of orders in and not necessarily knowing what order to tackle them in. The salad gets soggy under a heat lamp while the steak is raw, and Gordon Ramsay is kicking off somewhere about it.

But the IPPP pipeline acts as an executive chef standing at the pass and has a set number of steps to go through:

• The Application Optimization layer (APO): The kitchen has been trained on a small number of dishes, so when the chef sees a rush of orders for the “Wukong Special,” they know exactly what to do. So far, APO supports 21 game titles — shame that it’s so small, but only up from here.
• The Binary Optimizer Tool (IBOT): Instead of letting the cooks (the CPU cores) work randomly on different dishes, the Executive Chef intercepts the tickets and tells each cook to work on all of one thing at once. Machine code gets translated and streamlined in real-time to unlock better performance.
• Die-to-Die frequency boost (D2D): By increasing this by +900MHz, the executive chef just installed a faster conveyor belt in the kitchen, so the food moves quickly to the dining room (the GPU and RAM) without any bottleneck.

This is all opt-in, and I highly suggest you do to extract the most bang for your buck.

Intel Core Ultra 5 250K Plus & 7 270K Plus: The downs

But while Intel has royalty in its sights, there is one crucial place where it does fall down to AMD’s X3D silicon supremacy.

The 1% Low reality check

Testing game performance is much more than just telling you the average frames per second — you’ve got to look at the 1% Lows.

The intensity of graphics can turn on a dime from just looking at a plate on a table to a chaotic multi-person action scene, and while you can see the best FPS on average, measuring the average of the lowest 1% of frame rates shows you whether you’re in for a bit of stuttering in certain games.

For all the software trickery Intel’s done great work with here, it can’t overcome a pure hardware advantage. AMD CPUs have up to 3X the amount of memory cache on the chip, which forms a large hard-wired safety net for smoother gaming by putting essential instructions directly on the chip.

Meanwhile, the Core Ultra 5 and Core Ultra 7 suffer from a roughly 10-15% drop in 1% lows in comparison to AMD, leading to some small stutters that can be a problem in competitive titles.

Intel does have a couple of key things going on here to try to nullify this advantage — a fix to the latency with a 900MHz speed boost of Die-to-Die frequency (the speed at which data is moved around) and that software counter-attack with Intel’s Binary Optimization Tool (streamlining machine code in real-time to reclaim performance). This makes them much better than the Arrow Lake of old.

But Intel’s IBOT has a tiny bit of overhead, and if game logic gets chaotic, the instant response of a boatload of L3 cache is always going to win out. It’s not something that will overtly impact your gaming experience if you’re more of a single-player/cinematic player.

Intel Core Ultra 5 250K Plus & 7 270K Plus: Verdict

Intel aimed for its all-rounder era before and failed. Now, they’ve only gone and done it! Are they going to take the X3D’s crown for gaming? Not with those reduced 1% lows if you’re feeling competitive online. But for the prices that Team Blue’s charging, does it really matter? Not particularly.

Whether it’s in lower-priced pre-builts compared to AMD systems or in planning out your own DIY setup, this is a better balance of capabilities that makes these great CPUs for heavy work and cinematic play.

Windows PC makers are struggling through the ongoing RAM crisis, and a new report indicates it's not going to get better, thanks to an integral PC component getting more expensive.

The Korean outlet, ETNews is reporting that Intel will raise the prices of its CPUs by 10%, largely because of skyrocketing RAM prices due to AI greed. The publication says Intel informed its major clients of the price hike today (March 19) with increases expected to hit "most" of the company's CPU lineup.

It's suggested that the move is being made to secure profitability as the supply of consumer devices (ie gaming PCs, laptops, etc) shrinks due to the ever-growing demand of AI data centers.

Beyond RAM prices, Intel is facing stiffer competition from AMD and Intel. While Intel still dominates the CPU market, holding 60% of market share overall, AMD is catching up. According to charts from CPUbenchmark.net, Intel makes up 73% of Windows laptop share, but was surpassed by AMD last year in desktops at 51%.

This year, Qualcomm announced a new Snapdragon X2 Plus CPU that is supposed to be an "epic" upgrade. At CES 2026, AMD claimed its Ryzen AI CPUs are faster and better than Intel's Panther Lake for both gaming and AI.

Intel may be feeling the heat.

'Difficult to survive'

"There are concerns that if Intel CPU prices also rise, operating profits will shrink significantly, making it difficult to survive," an industry insider told ETNews.

Still, any price hike during RAMaggedon is going to adversely affect PC manufacturers who are struggling in the face of new competition from the $599 MacBook Neo and surging RAM prices.

Outside of the Neo, we may be seeing the end of the era of cheap PCs, with no relief insight until 2028, at the earliest. PC makers are trying different things to survive from seeking out smaller memory producers to focusing on premium PCs over budget offerings.

For now, the RAM crisis just keeps claiming victims. Even major companies like Samsung have put its mobile division into "emergency management" mode despite record pre-orders for the Galaxy S26 Ultra.

More from Tom's Guide

• Intel’s Core Ultra 200S desktop CPUs are 2x faster and $100 cheaper — a second chance at Arrow Lake's first bad impression
• Best graphics cards: I've tested every GPU to find the best bang for your buck
• I tested Intel XeSS 3 vs Nvidia DLSS 4: my laptop tests made me rethink the future of integrated graphics

As inevitable as day turns to night, the latest Apple silicon tests get leaked on Geekbench. This time around, it’s the M5 Max that has been exposed and as you probably expected, it’s an absolute monster.

But it’s easy to just see the M5 Pro and M5 Max MacBook Pros as simply “big number goes up” — I mean that’s exactly what my mates thought when I was trying to explain the new architectural changes while sharing a pizza.

“So, what’s the big deal then?” Ben asked (with an inappropriate amount of tomato sauce around his mouth). Then, I looked at the slice I was holding, and it hit me.

Apple calls it “Fusion Architecture,” and this is a pivotal moment where the company has stopped trying to make a bigger pizza, but build a buffet instead. This isn’t just a small change, it’s a generational leap that changes the way your computer thinks from having a single brain to a hive mind that can scale as big as the buffet allows.

Sounds weird, I know. But let me explain.

Why the monolith had to die

Intel names it “chips on a system” for its Core Ultra Series 3 chips — whatever it’s named, this is the direction everyone is heading in. And in trying to explain this to my friends, I went to the one dish I’m always thinking about.

Imagine you’re making a pizza. As more and more people love the pepperoni special you make, you have to go bigger. But you can only make a tasty ‘za so big before it won’t fit in the oven, or the middle stays raw while the crust burns.

Traditionally, Apple Silicon has always been a system on a chip (SoC): everything (the CPU, GPU and RAM) on one giant pizza. Great for efficiency, but limited in being able to scale up.

Chipmakers cannot print a single chip larger than a certain size without it becoming buggy and incredibly expensive. There’s a scaling issue (roughly 858mm-squared based on maximum exposure area of the lithography machines that make chips) and a quality yield issue:

• You could keep making the dough bigger, but your pizza oven is a fixed size.
• If you have a 5-foot monster pizza and a single hair falls on it, you have to throw the entire thing away.

So instead of trying to make one impossible pizza, Apple has decided to make four perfect 12-inch pizzas and “fuse” them together on one giant serving platter.

With M5 Pro and M5 Max, the Cupertino crew has gone back to the drawing board and rewrote the rules with chips on a system (CoS). Instead of just jamming it altogether, multiple specialized chips are fused together to act as one — tightly bonded so the software thinks it’s still just one chip.

• It’s much better for efficiency — if one pizza as a hair, you only need to toss that one small piece
• You still get that size advantage — you can still feed the masses a ton of pizza
• The Fusion Architecture magic — the “platter” these pizzas sit on is so fast that the person eating it (the software) can’t even tell they’re separate pizzas

Intel’s been the first big chipmaker to jump into this with Core Ultra Series 3, and not only has it made silicon production much more efficient for them, as you can see from numbers, it’s also brought a big performance/power efficiency boost with it.

And in Apple’s case, you can see how the floor of the M5 Pro chip has been raised, the ceiling of what a MacBook Pro can do has effectively doubled, and the architecture is now future-proofed with simplicity (all they need to do for an eventual “M5 Ultra” is just fuse more blocks onto it).

M5 Max: by the numbers

So how does this fusion architecture perform? We’ve got a solid idea from this Geekbench CPU test score.

That’s a 5.3% increase in single-core and 13% jump in multi-core speeds. Through this lens, it’s a minor jump, but then you take a step back and realize it’s actually faster than the M3 Ultra — Apple’s current workstation-class beast found in the Mac Studio.

Intel Core Ultra Series 3 is similar too, where the CPU speed increases are relatively small. But it’s in the GPU where you’ll see the biggest gains. Just like you saw with Team Blue, Apple’s touting some serious performance increases — all thanks to every single one of those 40 GPU cores having a neural accelerator in them:

• A 50% graphical performance boost
• 4x faster LLM prompting (making Siri and local AI feel instant)
• 3x faster video rendering in DaVinci Resolve (as the GPU is now AI-aware)

Also, shoutout to the 20% improved memory bandwidth to make things a whole lot faster, and the claimed 2x improvement in SSD speeds to 14.5 GB/s.

The modular future

At the beginning of this year, I went to an Intel event where they broke down the architectural shift of Core Ultra Series 3 by giving everyone a Lego set to build the chip. I thought this was a little weird at first (not a problem — love Lego), but the further we get into this year, the more I think it’s the perfect analogy.

We are entering an era of silicon Lego. This whole idea of waiting for a single giant chip is over, because the future is about how well you can glue smaller chips together. For the last few generations, Apple focused on making the “pizza” denser (moving from TSMC’s 4nm process to 3nm in chipmaking).

But we’ve reached a point where the ingredients can’t be made any smaller without things getting weird at a sub-atomic level. So by moving to chips on a system, M5 Pro and Max have broken the oven wall — not limited by how big a single pizza can be, but only by how many they can stick together.

Like I’ve been saying, 2026 is a huge year of change for laptops — with the biggest change happening in the brain of each system. And now with the ceiling virtually eliminated, the sky’s the limit for how far these performance gains can go.

Now, if you excuse me, I’m feeling hungry.

More from Tom's Guide

• MacBook Neo vs MacBook Air M1: There’s a clear (and colorful) winner
• MacBook Neo benchmarks leak — how does it compare to iPhone?
• Did you spot the MacBook Neo's clever message? It's hiding in plain sight

Nvidia GTC 2026 kicks off on March 16 and promises to deliver some big announcements from Team Green. While this is sure to be an AI-heavy show, we’re also expecting important hardware announcements, as CEO Jensen Huang recently teased that his company will unveil "several new chips the world has never seen before."

You may be wondering why you should tune into GTC, since it’s not as consumer-focused as CES. Even though there will be plenty of talk about data centers and AI, Nvidia is likely to drop hints or announce technology that could end up in the devices you use every day, like the best gaming laptops.

We’re covering all things GTC to keep you updated on Nvidia’s plans. Here are some of the biggest announcements we anticipate.

Nvidia’s mystery chip

Given Huang’s comments about a potentially revolutionary AI chip, we can expect a major reveal of a new inference-focused chip, possibly related to the Feynman architecture designed for faster AI tasks. This wouldn’t be a typical GPU, since it’s built for “agentic AI,” where systems act autonomously.

What does this mean for us? Imagine your phone or laptop running AI apps that can edit photos, summarize emails, or plan your day almost in real time. While these features exist, they can feel slow or unreliable. A chip like this could make AI more useful for everyday tasks, and do it all on-device instead of through the cloud. For gamers, it could lead to better upscaling tech, especially for lower- and mid-range PCs and laptops.

The Arm-based N1X CPU

Speaking about chips, there have been rumors about Nvidia’s N1X, an Arm-based CPU for the best computers and best laptops. But with N1X delays reported, GTC might only offer a sneak peek rather than a full launch. If this processor appears, expect discussions on integrated graphics that could challenge Intel’s Panther Lake.

Windows on Arm is now viable, thanks to Qualcomm’s Snapdragon X chips, so an Nvidia Arm chip would be huge. It could create a four-way race between Nvidia, Intel, Qualcomm, and Apple for who can make the fastest, most energy-efficient laptop chip. We might not get much N1X news, but it’s what I’m most excited about as a laptop reviewer.

Next-gen GPU tease

As we recently reported, Nvidia won't release new gaming GPUs in 2026, as they’re focusing on making chips for AI data centers rather than consumer GPUs. However, we might see some architecture teases.

I expect detailed looks at the Rubin GPU, which aims to provide five times the power of Blackwell for AI workloads. Feynman could be a significant leap and might use TSMC’s advanced 1.6nm process.

If you’re a PC gamer, pay close attention to what gets announced, as Nvidia’s data-center tech often influences what runs your PC. For example, Rubin could mean GPUs capable of 8K gaming or ray-traced worlds without overheating your system. Feynman’s efficiency could also improve laptop battery life. Plus, any new tech could lead to an upgraded version of DLSS, making games look and run better.

Physical AI and robotics

I expect Nvidia to also focus on “physical AI,” which combines simulation with real-world robotics. We saw some of this at CES, and I expect the company to expand on it at GTC. Look for sessions on AI factories and multimodal systems that combine vision, language, and action.

It sounds like sci-fi, but if you’ve seen videos of robots doing coordinated dancing, you know we’re getting closer to a Jetsons-like world. Maybe one day we’ll see affordable home robots that can clean, cook, or help those in need, all powered by Nvidia’s Omniverse platform. Nvidia will certainly paint an exciting picture, and I’ll be curious to see if any of it becomes real products.

Agentic AI

You’ll hear about agentic AI, where models go beyond chatbots to multi-agent systems that work together on complex tasks. GTC will likely include workshops on LLM augmentation, prompt engineering, and deploying inference at scale. Essentially, making AI more proactive rather than reactive.

While current LLMs like ChatGPT are useful, they’re not perfect or truly “intelligent.” Agentic AI could deliver genuinely smart AI systems that learn instead of just predict. We’ve heard promises of AI serving as virtual assistants, but these often need a lot of guidance, which defeats the purpose. I remain skeptical about seeing true agentic AI soon, but GTC might start changing my mind.

Outlook

Although Nvidia GTC 2026 will focus heavily on AI, it’s worth tuning in to see what Nvidia is working on and how it could benefit our work and gaming experiences.

Nvidia GTC runs from March 16 to 19, and we’ll be covering it all. The “GTC Live 2026” pregame show starts on March 16 at 11 am ET (8 a.m. PDT), before Jensen Huang delivers his keynote speech at 2 pm ET (11 a.m. PDT). I’ll be watching to admire his leather jacket and hear his insights on Nvidia’s latest plans. Stay tuned for more!

More from Tom's Guide

• It’s been 25 years since Nvidia GeForce 3 — and I think gamers accidentally built the AI era
• Resident Evil Requiem turned me into a path tracing believer — and Nvidia's DLSS 4 made it terrifying
• RAM prices keep going up — what is RAMageddon, and why is it getting worse?

Nvidia GTC 2026 is gearing up to be quite the show, with CEO Jensen Huang now hyping up the annual AI conference with "several new chips the world has never seen before." But what chips can we expect?

During an interview with media site Korean Economic Daily, Huang expressed that GTC 2026, taking place from March 16, will unveil "a chip that will surprise the world." While there's no indication of what this chipset could be, it's easy to think the upcoming Nvidia N1X for PCs is a likely candidate. However, this may not be the case.

Huang is teasing something big from Team Green next month, and while it's looking to be next-gen chips to power future AI data centers, it could be the beginning of the architecture that we'll see in RTX 60-series GPUs.

A mystery chip to 'surprise the world'

During the interview, Huang notes that "nothing is easy because all technologies are at their limits," but with the partnership between Nvidia and SK Hynix, one of the three semiconductor companies, "nothing is impossible."

This only makes Huang's statement of revealing "several new chips the world has never seen before" at GTC 2026 more bold, and while the long-anticipated Nvidia N1X and N1 chips for desktops and laptops are a good contender, looking to take on Intel's Panther Lake CPU lineup (especially in terms of integrated graphics performance), we've heard N1X is rumored to be delayed.

As with past GTC events, Nvidia generally focuses on the expansion of AI-driven software and chipsets, and we've heard about Team Green's new Vera CPUs and Rubin GPU that offers 5x more power than Blackwell to drive the AI data centers of the future at CES 2026.

This indicates we could see its new Rubin GPUs for AI take the spotlight at the show, as they've been in full production as of January 2026, as announced by Huang during CES 2026. However, Nvidia could throw a curveball by revealing its Feynman microarchitecture for next-gen GPUs.

That's considering Huang stating, "this team has worked incredibly hard to meet the great challenge of Vera Rubin and HBM4," referring to SK Hynix and its development of HBM4 (High Bandwidth Memory 4), designed for AI data centers and cutting-edge computing.

This remains to be seen, of course, but if Nvidia is set to unveil a world-surprising chip, then we can expect Team Green to bring the next evolution of implementing AI in the entire tech industry.

A glimpse of RTX 60-series?

With reports of Nvidia not releasing a new gaming GPU this year, this not only pushes back RTX 50 Super GPUs, but also RTX 60-series. And this is due to the RAM crisis. For consumers, this is disappointing, and having a new chip for AI data centers likely being the big surprise at GTC 2026 isn't ideal.

However, consider this. Nvidia Blackwell architecture (what you'll find in the latest RTX 50-series GPUs) was first designed for data centers, and then for consumer GPUs, around a year or two later. With this in mind, with the developments of Rubin now in place, we may see how this architecture comes to Team Green's next-gen graphics cards.

For now, we'll have to wait and see what Huang has up his (leather) sleeve, and it won't be much longer, seeing as Nvidia GTC 2026 is kicking off on March 16. Don't expect all-new RTX GPUs or N1X, but do expect a big reveal that could deliver a taste of Nvidia's consumer graphics cards lineup down the line.

More from Tom's Guide

• AMD just hinted the next-gen Xbox could ‘launch in 2027’ — here’s why I’m not convinced
• We just benchmarked the Intel Core Ultra X9 388H — AMD is officially on notice
• Panther Lake vs Apple M5 benchmarks — 'Intel has done the incredible'

Something interesting happened in AMD’s earnings call, and I’m not talking about the stock price dropping significantly. CEO Lisa Su started talking about its games console chip business, and let slip that “development of Microsoft’s next-gen Xbox featuring an AMD semi-custom SoC is progressing well to support a launch in 2027.”

She also mentioned the Steam Machine is still set for “early this year” too, but it’s that Xbox comment that piqued my interest, as this conflicts with what Microsoft has said in the past. So let’s break down what I predict will be in this next console, and why I’m skeptical about the launch year.

Predicting the next Xbox specs (way too early)

So as Xbox president Sarah Bond said, this is set to be a “very premium, very high-end, curated experience.” With a higher spec ceiling to reach (and a rumored cost of $1,000), that gives Microsoft (and AMD) a lot of room to work in to create what is set to be a console/PC hybrid.

I’m going to go through the rumors I’ve seen and given them a rating out of 10 on likelihood (in my opinion of course):

• AMD Zen 6 chipset architecture (7/10): This next gen CPU architecture is set to rollout in late 2026 — moving to TSMC’s new manufacturing processes to stuff more transistors on there for improved performance. In my mind, it seems likely.
• RDNA 5 GPU (6/10): Xbox Series X sports AMD’s RDNA 2 GPU architecture, which was Team Red’s graphics tech back in 2020. While I can see the logic here (RDNA 2 launched in early 2020 for PCs and came to consoles later that year), given the GPU squeeze at the moment for AI infrastructure, I can see the current-gen RDNA 4 coming instead.
• Up to 48GB of GDDR7 RAM (5/10): Casting serious doubt on this in the current RAM price crisis. Of course, I have my (rather selfish) prayer circle that this AI bubble bursts and components are reasonably priced again. But even with that in mind, I still anticipate 16GB of VRAM.
• AI and neural rendering (10/10): AMD has pivoted hard to AI trickery using FSR Redstone, and the PS5 Pro already does this. I would be shocked if the next-gen Xbox didn’t sport this.

Why I’m pressing ‘X’ to doubt

Well, my suspicions come from what Microsoft itself said. Back in the FTC vs. Microsoft court battle, documents revealed that the next Xbox (and PS6) were set for 2028. This “hybrid game platform” is set to combine local hardware and cloud computing.

Of course, this could’ve changed over the years since this court battle happened, and the “very high-end” comments from Bond does suggest that a pivot is on the table here. But let’s go deeper.

One thing that’s clear is RAMageddon is set to impact console gaming, and with next-gen systems, companies like Microsoft and Sony have a decision to make: Stick to this expected timeline and make it hurt, or wait it out for this DRAM drought to correct.

If I were a betting man, I’d go for the latter. But I’d love to know when you think the PS6 and next-gen Xbox will launch.

More from Tom's Guide

• 'The market environment is indeed challenging': Nintendo president sounds the alarm on 'pressure' caused by surging memory prices as Switch 2 forecast stays at 19 million
• AMD is cooked: Intel has ‘more planned’ for Core Ultra Series 3 in gaming handhelds — XeSS 3 multi-frame generation could be huge
• I tested Intel XeSS 3 vs Nvidia DLSS 4: my laptop tests made me rethink the future of integrated graphics

It's been a long wait for Nvidia's N1X and N1 CPUs, which have recently been confirmed by CEO Jensen Huang as a joint project with MediaTek. But we may be standing by a while longer until we see these chips finally arrive on PCs.

According to reputable leaker Moore's Law is Dead (MLID), Nvidia's N1X CPUs are suffering from bugs and software issues, leading to a delay in when we'll see Arm-based Windows laptops. Apparently, neither Microsoft nor Nvidia is rushing to make sure the SoCs (System on Chips) are fully supported on Windows.

Last year, rumors pointed to the N1X and N1 chips arriving in early 2026, but the recent development indicates that Team Green's CPUs won't launch on laptops until the second quarter of 2026, or even the summer.

Moreover, a previous leak claimed Microsoft's next-gen OS had caused an N1X CPU delay, so there may be ongoing compatibility issues with Team Green's Arm-based chip and Windows that both tech giants need to solve.

For now, this leak may specify why we haven't seen the launch of Nvidia's next-gen chip yet, especially during CES 2026. Initially, it was believed Team Green and MediaTek would announce the CPUs during Computex 2025. Now, it's looking like the reveal may happen during Computex 2026.

Nvidia N1X laptops are coming

While it's still unconfirmed when Nvidia will finally unveil its N1X and N1 chips in upcoming PCs, we at least have an idea of what laptops will come equipped with the all-new CPUs — if leaks are to be believed.

So far, we've heard that Dell's gaming brand Alienware will be among the first to launch new gaming laptops featuring the Nvidia and MediaTek CPU. The chips are expected to deliver the same performance as an RTX 4070-equipped laptop, after all, and we may see Alienware's new ultra-slim gaming laptop come with an N1-series chip.

According to tipster Huang514613 on X via VideoCardz, we should also see Nvidia N1X and N1 SoCs in upcoming Lenovo laptops, with product coding pointing to the launch of the chips. This includes the Lenovo IdeaPad Slim 5, Yoga Pro 7, Yoga 9 2-in-1 and Legion 7.

Not only does this indicate Team Green's N1X and N1 chips are both for laptops (previously, the N1X was indicated to be a CPU for desktops), but it also shows Nvidia's SoCs are for gaming and productivity laptops alike.

Of course, we'll have to wait and see when Nvidia and MediaTek's AI chips arrive, and the PCs we can expect them in. In the meantime, it's looking like we'll have to wait until summer before we see Nvidia enter the CPU market for consumers, and that gives more breathing room for Intel's Core Ultra 3 CPUs.

More from Tom's Guide

• RAM prices are rough right now, but I found 5 DDR5 kits that beat the crisis
• New leak claims Intel Panther Lake laptops could cost up to $2,400 — and my wallet is already crying
• I ditched my regular 'Pro' laptop for a gaming laptop for work — and my productivity just doubled

After months of Nvidia's N1X and N1 chips rumors, CEO Jensen Huang has finally confirmed its Arm-based CPUs designed for AI PCs in collaboration with MediaTek — and it's already taking aim at Intel Core Ultra Series 3 and AMD Ryzen AI 400.

In an interview during Nvidia's year-end party in Taiwan, Huang commented on work with MediaTek for the hugely anticipated next-gen N1X and N1 chips, which have been tipped to power AI-based desktops and laptops, respectively. According to Taiwan media outlet UDN, Huang states these CPUs have been specifically geared for PCs with "powerful AI capabilities."

The report notes that Nvidia's collaboration with MediaTek will offer a System-on-Chip (SoC) with "low power consumption but excellent performance," which recent rumors back up. Leaks indicate that N1X will be closely tied to Team Green's GB10 Superchip found in the DGX Spark, while the N1 for laptops may come in the form of a cut-down version of GB10.

With the joint project with MediaTek confirmed, Nvidia's N1X and N1 SoCs already look to stir up the CPU competition. We've heard that the N1 series will reportedly deliver the same performance as an RTX 4070-equipped laptop, but with improved power efficiency, taking integrated gaming performance in PCs even further.

While Jensen doesn't comment on when we'll see these Arm-based chips in upcoming PCs, reports have suggested an early 2026 launch, with wide availability later in the year. Other rumors believe the N1-series chips were delayed, pushing back the release to late 2026. We've heard that Nvidia's Arm-based SoC may power Dell's Alienware laptops down the line, and it could match the timing for Alienware's new ultra-slim gaming laptop.

However, with Computex 2026 coming up, this may set the stage for Nvidia and MediaTek to reveal their AI-focused CPUs.

Nvidia is coming for the CPU crown

Since benchmarking the Intel Core Ultra X9 388H, Team Blue's flagship laptop processor, we've been hugely impressed by the performance and power efficiency gains the chip delivers. Especially when it comes to Intel XeSS 3 and its integrated graphics for gaming — seeing Cyberpunk 2077 at High settings reach 217 frames per second is a highlight.

Without XeSS and its multi-frame generation or super resolution, Intel's CPU is closer to the performance of an RTX 4050. Still impressive, considering this comes from Arc B390 integrated graphics on the chip, but if rumors ring true and Nvidia's N1-series reaches RTX 4070 levels of performance, that's a significant gap to beat.

Rumors indicate that the N1 chip is expected to use 65W power to match the performance of a 120W RTX 4070 gaming laptop, and another source suggests the chip would offer a TDP (Thermal Design Power) of 80W to 120W. Intel's Core Ultra 3 chips have shown they thrive on low power consumption, too, with the Asus Zenbook Duo we tested hitting 45W.

For now, it's guesswork on what the N1X and N1 CPUs will deliver, but it appears to take the fight to Intel's impressive offering, along with AMD's Ryzen AI 400 series, according to the benchmarks we've seen in the Asus Vivobook 14 we tested.

The CPU wars are heating up this year, and if the ongoing RAM shortage doesn't get in the way, your next laptop or desktop is looking to see a significant upgrade.

More from Tom's Guide

• RAM prices are rough right now, but I found 5 DDR5 kits that beat the crisis
• AMD is cooked: Intel has ‘more planned’ for Core Ultra Series 3 in gaming handhelds — XeSS 3 multi-frame generation could be huge
• 'Google’s leaked Aluminum OS looks like the ChromeOS I've waited years for': why I might actually ditch Windows 11

We just tested Intel Panther Lake, and I’m here to tell you it’s the real deal. After years of Intel playing catch-up, this new architecture feels like a significant leap forward, offering the kind of raw performance and longevity we’ve long wanted from a Windows laptop. It's a night-and-day improvement over the already solid Lunar Lake, and shows that Intel isn't just treading water anymore.

Though Panther Lake has made great strides in performance and efficiency for the best Windows laptops, it has a mighty competitor—namely, Apple’s M5 chip. As I noted in my MacBook Pro M5 review, Apple’s silicon delivers stellar performance and phenomenal battery life. That’s not surprising, given how M-series chips have been the gold standard in those areas, but it makes for a formidable opponent for Intel to tackle.

We’ve officially tested the svelte Samsung Galaxy Book6 Pro with its Intel Core Ultra X7 chip and the dual-screen Asus Zenbook Duo (2026) with the beefier Core Ultra X9. In this piece, I’m going to compare our lab results for these Windows machines against our Apple M5 benchmarks.

Since we're comparing x86 architecture to ARM, I’ll also dive into what this means for app compatibility and your daily workflow. So, is Panther Lake a true M-series rival? Read on to find out!

Performance benchmarks

Let’s start with the raw numbers. I’m looking at our Geekbench 6 CPU scores and our Handbrake video transcoding results. Geekbench measures a chip’s overall "brain power," while our Handbrake test shows how long it takes to convert a 4K video into 1080p.

As you can see, both Panther Lake chips hold their own against the Apple M5 in the multi-core portion of the test. However, the M5 absolutely crushed Intel’s chips in single-core performance.

So, what does that actually mean? Think of single-core performance as having one world-class chef who can whip up a single complex dish really fast. Conversely, multi-core performance is like having a team of cooks working together to finish a big banquet. For daily tasks like browsing the web or opening Word, that high single-core score on the M5 makes everything feel just a bit snappier. But for heavy-duty multitasking, Intel is in the same league.

As with single-core performance, the M5 chip pulled ahead in the Handbrake test, clocking in almost a minute faster than the Core Ultra X9. In the real world, this is a big deal for anyone who does even light video work. If you're exporting a vlog for YouTube or shrinking a movie file for a long flight, the M5 will save you time.

Graphics benchmarks

Next up, let’s look at one of my favorite things: graphics. I’ve listed the results from 3DMark Steel Nomad and Solar Bay. Steel Nomad measures standard gaming performance at 4K, while Solar Bay focuses specifically on Ray Tracing — the tech that makes lighting and reflections in games look realistic. I chose these newer tests over the older Wild Life Unlimited Extreme ones because, frankly, if you’re buying a 2026 laptop, you want to know how it handles modern tech.

The Apple M5 chip scored the lowest here, though its Solar Bay results are still commendable since that test is notoriously heavy on Ray Tracing. Still, Panther Lake has a clear advantage over the M5 when it comes to gaming performance. That's a huge win for an integrated GPU (iGPU) that doesn't rely on a bulky, dedicated graphics card.

What’s really interesting here is that Intel’s Xe3 architecture is finally delivering on the promise of real gaming on a thin-and-light laptop. While the M5 is great for creative work, Windows still has the edge for gaming compatibility.

Seeing the Asus Zenbook Duo and the Galaxy Book6 Pro consistently pull higher frame rates means you can actually play modern titles at decent settings without things turning into a slideshow. Intel has essentially turned these productivity machines into "stealth" gaming rigs.

Battery life

Here’s where things get really fun: battery life! Please note that this comparison isn’t exactly "apples-to-apples" (pun very much intended) since we’re dealing with different hardware configurations. Also, keep in mind that both Panther Lake laptops we tested feature OLED panels, which are gorgeous but notorious battery hogs.

As you can see, the MacBook Pro M5 outlasted its Panther Lake-powered rivals by over three hours! These results might seem unfair, and I’d agree with you if these were typical Windows laptops that lasted for eight hours at best. However, getting nearly 15 hours of battery life from a Windows machine is insanely impressive. It’s a testament to Panther Lake’s endurance and proof that Intel has finally closed the gap that once made MacBooks the only real choice for all-day use.

Compatibility: x86 vs. ARM

To understand these results, you have to look at the engines under the hood. Intel uses x86 architecture, while Apple uses ARM. For years, ARM (originally designed for phones) had the advantage because it was built for efficiency — it stayed cool and sipped battery. x86 was the gas-guzzler: powerful, but hot and power hungry.

The advantage of x86 is its wide compatibility. Every app, game, and legacy program ever made for Windows runs natively on Panther Lake. The disadvantage has always been the heat and battery drain. However, with Panther Lake, Intel has finally figured out how to make an x86 chip act like an ARM chip, giving you that 15-hour battery life without sacrificing your ability to run any software you want.

ARM, on the other hand, is the king of performance per watt. It’s why the M5 can last 18 hours while still crushing single-core benchmarks. The downside? If an app isn't built specifically for ARM, it has to be "translated" (using software like Rosetta 2), which can occasionally cause a performance hit or compatibility glitches. For the average person, Apple’s ecosystem is seamless, but for those who need niche Windows software, the x86 architecture is still the safest bet.

Bottom line

So, is Panther Lake a true rival to Apple's M-series? Absolutely. For the first time in a long time, buying a Windows laptop doesn't feel like a compromise on battery life.

Intel has managed to pull off something incredible here: they’ve delivered a chip that can go toe-to-toe with Apple’s efficiency while still offering the raw graphics power that Windows users crave. Even though Windows 11 can still feel bloated compared to the streamlined macOS, Panther Lake makes the OS feel faster and more responsive than it would with a less powerful processor.

I’m genuinely impressed with how far Intel has come with its "M1 moment." The Samsung Galaxy Book6 Pro and Asus Zenbook Duo are not just awesome Windows laptops; they're great laptops, period. They set a high standard for all the other Panther Lake laptops I’m sure to review in 2026. Color me excited!

That said, Intel isn’t in the clear yet. While Team Blue has shortened the gap with Apple, I’m eager to see how Panther Lake stacks up against the Snapdragon X2 Elite. Qualcomm’s ARM-based Windows chips are the next big rival, and if they can beat Intel on battery while matching them on performance, we might be looking at a three-way silicon war. That’s a story for another day. But for now, Intel is officially back in the game.

More from Tom's Guide

• Forget Nvidia: Integrated graphics could be the future for gaming laptops
• Intel confirms XeSS 3 Multi-frame generation is coming to Arc B580 next month
• I just tried Apple Creator Studio — and I’m never going back to Adobe

The future of handheld gaming is shifting, with the new Intel Panther Lake and its impressive Core Ultra X9 388H CPU leading the charge. For gaming handhelds, this is looking to be a game-changer, and it's already beating AMD where it matters most.

With its next-gen XeSS 3 (Xe Super Sampling) and multi-frame generation tech, Intel has made significant gains for its Arc B390 iGPU — pushing integrated graphics to new highs on laptops. And, after benchmarking the Asus Zenbook Duo with Intel Core Ultra X9, the results are clear.

While seeing Cyberpunk 2077 running at 217 frames per second (FPS) at 1080p with Ultra settings impresses, the TDP (Thermal Design Power) it uses to reach these results will be massively beneficial for the future of gaming handhelds. In fact, it's seeing better performance than AMD's Strix Halo (Ryzen Max AI+) when it comes to low power consumption.

Not only does it outclass the AMD Ryzen Max AI+ in a Framework Desktop in single-core (see the table below), but recent benchmarks from YouTuber The Phawx have shown the Intel Core ultra X9 388H delivers stronger performance at lower TDP.

According to the results, the X9 CPU with its integrated Arc B390 graphics was able to hit 46.6 FPS at just a TDP of 15W in Black Myth: Wukong with 720p resolution at low settings. Compared to the AMD Strix Halo (featuring integrated Radeon 8060S graphics), this reached 39 FPS with the same configuration.

Not only this, but Team Blue's Core Ultra X9 388H even managed to reach 30 FPS at just 10W. Considering the ROG Xbox Ally X with its AMD Ryzen AI Z2 Extreme offers up 16.6 FPS at 35W TDP in Cyberpunk 2077 at 720p and Ultra settings, at 11.9 FPS at 17W, this shows the gains Intel has made with its CPU for handheld gaming.

This supports the benchmarks we've seen in Intel's Core Ultra 3 X9 chip in the Asus Zenbook Duo, with it using 45W as opposed to an Acer Predator Triton 14 AI with RTX 5070 using 140W. Of course, with its discrete GPU, Nvidia's graphics card offers better performance, but with lower TDP, Team Blue's chip still impresses.

That's a promising outlook for upcoming gaming handhelds, and Intel has previously confirmed it will be launching a suite of handheld gaming consoles boasting Intel Core Ultra Series 3 at CES 2026.

"They are the testament of our strategy to bring what the market requires," an Intel spokesperson told Tom's Guide. "A powerful CPU, a powerful GPU, packed with great efficiency within the same solution. With our 12Xe configuration, we deliver performance that approaches discrete GPU levels in thinner, lighter, and more portable devices. This meets the needs of customers and reflects our strategic commitment to rapidly advancing the capabilities of integrated GPUs."

Multi-frame generation is a big win

Aside from improved power efficiency, Intel's XeSS 3 is already knocking it out of the park with its frame generation and resolution scaling.

This would be a major perk for upcoming gaming handhelds, and while Intel confirmed to Tom's Guide that it's coming to the desktop GPU Intel Arc B580 next month (making it the only GPU we'd recommend given the RAM price crisis happening), we asked them about multi-frame generation coming to gaming handhelds.

"Gaming handhelds are an exciting and growing segment where we have more planned, featuring Intel Arc graphics. We’ll share more later this year."

Not much to go on, but considering these chips for gaming handhelds will "feature Intel Arc graphics," in other words, the latest XeSS technology we've seen in the Arc B390, it's looking likely we'll see huge gains in frame generation in gaming handhelds, too.

Combine this with the Core Ultra X9 388H's significant performance and power efficiency gains we've seen, and the future of the best gaming handhelds is looking bright. While AMD claimed its Ryzen AI CPUs beat Intel Panther Lake for gaming and AI, we'll see what Team Red has to offer when it comes to handheld gaming.

More from Tom's Guide

• Forget Nvidia: Integrated graphics could be the future for gaming laptops — here’s why
• I just tested the Samsung Galaxy Book6 Pro — and Panther Lake is the real deal
• ‘Chaotic’ RAM pricing won’t kill PC gaming, a CEO told me — and history backs him up

AMD is taking shots at Intel’s upcoming Panther Lake CPUs. Just ahead of the Intel Core Ultra Series 3’s launch, AMD is showing how its Ryzen AI MAX and Ryzen AI 400/300 mobile CPUs stack up against Panther Lake. As you’d expect, AMD’s infographics paint Team Red’s processors as the superior option.

In the chart below (which we received via email from AMD), you can see that AMD places its CPUs and Intel’s into four categories: Premium, Thin & Light, Mainstream, and Entry. Unsurprisingly, since this is an AMD deck, the company's processors effectively win across the board here for content creation, gaming, and AI.

AMD also takes time to debunk Intel’s claims, saying it is expected to have “the fastest processors for AI PCs.” For instance, the AMD Ryzen AI Max 395+ supposedly has 37% faster graphics and twice the processing threads than Intel’s Core Ultra X9 388H CPU.

AMD also said that Intel’s own data shows “almost no advantage” in performance and efficiency over Lunar Lake (Core Series 2), and that it expects “major processing and graphics wins” with its AMD Ryzen AI 400 series chips.

Since all of this is based on AMD’s internal testing and projections, we shouldn’t take it as gospel. Based on what I’ve heard, it seems that Panther Lake will be the real deal in terms of performance and efficiency. Though the jump from the original Core Ultra to Core Ultra 2 might not have been monumental, Series 3, or Panther Lake, could make Intel’s CPUs on par with the likes of Qualcomm’s Snapdragon X or even Apple’s M-series chips. That said, we also can’t rule out an AMD upset, especially if the company’s claims are true.

With the first Panther Lake laptops launching next week, we should expect reviews to paint a clearer picture. We at Tom’s Guide will conduct our own testing, so be sure to come back here to see what the results are. As always, stay tuned for more!

More from Tom's Guide

• Intel's Panther Lake chip graphics look 50% faster in early benchmarks
• I just got up close with the first Intel Panther Lake laptop
• Intel Panther Lake could power the future of handheld gaming

Asus pulled out all the stops with its lineup of next-gen gaming laptops at CES 2026, with its bonkers dual-screen Asus ROG Zephyrus Duo and a Death Stranding-inspired ROG Flow Z13. But one key device surprisingly didn't make an appearance: the Asus ROG Zephyrus G14 with Panther Lake.

In many ways, this is Asus' flagship machine that's one of the best laptops around, and while there were updated machines with RTX 50-series, the leaked integrated model was canned. Well, sort of, as we got a chance to see the ROG Zephyrus G14 with an Intel Core Ultra Series 3 chip during a hands-on demo in December, and there was no discrete GPU in sight.

Sporting an Intel Core Ultra X9 388H with integrated Intel Arc B390 graphics, this would be the first Zephyrus G14 without a GPU. As with the current Asus ROG Zephyrus G14 (2025), these laptops usually come with graphics cards now up to an Nvidia GeForce RTX 5080. However, with the power to game on an Intel Panther Lake chip, Asus was looking to shift gears.

For now, though, this wasn't meant to be. The Asus ROG Zephyrus G14 with an Intel Core Ultra Series 3 chip mysteriously vanished despite an upcoming reveal, but we now have some insight behind its apparent cancellation.

What happened?

During the lead-up to CES 2026, we got a preview of the ROG Zephyrus G14 with Intel Arc B390 power during an Asus hands-on event in December. Because of this, we were anticipating its big reveal during the CES showcase.

However, at the last-minute, Asus pulled the laptop from the show. It was briefly seen at Intel's demo booth, but was reportedly later removed from the area and wasn't seen again (as told by French outlet Les Numériques).

Many believe the ROG Zephyrus G14 was cancelled or at least postponed due to supply chain issues, with Asus later confirming a "change in strategy" for the laptop. Due to this, it is now looking to be cancelled.

After hearing about the Intel Core Ultra X9 G14's cancellation, we started asking around and found other companies confirming a struggle to nail down supply. So our own Jason England put the question to Intel themselves in an interview that will be published soon.

"A perfect combination of industry-level component shortages, significantly increased demand because of how good the chips are and also ramping up production, you're getting mixed signals depending on where the different partners are," Nish Neelalojanan, Senior Director of Product Management at Intel, commented.

"Good news is there are plenty of X9 designs available. So as a gamer, you will be able to go buy an X9 system from your favorite retailers. There will be availability."

While we will see laptops equipped with an Intel Core Ultra X9 chip for integrated graphics gaming performance, it won't be from the Asus ROG Zephyrus G14 right now. But perhaps this wouldn't have been the best fit after all.

Not the right fit?

During an interview with Asus, I asked whether we'd see integrated graphics on laptops, with Asus' Director of Technical Marketing Sascha Krohn stating: "I would say we're definitely getting into the territory where that becomes a possibility." However, the timing may not be right for the ROG Zephyrus G14.

It's clear that gaming on an Intel Core Ultra Series X9 or X7 processor has seen significant improvements in integrated graphics, being able to play the likes of Battlefield 6 at over 140 FPS at max settings without a discrete GPU. However, it still doesn't reach the heights of an RTX 50-series GPU, and the ROG Zephyrus G14 is known for its top-level performance in a compact design.

Asus may have changed plans to make sure its flagship gaming laptop still offered plenty of power to run the latest PC games at a high level of settings, but as Intel states, demand for the chip would mean a very limited supply for the G14, too.

However, this doesn't take away from the perks of an Intel Panther Lake chip being used in a laptop that isn't just made for gaming, but also for productivity, too. After testing Intel Panther Lake, we've seen major gains in power efficiency, and this should result in machines with gaming performance that have a much longer battery life.

At the very least, we will be seeing an Asus TUF Gaming A14 with an AMD Strix Halo chip, giving us powerful integrated graphics in the company's more budget lineup of laptops.

More from Tom's Guide

• I just went eyes-on with Nvidia G-SYNC Pulsar, and now my own gaming monitor feels obsolete
• 'Basically a photograph interacting with you at 500 frames per second': Nvidia CEO Jensen Huang shares his vision of the future of PC gaming, and how the company plans to tackle GPU pricing crisis
• I just tested Lenovo’s Legion Pro Rollable concept, and now my gaming laptop feels boring

Intel showed off how it raised the bar on integrated graphics in laptops at CES 2026, thanks to the arrival of its Intel Core Ultra Series 3 CPUs. Now, after seeing the latest PC games like Battlefield 6 pushing well past 120 frames per second (FPS) at max settings on a laptop not built for gaming, I'm blown away by the results.

No discrete GPU here, just a chip using Intel Arc B390 graphics (a.k.a. 12 Xᵉ cores) that features everything you'd expect from the latest graphics cards. That includes what Intel calls "Modern Rendering," with XeSS 3 tech featuring ray tracing, AI-based upscaling and multi-frame generation up to x4. If anything, it puts Nvidia's lower-end RTX 50-series GPUs on notice.

With Intel Core Ultra X9 and X7 processors, we're getting a 77% bump in faster graphics and 53% increase in faster AI compared to the previous Lunar Lake chips. That's a significant jump, and the end result is thin and light laptops being able to play the latest big-budget PC games without needing a dedicated graphics card — and at a high level.

I got a chance to try out many of the latest titles during a hands-on demo, and I'm already impressed by the high-level performance and smooth frame rates Intel's next-gen Core Ultra Series 3 chips bring to laptops. One thing's for sure: your next gaming laptop may not need a dedicated GPU after all.

A major leap for integrated graphics

As Intel boldly claimed during its big keynote, the Arc B390 in Core Ultra X9 and X7 chips will see Battlefield 6 at Overkill settings (the highest visual settings) and Super Resolution running at 59 FPS. Turn on its multi-frame generation tech, though, and this can be upscaled to a super-smooth 145 FPS.

That was fully on show during a brief demo I tried. Now, Battlefield 6 was already a well-optimized game when I tested it on an RTX 5070 gaming PC, where I was getting over 200 FPS at 4K with DLSS 4 switched on. However, this doesn't take away from how well a laptop sporting a Core Ultra Series 3 chip can handle it, as I was seeing detailed visuals and buttery-smooth framerates as explosions and gunfire went off all around me.

But there was more on show. I also got to try Dying Light: The Beast, where I saw framerates reaching 196 FPS at what appeared to be high settings. As expected, it was stunning running around, dropkicking zombies with beautiful visuals thanks to ray tracing.

Also on show was Clair Obscur: Expedition 33, Marvel Rivals and Painkiller, and while I couldn't see their frame rates or settings, Intel has confirmed that we can expect over 100 FPS with 1080p at high settings using XᵉSS Super Resolution.

Not only does it make playing competitive shooters on an Intel Core Ultra X9/X7-based laptop a breeze, but these new chips can also offer upscaled visuals in demanding PC titles — and that x4 multi-frame gen tech certainly helps.

Outlook

If there's one laptop I'm keeping an eye out for in 2026, it's one boasting an Intel Core Ultra X9 or X7 processor. Team Blue has pushed integrated graphics on its next-gen chip even further with Intel Arc B390, and from the games I've played, it's looking to kick off a new era of PC gaming on thin and light laptops.

Of course, we'll have to put these Intel-based machines through their paces in our testing labs once we get our hands on them, but it's looking like we could see these laptops creep into our list of the best gaming laptops — especially for gamers on the move (and if the price is right, no less).

Speaking of, with Intel also announcing an entire handheld gaming platform with Panther Lake chips, gamers will have a lot to look forward to this year.

More from Tom's Guide

• Dell XPS is back! I went hands-on with the new XPS 14 — and it's the best kind of apology
• I just saw Alienware's new ultra-slim gaming laptop at CES 2026 — and it's nearly as thin as a MacBook Air
• Qualcomm announces Snapdragon X2 Plus CPUs at CES 2026 — and it promises an ‘epic’ upgrade

2026 is the year of the laptop, and the CPU wars are critical to that. I’ve tested Intel’s Core Ultra 300 Series chips and Qualcomm Snapdragon X2 Elite, but there’s one more player in the Windows laptop space.

AMD is back and the rumors are true — Ryzen AI 400 Series chips (codenamed “Gorgon Point”) have just been made official at CES 2026. Here’s everything you need to know about the new silicon from Team Red.

AMD Ryzen AI 400 Series Processor specs

The same, but much better

With Ryzen AI 300 Series chips (codenamed “Strix Point”), AMD tore up the rulebook and overhauled its architecture to point their laptop chips at zippier performance, better local AI, improved graphics and power efficiency.

And now, Ryzen AI 400 Series (“Gorgon Point”) looks set to take that to a new level with a suite of chips that you’ll find in budget, mid-range and premium notebooks that focus on three key things — faster multitasking, better graphics and AI, and improved power efficiency.

Starting with the first, that CPU boost clock has been increased up to 5.2 GHz and RAM speed support has been upped to 8,533 Mega Transfers per Second (MT/s). So faster CPU cores and zippier RAM.

Second, those RDNA 3.5 GPU cores now run at a faster boost clock of up to 3.1 GHz, and the next-gen NPU on the chip die contains up to 60 TOPs of AI performance. And finally, the low power architecture on the chip has been optimized to allow for not just faster unplugged performance, but improved power efficiency too.

By the numbers

Much like everyone else reporting on these chips, we won’t have specific numbers until we are able to give it a full test in our labs. However, there have been several leaks — all of which I’m keeping an eye on in my CPU wars report.

This can give us context as to whether AMD’s numbers can be backed up, and when we take our Geekbench tests of laptops we reviewed over the past 18 months and compare, Team Red’s claims are legit.

But of course, this is looking at it through a vacuum of comparing old to new. Seeing the numbers when compared to the competition of 2026 shows that it’s neck and neck with Intel, whereas the Arm-based likes of Snapdragon X2 Elite have a considerable lead.

That being said, there’s compatibility on AMD’s side here running on x86 architecture and not needing an emulation layer. For tests on the likes of the company’s battery life, gaming and NPU claims, we’ll have to wait and see when we get to properly test these.

Outlook

Geekbench is only half the story here when it comes to chips, but an effective part in showing that AMD has stepped up its game performance-wise for the next generation of laptops and mini PCs coming this year.

But it's also the attention paid to tackling the issues with x86 that make people go for the Arm likes of a MacBook that could pay off for them in the long run — faster performance when unplugged, speedier content creation workloads and multitasking, and of course that overhauled low power architecture for multi-day battery life.

Whether these claims will be backed up, we’ll see when laptops from the likes of Acer, Asus, Dell and more launch over the next few months. But signs point to optimism.

More from Tom's Guide

• I went hands-on with the redesigned MSI Stealth 16 AI+ — it’s actually stealthy now, and gunning for the MacBook Pro
• Gaming on Snapdragon X Elite laptops just got a whole lot better — I tested the latest update and saw 40% faster gameplay
• Intel Panther Lake vs Snapdragon X2 Elite vs AMD Gorgon Point — one chip is already pulling ahead in 2026's CPU war

So far, my team and I have been blown away by the AMD Ryzen AI Max+ 395 (codenamed Strix Halo) — packing integrated graphics so powerful they match up to dedicated Nvidia laptop GPUs. In fact, it’s one of the more crucial chips to 2026 being the year of the laptop.

And in response, Team Red is bringing this prowess to more mid-range options with two new Strix Halo chips. These reduce the total CPU cores and the clock speed, while giving you that same 40-core GPU, and I’m so excited to see these capabilities come at better prices.

Meet the AMD Ryzen AI Max+ family

A gaming powerhouse

Over the course of 2025, one thing became abundantly clear — that AMD Radeon 8060S GPU built into the Ryzen AI Max+ chip is the key reason why “integrated” is no longer a dirty word in PC gaming.

Whether it’s the Asus ROG Flow Z13 or the Framework Desktop, we’ve been stunned at just what that RDNA 3.5 architecture can pull off in AAA games running at buttery smooth framerates. Turn on FSR and that potential gets even higher.

And this year, by compromising a little on the CPU side, that same fully-loaded GPU can come to cheaper systems. Price was always one of the main obstacles here, so it’s great to see the company tackle this head on.

A true workhorse

And one of the key things that grew in the background of all this is the ability to support huge offline large language models via this chipset’s GPU cores. In fact, it can run ChatGPT-OSS 120B 1.7x faster than Nvidia’s own DGX Spark AI mini supercomputer (according to AMD’s own testing).

In a more apples-to-apples comparison, it has 1.4x faster AI performance than an M5 MacBook Pro, and 1.8x faster content creation performance to boot.

Put simply, Strix Halo drove a lot more of the AI PC era than any chip’s NPU ever did, in terms of actually getting stuff done. And I’m excited to see the systems this chip ends up in from the likes of Asus, Acer, HP, Lenovo and more.

More from Tom's Guide

• AMD Ryzen AI 400 Series might finally make charging anxiety a thing of the past — 1.7x higher unplugged performance and big multitasking upgrades
• Gaming on Snapdragon X Elite laptops just got a whole lot better — I tested the latest update and saw 40% faster gameplay
• Intel Panther Lake vs Snapdragon X2 Elite vs AMD Gorgon Point — one chip is already pulling ahead in 2026's CPU war

The team at AMD brought some new silicon to CES 2026 in Las Vegas this week, and one of the most exciting is a new addition to the company's Ryzen lineup of desktop CPUs.

The new AMD Ryzen 7 9850X3D looks like a slightly souped-up version of what's arguably AMD's best desktop chip to date, the AMD Ryzen 9800X3D. This is a big deal because the 9800X3D is generally regarded as the best CPU for gaming PCs right now thanks to its winning combo of killer speeds and a hefty L3 memory cache for rapid loads.

In fact, the current AMD Ryzen 9800X3D ($479) is so good for gaming that it's what we put in the RTX 5090 gaming PC we built and it's what our old contributor Dave Meikleham put in the high-end gaming PC he built last year. So the fact that AMD just released a slightly better version at CES 2026 is a big deal.

I'm afraid AMD announced the specs of the Ryzen 7 9850X3D and the general release window (first quarter of 2026), but the company hasn't yet confirmed pricing for the new CPU. We'll update this article with that fact as soon as we have it, but in the meantime let's dive into the specs and talk about what's new!

AMD Ryzen 7 9850X3D specs and where it fits in the Ryzen lineup

The best Ryzen desktop CPU for gaming just got an upgrade

As you can see from the specs chart, there isn't a huge difference between the 9800X3D and the new and improved 9850X3D.

The big upgrade is the clock speed, as AMD has juiced the max boost up from 5.2 Ghz on the old chip to 5.6 Ghz on the new CPU. According to performance testing conducted by AMD that appears to deliver a small but noticeable improvement over the old 9800X3D when benchmarking games at 1080p, though of course we'll need to wait for data from use in the wild before we know for sure.

And since the AMD Radeon 7 9850X3D is slated to ship in the first few months of this year, I don't expect we'll need to wait too long.

Outlook

Knowing what we do now about AMD's new desktop CPU, I think there are only two major questions left to answer: what will it cost and how much better is it, under real-world conditions, than the 9800X3D?

Because right now as I look at the tiny performance improvements that AMD claims its new CPU offers over the old top model, I don't see much reason to buy this over the 9800X3D unless they're very close in price.

On the other hand, after years of watching PC component prices bounce up and down due to the intricacies of global commerce I think it's great that AMD is expanding its desktop CPU line around presumably the most popular product, so having another attractive Ryzen 7 gaming CPU option is a good thing for anyone building a gaming PC in the years ahead.

Since we're in Las Vegas covering CES 2026 live all this week I hope we'll have a chance to go more hands-on with an AMD Ryzen 9850X3D system and tell you all about it—stay tuned!

More from Tom's Guide

• I went hands-on with the redesigned MSI Stealth 16 AI+ — it’s actually stealthy now, and gunning for the MacBook Pro
• Gaming on Snapdragon X Elite laptops just got a whole lot better — I tested the latest update and saw 40% faster gameplay
• Intel Panther Lake vs Snapdragon X2 Elite vs AMD Gorgon Point — one chip is already pulling ahead in 2026's CPU war

Monday is always a big day for press conferences at the annual Consumer Electronics Show, and today was no exception. I know because we're on-site at CES 2026 in Las Vegas all week, and we just watched Intel's big keynote shortly after Nvidia's and not long before AMD is set to take the stage.

You can check out our Nvidia CES 2026 keynote coverage for more on that event, which was entirely about AI and robotics. Intel, on the other hand, brought a big box of new Intel Core Ultra Series 3 laptop chips to show off.

Codenamed Panther Lake, these are the same Intel chips we were first introduced to back in October. Back then, we trouped down to Intel to check out the new chips firsthand and test them ourselves, admittedly under the watchful eye of company representatives.

So while we couldn't put these new chips through our full gauntlet of performance tests, we were able to verify that under controlled conditions, these new Core Ultra 3 CPUs are faster and more power-efficient than their predecessors, which should mean better gaming performance and battery life for the best Windows laptops packing these chips.

Since I can't share any independent testing results yet, let's dive into the most notable new features of these chips as well as the specs of the full lineup. Let's dig in!

Intel Core Ultra 3: Biggest upgrades at a glance

• Core Architecture: Panther Lake is built on the new Intel 18A node and employs a "System of Chips" design featuring up to 16 cores and fabricated out of smaller "tiles," which can be moved and swapped to fine-tune performance and efficiency.
• GPU Changes: The GPU built into these new Core Ultra 3 chips introduces a new, more efficient Xᵉ3 graphics architecture and scales up to 12 Xᵉ cores and 12 Ray Tracing Units, delivering better performance and featuring an increased 16 MB L2 Cache in the high-end configuration.
• Multi-Thread Performance: Multi-thread performance is boosted by more than 50% versus Lunar Lake and Arrow Lake (a.k.a. Intel Core Ultra Series 200 and Series 100 chips) when operating at similar power levels.
• Single-Thread Power Efficiency (vs. Arrow Lake): It delivers similar single-thread performance with more than 30% less power compared to Arrow Lake, to boot.
• GPU performance: The new Xe3 architecture reportedly delivers up to 50% more GPU performance compared to Lunar Lake, and achieves 40% higher performance per watt compared to Arrow Lake H. Plus, new XeSS 3 tech introduces multi-frame generation for laptops with integrated graphics — a potential game-changer that could deliver seriously better framerates.

Intel Core Ultra Series 3 chips announced at CES 2026

Outlook

Since we already had a look at the new Panther Lake chips and knew they'd be launched at CES 2026, what I was really hoping to learn today is their specs and how much they'll cost.

So I got half of what I wanted, and for now, that's enough. I expect we'll start to see pre-order prices for laptops packing the new Intel Core Ultra Series 3 going live this week, and since the first laptops are slated to hit store shelves by January 27, we'll have a much clearer view of the pricing by the end of the month.

Given how volatile prices on laptops and components have been over the last year, I don't have high hopes that we'll see low prices on Panther Lake silicon. But I am hopeful that we're going to see some meaningfully better performance across a wide variety of Windows laptops, from dedicated gaming laptops sporting the Intel Core Ultra X9 and X7 chips to ultraportable business laptops that can achieve 20+ hours of battery life in our tests.

Of course, we'll have to wait until we can get a few new Panther Lake laptops into our testing lab — stay tuned!

More from Tom's Guide

• I went hands-on with the redesigned MSI Stealth 16 AI+ — it’s actually stealthy now, and gunning for the MacBook Pro
• Gaming on Snapdragon X Elite laptops just got a whole lot better — I tested the latest update and saw 40% faster gameplay
• Intel Panther Lake vs Snapdragon X2 Elite vs AMD Gorgon Point — one chip is already pulling ahead in 2026's CPU war

Qualcomm has just announced its latest Snapdragon X2 Plus CPU at CES 2026, further expanding its lineup of Snapdragon X Series processors for upcoming laptops this year.

The Snapdragon X2 Plus follows the same pattern as the tech giant’s last-gen chipsets, being a step down from the already revealed Snapdragon X2 Elite (and X2 Elite Extreme). However, a significant jump in performance, battery efficiency and AI power compared to its predecessor.

Just how much of an improvement? Thanks to the third-gen Qualcomm Oryon CPU with up to 10 cores offering up to a 35% faster CPU single-core performance, 43% better power efficiency for longer battery life and 80 TOPS (trillions of operations per second) in the Hexagon NPU, expect a boost in a wide range of mid-range laptops.

Arriving in Copilot+ laptops in the first half of 2026, let's dive into what we know about the Snapdragon X2 Plus so far.

Snapdragon X2 Plus: What we know

Like the Snapdragon X2 Elite chips, the X2 Plus brings advanced performance to Arm-based laptops and aims to achieve multi-day battery life along with native built-in AI features. Think of the Plus as the mid-range to Elite’s (and Extreme’s) premium lineup.

There are two models available: the Snapdragon X2 Plus X2P-64-100 and the Snapdragon X2 Plus X2P-42-100. The former comes with a 10-core CPU, while the latter sports a 6-core CPU. Here’s a quick rundown of the specs and how they compare to their Elite counterparts:

Impressively, the Snapdragon X2 Plus also features the same Wi-Fi 7 connectivity, integrated Adreno GPU (up to 1.7GHz) with a near-30% improvement over its predecessor and the same TOPs to make it the “world’s fastest NPU for laptops.”

“Modern professionals and creators want to do more, create more, and push the limits of generative AI and all-day performance,” says Kedar Kondap, SVP and GM of computing and gaming at Qualcomm.” Snapdragon X2 Plus platform delivers the power, efficiency and intelligence to surpass their ambitions, making each experience more responsive and personal.”

That should bring a significant step up compared to the Snapdragon X Plus chip, and at a more affordable price than the Elite series. We can expect to see these chips in laptops arriving in 2026, and who knows, they may just make our list of the best laptops this year once we put them through their paces.

More from Tom's Guide

• I just held the world's thinnest tablet at CES 2026 — and it makes the iPad Pro look bloated
• Intel Panther Lake vs Snapdragon X2 Elite vs AMD Gorgon Point — one chip is already pulling ahead in 2026's CPU war
• I saw Samsung’s 2026 flagship OLED and Micro RGB TVs side by side — here's the one I like more

Already, I'm expecting a big year ahead for AMD, and it's all set to kick off during its CES 2026 keynote on January 5. With CEO Lisa Su taking to the stage to spotlight the future of its CPUs and AI solutions, it's all pointing to the rumored AMD Gorgon Point — a.k.a. Ryzen AI 400 series and a Strix Halo refresh.

Taking place in Las Vegas, Team Red clearly has a big reveal up its sleeve, with "advancements in AI PCs, gaming" and more taking the spotlight during the keynote. If next-gen chips are coming our way, we may see these Ryzen CPUs come fitted in the best laptops coming in 2026. Find out how to watch the keynote and what to expect, and for everything else, check out our CES 2026 live blog.

How to watch AMD keynote

AMD will be holding it's keynote at CES 2026 in Las Vegas on Monday, January 5 at 6.30 p.m. PT / 9.30 p.m. ET / 2.30 a.m. GMT.

We'll see CEO Dr. Lisa Su take to the stage to make big announcements, and you can catch them all over on the livestream on AMD's official site or on YouTube (see the video above).

What to expect from AMD at CES 2026

As per AMD's page, Su will "highlight the AMD vision for delivering future AI solutions – from cloud to enterprise, edge and devices." For laptops, that should include AMD Gorgon Point (Ryzen AI 400 series/Ryzen AI HX 400 series) and possibly an AMD Strix Halo (Ryzen AI Max series) refresh.

For now, all we've heard of is from leaks. Given it’s leaked on Geekbench, the AMD Ryzen AI HX 400 series looks set to take the current core configurations you see in AMD laptops today, but give them a turboboost, specifically 15-20% higher than the previous generation.

Not only that, but the leak reported by VideoCardz seemingly confirms a new generation of integrated graphics, too, that will support RDNA 3.5 — specifically the Radeon 890M.

From what we've seen in our own benchmarks and leaks, here's how AMD's expected chips stack up against the competition.

The latter is still a possibility at the moment, but the plan seems to be a Strix Halo refresh that broadens the amount of chips available to target both mid-range and high-end offerings with that diabolical integrated GPU. WCCF reports that these will also bring faster memory support.

In what laptops will we see these chips? If it's anything like the upcoming Intel Core Ultra 3 processors and Snapdragon X2 series chips, you'll find AMD's CPUs in machines from Lenovo to Asus. So, expect faster clock speeds, higher TDPs and improved AI performance. Oh, and even greater integrated graphics in Ryzen AI Max.

Regardless, it's only a matter of time until we see Team Red's game plan for the year, and we'll be on the ground covering the latest in Las Vegas. Stay tuned!

More from Tom's Guide

• Gaming on Snapdragon X Elite laptops just got a whole lot better — I tested the latest update and saw 40% faster gameplay
• Intel Panther Lake vs Snapdragon X2 Elite vs AMD Gorgon Point — one chip is already pulling ahead in 2026's CPU war
• I think 2026 will be a breakout year for laptops and a brutal one for buyers

Months after officially announcing its Snapdragon X2 Elite chips, Qualcomm's next-gen mobile CPUs are set to be back on show at CES 2026. And from what I've already seen, laptops are about to get a serious upgrade in performance and battery.

After benchmarking the Snapdragon X2 Elite Extreme, I've seen how far Qualcomm has come from its previous-gen chips, the Snapdragon X Elite series. The latter were fitted in some of the best laptops around, and that's a good sign for what the X2 series will bring to the table — even compared to the latest Apple M5 processor.

But what else is in store for Qualcomm at CES 2026? As we gear up for the biggest tech event of the year, here's what you need to know about all things Snapdragon X2 series and what to expect.

Will Qualcomm have a keynote?

Qualcomm had its own special keynote over in Hawaii at its Snapdragon Summit 2025 in September, giving us a hands-on look at its upcoming Snapdragon X2 Elite and Snapdragon X2 Elite Extreme chips coming to laptops in 2026.

In light of this, at CES 2026, Qualcomm won't be holding a keynote like its competition, including the Intel Core Ultra Series 3 keynote and Nvidia's big announcement. However, Qualcomm will still have plenty up its sleeve, including a special guest appearance from CEO Cristiano Amon at Lenovo Tech World on January 6 at 5 p.m. PST / 8 p.m. EST / 1 a.m. GMT, fireside chats and more.

We'll be on the ground covering the latest Qualcomm has on show, so stay tuned for all the updates.

What to expect from Qualcomm at CES 2026

One thing's for sure: Qualcomm will be following up on its Snapdragon X2 series promise in this year's upcoming laptops, whether that be from Dell, Asus, Lenovo, Microsoft and more. So, expect many laptops to be announced sporting the tech giant's latest and greatest.

We do know the performance we can expect from the Snapdragon X2 Elite Extreme on sample machines, and how it compares to other current CPUs on the market. As you'll see below, it's quite the leap compared to the Snapdragon X Elite series.

With Amon taking to the stage at Lenovo Tech World, it's safe to say that we'll see Lenovo laptops sporting Snapdragon X2 Elite chips. A leak has confirmed that we can expect a Lenovo Yoga Slim 7x, IdeaPad 5x 2-in-1 and both 13- and 15-inch IdeaPad Slim 5x laptops, and I'm sure that's only the tip of the iceberg.

The latest Snapdragon X2 Elite processors have impressed, and are sure to give Windows Copilot+ PCs a major boost in CPU, GPU and AI performance, according to the benchmarks we've seen (it's even better for gaming). So, prepare for Arm-based laptops ready for improved performance and extended battery life being revealed at CES 2026.

So, any other surprises from Qualcomm? While we're sure to see reveals about its AI advances (similar to last year), we also might see another entry to the Snapdragon X2 series. During CES 2025, Qualcomm announced its entry-level Snapdragon X chip, and we may see a similar trend happen this year.

We've seen the tech giant introduce its Snapdragon Elite, Plus and X series chips before, so there's reason to believe Qualcomm will follow suit with its next-gen X2 series. Will CES 2026 be the stage where it's all announced? It won't be long until we find out.

More from Tom's Guide

• Gaming on Snapdragon X Elite laptops just got a whole lot better — I tested the latest update and saw 40% faster gameplay
• Intel Panther Lake vs Snapdragon X2 Elite vs AMD Gorgon Point — one chip is already pulling ahead in 2026's CPU war
• I think 2026 will be a breakout year for laptops and a brutal one for buyers

As soon as I was starting to get used to the codename Panther Lake, Intel Core Ultra Series 3 is set to be made official at a CES 2026 keynote on Monday, January 5. After a couple of years of disappointing laptop silicon, this is Team Blue’s fightback moment, and after testing the chips myself, they’re coming in heavy with a big generational leap.

And you can expect to see them in so many laptops announced in Vegas. Here’s how you can watch the announcement live with us, and specific details you can expect Intel to cover.

How to watch Intel at CES 2026

The Intel keynote is set to kick off on Monday, January 5 at 3pm PT / 6pm ET / 11pm GMT. There is a livestream page with a countdown timer over on Intel’s website. But currently beyond that, there’s no YouTube livestream available for this yet. I’ve no doubt that one will be shared on Intel’s YouTube page soon enough

What to expect from Intel at CES 2026

This is the most obvious one to predict, because there’s no prediction whatsoever. As Intel itself says on the event page, Senior VP Jim Johnson is going to talk to us about Intel Core Ultra Series 3 processors — codenamed Panther Lake.

I’ve already had the chance to test these quite a lot last year, and to put it simply, you’re in for a treat. These chips pair the raw performance of Arrow Lake with the power efficiency of Lunar Lake, and combines them into one chunk of silicon for quite the generational leap in laptop prowess.

In practice, this leads to a 50% improvement in CPU and GPU performance, a 40% reduction in the power needed to achieve those highs, and NPU improvements that aim to make on-device Agentic AI a real thing this year.

It gets even more tasty when you look at the X-series chips (rumored name) — the top of the range ones I tested that sports that mighty 12-core GPU on the chip die. With Intel’s Xe3 technology driving the latest iteration of Team Blue’s XeSS AI trickery in gaming, I was able to play AAA titles in excess of 120 FPS on integrated graphics.

Outlook

Intel is one of three companies jumping headfirst into the CPU wars of 2026 with a CES keynote — Qualcomm and AMD are also stepping up too. It will be interesting to see how the company’s innovations here from the new 18A architecture to putting transistors on the backside of the chip will help the big I in this war.

Look forward to all the latest announcements as they happen live in our CES 2026 coverage, alongside plenty more hands-on coverage with systems sporting these chips.

More from Tom's Guide

• Intel Panther Lake could power the future of handheld gaming — here's why AMD should be nervous
• Intel Panther Lake vs Snapdragon X2 Elite vs AMD Gorgon Point — one chip is already pulling ahead in 2026's CPU war
• I think 2026 will be a breakout year for laptops and a brutal one for buyers

Saying that Apple’s M-series processor shook up the computing industry would be an understatement. Five years after Tim Cook announced that the company was ditching Intel for its own silicon for Macs, M-series chips from the original M1 to the current M5 continue to set new standards for performance and efficiency. It has made everyone rethink what’s possible with Arm-based architecture.

While the success of the M-series seems inevitable in hindsight, it wasn’t guaranteed. After all, Macs powered by Intel chips were successful for years, and switching from x86 architecture to ARM could have proved disastrous. However, when Apple’s internal testing showed how much faster M1 was over Intel, the company knew it had made the right decision. The fact that the five-year-old M1 MacBook Air can still hold its own against competitors is a testament to the M-series’ strengths.

For the fifth anniversary of the M-series chip, we sat down with Tim Millet, VP of Platform Architecture, and Tom Boger, VP of Mac and iPad Product Marketing, to gain further insight into how the company developed the chip. We also spoke with Mac users about how M-series Macs have changed their workflow, and chatted with industry analyst Avi Greengart for his thoughts on the M-series.

Genesis of the M-chip: Moving Away from Intel

Apple switched from PowerPC to Intel in 2006. That particular relationship lasted until 2020, when Apple began phasing out Intel Macs for computers running on M-series chips. Given the length of the partnership with Intel, it wasn’t an easy decision to make.

“It was definitely not a step Apple took lightly,” said Tim Millet. “We had a very strong relationship with Intel. They were great silicon partners. I was here when we transitioned in 2006 to Intel, and that was a fantastic transition, and I think it definitely injected a lot of energy into the Mac.”

Though potentially risky, Apple already had experience developing chips for the iPhone and iPad. This gave the company the knowledge and confidence to develop its own processors for Macs.

“We were building chips for phones and building a team and building the capability, and building the relationships with the foundries,” said Milet. “We got a lot of practice building that up, and we got good at it. And as we extended our efforts into iPad, it allowed us to scale things a little bit.”

Apple’s experience with the iPad Pro allowed the company to imagine a processor that would work well for Macs. There were talks with the software teams to figure out how they could pull this off effectively. Apple didn’t just want to create an ultra-powerful chip, but also one that was energy efficient.

“This took us time,” said Millet. “It took us time to convince ourselves and to convince the executive team and the marketing teams to say: ‘Hey, we think we can do this, and we think it might be something special.’ And we were paying very close attention to what really mattered for the Mac.”

Fortunately for Apple, they had Mac experts with 30 years of experience who could tell the team everything that was important for the Mac. Tim Millet believes that was the secret weapon that helped the team build the M1 chip.

“As a silicon architect for most of my life,” said Millet, “having the ability to really engage directly with the software and system team side by side as you build the chip makes all the difference, because you don't have to guess. You know exactly what you're building. You know exactly how it's going to get used. You double down on the things that matter the most.”

The “a-ha” moment

With the idea of what would become the M1 chip in place, Apple soon developed a prototype. The team had an idea of what to expect, but even they were amazed by the initial results.

“It was like being introduced to the Mac again for the first time,” said Tom Boger. “Opening up the system and seeing [that] your battery life indicator didn't budge, and wondering if the software was working correctly. I'll never forget walking around with that first system. I could not wait for the world to experience the Mac again.”

“We started exploring prototypes pretty early,” said Tim Millet, “and the battery life was certainly one of those moments. And because we were building up this engine from the phone, all the blocks that we were putting in, there were extensions of this super energy-efficient focus that we [implemented].”

In hindsight, it seems obvious to take the building blocks of the iPhone’s battery and put them in a Mac. Apple knew the battery life would be fantastic, given the size of a MacBook’s battery. As Milet said: “It's like you got this ocean of energy using this thing that was meant to operate in this very tiny pond of energy.”

The other revelation was the performance the team was seeing from the M1’s CPU, especially since Apple felt performance-per-watt had stalled on Intel. They were able to get more from the M1 at a “dramatically” lower frequency.

“It was really the energy efficiency focus that helped make a huge difference, not just in the battery life,” said Millet. “Being able to deliver amazing compute within the thermal constraints of the devices that we wanted to build (and were already building) into that beloved, classic MacBook Air enclosure that we shipped it with… I think it completed the story for the MacBook Air. This is what the MacBook Air dream was when Steve pulled it out of the envelope.”

With the MacBook Air and iPhone both using Apple-developed silicon, it was easier for the software teams to focus on a unified platform. It allowed the teams to enable many of the applications that folks were already using on their iPhones. This helped everything feel familiar to iPhone users who were also Mac users. “That really is part of the magic,” according to Millet.

Another “a-ha” moment for Tom Boger was opening the MacBook Air M1 for the first time and seeing it instantly wake up. The overall snappiness of using the laptop was also memorable. “It just gave me an experience on my Mac that I hadn’t had before.”

Unified architecture

The M-series’ unified architecture is its defining feature. Integrating the CPU, GPU, Neural Engine and memory into one die allows for the astonishing performance and power efficiency that Mac users have come to enjoy over the past five years. Thanks to this architecture, developers aren’t as constrained as they would be on traditional hardware with separate CPU, GPU, and RAM.

“On iPhone and iPad,” said Millet, “[the] memory system has enough bandwidth to feed the GPU. But we needed the capacity for the CPU as well. So when we introduced [the M1] on the Mac, it allowed internal developers to say ‘Oh my gosh, I can now operate in the GPU, move these buffers back and forth between the CPU and the GPU.’” These developers are also able to move those buffers back and forth between the video accelerators and media accelerators.

This is a fundamental architectural shift that enables developers to really operate without restrictions. For example, you don’t have to worry about generating images in one place and then running them through a PCI Express to get to the other memory system. There's one big pool of memory.

Tim Millet said that bringing this unified architecture to the Mac introduced it into mainstream computing in a way that had never been done before. While the M1 wasn’t the first to operate this way, having this architecture in such a popular laptop gave it more impact.

“I would say [the unified architecture] continues to pay dividends even today, especially when you’re speaking about AI,” said Tom Boger. “With large language models running on a device, we can tap into that whole pool of memory, and you can take a MacBook Air and run LLMs with tens of billions of parameters all the way up to an ultra chip there, where you can have hundreds of billions of parameters in an LLM on device, and that's all because of the magic of having a unified memory architecture.”

The Neural Engine

Large Language Models (LLMs) like ChatGPT have dominated headlines for the past two years. This has given rise to “AI laptops,” which utilize an NPU (Neural Processing Unit) to perform AI-driven tasks.

While computers with an NPU might sound novel, Apple’s M-series chip has had a Neural Engine since the beginning. Though Apple initially didn’t discuss the Neural Engine as much as it did general performance and efficiency, the company is now stating that the M5 processor is the most powerful AI chip ever. Clearly, AI is an important feature of the M-chip.

“We introduced our Neural Engine in the iPhone back in 2017, but it was really about computational photography,” said Tim Millet. “You didn't have all these chatbots and LLMs and things.”

For the M1 chip, Tim’s team said they needed to re-architect the Neural Engine for the processor. He wasn’t sure why it would matter yet, but he told the team to go forward. Because of that decision, M1 (and the M-series in general) was ready for the AI wave we’re currently experiencing. While the Neural Engine exists to aid with AI-driven tasks, Apple isn’t limiting said tasks just to the engine.

“The neural engine is there when we absolutely need the most energy-efficient solution, but we have this great GPU as well, and we can run [it] at the same time,” said Millet. “It's a super convenient way for [developers] to invest in extending neural compute in the applications that are media-driven.”

The M-series chip ‘unleashed Apple’

Apple’s shift from Intel to M-series chips on the Mac didn’t just have an impact on the computing industry. You could argue that it was also transformational for the company itself.

"It's really unleashed Apple,” said Avi Greengart, consumer technology analyst and founder of Techsponential, “It's given them the control they've always wanted, along with the flexibility to allocate resources to specific parts of the chip. Some years they've focused on CPU, and other years on GPU. The architecture itself provides advantages with memory being part of the package, allowing it to be shared across both CPU and GPU rather than having the GPU rely on separate memory.”

This flexibility has given Apple a dual advantage: technical performance and strategic independence. No longer beholden to Intel’s slower roadmap, Apple differentiated itself in a way it hadn’t since the Motorola era. The narrative shifted from “we have better computers because our software is better to ‘we have better computers because our software is better and our hardware is better too,” said Greengart.

Apple’s focus on performance per watt has been equally transformative. Whereas the industry used to focus on raw power, Apple decided to emphasize efficiency. Thanks to that, MacBooks can now deliver impressive performance without being tethered to a power supply. “Being able to achieve a reasonable fraction of high-end GPU performance while not recharging your laptop for 15 hours is incredibly impressive — and it wasn’t a user experience we had until the M series,” says Greengart.

This shift has forced competitors like AMD, Intel, and Qualcomm to rethink their own efficiency strategies. “Apple’s move to its own silicon has turned the industry on its head, opening the market for Qualcomm Snapdragon to be more broadly adopted by PC OEMs who need something competitive. It’s also pushed AMD and Intel to ramp up the performance per watt of their products.”

One of the things that few expected was that, thanks to Apple’s economies of scale and vertical integration, the company also gained cost controls that others don’t. Apple has long been associated with premium pricing, but the M-series has changed that perception (at least to some extent).

“We’re seeing super high-end computers being sold for under $1,000 from Apple — something not everyone predicted,” says Greengart. “People often think of Apple as a premium brand because it primarily plays at the high end of the market. But the M series, especially when you use [older chips like] the M1, allows Apple to put powerful processors into laptops sold to high school students. At the same time, you can get an iMac with the latest chip — I had an M4 iMac the day they announced the M4, and that was complete overkill.”

The analyst told us that even if you’re just planning to use an M-series MacBook for Facebook and email, you’re guaranteed it will remain powerful and fast for a very long time.

M-series impact on Mac users

Apple’s transition to its own silicon has reshaped the computing landscape in ways few anticipated. It has also made users’ lives easier. Two such users are Tyler Stalman, known for his popular YouTube channel, and David Stout, Co-founder and CEO of WebAI. Each shared their thoughts and experiences using Macs with M-series chips.

“There’s obviously an enormous jump at the M1 era,” said Tyler Stalman. “Just before that, you'd have to be careful about which codecs you're working with on your laptop versus on a desktop. Different machines could handle different levels of 4K and compression. And now we're at a point where, as long as I've got my MacBook Pro with me, I can throw anything at it, and it just works. Nothing ever slows down until you start having layers of effects or [you throw] AI at it. But all of the basic editing now works on every single Mac that I work with.”

Regarding how M-series MacBooks changed or improved his workflow in comparison to other laptops, Tyler says one of the biggest differences is that he no longer has to choose between a desktop or a laptop to get the performance he needs. Before, he would do lighter work on a laptop when on the road, and save the heavy lifting for his desktop. With a 16-inch MacBook Pro, not only does he get all the processing power he needs, but a large display that’s suited for video editors.

“Some of the biggest quality of life improvements for video editing came with the additions of more onboard encoder and decoder cores on the M chips,” said Tyler. “As [Apple] has updated the M-series, they’ve added more dedicated encoding and decoding chips that just handle video. They offload everything from the CPU and GPU to this one chip that’s handling everything, and that’s really sped things up.”

David Stout’s WebAI created an AI library called webFrame that is designed to work directly with Apple silicon. Because of that, his company can leverage the unified memory on the chip. “[There] is less copying from memory because it's shared, so when you cut that out, you have much faster inference time, as well as leveraging the architecture's processing capabilities of these models. The effectiveness in performance you get with Apple silicon is one of the ways we've been optimizing.”

Iterative generations

The leap from Intel to the M1 was monumental. However, there has been a perception that subsequent generations have just been iterative at best. As we get to 2nm chips, some believe Apple and other manufacturers might hit a wall. However, Apple believes there is still more that the M-series can do.

“I would point out that the success of generations has been relentless every year, right?” said Tom Boger. “We didn't just bring out M1 and rest on our laurels. We've kept our foot on the gas. And so every year we move the ball forward when it comes to Apple silicon and the M series of chips.”

Boger also said that there has been six times the AI compute for the GPU between M5 and M1. In addition, there is also the scope of the M-series product line, which includes the M5 chip in the latest iPad Pro and 14-inch MacBook Pro. “We've scaled Apple silicon even further,” said Boger. “If you look at where we were with them, we've continued to push every year.”

This is a sentiment Tim Millet shares, saying that his team isn’t dwelling on the past. “We have our foot on the accelerator, and we're going to stay on the leading edge of the process nodes. There's nothing relaxed about our posture for what we want to do for our chips to enable these amazing products, including the Mac. We’re happy about what we’ve done.”

The future of MacBook design

In 2021, Apple revealed a new utilitarian and slimmer design for the MacBook Pro. The MacBook Air would eventually adopt this design in 2022. Nearly three years later, MacBooks have retained this same basic look. Naturally, this has raised the question of when Apple might redesign its MacBooks, and if the M-series chips influence design choices.

“We sit side by side with our design and software teams to imagine possibilities,” said Millet. “When we solidify on something like a new display or new capability, all of that is done with silicon enablement. We try to figure out how to enable the experience we want to deliver without burning the battery quickly.” Tom Boger put it succinctly: “[Tim] is designing his chips to skate where the puck’s going to be.”

The future of the M-series chip

Looking forward, there are several things Apple can do because it controls its silicon design, according to Avi Greengart. One depends on its AI strategy and partnerships. If Apple moves forward with better models and enables more on-device processing, it could enhance its privacy and data security while giving users capabilities they’d otherwise need the cloud for.

“The fact that Apple can choose how large the NPU is while simultaneously building the software to take advantage of it gives them a different kind of advantage compared to Intel or Qualcomm,” says Greengart. “Qualcomm’s NPUs are huge, as are MediaTek’s, but if Microsoft or ISVs don’t leverage that power, some of it is wasted. Apple, by contrast, only includes what it knows it will use.”

Greengart also says that another area to watch is cellular integration. Apple already has a modem, and he expects the company will eventually integrate it onto the die. “The capability of making every Mac a cellular Mac is very interesting. Apple would only pursue that if it had the right connectivity relationships, similar to what it has with the iPhone and iPad. These are the kinds of capabilities that come from designing your own silicon, even if Apple doesn’t manufacture it themselves."

Outlook

Apple’s M-series processors have done more than boost the performance and power efficiency of Macs. The company’s chips have redefined what you can expect from a computer. By combining efficiency with raw computing power, Apple’s silicon allows for laptops with astonishing battery life and that are as mighty as a traditional desktop.

Five years later, we’re still feeling the effects of Apple’s disruptive processor. Competitors like Intel, AMD, and Qualcomm are now scrambling to match Apple’s focus on performance-per-watt. Average users have also benefited from features like unified memory that help them get work done faster and more efficiently.

Moving forward, we could see even deeper AI integration, new designs, and perhaps even cellular-enabled Macs. It’s also possible Apple will surprise us all with something we hadn’t considered. Regardless, the M-series chip gave the industry a major shake-up, and it will be interesting to see what Apple has in store for the next five years and beyond.

As CES 2025 approaches, we're preparing for the expected launch of next-gen Intel Panther Lake-equipped laptops. And so are brands like HP and MSI, as the first Core Ultra series 3 machines have now appeared at retailers — including their prices.

The first laptops with Intel Panther Lake CPUs are now listed at multiple stores, including Walmart. As spotted by tipster @momomo_us on X, there's now an HP OmniBook X Flip 16 with an Intel Core Ultra 7 355 CPU, and it's priced at $999.

Since Intel has yet to officially launch its upcoming chips, the laptop isn't available to purchase yet. This is just a placeholder as the retailer prepares a new lineup of next-gen notebooks, but for now, it gives us an idea of the cost we can expect. And, as it turns out, it's not the major price bump we'd expect.

With every new generation of laptops sporting the latest specs, prices are always expected to be set at a premium, especially over last-gen products. Plus, the RAM crisis will only see costs soar higher. According to this listing, though, this Intel Panther Lake laptop is at the same price as the previous-gen model.

You can find the HP OmniBook X Flip 16 with an Intel Core Ultra 5 226V CPU priced at $999 over on the HP Store, even though it's currently reduced to $649 thanks to a sale. Arguably, the latest OmniBook with a Core Ultra 7 processor is the better deal at full price.

Along with its Core Ultra 7 355 CPU, the next-gen model of HP's OmniBook X Flip 16 also comes with Intel integrated graphics, 16GB of DDR5 RAM, 512GB of storage and a 16-inch (1920 x 1200) LED touchscreen (not the 2K the listing mentions).

Since this is a listing that's published before the official launch of the laptop, pricing may change. But it's a good indication of costs to come, especially on machines sporting the upgraded Intel Core Ultra 300 processors.

Gaming laptops with Intel are a little high...

It's not unheard of to see MSI laptops set at premium prices, with the monstrous MSI Titan 18 HX priced from a whopping $5,000. But these new Intel Panther Lake laptops appearing on EU retailer sites (via Wccftech) seem a tad overpriced. Or at least, hopefully.

For example, there's an MSI 16I with an Intel Core Ultra 7 355 (same as the HP OmniBook) priced at €2,611 at French retailer PC21, so around $3,061. Despite it also boasting a 2.8K resolution OLED display and 32GB of DDR5 RAM, that seems awfully expensive.

There's also an MSI 16I with a much stronger Intel Core Ultra X9 388H CPU for €3,006 at PC21, making it around $3,524. There's no GPU in sight here, with this chip coming with Xe3 graphics architecture to deliver onboard graphics performance for gaming, albeit not to the level of a discrete RTX 50-series GPU.

Those prices seem strangely high for these kinds of laptops, even with next-gen Intel CPUs inside. These could be random placeholder numbers for these retailers, as again, these laptops aren't available to purchase yet.

We've seen how Intel Panther Lake has the gaming chops to power the future of handheld gaming, and that's also a good sign for thin, lightweight laptops to play the latest AAA games with just an Intel chip. The question is, how pricey will these machines be? For now, if Walmart's listing price on the HP OmniBook X Flip 16 is accurate, it may not be at such a high price.

More from Tom's Guide

• I just tested my favorite new gaming laptop, and it’s not from Alienware, Asus or MSI
• Asus ROG Zephyrus G14 vs Razer Blade 14: Which gaming laptop should you buy?
• Intel Panther Lake vs Snapdragon X2 Elite vs AMD Gorgon Point — one chip is already pulling ahead in 2026's CPU war

The stage is set for a huge CPU battle in 2026, and it’s already started as Intel and Qualcomm have fired away ahead of Panther Lake and Snapdragon X2 Elite debuting early next year (read: CES 2026).

We’re also seeing leaks related to AMD and will bring to the party with Gorgon Point, and there are a lot of moving parts. So to keep track of everything, I’ll keep this hub updated with everything you need to know about the next generation of laptop and desktop chips.

Meet the competitors

• Intel “Panther Lake” (Core Ultra 300 Series) laptop chips - CONFIRMED
• Intel “Arrow Lake-S” desktop chip refresh - PREDICTED
• Qualcomm Snapdragon X2 Elite - CONFIRMED
• AMD “Gorgon Point” (Ryzen AI 400) laptops, “Strix Halo” (Ryzen AI Max+) laptop refresh and Ryzen desktop CPUs - EXPECTED

Three big players are entering the arena: Intel, Qualcomm and AMD. They are coming in heavy with a huge suite of new chips. Let’s break it down into a battle on two fronts.

Laptops

2026 will be the year of the laptop, and we already know it's going to be a silicon bloodbath. In one corner, you’ve got Intel Panther Lake — built on the company’s 18A platform, which is a new technological foundation that brings the best of Intel’s faster chips and the more power-efficient chips into one piece of silicon.

As we’ve tested, this is looking to provide up to 50% faster CPU performance, a 50% faster GPU, up to 1.2x faster AI performance, and up to 40% higher performance per watt (better power efficiency).

In the other corner is Qualcomm. We’ve got to test the Snapdragon X2 Elite and X2 Elite Extreme, which follow a lot of the same formula as the original X Elite, bar one difference — memory (RAM) can be embedded directly onto the chip. Taking notes from Intel and AMD, the company’s now adding shared memory between the CPU, GPU and NPU.

This rethink and generational uplift is showing up to 39% faster single-core and 50% faster multi-core speeds, 69% higher memory bandwidth, 78% faster AI performance, and up to 2.3x faster graphics performance.

And in the final corner (shrouded in secrecy because this is based on leaks and rumors at the moment) is AMD. Team Red has the headline keynote slot at CES 2026, so they have to be bringing something! And that much seems true with two new laptop chip lines: Gorgon Point and a Strix Halo refresh.

Given it’s leaked on Geekbench, the AMD Ryzen AI HX 400 series (Gorgon Point) looks set to take the current core configurations you see in AMD laptops today, but give them a turboboost — specifically 15-20% higher than the previous generation. Not only that, but the leak reported by VideoCardz seemingly confirms a new generation of integrated graphics, too, that will support RDNA 3.5 — specifically the Radeon 890M.

The latter is still a possibility at the moment, but the plan seems to be a Strix Halo refresh that broadens the amount of chips available — to target both mid-range and high-end offerings with that diabolical integrated GPU. WCCF reports that these will also bring faster memory support.

Desktops

Starting with Intel, Team Blue is taking another crack at its Arrow Lake desktop chips with a refreshed range. The original launch was quite the stumble, with problems in power management, performance limitations, and some BIOS bugs, so Intel wants to take another crack at it.

Looking at leaks, it seems the company’s getting its act together with higher clock speeds, more efficiency cores, and better RAM support to overcome memory latency issues. Basically, this is a much more optimized version of the same chips.

Meanwhile, AMD seems to be all in on launching two new CPU lines at CES — the souped-up gaming-centric CPUs in the Ryzen 7 9850X3D and the Ryzen 9 9950X3D2. X3D chips are finely tuned to play super nicely with dedicated GPUs, and it seems to be more of the case here.

Second is a new Ryzen 9000G desktop APU, which is more of an all-in-one approach that updates the architecture for newer graphics standards like RDNA 3.5.

Codename translator: what will they actually be named?

• Intel Panther Lake = Core Ultra 300 Series (according to leaks)
• AMD Gorgon Point = Ryzen AI 400 series/Ryzen AI HX 400 series
• AMD Strix Halo = Ryzen AI Max series

This can get confusing. You're reading me flipping between calling the latest Intel chips "Panther Lake" and "Core Ultra 300 Series." So what will you actually be buying in retail stores in early 2026? AMD is just as guilty of this with calling their chips "Gorgon Point" and "Strix Halo."

Luckily, Qualcomm's much more straightforward with the names "Snapdragon X2 Elite" and "Snapdragon X2 Elite Extreme" (even though the latter is getting a little too close to what I'd name an energy drink back in the 90s). But to help you out, let me translate the codenames here.

Intel Panther Lake

We know three levels of Panther Lake chips exist (I've tested them). But here's what the Geekbench leaks point to these being called.

• Intel Core Ultra 300V: This would be the base configuration with 8 CPU cores and 4 GPU cores
• Intel Core Ultra 300H: This name may designate the chip that sports 16 CPU cores, but ditches the GPU (the one you will probably see in gaming laptops)
• Intel Core Ultra X 300H: That X looks set to identify the top-of-the-range model with 16 CPU cores and that full-blown 12-core GPU. Geekbench results just dropped on the Ultra X9 388H.

AMD Gorgon Point

AMD's not messing too much with the formula here, and just upping the number by one. So far, leaks have shown the following chip names.

• AMD Ryzen AI 400 series: Based on the past generation, these will be the base-to-mid-range chipsets.
• AMD Ryzen AI HX 400 series: That HX stands for higher performance, so is reserved for their upper-level chipsets with faster clock speeds, higher TDPs and improved AI performance.
• Strix Halo = Ryzen AI Max+: Given the rumors of more Strix Halo chips coming to CES, it's worth me noting that for the all-in-one chips with crazy good integrated graphics, you're going to want to find the name "AMD Ryzen AI Max" followed by a number.

How they stack up (according to our tests and leaks)

• Snapdragon X2 Elite has established an early lead in CPU speeds
• Intel Panther Lake has overhauled graphics performance and power efficiency
• But AMD leaks show Gorgon Point may crash Intel’s party

To get into this, I need to break something down here — the difference between Arm and x86 computing is significant, because they go about completing tasks very differently.

So this is not just a battle of companies, it’s a battle of principles, and this shows in the performance numbers. Arm usually gets a lead on pure CPU speed and battery life, but x86 gets its advantage in graphics and compatibility.

There are going to be many more chips than what we've tested/seen leaked so far. For example, we don't have benchmarks for the top-of-the-range Panther Lake chip with a 12-core CPU. But these are a good baseline of what to expect.

When we get more benchmarks, I'll update this chart over time!

UPDATE: New leaked Geekbench benchmarks just dropped for some of the mid-range AMD and Intel chips.These numbers are (mostly) based on the leaks we’ve seen. I did get to benchmark the Snapdragon X2 Elite and Elite Extreme myself, so these numbers are accurate based on my own testing. Other numbers came from leaks courtesy of VideoCardz and PC Gamer.

Geekbench CPU benchmark results

So in terms of Geekbench leaks and testing, Qualcomm has the early advantage, but that’s only half the story. Intel’s fully gone in on turboboosting those Xe3 GPU cores, and adding a whole lot of AI-infused trickery to boost gaming. In fact, it’s looking like you could get something similar to a dedicated Nvidia laptop GPU in integrated graphics.

We saw this start to happen throughout the course of this year, and while I was a little too dramatic in saying Nvidia’s days on top were numbered back on New Year’s Day, the crux of the piece is still true — integrated graphics are really taking the fight to dedicated GPUs.

That much was true with AMD Strix Halo, and the company seems to be taking another step forward performance-wise, while also broadening its range of chips available to make them more affordable, too. One step forward with Gorgon Point, while giving you more Strix choices to pick from.

What about Apple?

• M5 is already offering similar performance to what we see in Snapdragon X2 Elite
• M5 Pro architecture rumored to be altered to offer vastly improved performance

This is a big unknown heading into 2026. Apple Silicon runs on an Arm architecture, and one thing we know for sure is that the latest M5 chip in the MacBook Pro is a power-efficient screamer.

But what about the M5 Pro and M5 Max? It’s being rumored that the architecture may be getting a big switch-up from a system on a chip (everything on one die) to a system of chips (separate chiplets for each thing). As we’ve learnt from Intel Panther Lake and AMD’s Ryzen AI Max+, doing so makes for some serious performance upgrades.

We won’t know for sure until they are officially in our hands, which is looking likely to arrive in Spring 2026.

More from Tom's Guide

• RAM prices are exploding — here's why and everything you need to know about surviving RAMageddon
• Lenovo leak confirms first rollable gaming laptop with ultrawide OLED display coming in 2026 — here's what we know
• M5 MacBook Air reportedly coming Spring 2026 — plus the missing MacBook Pros and a new Mac Mini

Just a week or so after Intel’s new flagship Panther Lake chip leaked on Geekbench and had its benchmarks shared for all to see, it appears we’ll be able to get our hands on one as soon as early January.

While we expected a 2026 release for the new chips, Asus may have jumped the gun by kicking off an event with the subtitle “Ubiquitous AI. Incredible Possibilities.” On January 6, just hours after Intel’s event is expected to reveal the chips themselves as part of CES 2026.

Panther Lake, here we come

As Asus explains via its CES site, it “will showcase our latest solutions, including a full lineup of advanced AI PC products and other innovations — all designed to empower everyone, everywhere.”

“See how our vision, "Ubiquitous AI. Incredible Possibilities," drives AI transformation for the workspace, creators, and everyday life!”

As for how the new Core Ultra 9 386H chip will perform, there’s a 2.1 GHz base clock speed, while boosted clock speeds reach 4.724 GHz — lower than the anticipated 4.9Ghz.

It already beats Arrow Lake chips, with 9% faster single-thread speeds and 4% faster multi-thread speeds.

It’s also curious to see Asus touting AI-based laptops while the Core Ultra 9 386H is more of a gaming CPU that won’t pair with integrated graphics. Is Asus about to push further in that direction?

Whatever the answer, we really don’t have long to find out — we’ll be all over the coverage of Intel’s event on January 5, and the Asus one the next day.

A new challenger approaches

Of course, CES isn’t an all-Intel affair, because Snapdragon’s X2 Elite chips are a-coming, too.

Our own Jason England benchmarked Qualcomm’s latest and said the latest revision should have Intel, Apple and AMD sweating.

They’re expected to be a big focus at CES this year, suggesting we could be weeks away from a processor war the likes of which we’ve not seen in a while.

As for the numbers, the Oryon CPU prime core of the Snapdragon chip offers up to 39% speed increases on single core and 50% faster in multi-core than the last generation, while there’s 69% more memory bandwidth since RAM is on the chip itself.

Finally, its AI credentials are the new Hexagon AI engine, which is 78% faster in AI tasks, while the GPU is 2.3 times faster, too. There’s a whole new era of laptops coming, and it won’t be long until we see if they make the cut for the best laptops around.

More from Tom's Guide

• Brace yourself: PCs, laptops and phones may jump in price as early as December thanks to a '500%' RAM/SSD surge
• Intel Panther Lake could power the future of handheld gaming — here's why AMD should be nervous
• I’m shocked at how well Snapdragon X2 Elite can play games with Cyberpunk hitting over 75 FPS in my testing

If you’re looking to pick up the latest and greatest gaming processor, it might be worth holding fire for a little while.

The new Panther Lake processor, the Intel Core Ultra 9 386H, has popped up on Geekbench, packed into an Acer Predator gaming laptop, and is expected to be a flagship chip for Intel’s 2026 lineup.

The chip, which includes 4 Xe3 cores instead of the 12 of the Ultra X9, is focused on gaming, and its appearance here lines up with an expected appearance at CES.

Panther Lake benchmarks are in

There are still some caveats to cover before we get into the raw numbers. For one, this is in-development hardware, so there’s every chance it’s not hitting the numbers it could in a few months.

Then there’s the fact that the Core Ultra 9 386H doesn’t pair with integrated graphics, and we don’t know which GPU it’s attached to in the new Acer Predator model.

With all that said, Videocardz has shared the numbers, and it’s looking mighty impressive already. There’s a 2.1 GHz base clock speed, while boosted reach 4.72 GHz - lower than the anticipated 4.9Ghz.

As the report points out, this processor already eclipses its predecessor, the Arrow Lake-H flagship Core Ultra 9 285H, with 9% faster single-thread speeds and 4% faster multi-thread speeds. While it’s behind the Core Ultra X9 388H, that chip is the uber top of the line offering with 24 cores with its own integrated GPU.

Outlook

While 9 or 4% might not feel like a significant jump, chip development has reached a pretty swift release cadence, and according to Videocardz, we're still getting the same TDP.

That means we're getting additional performance at the same power efficiency, which should mean games on upcoming laptops will be more performative without draining quite as much battery charge as older generations.

Expect the new Panther Lake chips to be shown off at CES, along with the models we'll find them in.

More from Tom's Guide

• Intel Panther Lake could power the future of handheld gaming — here's why AMD should be nervous
• I just tested the new Snapdragon X2 Elite laptop chips — and Apple, Intel and AMD should be worried about 2026
• You can get a Chromebook for just $49 — go check your wallet, we're serious about this deal

Thinking of buying a Mac Studio? You may want to hold off, as the M5 Ultra is rumored to make its debut on what will be Apple's most powerful Mac desktop yet.

According to Bloomberg's Mark Gurman, Apple is skipping the M4 Ultra and going straight to an M5 Ultra chip to power its upcoming Mac Studio. Considering the impressive M5 chip benchmarks and how the M-series Ultra processors link two Max CPUs using UltraFusion architecture to double performance, this will be the Cupertino tech giant's highest-performing chip to date.

Currently, the latest Mac Studio comes with mismatched M4 Max and M3 Ultra chip options. Apple made a point that not every generation will get an Ultra chip variant (as per Ars Technica), with the last being the M3 Ultra, which features a 32-core CPU, an 80-core GPU, 512GB of unified memory, and a 32-core Neural Engine.

Gurman claims the M5 Ultra will arrive in 2026, following the launch of the M3 Ultra in March earlier this year. This was announced for the latest Mac Studio, and as per the report, Apple is looking to follow suit with its next Mac Studio. However, it's unclear when the M5 Ultra will officially make its debut.

As for the Mac Studio itself, it isn't expected to see major design changes, as the chip will be the focus for the update. What we do know is that it won't come cheap, seeing as the M3 Max Mac Studio is priced from $3,999.

M5 Pro and M5 Max come first

With Apple's latest silicon arriving on the M5 MacBook Pro and M5 iPad Pro, we now look towards the M5 Pro and M5 Max chips to arrive — expected to be sometime in early 2026. If this is the case, the M5 Ultra will likely launch after these silicon are released.

In the past, Apple launched its M1 Ultra and M3 Ultra in March, with the M2 Ultra chip being announced in June. This gives us a good idea of when the M5 Ultra will be released, as the Ultra variants typically launch after the Pro and Max chips. If this is the case, it may be sometime during the first half of 2026.

Hopefully, the M5 Ultra will arrive sooner rather than later in the year, seeing as a redesigned MacBook Pro with a touchscreen OLED panel is set to kick off the M6-series chips.

Interestingly, the tech giant is will reportedly debut a new Studio Display is gearing up to arrive by 2026, which could coincide with the launch of the M5 Ultra Mac Studio. In any case, time will tell when we'll see Apple's most powerful chip arrive.

That's not all that's rumored to be in the works, as Apple is reportedly planning to reveal a budget-friendly MacBook with an iPhone processor sometime in the first half of 2026. So, expect an array of Mac models coming next year.

More from Tom's Guide

• The new Siri will reportedly 'lean' on Gemini — here's what that means for you
• Samsung Galaxy XR vs Apple Vision Pro vs Meta Quest 3: Which headset is right for you?
• I'm a clumsy MacBook Pro owner — here are the best MacBook pro cases I personally recommend

Now that Apple's M5 chip has arrived in the latest MacBook Pro and iPad Pro, we finally see the impressive performance gains it offers. But it's already found a powerful Arm-based Windows rival that's set to arrive soon: the Snapdragon X2 Elite Extreme.

With Qualcomm's upcoming X2 Elite chipsets coming to power a new generation of Windows Copilot+ laptops in the first half of 2026, Apple will already have a fight on its hands as the CPU race heats up (not to mention Intel Panther Lake).

Now, after benchmarking the M5 MacBook Pro in our testing labs, and testing the high-end Snapdragon X2 Elite Extreme, we can see just how well these chips perform — and there are some surprising results.

There's still more testing to be done with the Snapdragon X2 Elite Extreme, as this was tested on a reference laptop with 48GB of RAM and a 1TB SSD. Still, it gives us a good idea of the performance to expect in upcoming machines.

So, which CPU comes out on top? Let's dive into the numbers.

Apple M5 vs Snapdragon X2 Elite Extreme: CPU tests

With Qualcomm boasting a 50% increase in CPU performance over its original Snapdragon X Elite chip and Apple seeing a 15% jump in power over the M4, there's clearly a drive to deliver the biggest gains over its last-gen processors.

For the most part, both tech giants achieve this, leading to some of the highest scores we've seen when it comes to CPU performance. That shows in the Geekbench 6 benchmark, which measures how fast a CPU is through handling tasks.

Just for reference, I've gathered results for both Apple and Qualcomm's previous-gen chips to show the jump in generational performance below.

When it comes to single-core performance — which is best for handling basic tasks on your laptop — Apple's M5 chip comes out on top with a 4,288 score. That's actually one of the highest scores we've seen when it comes to single-core power. However, the Snapdragon X2 Elite Extreme X2E-96-100 (the highest chip from the X2 Elite series) isn't far behind with a 4,074 score.

With the M5 chip already beating the Snapdragon's unreleased processor, it goes to show how far Apple has come with its M-series chipsets. But this all changes when it comes to multi-core performance, which performs best at multitasking and handling demanding apps.

The X2 Elite Extreme scored a whopping 23,449, which even outshines the powerful M4 Pro MacBook Pro's score. The M5 chip can't quite keep up, but considering the leaps it's made over its predecessor, it should still be applauded.

It's clear that the Snapdragon X2 Elite Extreme is dishing out some majorly advanced CPU performance. However, let it be known that this is Apple's base M5 chip, as we're expecting to see the M5 Pro and M5 Max make an appearance early next year.

Apple M5 vs Snapdragon X2 Elite Extreme: GPU tests

We're now seeing SoCs delivering extremely impressive graphics performance, without the need for a dedicated GPU. And if the M5 chip is anything to go by, with frame rates up to 55 FPS at 1080p resolution in Resident Evil 4 Remake, and up to 120 FPS with MetalFX enabled, we're all in for a treat.

But does the Snapdragon X2 Elite Extreme also have the gaming chops? It sure does, with the chip bringing up to 2.3x faster performance compared to the Snapdragon X Elite when playing games like Cyberpunk 2077 and GTA V.

When it comes to GPU testing, I only got to check out a few 3DMark benchmarks: Solar Bay, Wild Life Extreme and Steel Nomad Light. That's without trying to play popular PC titles, but it gives us a good comparison to measure against the M5 chip.

As it turns out, the graphics performance in several 3DMark benchmarks for both the M5 chip and Snapdragon X2 Elite Extreme aren't too far apart! Although, the M5 takes the overall win here, with up to 90 FPS in Solar Bay and 73 FPS in Wild Life Extreme. Still, in Steel Nomad Light, the X2 Elite Extreme takes the win.

Both chips are well above the scores of their predecessors, and considering the results, we should be seeing the latest games run at over 60 FPS at this point. Apple has the upper hand, though, with MetalFX being able to deliver even high frame rates.

In any case, the Apple M5 with its advanced 10-core GPU and Snapdragon X2 Elite Extreme's next-gen Adreno GPU are seeing an uplift in the graphics department, and that's more than a welcome sight in slim, lightweight laptops.

Apple M5 vs Snapdragon X2 Elite Extreme: AI tests

As with all the latest chips, AI performance plays a major role in powering the latest laptops. Apple and Qualcomm look to double down on this, with the Hexagon NPU in the Snapdragon X2 Elite Extreme featuring 80 TOPS (Trillions of Operations Per Second) and the M5 chip delivering up to 3.5x the AI performance over the M4.

So, how can these chips handle juggling AI tasks on PCs? Well, Qualcomm was right to claim it's offering the "world's fastest NPU for laptops."

Compared to its M5 competitor, the X2 Elite Extreme wiped the floor with its 88,615 score. Despite its strong improvements, the M5 chip still falls far behind. It's actually more than double what the previous Snapdragon X Elite could offer up.

This goes to show how much effort Qualcomm has put into its latest chipsets to achieve its superb AI performance results, so much so that it aims to continue to take a helpful load off the CPU with doing multiple tasks at once.

Apple M5 vs Snapdragon X2 Elite Extreme: Outlook

With Qualcomm's new chip set to be fitted in next-gen laptops in 2026, we won't have full test results until we finally get our hands on them. As you can see, though, the Snapdragon X2 Elite Extreme and the rest of its X2 lineup are a beast to compete with.

Will this make Apple sweat with its latest M5 chip? Doubtful. Sure, while the X2 Elite Extreme offers up some exemplary CPU performance, especially when it comes to multitasking, this is Qualcomm's strongest chip. And Apple has yet to reveal its full hand, with M5 Pro and M5 Max expected to be coming down the line.

Still, the latest Snapdragon X2 Elite processors impress, and are sure to give Windows Copilot+ PCs a major boost in CPU, GPU and AI performance. But Apple's M5 is already turning out to be a mighty chip for productivity and gaming, and this time next year, we may already have its M6 chip in an OLED MacBook Pro with a touchscreen.

More from Tom's Guide

• I tested the iPad Pro and the MacBook Pro — and this is the M5 device I’d buy
• Intel Panther Lake could power the future of handheld gaming — here's why AMD should be nervous
• I just tested the new Asus TUF Gaming A14 — and it's no longer my favorite gaming laptop

It's hard to imagine some of the best gaming laptops without a discrete GPU like Nvidia's RTX 50-series graphics cards, but the upcoming Intel Panther Lake CPUs sporting integrated graphics may change the playing field.

A next-gen Intel Core Ultra 300 series CPU has shown up in Geekbench, showing benchmark results for an Intel Core Ultra X7 358H processor in an Asus ROG Zephyrus G14. Not only does this reveal that we can expect Panther Lake to show up in one of the most popular gaming laptops we've tested, but it also shows its integrated graphics performance — without a dedicated GPU.

Showing its OpenCL result, which measures a GPU's power, it achieved a score of 52,014. As we've seen when testing Panther Lake, the ROG Zephyrus G14 configuration showed the processors with 4 P-cores (performance) and 12 E-cores (efficiency), along with a hint of the 12 Xe3 graphics cores as suggested by the 96 compute units. Plus, it even noted the integrated Arc GPU delivering 16GB.

Now, there's no telling what this test was really for, and it may not show the true strength of the integrated graphics performance of the CPU, considering Panther Lake is set to make its way to laptops in 2026. However, it tells us two key details:

• We'll likely see Intel Panther Lake power upcoming Asus ROG gaming laptops: With the Intel Core Ultra X7 358H in the Zephyrus G14, that's a good sign that testing is already being done for next year's Asus laptops.
• There may not be a need for a discrete GPU: There's no dedicated Nvidia GPU in sight in this benchmark unit, showing that Asus may be considering laptops without Nvidia's latest graphics cards.

Of course, this may not be the case at all. As per the result, while the 52,014 OpenCL score beats an RTX 3050 laptop GPU's Geekbench score of 50,918, which is an impressive feat for integrated graphics, it still falls behind an RTX 4050's 75,228 benchmark. We're now on entry-level RTX 5050 graphics cards, and even the latest Asus ROG Zephyrus G14 we tested comes with up to an RTX 5080.

We've seen how Intel's upcoming Panther Lake CPUs aim to deliver 50% more GPU performance compared to Lunar Lake, thanks to its advanced Xe3 architecture, and this certainly appears to be the case, according to another recent early benchmark showing 3DMark results for a Core Ultra X9 388H processor.

More integrated, less dedicated

All-in-one chipsets with integrated graphics are having their moment. Not only is Intel Panther Lake gearing up to be a fitting processor for gaming handhelds, but with Intel and Nvidia partnering up to deliver x86 RTX chips and Qualcomm making strides in gaming with its upcoming Snapdragon X2 Elite Extreme chip for laptops, we may see a new lineup of slim, portable gaming laptops with formidable performance in the next year.

This brings into question whether we'll see less dedicated GPUs like an RTX 50-series in upcoming gaming laptops. Even Apple's recent M5 MacBook Pro saw frame rates up to 55 FPS at 1080p resolution in Resident Evil 4 Remake, and up to 120 FPS with MetalFX enabled.

With AI tech such as Nvidia's DLSS 4 and AMD's FSR 4 offering upscaled visuals and higher frame rates, and this being put into action in Intel's latest Panther Lake chips with XeSS 3 tech introducing multi-frame generation with a 1.5x improvement, there may be less of a need for discrete graphics cards in laptops.

But not just yet. As we've seen time and time again, the latest lineup of GPU units in laptops delivers high-end gaming performance that integrated graphics in a CPU just can't match. Take the Alienware 16X Aurora with an RTX 5060 I tested, reaching over 160 FPS at high settings with DLSS 4 turned on.

Either way, there's a new era of gaming laptops on the horizon, and you may not find a dedicated GPU in them.

More from Tom's Guide

• Intel Panther Lake could power the future of handheld gaming — here's why AMD should be nervous
• I benchmarked the Snapdragon X2 Elite Extreme — here's how it compares to Apple M4, Intel Core Ultra 9 and more
• I spent 5,000 miles gaming on the ROG Xbox Ally X — it’s incredible, but something doesn’t add up

The Apple M5 chip is set to debut on three new Apple devices this fall. Utilizing an advanced 3nm process for enhanced performance and improved battery life, the new chip is expected to deliver a noticeable boost to the devices it powers.

The new devices in question are the M5 MacBook Pro, M5 iPad Pro, and, perhaps surprisingly, a new Apple Vision Pro. Apple only announced the base M5 chip, with no word on potential M5 Pro and M5 Max chips. Those might come later, but for the moment, the M5 chip boasts some impressive gains, according to Apple.

Here’s everything you need to know about the Apple M5 chip.

Apple M5 chip: Release date and price

The new MacBook Pro, iPad Pro and Apple Vision Pro with M5 will be released starting on October 22.

Right now, you can pre-order the M5 MacBook Pro from $1,599, the M5 iPad Pro from $999, and the Apple Vision Pro from $3,499.

Apple M5: Graphics

Thanks to the new GPU and enhanced shader cores in M5, devices featuring the chip will have increased graphics performance, with up to 30% faster performance compared to M4 and 2.5x faster performance than M1.

The M5 chip also has a third-gen ray tracing engine, which Apple says can provide up to 45% graphics uplift in apps that use ray tracing, such as video games. When combined with the rearchitectured second-gen dynamic caching, the GPU promises to deliver smoother gameplay and more realistic 3D visuals. The chip will also allow for faster rendering times for video projects and video games.

Apple M5: Enhanced AI

Apple says every compute block of the M5 chip has been optimized for M5, thanks to the chip’s new GPU architecture.

The 10-core GPU features a dedicated Neural Accelerator in each core, which can deliver over 4x peak GPU compute compared to M4 and over 6x peak GPU compute for AI performance compared to M1, according to Apple. These advancements will enable the new 14-inch MacBook Pro and iPad Pro to have accelerated processing for AI-driven workflows right on the devices.

The new 16-core Neural Engine promises to deliver better and more energy-efficient AI performance, which will complement the Neural Accelerators in the CPU and GPU to make the chip better optimized for AI workloads. Naturally, the Neural Engine will also enhance performance for Apple Intelligence. You’ll be able to use apps like Image Playground right on the device, with better performance than in M4.

Apple M5: Enhanced memory

Apple says the M5 chip has unified memory bandwidth of 153GB/s, which is a nearly 30% increase over M4 and more than twice that of M1. The new architecture lets the entire chip access a large single pool of memory. This lets the chip run larger AI models right on the device. This also powers faster CPU, GPU, and Neural Engine, allowing for higher multithreaded performance in apps, such as video editors and games.

With up to 32GB of memory capacity, the M5 chip can also help you run demanding apps like Adobe Photoshop and Final Cut Pro at the same time, while you’re uploading large files to the cloud in the background. If you’re a professional creative, this will certainly speed up your workflow.

Apple M5: Outlook

More from Tom's Guide

• Apple TV+ is getting a rebrand, and this could get confusing
• iPhone 17 vs iPhone Air: Which new iPhone should you buy?
• I just discovered what’s really draining your iPhone battery — and it’s not what you think

With Intel’s Panther Lake chip ready to deliver a major uplift in performance to laptops in 2026, we've now seen early graphics performance tests for what this CPU can deliver — and it already looks to crush Lunar Lake.

Benchmarks for Intel's flagship Panther Lake Core Ultra X9 388H processor have been shared by tech site LaptopReview, showcasing a 50% increase in graphics performance compared to Lunar Lake. According to its 3DMark results, it achieved a TimeSpy score of 6,300, with Lunar Lake reaching 4,000, as per the report.

This is a substantial improvement over Intel's Lunar Lake (Core Ultra 200V series) and Arrow Lake (Core Ultra 200H and HX) chips, and is on point with the tests we've seen when going hands-on with Panther Lake. This already shows that the next-gen CPU should deliver massive gains on PCs, but also power the future of gaming handhelds.

As noted in the report, Panther Lake's TimeSpy score even exceeds an RTX 3050 GPU. In our lab tests, we've seen an RTX 3050 Ti achieve a score of 4,637, so this checks out. However, it's still behind the likes of an RTX 4050, with the Acer Nitro V 15 we tested scoring 7,415.

We got a first look at Intel's Panther Lake series, and were impressed with the SoC's GPU onboard offering the new Xe3 graphics architecture and up to 12 Xe cores and 12 Ray Tracing Units. Over its Lunar Lake predecessor, this already aims to deliver big performance gains.

We've already expected 50% more GPU performance compared to Lunar Lake, and 40% higher performance per watt compared to Arrow Lake H series. Plus, its advanced XeSS 3 tech offers multi-frame generation for even higher frame rate potential in games. That's similar to Nvidia's RTX 50-series and DLSS 4, bringing Multi Frame Generation.

AMD may still have the upper hand

Despite the impressive benchmarks, the high-end Panther Lake Core Ultra X9 388H may fall behind its competition in AMD's Strix Halo APU.

We tested AMD's Ryzen AI Max+ 395 processor in the Framework Desktop, and in the 3DMark TimeSpy benchmark, it scored 11,530. There's clearly a significant gap in graphics performance, but considering this is a desktop, whereas Panther Lake is set to power laptops, this isn't a huge surprise.

It's still early days, and it's worth noting that Intel's Panther Lake runs on 45 watts compared to AMD's Ryzen AI Max+ 300 series chips can run up to 120 watts. There are still more tests to be done, but from the improvements these benchmarks show, Panther Lake is looking to be one powerful chip to give gaming performance a boost in notebooks.

There's plenty more to find out about Team Blue's latest Panther Lake chips, and how Nvidia and Intel's partnership will see these chips see even further leaps in performance.

More from Tom's Guide

• Intel Panther Lake could power the future of handheld gaming — here's why AMD should be nervous
• I benchmarked the Snapdragon X2 Elite Extreme — here's how it compares to Apple M4, Intel Core Ultra 9 and more
• Wi-Fi 8 just got its first-ever demo — here's what makes it different

This is a pivotal moment for Intel. Lifelines may have come from the U.S. government and Nvidia, but Team Blue needs to be in fight back mode — which is exactly what the company’s done with Panther Lake, the new chip that will eventually (probably) be known as Intel Core Ultra series 3.

And looking back, the timeline has led up to this moment. Arrow Lake (Intel Core Ultra 200H and HX) brought the performance without the battery life, whereas Lunar Lake (Intel Core Ultra 200V) went hard on battery life at the sacrifice of performance.

Panther Lake brings both of these together, while drastically amping up performance by more than 50% in CPU and GPU prowess, while bringing insane graphics tech to the table. Let me tell you about it.

Intel Panther Lake: Quick takes

• Core Architecture: Panther Lake utilizes a modular "System of Chips" design built on the Intel 18A node, featuring up to 16 next-generation Cougar Cove P-cores and Darkmont E-cores, clustered for optimized performance and efficiency.
• GPU Changes: The GPU introduces the new Xᵉ3 graphics architecture and scales up to 12 Xᵉ cores and 12 Ray Tracing Units, delivering big performance and featuring an increased 16 MB L2 Cache in the high-end configuration.
• Multi-Thread Performance: Multi-thread performance is boosted by more than 50% versus Lunar Lake and Arrow Lake when operating at similar power levels.
• Single-Thread Power Efficiency (vs. Arrow Lake): It delivers similar single-thread performance with more than 30% less power compared to Arrow Lake.
• GPU performance: The new Xe3 architecture results in 50% more GPU performance compared to Lunar Lake, and achieves 40% higher performance per watt compared to Arrow Lake H. XeSS 3 tech introduces multi-frame generation, which is huge for integrated graphics — 1.5x improvement.

Intel Panther Lake: Specs

Meet 18A

18A is not just another manufacturing step, it's the most advanced chip-making process currently in production in the United States — positioning Intel for a huge leap in CPU performance and efficiency. This 2-nanometer class node is the technological foundation for upcoming CPUs like Panther Lake, and it relies on two revolutionary changes.

This gets technical. We’re talking about the very fabric of how these process nodes are made, because as I’m about to get into, Panther Lake is really all about the numerous overhauls Team Blue has made to deliver big, but let me get into two key pillars of how it's built.

RibbonFET (The ultimate transistor)

• Transistors on a chip are always in two states: on or off. These are all controlled by gates, and in older Intel chips, these only covered some sides of the transistor — leading to power leakage.
• RibbonFet wraps the gate around all sides. Imagine trying to keep water from leaking out of a pipe by only squeezing the top. It still leaks from the sides. RibbonFET is like completely wrapping the pipe from all sides with the control mechanism. This perfect, "all-around" gate control ensures zero leakage.
• That allows for greater control of transistors, and allows Intel to shrink them even further without losing power — specifically a 15% performance-per-watt increase with a 25% power reduction.

Speaking of power…

PowerVia (Fixing silicon traffic jams)

• Instead of taking the traditional route of chip production, which is putting all the signal and power lines on one side of the chip, Intel is moving power delivery to the backside of the chip.
• Think of a multi-story city where power lines and data cables (signal lines) are all running on the same crowded street (the front side of the chip). This creates traffic jams and bottlenecks. PowerVia moves all the heavy power lines to a separate, dedicated highway underneath (the backside of the wafer).
• By making this move, the signal lines up top are now free to spread out, which dramatically reduces bottlenecks and allows Intel chips to be a lot more dense with cores and engines — bumping the chip density up by 1.3x.

Panther Lake takes full advantage of 18A (as you can see in the specs table), and the synergy of these two pillars translates to massive generational improvements.

Best of both worlds

Panther Lake makes a move from being a system on a chip (all the elements threaded together) to a more modular system of chips. You can see how Nvidia could insert itself into this conversation in future silicon with a GPU chiplet of its own!

But that's way further down the line (more on graphics later). Right now, what we have on the compute side of it is a familiar mix of performance and efficiency cores. The Cougar Cove performance cores have been engineered to drive up single-threaded performance, and deal with the high-demand work like games and content creation.

Meanwhile, the Darkmont efficiency cores are on hand and integrated alongside those performance cores to contribute significantly to multi-threaded performance. And for those low-demand tasks that fall into the casual productivity space, that’s where the low power E-cores come in.

There’s also a lot of software trickery at play here — including what’s called the Intel Thread Director. Basically, your Panther Lake chip has a boss that is really smart at looking at the tasks coming in and assigning cores to it.

And all these things lead to two key things: faster performance both when being strained and in more power-efficient moments. Single-threaded performance is more than 10% faster than Lunar Lake and Arrow Lake at similar power levels, and more than 50% faster multi-threaded.

Agentic AI at its core

While the Snapdragon X2 Elite is aimed at packing more TOPS into its NPU (80 to be specific), Intel is focused more on making better use of the surface area.

The 5th generation NPU here has a 40% improvement in TOPS in the area compared to Lunar Lake — combined with the CPU and GPU, that’s a total of 180 TOPS to help with Agentic AI capabilities coming down the line.

And efficiency is the name of the game, as the NPU consumes 10% less power than Lunar Lake and 40% less than Arrow Lake, while also being optimized for the latest ways AI workloads are conducted, such as Floating Point.

Huge GPU gains

Now for the fun part. The GPU has undergone a huge upgrade here, as it is now built on the Xe3 core architecture — driving both drastic improvements in efficiency and performance. Specifically, we're talking more than 50% faster than Lunar Lake, and 40% better performance-per-watt when compared to Arrow Lake.

These are significant numbers for GPU-intensive workloads like content creation, but the magic really starts to happen when you go gaming with it.

Intel was already on a great track with Lunar Lake (I loved the most recent driver update that unlocked more performance in the MSI Claw 8 AI+). But with Panther Lake, Intel is promising a 1.5x uplift in gaming performance, and that is Intel being “conservative” in their own words.

XeSS tech has also gotten an upgrade, too, and multi-frame generation is coming. If you’re not familiar with this tech since Nvidia launched it with RTX 50 series, the GPU will render one frame and use AI to predict the next three — essentially offering far smoother frame rates.

To have this in integrated graphics is, put simply, a bonkers power-up. On the demo units running at 45 watts, I saw Painkiller running at 220 FPS and Dying Light 2 at 140 FPS. Then, with 12 ray tracing units, these could very well overcome Intel’s old weakness in ray tracing.

It’s worth noting that since there isn’t any competitive testing, we aren’t able to confirm how Intel’s new GPU tech fares against AMD’s Strix Halo platform (the chip you’d find in the Framework Desktop).

In my mind, given the ludicrous specs of the Ryzen AI Max+ 395, Team Blue’s Panther Lake isn’t stretching to compete with this — that’s where the Nvidia partnership will come in. What we have here is more of a value unlocker to give you an improved price to performance.

The nice-to-haves

Then there’s the new image processing unit, which Intel is calling IPU 7.5. Basically, this covers what I call “just laptop things” — the things you don’t even think about, but also massively appreciate when they’re done well.

For example, better webcam support (you've got support for up to 16MP picture quality, and 1080p running at 120 FPS), AI-driven tone mapping for more lifelike color and contrast, and AI noise reduction to smooth that image.

But probably most eye-catching (literally) for me is the enhanced HDR. Processing the signals for that kind of content can take a toll on the chip’s power, but through hardware accelerated staggered HDR, the system can do it far more efficiently at a fraction of the power.

Once again, another series of small changes that make for a big difference.

Made in America

More from Tom's Guide

• I benchmarked the Snapdragon X2 Elite Extreme — here's how it compares to Apple M4, Intel Core Ultra 9 and more
• M5 MacBook Pro — all the rumors we’ve heard so far
• Nvidia N1X CPU: Everything we know so far

Qualcomm's latest Snapdragon X2 Elite Extreme processor has finally been unveiled, and we now have a look at the performance we can expect in upcoming Windows PCs. It's already one powerful chip, and I have the benchmarks to prove it.

During the Snapdragon Summit 2025 in Hawaii, I obtained results for Qualcomm's most powerful offering, the aptly named Snapdragon X2 Elite Extreme (X2E-96-100) CPU. The company promised numerous upgrades, including the "world's fastest NPU for laptops," along with up to a 50% increase in CPU performance, double the GPU power, and more.

From a look at the results, the X2 Elite Extreme delivers on its promise. Not only does it significantly surpass the Snapdragon X Elite, but also the latest processors from Apple, Intel and AMD in the best laptops on the market today.

The benchmarks were taken on a reference laptop model powered by the X2 Elite Extreme, along with 48GB of RAM and a 1TB SSD. Once the chip arrives on upcoming Windows Copilot+ laptops in the first half of 2026, we'll be able to test these laptops in our own labs (along with the other Snapdragon X2 Elite CPUs).

For now, I've compiled all the testing we've done on competing chips in the best Windows laptops, as well as Apple's M4 MacBook Air and M4 Pro MacBook Pro, against the X2 Elite Extreme's benchmarks.

Ready for some numbers? Let's dive in.

Snapdragon X2 Elite Extreme: CPU tests

It's clear Qualcomm has wanted to make a big leap in generational performance in its Snapdragon X2 Elite series compared to its first-gen chips, with its third-gen 3nm Oryon CPU boasting up to 18 cores. This is said to offer a 39% increase in single-core peak performance and a 50% boost in multi-core peak performance.

As expected, this appears to be the case when it comes to Geekbench benchmark results. In both single-core and multi-core scores, the X2 Elite Extreme delivered a massive jump in numbers. Better yet, it's completely outshining its competition from Apple's M4 chips, Intel's Core Ultra 9 Series 2 processors and AMD's Ryzen AI 9 300 series CPUs.

Receiving a 4,074 single-core score and a whopping 23,449 multi-core result on Geekbench 6.5 is quite the feat, especially considering the power its competition brings to the table.

It isn't surprising that the X2 Elite Extreme wiped the floor with the Snapdragon X Elite in one of our favorite laptops, the Dell XPS 13, seeing as this is the next-gen chipset, but it's still exciting to see the gap between the two.

However, even when compared to an Intel Core Ultra 9 285H in the Asus Zenbook Duo and AMD Ryzen AI 9 HX 365 in the MSI Creator A16 AI+, the X2 Elite Extreme delivers some massive gains over the competition. The same goes for the M4 MacBook Air, and that's especially apparent when it comes to the multi-core score.

What makes the Snapdragon X Elite Extreme even more impressive is its score over the M4 Pro MacBook Pro. Apple's M4 Pro is no slouch, as it's one of the most powerful laptops you can buy, but the X Elite Extreme actually gains the upper hand — even if it isn't by a massive amount. That's impressive stuff, especially for a processor designed for Windows machines.

Now, compared to a chip like the M4 Max (as we tested on the Mac Studio), the X2 Elite Extreme still falls slightly behind. The M4 Max delivered a single-core score of 4,113 and a multi-core result of 26,966. That doesn't mean the Extreme CPU isn't a mighty chip, though, as it shows that it's up there with one of the strongest processors around (plus, the M4 Max will set you back $2,000).

Overall, the Snapdragon X2 Elite Extreme is setting itself up to be one of the most powerful chips on the market. Now, we'll have to wait and see how Apple's rumored M5 chip handles the pressure.

Snapdragon X2 Elite Extreme: GPU tests

Now, when it comes to GPU testing, I only got to check out a few 3DMark benchmarks: Solar Bay, Wild Life Extreme and Steel Nomad Light. However, I did get a glimpse at Steel Nomad, too, which requires more demanding graphics power.

The Snapdragon X2 Elite Extreme aims to take PC gaming seriously, and the results show it can pull off some impressive performance. It's expected to offer 2.3x faster performance compared to the Snapdragon X Elite when playing games like Cyberpunk 2077 and GTA V, and the results show it can do this better than other mobile chips on the market.

The Snapdragon X2 Elite Extreme received a score of 23,157 (88.05 FPS) in the Solar Bay benchmark, 5,628 (41.69 FPS) in Steel Nomad Light and 11,530 (69.04 FPS) in Wild Life Extreme. In Steel Nomad, I saw a score of 1,306 (13.06 FPS), but take note that this is a higher-demanding benchmark that pushes GPUs to their limits.

The Adreno GPU is being pushed even further in the X2 Elite Extreme. With these cores, I could definitely imagine seeing games like Cyberpunk 2077 hit over (or around) 60 FPS with lower settings. That's an extremely impressive result for a CPU with an integrated GPU, even though it won't match the power of a full-on RTX 50-series laptop GPU.

Snapdragon X2 Elite Extreme: NPU tests

Finally, we have tests for the Hexagon NPU in the Snapdragon X2 Elite Extreme. Featuring 80 TOPS (Trillions of Operations Per Second), this blasts past the previous X Elite chip at 45 TOPS, but also processors from Intel (13 TOPS) and Apple (M4 at 38 TOPS).

During the benchmarking session at the summit, the Procyon AI Computer Vision (and Geekbench AI) benchmark, as shown by Qualcomm, revealed a significant gap in the X2 Elite Extreme's AI processing performance, which is claimed to be the "world's fastest NPU for laptops." According to these results, this rings true.

Compared to virtually all the CPUs it went up against, the X2 Elite Extreme showed over double the score by achieving 4,151 in the benchmark. It completely put the Intel Core Ultra 9 285H out of the race (understandably), but also topped the Apple M4 chip with its score of 2,121.

This was also seen in the Geekbench AI score, with the Apple M4 chip still being far behind despite being in second place in these benchmark scores.

The Snapdragon X2 Elite Extreme is one powerful beast in all departments, and I can't wait to check it out when it finally arrives on upcoming Windows PCs in 2026. Then, we'll see these results in action (and test them ourselves on consumer laptops in our labs).

Performance isn't the only feature I'm looking forward to when it comes to Qualcomm's latest CPU offering, as there are a few features I can't wait to try on the Snapdragon X2 Elite chips.

More from Tom's Guide

• Snapdragon X2 Elite: 5 key features I can't wait to try
• 5 things I wish I knew about Windows 11 when I upgraded from Windows 10
• I benchmarked the Snapdragon 8 Elite Gen 5 — the iPhone 17 Pro has already met its match

Qualcomm's Snapdragon Summit 2025 unveiled the highly anticipated Snapdragon X2 Elite CPUs, ringing in the next-gen era of Windows PCs and the "world's fastest NPU for laptops," featuring a leap in AI processing power.

Set for release in the first half of 2026, the latest series of Qualcomm's Arm-based processors already looks to deliver a massive jump in CPU performance, over double the GPU power, multi-day battery life, and, of course, a significant boost in all things AI.

That's all compared to the original Snapdragon X Elite CPUs that kicked off the company's venture in PCs, which is quite the statement to make, considering these CPUs can still be found in the best laptops on the market today. What's more, even compared to its competition in Intel, AMD and Apple (and now even Nvidia), X2 Elite processors have a lot to show off.

There are a few Snapdragon X2 Elite features to look forward to, but how do these next-gen CPUs compare to their first-gen Snapdragon X Elite counterparts? We still need to do our own benchmark testing, but in the meantime, let's dive in.

Snapdragon X2 Elite vs Snapdragon X Elite: CPU Performance

So far, Qualcomm has announced three new Snapdragon X2 Elite processors, which include the Elite Extreme model (which is just a way of saying this chip delivers peak performance compared to the rest). Just from the specs on paper, there is a clear uplift in power that can be seen from even the strongest Snapdragon X Elite CPU.

Here's a look at the specs in the Snapdragon X2 Elite Extreme (X2E-96-100) and Snapdragon X2 Elite (X2E-88-100) compared to the Snapdragon X Elite (X1E-84-100).

From the number of CPU cores and max boost frequency to processor cache, there's a clear major jump in specs, even without the "Extreme" moniker. As noted by Qualcomm, this translates to a 39% increase in single-core peak performance and a whopping 50% uplift in multi-core peak performance.

While we'll have to wait and see how these improved specs perform in the next-gen Oryon CPU (based on a 3nm process), we can still gauge how much stronger it will be compared to a Windows laptop running a Snapdragon X Elite processor.

Taking a look at the Snapdragon X Elite (X1E-80-100) in the Dell XPS 13 (2024), which achieved some of the highest scores when first benchmarking Snapdragon X Elite laptops, the gains here should be jaw-dropping.

With the X Elite scoring 2,797 single-core and 14,635 multi-core in Geekbench, the expected improvements Qualcomm claims should see the Snapdragon X2 Elite (at least the Extreme model) see scores skyrocket to 3,887 single-core and 21,952 multi-core. That's nearly on par with the MacBook Pro with M4 Pro (3,910 and 22,822, respectively).

While the Snapdragon X Elite offered impressive performance at the time, even surpassing the likes of an M3 MacBook Air, the Snapdragon X2 Elite CPUs aim to deliver a next-gen leap in CPU power.

Snapdragon X2 Elite vs Snapdragon X Elite: GPU Performance

I've written about how the Snapdragon X2 Elite shows that Qualcomm is getting serious about PC gaming, and from the numbers the company boasts, this should prove to be true compared to the Snapdragon X Elite chips.

The upgraded Adreno GPU in the X2 Elite series brings a boost in frequency rates, which aims to bring up to 2.3x graphics power when playing the latest demanding PC games.

Interestingly, the Snapdragon X2 Elite processors also come with support for 5K resolution at 60Hz across three displays. The X Elite can only support dual 5K displays at 60Hz, or three screens with 4K resolution at 60Hz.

This makes the X2 Elite's support for 4K (3840 x 2160) at 144Hz all the more impressive, and that's a high bar for high-end gaming. Now, the question is whether the Snapdragon X2 Elite can deliver the boosted gaming performance. Here's a look at the numbers Qualcomm claims the next-gen chips will offer in the latest games:

• GTA V: 1.6x
• Street Fighter 6: 1.8x
• Far Cry 6: 1.9x
• Dirt 5: 2.0x
• Shadow of the Tomb Raider: 2.0x
• Cyberpunk 2077: 2.1x
• Read Dead Redemption 2: 2.1x
• Black Myth Wukong: 2.1x
• Hitman World of Assassin: 2.2x

As I've previously noted, we tested Cyberpunk 2077 and GTA V when gaming on a Snapdragon X Elite laptop, with frame rates at 30 FPS and 57 FPS at around low-to-medium settings, respectively. From what Qualcomm states, we will see these games shoot up to 60 FPS in Cyberpunk 2077 at possibly higher settings, and close to 100 FPS in GTA V.

That's exciting stuff, but we won't know until we get some real-life testing done on upcoming laptops sporting Snapdragon X2 Elite processors.

Snapdragon X2 Elite vs Snapdragon X Elite: AI Performance

The Hexagon NPU continues to be one of Qualcomm's biggest selling points, with the first wave of Snapdragon X Elite processors topping (no pun intended) at 45 TOPS (Trillions of Operations Per Second). This component alleviates pressure from the CPU and GPU when put to work, leaving more space for the Oryon and Adreno units to work at maximum potential.

So, with the Snapdragon X2 Elite's 80 TOPS, this should kick on-device AI processing (no need for the cloud) to an even higher gear. That makes it (currently) the most powerful NPU in laptops, nearly doubling what the Snapdragon X Elite processors can handle.

In fact, Qualcomm states we can see a 78% increase in peak NPU performance compared to the previous generation, and according to the Procyon AI Computer Vision benchmarks the company showed during the keynote, we can expect up to 5.7x faster AI processing than some of the strongest silicon on the market today.

Time will tell how much stronger the X2 Elite chips are over the first-gen X Elite processors, but considering the level-up in TOPS and the comparisons made with 2025's lineup of silicon, we're sure to see a worthwhile jump in AI performance.

Snapdragon X2 Elite vs Snapdragon X Elite: Features

As it turns out, there aren't too many differences between the Snapdragon X2 Elite and X Elite processors when it comes to features, as these share Wi-Fi 7 and 5G connectivity (with the X2 series coming with Bluetooth 5.4). But there are still a few features worth noting.

As claimed, the Snapdragon X2 Elite will come with "multi-day" battery life. That's a bold claim, but one that has some legs to stand on. The Dell XPS 13 with a Snapdragon X Elite is one of the best laptops for battery life at nearly 20 hours, which is only an hour away from the M4 Pro MacBook Pro at just under 21 hours.

With a multi-day battery, we're hoping to see the chip's power efficiency boost the longevity of Windows laptops by a significant margin — likely over the claimed 27 hours of battery life in the Dell Premium 16. Of course, it depends on the laptops that manufacturers develop, too.

Another interesting feature coming with X2 Elite chips is Snapdragon Guardian. This PC management tool allows users to locate, lock and wipe data from their device via a Snapdragon Guardian app. So, no matter where you are in the world, you just need a cellular connection to find your misplaced laptop, lock it from a remote location or wipe it clean before threat actors get to it if it's stolen.

Even when these devices are "offline, powered down or unbootable," the built-in cellular modems, 4G, 5G, and Wi-Fi 7 connectivity allow them to be managed at any point.

There are sure to be more features that will utilize the next-gen power of Snapdragon X2 Elite processors from third-party developers, but that isn't to say the original Snapdragon X Elite CPUs won't get the same treatment (they're in the same family, after all).

For more on Qualcomm's latest offerings, check out all you need to know about its next-gen Snapdragon 8 Gen 5 chip for smartphones.

More from Tom's Guide

• All-new Humain Horizon Pro AI laptop announced at Snapdragon Summit 2025 and it comes with its own OS — here's what we know
• How to safely dispose of your Windows 10 laptop after upgrading to Windows 11
• Logitech just unveiled a solar keyboard that could last for a literal decade — here's my take after testing it

Now that Qualcomm has finally announced its Snapdragon X2 Elite CPUs, including an all-new X2 Elite Extreme chipset, we have a better idea of what these next-gen Arm-based processors can bring to upcoming Windows PCs. And it's already looking mighty.

Throughout the Snapdragon Summit 2025, I've been hearing how the X2 Elite series will offer up the "world's fastest NPU for laptops." From the brief comparisons shown at the keynote, it's made abundantly clear that Qualcomm will deliver on this promise.

Being up to 5.7x faster than its competition is a bold statement, considering the company specifically calls out some of the strongest CPUs in the best laptops right now, such as the Intel Core Ultra 9 285H, AMD Ryzen AI 9 HX 370 and Apple's M4 chip.

That Hexagon NPU with 80 TOPS (Trillions of Operations Per Second) of AI performance is already one major feature to look forward to, but the Snapdragon X2 Elite has other tricks up its sleeve.

While we wait for the first set of laptops (and other platforms) with Snapdragon X2 Elite chips to arrive sometime in spring 2026, here's a look at what else Qualcomm's next-gen processors are bringing to the table.

Faster CPU performance

When we benchmarked the first Snapdragon X Elite CPUs, we were majorly impressed with how they measured up to their competition at the time. Now, Qualcomm appears to be at it again, with a significant uplift in CPU performance.

Now, as always, we still need to do our own testing, but it's already looking like there will be a clear difference in the speeds we get in Windows PC sporting X2 Elite — and especially X2 Elite Extreme — processors. First, we know there will be a claimed 39% increase in single-core performance, and up to a more impressive 50% boost in multi-core speeds, compared to its first-gen chipsets.

Comparing this to the Dell XPS 13 with a Snapdragon X Elite X1E-80-100, which scored 2,797 single-core and 14,635 multi-core in Geekbench, it means we could see scores reach 3,887 single-core and 21,952 multi-core. That's just pure speculation, but those are some crazy numbers compared to, say, an M4 MacBook Air.

That should be all thanks to the 18 CPU cores (12 prime cores, 6 performance cores) fitted in its third-gen Oryon CPU. And take note, this will be on two of the three chips announced — not just the X2 Elite Extreme processor.

While the AI power is seeing major gains, so too is CPU performance, and it will be great to see how the next suite of Windows PCs will leverage this power.

PC gaming gets a boost

I'm happy to see that Qualcomm is taking PC gaming seriously with its X2 Elite offerings, as it was one of the main talking points during the big reveal. As in, up to a claimed 2.3x faster graphics performance in its advanced Adreno GPU.

This may not mean much, considering gaming on a Snapdragon X Elite laptop wasn't the main selling point, but what we do know is how much better it will handle even the latest, highly demanding PC games.

For instance, we tested Cyberpunk 2077 and GTA V on a Snapdragon X Elite laptop, with frame rates at 30 FPS and 57 FPS at around low-to-medium settings, respectively. From what Qualcomm states, we will see these games shoot up to 60 FPS in Cyberpunk 2077 at possibly higher settings, and close to 100 FPS in GTA V.

• GTA V: 1.6x
• Street Fighter 6: 1.8x
• Far Cry 6: 1.9x
• Dirt 5: 2.0x
• Shadow of the Tomb Raider: 2.0x
• Cyberpunk 2077: 2.1x
• Read Dead Redemption 2: 2.1x
• Black Myth Wukong: 2.1x
• Hitman World of Assassin: 2.2x

More upcoming PC titles will need to be optimized for the Snapdragon X2 Elite platform, but the company has stated it's working on just that. And it's bringing in a few more gamer-ready features, such as anti-cheat software support from Epic, Razer's Synapse app to bring more to a gamer-ready setup and an on-device AI experience in Alien: Rogue Incursion Evolved Edition using the X2 Elite's NPU.

Improved gaming on ultra-thin laptops? That's what we've been wanting to see from the Nvidia and Intel partnership in its upcoming x86 RTX chips. But Snapdragon CPUs are now also in the mix, and we'll have to see what this improved gaming performance delivers.

Going big on battery life

According to Qualcomm, battery life is the biggest factor for consumers choosing their next laptop. Well, it certainly helps, and if there's one feature I'd want to see these Snapdragon X2 Elite chips offer, it's the "multi-day" battery life that the company is touting.

Already, Qualcomm has done well in this department. The Dell XPS 13 with a Snapdragon X Elite is still one of the best laptops for battery life at nearly 20 hours, going head-to-head with beasts like the M4 Pro MacBook Pro at just under 21 hours.

As the company has stated time and time again, Snapdragon X-equipped laptops still deliver maximum performance when unplugged, with the X2 Elite series claimed to bring even better power efficiency, while its competitors comparatively require "222% more power" to reach their peak performance.

The real question, though, is whether we will see this "multi-day" battery life. We've heard of laptops expected to offer up to a whopping 27 hours in the Dell Premium 16, but since the claimed 20-hour battery in the Dell Premium 14 didn't pan out (tested at just over eight hours, actually), I'll believe it when I see it.

In any case, all signs are pointing towards a huge boost in battery life in Windows laptops rocking a Snapdragon X2 Elite, and that's a huge boon for workers on the move.

Remote security

Nobody likes having their personal items stolen, especially when it comes to a pricey laptop. It was made apparent in the keynote that 12,000 laptops are lost weekly in airports (although it appears this figure can be traced back to research done in 2008), but Qualcomm is bringing a new way to manage security on X2 Elite laptops.

With its Snapdragon Guardian, users have the ability to locate, lock and wipe data from their device via a Snapdragon Guardian app. With just a cellular connection, you can find your misplaced laptop, lock it from a remote location or wipe it clean before threat actors get to it if it's stolen.

That's a major win for security, and I saw this in action during a demo. What's more, even when these devices are "offline, powered down or unbootable," the built-in cellular modems, and 4G, 5G and Wi-Fi 7 connectivity, allow them to be managed at any point.

Hopefully, it's something that won't be used as much, but for security purposes, this is a feature that will be hugely beneficial.

Not just laptops

While Snapdragon X processors have become synonymous with laptops, the Snapdragon X2 Elite chips will be made for more than just mobile computing. While only showcased during the event (and in lovely red styling), we should see these CPUs come in mini PCs and AICs (all-in-one), too.

Mini PCs have become a popular option for many, bringing the performance of a desktop in a much smaller form factor that's even easy enough to travel around with. Plus, these have become more affordable options, with the Mac mini M4 being a standout. Bringing Qualcomm's next chips to more computer platforms is a step in the right direction, as it expands what these processors can accomplish beyond just laptops.

We'll have to see how these perform, especially considering they won't need the advantage of battery life. But I'm more excited to see what other form factors we'll find Snapdragon X2 chips in, even if that means a push towards gaming handhelds.

More from Tom's Guide

• All-new Humain Horizon Pro AI laptop announced at Snapdragon Summit 2025 and it comes with its own OS — here's what we know
• 6G is coming sooner than you think — Qualcomm is ready to have 6G devices 'as early as 2028'
• Logitech just unveiled a solar keyboard that could last for a literal decade — here's my take after testing it

While Qualcomm has been bragging about its all-new Snapdragon X2 Elite CPUs being equipped with the "world's fastest NPU for laptops," I've been focused on another key aspect that should make these chips for upcoming Windows PCs a certified win: PC gaming.

Being able to play the latest PC games on ultra-thin Windows laptops is finally taking shape, and slowly, we'll start to see less of a need for full-blown discrete GPUs like Nvidia RTX 50-series graphics cards. I mean, if the best gaming handhelds can play demanding titles with a single chip like the latest AMD Ryzen Z2 Extreme, what's stopping laptops?

Now, I've been wanting this to happen in the form of Nvidia's rumored N1X chips, and in the recent Nvidia and Intel partnership for Intel x86 RTX silicon. But with Qualcomm's Arm-based Snapdragon X2 Elite (and Elite Extreme) chipsets, PC gaming is taking priority.

In fact, the X2 Elite Extreme is specifically "Snapdragon's biggest advancement in PC gaming," as noted by Qualcomm's SVP of Compute and Gaming, Kedar Kondap, during the keynote. With the gains over its previous-gen processors, I'd believe it, but there's still some work to be done.

In any case, Qualcomm is finally stepping up to the plate by bringing something that many Windows users all secretly (and no-so secretly) want, and its Snapdragon X2 Elite brings the gaming goods.

Doubling performance

"X2" is a fitting name for Qualcomm's latest Snapdragon chips, as it also appears to mean the multiplied performance we can expect across the board when it comes to PC gaming on these machines.

With an 18-core Oryon CPU (to put that into perspective, Apple's M4 Max comes with a maximum of 16 CPU cores) split between prime and performance cores, an advanced Adreno GPU and max boot frequencies of 5.0GHz (1.85 GHz for the GPU), it's hardly a surprise that X2 Elite Extreme claims to bring up to a 50% performance boost.

But it's also coming with 2.3x faster peak GPU performance compared to last-gen processors. That translates to higher graphics settings and faster frame rates, which (I hope) is a huge win for playing the latest AAA PC games on ultra-thin laptops.

Just how much faster? According to Qualcomm, this includes the following:

• GTA V: 1.6x
• Street Fighter 6: 1.8x
• Far Cry 6: 1.9x
• Dirt 5: 2.0x
• Shadow of the Tomb Raider: 2.0x
• Cyberpunk 2077: 2.1x
• Read Dead Redemption 2: 2.1x
• Black Myth Wukong: 2.1x
• Hitman World of Assassin: 2.2x

Okay, that isn't much to go on. But we have tested gaming on a Snapdragon X Elite laptop, including Cyberpunk 2077 and GTA V. In Cyberpunk, we saw an average frame rate of 30 FPS with 2880 x 1620 resolution and low settings, and 57 FPS in GTA V at 1080p resolution at low-to-medium settings.

With all this in mind, that means we can expect to see over 60 FPS in Cyberpunk 2077 at possibly higher settings, and close to 100 FPS in GTA V. Does that match the heights of the best gaming laptops? No, but that's still a surprisingly capable machine that's built to be thin and light, and with long-lasting battery life.

It's not just for laptops either, as the Snapdragon X2 Elite chipsets will be coming in different form factors, such as mini PCs and all-in-ones. That means gaming on these platforms will also see a boost in gaming performance, which is hugely welcome!

Yes, the competition like AMD's Ryzen AI Max 395+ in the Asus ROG Flow Z13 already shows the type of gaming performance that can be had out of an APU, but that's one pricey machine (over $2,000), and doesn't reach the heights of a battery-efficient laptop, unlike a Snapdragon X Elite device (think nearly 20 hours of battery life in the the Dell XPS 13).

Now, we won't know how the latest Snapdragon chips will perform until we get our hands on them in spring 2026 (stay tuned for benchmarks). But so far, gaming with these chips already looks promising.

Building past casual gaming

Qualcomm is finally making its stamp on gaming with its Snapdragon X2 Elite CPUs, but it still has a way to go.

It's nice to see that it's supporting anti-cheat software from Epic to bring more competitive shooters (Fortnite, of course) into the ring, and that Qualcomm is partnering with Razer to optimize its Synapse app on X2 Elite systems. It's even bringing an on-device AI experience in Alien: Rogue Incursion Evolved Edition using the X2 Elite's NPU.

However, these are still stepping stones in pushing PC gaming on Snapdragon devices further. As noted by Qualcomm's Group GM of Mobile, Compute and XR, Alex Katouzian, Snapdragon X series was "never advertised as a gaming platform, ever... until now." That's true, seeing as its chip was made to drive through productivity and creative work.

"It's a really, really good casual gaming platform," Katouzian said. "The more we attract from developers, we're going to see an increase in development of graphics, and take it step by step."

I mean, it's clear that X2 Elite, and especially the Elite Extreme model, will be capable of playing the best PC games, and at much higher performance. But there's still a long way to go to reach the heights of a discrete GPU. That said, for those of us who mainly use a laptop for work and like gaming when they get the chance, this type of performance won't be sneezed at — and certainly not by me.

The takeaway

We may already have stunning, sleek gaming laptops like the Asus ROG Zephyrus G14 and Razer Blade 14, but combining AI power, battery efficiency and gaming performance in a single Arm-based chip in the Snapdragon X2 Elite series already looks to be a compelling choice for PC gamers — even if they err more on the casual side.

We know big manufacturers like Asus, HP and more will be delivering laptops (and possibly other platforms) equipped with Snapdragon X2 Elite and Elite Extreme chips on devices starting to arrive in the first half of 2026. That's when we'll know for sure just what the laptops are capable of gaming-wise.

Will the Nvidia and Intel partnership see its all-new processors with an Nvidia GPU built in surpass Qualcomm's next-gen chip? Only time will tell, but if we're seeing a new era of PC gaming on thinner, lighter laptops, then you won't see me complaining.

More from Tom's Guide

• Logitech just unveiled a solar keyboard that could last for a literal decade — here's my take after testing it
• Nvidia, Intel team up to build hardware: Here's why that's huge for PC gaming
• Intel promises Arc GPUs will survive the Nvidia deal — I'm not so sure about that

The leaks of Snapdragon X2 CPUs have finally been realized, as Qualcomm has revealed its next set of CPUs to power the latest PCs — Snapdragon X2 Elite and a whole new set of Snapdragon X2 Elite Extreme chips are here.

During the Snapdragon Summit 2025, Qualcomm finally announced its next Snapdragon X2 Elite Extreme and Snapdragon X2 Elite CPUs, its next-gen silicon to boost performance and efficiency in lightweight Windows laptops, and beyond (mini PCs included).

Not only can we expect up to 75% faster CPU performance in its third-gen 3nm Oryon CPU, but the Snapdragon X2 Elite Extreme also delivers a 2.3x uplift in graphics power and efficiency in its upgraded Adreno GPU, while also featuring up to 80 TOPS (Trillions of Operations Per Second) in its Hexagon NPU to bring the "world’s fastest NPU for laptops."

Qualcomm was the first to bring AI performance to Windows Copilot+ PCs, and it aims to boost performance, power efficiency and AI power with its next-gen chips — this time with an Extreme CPU contender.

The first devices with Snapdragon X2 Elite chips are expected to launch in the first half of 2026. In the meantime, here's what we know about Qualcomm's next-gen chips.

Snapdragon X2 Elite Extreme: Specs

Snapdragon X2 Elite Extreme: What we know

Just from their specs, the Snapdragon X2 Elite CPUs already show a major boost over the last Snapdragon X Elite chips. This includes up to 18 CPU cores (12 prime cores, 6 performance cores), which is up from the Snapdragon X Elite's 12 cores, 1.85 GHz max frequency in its next-gen Adreno GPU and 80 TOPS for improved AI performance — nearly double the amount of the original X Elite's 45 TOPS.

So far, three CPUs have been announced, including the Snapdragon X2 Elite Extreme and two other X2 Elite chips. Compared to its previous-gen processors, Qualcomm claims its latest chipset delivers 31% faster performance at ISO power and only needs 43% less power, translating to greater battery life.

What really brings the heat is the Snapdragon X2 Elite Extreme, set for "ultra-premium PCs." This chip will take full advantage of agentic AI, and aims to offer high-end performance for photo and video editing, data analytics and, of course, gaming in thin-and-light Windows PCs.

As for the upgraded Adreno GPU, the Snapdragon X2 Elite series claims to boost graphics quality and frame rates (including ray tracing), all while making the most of power efficiency to improve battery life.

The Snapdragon X2 Elite-series chips also offer Wi-Fi 7 and Bluetooth 5.4, while support for Snapdragon X75 5G Modem also means the latest in 5G connectivity.

While the boost in performance is sure to impress, Qualcomm also points out that we'll see multi-day battery life. Snapdragon chips are known for delivering the best battery life in laptops, with the Dell XPS 13 boasting nearly 20 hours of battery. With Snapdragon X2 Elite chips, we expect to see huge improvements.

Can it game?

As we've seen when gaming on a Snapdragon X Elite laptop, Qualcomm's chips are capable of playing the latest PC games while saving battery. However, X2 Elite looks to deliver even more.

In fact, PCs with Snapdragon X2 Elite chips will offer up to 4K (3840 x 2160) resolution with 144Hz refresh rates on their displays. That's on par with some of the best gaming laptops, all without a discrete GPU to deliver ultra-thin PCs. Of course, we'll have to see how these CPUs perform once we get hands-on time with them (more on that soon).

“Snapdragon X2 Elite strengthens our leadership in the PC industry, providing legendary leaps in performance, AI processing and battery life to enable the experiences that consumers deserve,” says Qualcomm's SVP and GM of compute and gaming, Kedar Kondap.

Kondap continues: “We continue to push the boundaries of technological innovation, introduce breakthrough products that set new industry standards and redefine what’s possible for PCs.”

That's not all that was announced during Qualcomm's annual Snapdragon Summit, as CEO Cristiano Amon announced that 6G connectivity will arrive in devices as early as 2028.

More from Tom's Guide

• Nvidia, Intel team up to build hardware: Here's why that's huge for PC gaming
• Nvidia and Intel's major deal puts the future of gaming consoles and handhelds at stake — and AMD should be worried
• Intel promises Arc GPUs will survive the Nvidia deal — I'm not so sure about that

It's official: Nvidia and Intel are partnering up, and is set to bring a fusion of Intel CPUs and Nvidia's RTX GPU power into a new breed of SoC (System on Chip): Intel x86 RTX silicon.

Two rival juggernauts of the industry are now in alliance, with Nvidia throwing $5 billion at Intel to kick-start a new era of chips in PCs. With Team Blue's massive share of laptops in the market and Team Green's dominance as a GPU maker, this is arguably the dream scenario not just for both companies, but for the rest of us, too.

Having an Intel SoC fitted with an Nvidia RTX graphics chip should deliver more thin, lightweight Windows laptops made for productivity and gaming. It truly would be a leap forward for PC gaming, even if it may be putting Intel's own Arc GPUs on thin ice.

But there's one player who may feel the sting of this partnership — AMD. Team Red has been gaining momentum with its latest suite of CPUs and GPUs, showing off incredible performance with its AMD Ryzen AI Max 390 in the Asus ROG Flow Z13, which shows off its integrated graphics prowess.

That's the kind of might Nvidia and Intel appear to be striving towards with their new investment into SoCs, but it isn't just desktops and laptops that are on the line here. With AMD's hand in powering the best gaming handhelds and the latest consoles, including the PS5 and Xbox Series X, we could see a change in the future of gaming consoles, and AMD may have a serious fight on its hands to remain the go-to chip maker.

Gaming laptops and beyond

Despite many of the best gaming laptops today coming equipped with Intel's latest CPUs, they only manage to be top-of-the-line machines thanks to discrete GPUs, whether it be from Nvidia or AMD.

Laptop manufacturers can do their best to make them more portable, with the Asus ROG Zephyrus G14 or Razer Blade 14 being fine examples, but they can't match the ultra-lightweight design or power efficiency of, say, a MacBook Pro sporting the latest M-series silicon that, for the most part, acts as a work laptop by day and a gaming machine by night — with a battery that lasts.

And so, having Nvidia and Intel's next venture into a CPU with integrated graphics would potentially solve that problem. But AMD has already been chipping away at this for a while. In fact, AMD put the scare in Intel with its Ryzen Z2 Extreme chip, surpassing Team Blue's Lunar Lake chip (the Intel Core Ultra 7 258V) in gaming handhelds.

That's why you'll find AMD's Ryzen Z2 Extreme in the most popular gaming handhelds arriving, including the ROG Xbox Ally X and MSI Claw A8. It simply pulls ahead of its competition, and there's a very good reason behind Sony and Microsoft continuously using a custom AMD chip to power their consoles.

The PS5 Pro managed to become the most powerful console on the market thanks to AMD, and as we learned from lead architect Mark Cerny, it will continue to push ahead with FSR 4 (FidelityFX Super Resolution) and Project Amethyst, the multi-year partnership with Sony and AMD.

But what if Nvidia and Intel's newfound partnership disrupts AMD's current journey into different platforms? Nvidia has an edge in the GPU market, with its GeForce RTX 50-series graphics cards bringing DLSS 4 (Deep Learning Super Sampling) upscaling tech, and it's taken the lead on how games can perform on PCs, whether it be on a desktop or laptop.

Since Nvidia and Intel's x86 RTX silicon should deliver more powerful and efficient performance for gaming and AI on laptops (as tech analyst Avi Greengart told Tom's Guide), if this expands onto other platforms like handheld PCs and consoles, then who's to say tech giants won't make a shift in chip makers?

Now, when this upcoming Nvidia/Intel chip will arrive is anyone's guess, and AMD already appears to have plans in place for its next-gen SoC in the rumored AMD Medusa Ryzen CPUs tipped for 2026. What's more, the long-rumored PS6 is expected to be powered by an AMD chip capable of being 8x faster than the PS5, with 4K at 120 FPS.

But with this shift in the tech industry, the future of gaming on different platforms could see a new name powering these devices, and Team Red may not be the first choice much longer.

Nvidia has its own card to play

During the chat to confirm the deal, Huang made it clear that the partnership has nothing to do with President Trump (as pointed out by The Verge) or Nvidia's own chip plans. If anything, that means Team Green is still pushing ahead on its expected Nvidia N1X chip.

Interestingly, that appears to be another deal Nvidia is striking up with MediaTek, which hints at bringing Arm-based CPUs for desktops and laptops. Again, that means slimmer, more powerful gaming laptops thanks to the might of Nvidia's latest graphics cards. Apparently, this SoC could arrive sometime in 2026, and it's looking to leverage the power of an RTX GPU.

There have been hints that the N1X will come with RTX 5070 desktop GPU power, boasting a 20-core CPU and 6,144 CUDA cores. This is exciting to hear, considering this is just from a single chip. If this is the performance we might expect, then Nvidia is setting itself up to shake up the industry.

Of course, without an official announcement, we won't know what kind of consumer-grade SoC we can expect from Nvidia until it's revealed. The performance of Nvidia's new lineup of silicon is still up in the air, but the implications of the top GPU maker venturing into the CPU world should put AMD and other chip makers on their toes.

The takeaway

It's not like Nvidia hasn't made its own custom chip for consoles before. The Nintendo Switch 2 is powered by a custom Nvidia processor featuring a GPU with dedicated RT Cores and Tensor Cores. But an all-new SoC with Intel by its side could drastically shift the type of chips we see in not just PCs, but handhelds and consoles, too.

Sure, it's not like we haven't seen Arm chips for Windows in action before, with Snapdragon X Elite laptops impressing with their long battery life and fast speeds. That said, we've tested Snapdragon X Elite PCs for gaming, and while impressive, they aren't quite built for demanding titles.

Still, there are other plans from Intel and Nvidia's competitors underway. For one, the AMD Strix Halo APU already delivers close to RTX 4060 desktop GPU power, and Qualcomm's Snapdragon X2 Series chip is set to arrive soon.

In any case, we're sure to see some exciting developments in the CPU market over the next few years, and AMD has now found an even bigger rival in a combined effort between Nvidia and Intel — and I can't wait to play on whatever devices come from this battle.

More from Tom's Guide

• I just tried the upgraded GeForce Now cloud gaming — and I'm seriously impressed
• Nvidia's RTX 50 SUPER GPUs looking to skip 2025 entirely — here's when they could launch
• Here's how to update your Nvidia GPU drivers in 5 easy steps

Nvidia and Intel have teamed up to build hardware together as Nvidia invests $5 billion in Intel, taking a nearly 5% stake in the company and potentially changing the future of PC gaming.

We're still unpacking what this Nvidia/Intel deal means for the future, and we got some clues this morning when Nvidia CEO Jensen Huang and Intel CEO Lip-Bu Tan conducted a webcast press conference to announce the deal and answer questions.

I sat in on that chat, and while much of the conversation was about AI and data centers there was some intriguing talk about how the two titans will collaborate on new chips that integrate Nvidia RTX graphics tech directly with Intel's CPUs.

That means we can expect a future in which thin-and-light gaming laptops and handheld gaming PCs are running on a new breed of Intel SoCs (System-on-Chip) with an Nvidia RTX graphics chip built right in.

A much-needed shot in the arm for Intel

I'm no business expert, but it's clear that Intel's fortunes have been falling for some time as layoffs rake the company while AMD, Apple, Qualcomm and even Nvidia offer serious competition in multiple markets. The U.S. government invested nearly $9 billion in Intel just last month to take a nearly 10% stake in the company, and now Nvidia has thrown in another $5 billion.

This generally seems like good news for Intel, and indeed the company's stock price climbed siginificantly in the wake of the announcement. However, it remains to be seen if this will ultimately be more to Intel or Nvidia's advantange.

For the rest of us, this could be a huge deal because there's still a lot of room to improve the experience of gaming on thin-and-light Windows laptops and handhelds. It's startling how quickly AMD and Qualcomm have made strides in building smaller, more capable SoCs, but in my experience you still can't expect the same level of performance as you can from a premium CPU paired with a powerful discrete laptop GPU.

And while Intel's consumer-grade CPUs power many of the best gaming laptops, in those cases they're partnered with discrete AMD or Nvidia laptop graphics cards that contribute to those laptops being heavy, power-hungry beasts. If you actually want a Windows laptop that plays games well yet is light enough to comfortably carry all day and capable of lasting longer than 4-5 hours on a full battery, you're going to have a rough time of it.

As much as I prefer Windows over macOS when it comes to gaming, a laptop like the MacBook Pro M4 is actually better for someone who wants a portable PC that can play games well and last through a full day of work or school. Sure, macOS doesn't support as many games as effectively as Windows, but you can play Diablo IV unplugged on your MacBook for a few hours and still count on it getting you through a day of classes. I can't say the same about almost any Windows laptop I've ever tested.

Part of the reason is that Apple silicon is remarkably efficient, which might have something to do with the physical design of its M-series chips. See, Apple employs a unified memory architecture that basically lets both the CPU and GPU on an M-series chip pull from the same unified pool of RAM, in a way that's slightly faster and more efficient than if the RAM was separate.

So when I read Tom's Hardware reporting that the forthcoming Intel x86 RTX SOCs will have a similar design, I got really excited. See, the folks at our sister publication were reportedly told that future consumer-grade Nvidia/Intel SoCs will have uniform memory access (UMA) so that both the CPU and GPU components of the chip will be able to draw from the same shared memory pool.

That could be a revolutionary step forward for Intel that helps secure its shaky place in the market, as well as afford Nvidia a stronger foothold in segments of the PC gaming market that can't afford its high-priced graphics cards. When I reached out to tech analyst and Techsponential founder Avi Greengart for his thoughts, he said something similar.

"The Intel / NVIDIA deal should mean more powerful and efficient silicon, especially for gaming and AI on laptops," said Greengart. "You can already put an Intel CPU and an NVIDIA GPU on the same motherboard, but there should be significant gains by moving to a chiplet architecture within an SoC. Intel insists that this doesn't change anything for its Arc GPU business, but given Intel's cash crunch as it builds out its Foundry business, it's hard to believe that Arc won't be impacted, at least for the discrete cards and higher-end laptop processors where NVIDIA now slots in. "

That makes a lot of sense, and it highlights that this deal isn't going to singlehandedly save Intel—the company still appears to be behind the ball relative to the competition.

But for folks like you and me, it could mean that the state of ultra-light gaming laptops and handhelds is about to get significantly more exciting.

Bottom line

Intel and Nvidia's announcement is still fresh and it's way too early to say definitively what impact their partnership will have on the PC market. However, I can say for sure it will definitely shake things up.

Frankly, that seems like exactly what Intel needs right now. The company has appeared to be in danger of crumpling for about a year now, but recent investments from Softbank, the U.S. government and now Nvidia have helped it survive long enough to lay off lots of people and try some fresh approaches. Hopefully this one pays off.

But really it seems like Nvidia comes out on top here, because it gets to take a stake in Intel, get some new x86 CPU hardware for data centers and new x86 SoCs with Nvidia RTX GPU tech built in. Plus, it comes out looking like a stand-up partner that's supporting a struggling U.S. chipmaker right as the U.S. government is making a lot of noise about buying and building local.

What remains to be seen is how impactful the resulting Intel x86 RTX SoCs will be when they eventually arrive on the scene. I doubt we'll see them anytime before 2026 at the absolute earliest, but I'm going to be counting the days because I want to see what Intel's SoCs can do after they get an injection of Nvidia's RTX graphics tech.

Maybe I'm reading too much into the Tom's Hardware claim that these upcoming chips will have uniform memory access, but I want to believe that Intel and Nvidia can do for portable PC gaming what Apple and Qualcomm have done for productivity laptops: upend expectations and fundamentally raise the bar for what we expect from our tech.

More from Tom's Guide

• Nvidia N1X CPU appears in new benchmark — but it doesn't show its true potential
• I played games for 100 hours on Intel Arc B580 — it proved to me that GPUs are Intel's way back from the brink
• Intel’s in trouble — AMD’s Ryzen Z2 Extreme pulls ahead where it counts for handheld gaming

I always love a “when enemies become friends story,” but Nvidia and Intel striking a deal to work together was not on my bingo card. In a game-changing announcement reported by Tom's Hardware, Team Green and Team Blue will be working together to develop chips.

These include data center CPUs for business and (most interestingly to me) Intel CPUs with Nvidia RTX graphics built into them for consumer PC gaming. On top of that, Nvidia will buy $5 billion in Intel stock, which has caused Intel’s stock price to skyrocket by over 30% today.

During a press conference, both Nvidia's Jensen Huang and Intel's Lip-Bu Tan stated their excitement for the partnership and how it will benefit both companies and the tech industry overall. They steered clear of answering questions about how and where their new SOCs will be manufactured, but we should learn that in due course.

While today's press conference is over, this is still a developing story with a lot of moving parts and big questions to ask: what does this mean for Nvidia’s own N1X CPU project? How will the likes of AMD and Qualcomm respond? So we’re covering every new development live.

What you need to know about the Nvidia-Intel deal

• Nvidia and Intel will work together to jointly develop new generations of x86 products over several years.
• Alongside the enterprise CPUs, this includes “Intel x86 RTX SOCs” — Intel CPUs infused with Nvidia RTX graphics.
• Nvidia will also buy $5 billion in Intel stock at $23.28 per share.
• This partnership is in the very early stages, and any product specifics will be disclosed at a later, unspecified date.

LIVE UPDATES

Welcome to the live blog!

There are a lot of moving parts here, so let's go quickly into how this is big news for you. Besides the business talk and the many different machinations (which I will cover), the big thing here is the Intel x86 RTX SOCs for the PC gaming market.

Up until this point, the integrated graphics in Intel Core Ultra CPUs have been pretty impressive. But it's clear that Nvidia has led the way on GPU performance, so to see a tight integration like this — an Intel CPU with Nvidia RTX chiplet for gaming performance — is mightily exciting.

Merging these into one package could make this a huge rival to AMD's own APUs that sport both compute and graphics in one package.

Press conference at 1pm ET / 10am PT / 6pm BST

The CEOs of Nvidia and Intel will be conducting a webcast press conference later today at 1pm ET. We'll be listening in to what is discussed about this seismic deal and what this means for the future of PC CPUs!

What will Intel and Nvidia make together?

This partnership looks set to be far-reaching through both enterprise and consumer computing, so let's break down what products look set to be made.

First of all, this isn't just a "one and done" situation, the companies will collaborate to "jointly develop multiple generations of custom data center and PC products." That's a multi-year joint venture! In terms of the specifics.

• Data centers: Intel will build Nvidia-custom x86 CPUs to be integrated into its AI infrastructure for enterprise use.
• PC gaming: Most enticing to me is the Intel x86 RTX SOCs. Simply put, rather than Intel's own integrated graphics on its Core Ultra CPUs, these chips will get Nvidia RTX GPU chiplets instead. That could open the door to a huge performance boost!

Let's talk about that Intel x86 RTX SOC

This is the most intriguing bit to me, so let's go into what this could be. Intel's current CPUs pack both the compute power and the company's own Arc GPU tech, and they've been pretty good.

What this Nvidia partnership will do is essentially perform a GPU brain transplant — taking that Arc GPU out and replacing it with an RTX graphics chiplet. This super tight integration brings huge gaming performance into a package on one chip, which could be huge competition for AMD's all-in-one chipsets too.

For PC gaming, this could knock it out of the park with access to DLSS 4, neural rendering technologies and more.

Handheld gaming is about to change

And while I could sit here and talk about the Intel x86 RTX SoC itself (and I will when we get more detail), I want to talk about the doors a partnership like this opens.

Handheld gaming has been largely dominated by AMD's chips like the Ryzen Z2 Extreme in the ROG Xbox Ally X. But the MSI Claw 8 AI Plus has been the one to stick with Intel and see some solid framerates because of it.

Now, with Nvidia DNA, PC handheld gaming could be incredible. DLSS tech has been outstripping AMD's FSR equivalent for a while now, and I'd just love to see this come to handhelds.

A "historic collaboration"

Nvidia founder and CEO Jensen Huang has been sharing his excitement over this new deal with Intel. “AI is powering a new industrial revolution and reinventing every layer of the computing stack — from silicon to systems to software. At the heart of this reinvention is NVIDIA’s CUDA architecture,” he commented.

“This historic collaboration tightly couples NVIDIA’s AI and accelerated computing stack with Intel’s CPUs and the vast x86 ecosystem — a fusion of two world-class platforms. Together, we will expand our ecosystems and lay the foundation for the next era of computing.”

"Innovating across our portfolio"

Heading over to Team Blue, Intel CEO Lip-Bu Tan has also been sharing his excitement for this partnership with Intel (unsurprising given the stock price has shot up over 30% since the announcement)!

“Intel’s x86 architecture has been foundational to modern computing for decades — and we are innovating across our portfolio to enable the workloads of the future,” he commented.

“Intel’s leading data center and client computing platforms, combined with our process technology, manufacturing and advanced packaging capabilities, will complement NVIDIA’s AI and accelerated computing leadership to enable new breakthroughs for the industry. We appreciate the confidence Jensen and the NVIDIA team have placed in us with their investment and look forward to the work ahead as we innovate for customers and grow our business.”

"Seismic for Intel"

For more thoughts, I've reached out to Paul Alcorn, Editor-in-chief of Tom's Hardware who originally broke the story. This is a pivotal moment and a real lifeline for Intel, and Alcorn agrees.

"The announcement is nothing short of seismic for Intel, which already counts the United Sates of America as its largest shareholder, and now counts the world's largest company, Nvidia, as one of its other largest shareholders."

And in terms of their personal computing aspirations, AMD should be very worried by this announcement.

"Intel's new x86 RTX CPUs will compete directly with AMD's APUs by design. For AMD, that means it faces intensifying competition from a company with the leading market share in notebook CPUs (Intel ships ~79% of laptop chips worldwide) that's now armed with GPU tech from Nvidia, which ships 92% of the world's gaming GPUs."

AMD should be worried

As you know, I've been blown away by the AMD Ryzen AI Max 395+ chip found in the Asus ROG Flow Z13. It's integrated graphics performance is simply mindblowing and can rival a lot of dedicated GPU gaming laptops.

APUs like this have been Team Red's bread and butter, but to have Intel and Nvidia coming together — two titans in note CPUs and gaming GPUs — to build its own SoCs is going to bring a massively intense competition to AMD.

DLSS4 is already quite a ways ahead of any FSR capabilities, and the game support for Nvidia's neural rendering smarts is far larger too. This seems like a slam dunk.

The $5 billion deal

At $23.28 per share, Nvidia is now one of the largest shareholders in Intel behind the U.S. Government. Part of me is a little curious about whether there is another intention to this deal — but that will be all dependent on whether Nvidia goes on a shopping spree and buys more. The agreement is clear in exactly how much right now.

Wait...this sounds familiar

As Paul highlights in his report for Tom's Hardware, this isn't the first time Intel has joined up with another company for its GPU tech. Back in 2017, there was a Kaby Lake-G chip that was fused with an AMD Radeon GPU chiplet.

This is sort of similar in concept to what you'll see between Intel and Nvidia, but there's some significant differences. On paper, it's the same principal — take Intel's CPU and put and AMD GPU on the same die.

But in this upcoming SoC, the RTX GPU will be connected by the vastly faster and more efficient NVLink interface, and there will be uniform memory access — meaning the CPU and GPU will be able to access the same stock of memory intelligently.

What does this mean for Arc graphics?

It's a tough question. Intel moving into this partnership with Nvidia does bring the planet's biggest GPU-maker in, and that could be curtains for Arc GPU.

That's a shame — especially given the huge gains the company has made in this space (particularly with the Intel Arc B580), but given how far ahead Nvidia actually is, this wouldn't be a surprising conclusion.

Ming-Chi Kuo weighs in

You'll know that name as the analyst who is usually pretty bang-on with Apple rumors and leaks! But the attention has been turned to this Nvidia-Intel deal. Here's what we've learned from this analysis:

• Nvidia N1X is up in the air: Nvidia has been rumored to be working on its own ARM processor, but that now "carries high uncertainty." Intel was facing "rapidly enhancing competitiveness in the GPU sector," so to tie these two together makes the world of sense.
• High synergy for enterprise: Intel has the distribution channels for x86 servers, while Nvidia holds all the cards technologically (AI chips, NVLink and CUDA to name only 3), so to bring them together makes for a huge opportunity to "significantly benefit from substantial potential demand"
• TSMC remains unaffected for now: There is the question of using Intel's foundry for producing more chips in this partnership. But with the "advanced process advantage" of TSMC chip making expected to "persist at least until 2030," this agreement doesn't look set to rock the boat here. More importantly though, it could impact the market share of competitors like AMD and Broadcom.

What happened to Intel's stock price?

So it's been no real shock that Intel's been struggling in recent months — reception to CPUs has been tepid as rivals move ahead. The financial results and redundancies prove this.

But this Nvidia partnership has been such a lifeline for the company, as the stock price jumped up by over 30% before opening, and is now at a 26.8% increase over its previous price.

What is Nvidia's main goal here? Here's my take

So I've talked about this being a lifeline for Intel's hopes of competing on the world stage of chip production, but what will Nvidia get out of this deal? Put simply, in my opinion, it's the license to build x86.

AMD's Strix Halo (Ryzen AI Max 395+) is in a world of its own in terms of an uber-powerful APU with integrated graphics that runs on the x86 platform. Nvidia doesn't have an x86 license, so they have to make their own Arm CPUs. The biggest example of this is definitely the Tegra chip you see in the Nintendo Switch 2. Another example is the DGX Spark mini AI PC.

But now the tie up with Intel is happening, Nvidia has the chance to create a DGX Spark-like product with an x86 CPU — that is a significant tectonic shift in the chipset market.

A monumental announcement

Hey, all! Tony Polanco here to take over this blog for the next few hours. That means I'll be posting during the event, which, as we've said before, is audio-only.

As stated before, this news is easily one of the most significant for all of tech. It's not every day that two giants of the industry have a partnership like this. The fact that it's the world's biggest CPU maker and the world's biggest GPU maker makes it all the more monumental.

We likely won't see the fruits of this partnership for some time, but the prospect is exciting nonetheless. You don't want to miss a minute of the presentation, so make sure you stay here for all the latest updates.

It's almost time!

In just 15 minutes, the CEOs of Nvidia and Intel will be conducting a webcast press conference. Again, we'll be listening in and posting what's discussed, so you definitely want to keep it locked in here!

5 minutes left!

It's almost time! Again, this is going to be an audio-only press conference. But don't fret, as I'll make sure you don't miss any of the important details.

And we're live!

Here we go! The CEOs of Nvidia and Intel have begun the press conference.

A historic partnership

Jensen Huang begins by discussing the history of computing. This includes Nvidia's involvement with developing the world's first GPU.

Huang then discusses the partnership between Nvidia and Intel, which aims to push the industry forward. This will bring x86 into Nvidia's systems and thus fuse the best aspects of Intel CPUs and Nvidia GPUs.

Intel's CEO speaks

Intel CEO Lip-Bu Tan expresses his excitement in building "a new era" for both companies, a "game-changing" moment.

This collaboration builds on the core strengths of each company. Tan says this collaboration will be good for the entire tech industry. Customers will also benefit thanks to this partnership.

Time for questions

CNBC's Jim Kramer is the first person to ask a question.

When asked how this partnership will benefit Intel and Nvidia, Jensen Huang says that for Nvidia, having access to x86 gives the company the flexibility to scale its supercomputers further than it could on its own.

As for the gaming market, the companies hope to create an SOC that fuses the two processors (CPU and GPU) to create a new class of integrated laptop chip. Tan reiterates how these chips will have the best aspects of the company's respective technologies.

Kramer then asked if Nvidia will use Intel's foundry, but Huang said today's news is about CPUs, not exactly how (or where) these chips will be manufactured. Nvidia says it will become a large buyer of Intel chips, as long as a supplier.

When will we see devices?

Bloomberg's Ian Cain asks the next question, which is about when we'll see devices that utilize the companies' chip. Huang didn't say specifically, though he stated his announcement.

Another question is about the Trump administration's involvement, to which Huang said the American government has no involvement with today's partnership. Huang then stated his admiration for TSMC and what it has done for the industry with its chip manufacturing.

Huang says today's announcement is purely about the custom chips Nvidia and Intel will produce. This involves chips for data centers and laptops.

The manufacturing process

Yahoo Finance's Laura Bratton asks about the manufacturing process the chip will use.

Lip-Bu Tan says that's something they will discuss at a later time. As Huang previously stated, today's press conference is about discussing the partnership. He also says Nvidia is committed to ARM. The company's ARM roadmap isn't affected by Intel and Nvidia's partnership.

Nvidia's investment in Intel

Reuters inquires about the investment side of the partnership.

Jensen Huang says Nvidia is delighted to invest in Intel and thinks it will be "fantastic" for Intel. Lip-Bu Tan says he wants to strengthen the company's balance sheet and that the partnership will benefit both Intel and Nvidia.

Enterprise vs. consumer focus

When once again asked how the chips will be manufactured, Intel's Tan reiterates that today is about discussing the partnership. As far as which sector these chips will primarily focus on (enterprise vs. consumer), Huang said both are important.

The economics of x86

Bloomberg asks about the economics of x86 architecture. Jensen Huang says CPUs are sold as separate servers that are then fused with Nvidia GPUs in its data centers. Now, the company can buy the CPUs from Intel and integrate them. Nvidia will sell its GPU chiplet in a passthrough way with Intel or to Intel. He says this will expand the market for both Nvidia and Intel.

Regarding the culture at Intel after today, Tan says there will be a lean and fast-moving culture within Intel. He wants to match Nvidia in that regard.

Another question about ARM

When asked about how this partnership will affect Nvidia's ARM plans, Huang once again says nothing is changing on that end.

Target customers

Regarding customers, Huang says the x86 architecture is still large, so today's partnership will allow the company to produce more chips for both enterprice and consumers. Since the majority of enterprise is x86-based, they'll benefit from Nvidia's AI technology in the upcoming Intel/Nvidia chips.

Closing remarks

Jensen Huang once again reiterates his excitement for today's partnership with Intel and how it will drive the industry forward. Lip-Bu Tan says it thanks Nvidia for its trust and support, and says this collaborating will deliver a new compute platform that opens new doors.

Press conference impressions

We didn't hear a lot of information that we didn't already read in Tom's Hardware's report. Several reporters asked about how and where Nvidia and Intel plan to manufacture their chips, but were denied an answer every time. It's also unclear when we might see products utilizing this chip, though I wasn't expecting a solid answer.

Today's presser was really a chance for the CEOs of Intel and Nvidia to share their excitement about the partnership.

No device announcements

Like I said, we didn't get any announcement of potential devices that will utilize Nvidia and Intel's SOC. While that's not surprising, it's somewhat disappointing that we didn't get word about partners.

I know it's entirely too early to discuss that, no doubt, since deals are probably being inked, but knowing that manufacturers like Lenovo, Asus, MSI, or others would use the new SOC would've been nice. I guess we'll just have to wait a little longer for that news to drop.

Handheld potential

Even though we didn't get word on which devices would use Nvidia and Intel's chip (it's still weird writing that!), it doesn't take a genius to figure out the SOC will power laptops. That, in turn, opens the door for handheld manufacturers to use this chip.

Right now, AMD rules the roost when it comes to processors for handhelds, with devices like the Steam Deck and Asus ROG Ally X using Team Red's chips. In the future, we may see handhelds that utilize Nvidia and Intel's chips.

This would be huge, especially since the Switch 2 is the only handheld that uses an Nvidia chip. And in a strange twist of fate, Intel, which hasn't had a great track record with its Arrow Lake CPU in handhelds, could gain serious ground in this market via this partnership.

If you're a fan of handhelds like I am, then the prospect of a chip to truly rival AMD is huge.

Who will manufacture?

One of the big questions about Nvidia and Intel's chip is who will manufacture it. Both CEOs declined to answer every time they were asked, but it's a safe bet that this is something they're discussing behind closed doors.

Intel has long fabricated its own chips, though it had begun partnering with TSMC in recent years. The latter is currently the leading chip manufacturer for companies like Samsung and Apple (among many others). When asked how Nvidia and Intel's manufacturing plans could interfere with TSMC, Jensen Huang was quick to state his admiration for TSMC.

So, will Intel produce the new chips, or will it be TSMC? It's hard to say at this time, but the potential for a true TSMC rival in this field would be massive.

The end of Intel Arc graphics?

Jason discussed this earlier, but it's worth bringing up again since it's so crucial. Intel's latest CPUs have the compute power and the company's Arc GPU tech, which have made their latest Intel Core Ultra series generally good.

However, if Intel chips effectively ditch the Arc GPU and replace it with an RTX graphics chiplet, it would spell the end of Intel-based graphics tech like XeSS. However, given how Nvidia is far ahead of everyone when it comes to GPUs, perhaps this loss won't be as devastating. Having DLSS on an Intel chip would certainly be interesting.

AMD is on notice

I'm wrapping up my posts for the day, but before I go, I wanted to discuss one last thing. And that is that the Intel x86 RTX SOCs for the PC gaming market can't be understated.

As we've said before, Intel's integrated graphics have improved, especially with the latest Core Ultra chips. That said, Nvidia has led the charge on GPU performance. Because of that, Nvidia's partnership with Intel is very intriguing. In short, this chip will combine an Intel CPU and an Nvidia RTX chiplet for gaming performance.

Combining these two into one SOC could make this a major rival to AMD's APUs that also pack compute and graphics.

It's been over a year since Qualcomm unleashed its Snapdragon X Series CPUs on the first lineup of Copilot+ PCs, and now it's time for next-gen processors — with the rumored Snapdragon X2 CPUs expected to be just around the corner.

With Qualcomm's range of Snapdragon X chips, including the Snapdragon X Elite, Snapdragon X Plus and Snapdragon X, the company gave Windows laptops the performance and energy efficiency to rival its competitors. Even giving Apple's M-series CPUs a run for their money.

In fact, one of the best laptops still around is the Dell XPS 13 with a Snapdragon X Elite processor, boasting nearly 20 hours of battery life and performance nearly on a par with the MacBook Air M3. However, that was over a year ago, and new Apple, Intel and AMD chips have arrived to deliver AI-powered improvements (even a rumored Nvidia N1X CPU is expected to be in the works).

But this could all change very soon with the arrival of Snapdragon X2 CPUs. With Qualcomm's Snapdragon Summit 2025 announced and rumors indicating an all-new chipset with incredible specs, all signs are pointing to a fresh slate of processors. And soon.

Let's dive into what we know so far.

Snapdragon X2: Rumored release date

This should be an easy one. Qualcomm is holding its annual Snapdragon Summit on September 23, 2025, taking place in Hawaii. That should set the stage for a Snapdragon X2 Series announcement.

The keynote is expected to take place on September 23, with CEO Cristiano Amon taking the stage to hopefully announce the latest CPUs. Of course, we won't know for sure until the event takes place, but there's a good chance this will be where the tech giant spills the beans on its next laptop processors.

With the Snapdragon X Elite chip announced during the Snapdragon Summit 2023, we may at least see a Snapdragon X2 Elite arrive. It wasn't until April 2024 that we saw the Snapdragon X Plus, and then CES 2025 when we saw the Snapdragon X CPU.

But many believed Qualcomm would announce its Snapdragon X2 lineup at Computex 2025, which didn't turn out to be the case. Amon specifically said: "There will be no focus on chip announcements" at the event. So, it's still unclear if we will see Snapdragon X CPUs by the end of September after all.

That said, right at the end of the Qualcomm keynote, Amon confirmed that "the revolution continues at Summit 2025." Now, that's heavily implying that Snapdragon X2 is set to be revealed in September.

As for when the Snapdragon X2 chips will launch in laptops, this may not be until early 2026 — if the X Elite laptop launch is anything to go by.

Snapdragon X2: Performance

If there's one thing that's certain about Qualcomm's chips, it's that the Snapdragon X Series delivers an exceptional balance of performance and power efficiency.

The tech giant even slammed Intel for being able to run a laptop at "max performance when unplugged." This was made apparent when I saw how a Snapdragon X Elite could outperform an Intel CPU. But, of course, it depends on the laptop they're powering, too.

With this being the case, what performance can we expect in the rumored Snapdragon X2 processors? According to known leaker Roland Quandt (via TechPowerUp), the Snapdragon X2 Elite looks to be a powerful chip. It's expected to be an 18-core CPU based on Oryon V3 CPU architecture.

SC8480XP aka SD X2 Elite in testing with 64GB RAM... looking like 18 core thing more and more.

— @rquandt.bsky.social (@rquandt.bsky.social.bsky.social) 2025-09-05T20:15:51.102Z

Interestingly, the tipster notes that it's being tested on 64GB configurations, showcasing the type of laptops and possibly even desktops that will best suit the performance of the chip.

Compared to the Snapdragon X Elite, which features 12 cores, a base clock of 3.4 GHz, and 45 TOPS (Tera Operations Per Second), the X2 Elite appears to have a 50% increase in CPU cores. There have been murmurs that this would see up to a 30% increase in single-thread performance, and an even bigger increase in multi-thread performance.

All of this translates to better performance in PCs, which may outclass the current roster of CPUs, including the Intel Core Ultra 2 series, AMD Ryzen AI 300 and M4 chips in MacBooks.

We'll get a proper gauge of what to expect once Qualcomm officially announces its Snapdragon X2 Elite processor, and it's likely X2 Plus and X2 chips. But it may not just be for laptops this time around.

Snapdragon X2: More than laptops?

As mentioned, the Snapdragon X2 Elite looks to come packed with plenty of performance power, and rumors (via Tom's Hardware) suggest Qualcomm may be looking to bring its next-gen CPU lineup to desktops, too.

This comes from a leak indicating that Qualcomm tested its X2 Elite with 120mm AIO cooling solutions. This is an all-in-one cooling system usually found in desktop towers to keep components nice and cool.

It wouldn't be out of the realm of possibility, as Qualcomm's range of Snapdragon chips can be found in a majority of devices. That includes smartphones like the Samsung Galaxy S25, tablets including the RedMagic Astra for gaming, the Meta Quest 3 and more. It's even taking the lead with new tech in XR glasses.

These are all different SoCs for different platforms, but seeing how it's expected that Snapdragon X2 chips will be in more than just laptops, it wouldn't be completely shocking.

Snapdragon X2: Worth the wait?

More from Tom's Guide

• 5 things to look for when buying a laptop, from someone who reviews them for a living
• I just got up close with the first Intel Panther Lake laptop — and it looks incredible
• 3 expert tips for buying a student laptop that will actually get you through school

Next-gen CPUs are always on their way, but now we've got a taste of what Intel and AMD may be cooking up, and when we may see these new processors.

Reliable leaker momomo_us on X has revealed a roadmap of Intel and AMD mobile CPUs set to arrive until 2027, showcasing the processors we can expect to see in upcoming laptops. According to VideoCardz, this comes from an unknown laptop manufacturer, with the roadmap dated back in May.

As per the leak, we can expect Intel Core Ultra 300 CPUs, also known as Panther Lake, and AMD Medusa Ryzen CPUs, in Q2 of 2026. This puts the launch window around April, but they could come earlier, seeing as Intel already had demos of Panther Lake CPUs at Computex 2025.

Interestingly, we also get a glimpse of what Intel has planned for 2027. If accurate, we can expect Intel Core Ultra 400 CPUs — codenamed Nova Lake — just a year after Intel launches its Panther Lake SoC on laptops.

There has been plenty of talk on Intel's Nova Lake, including CPU spec leaks showcasing double the performance and efficiency over Arrow Lake. We also saw Nova Lake-HX, H and U processors in this leak, although this was expected to be for next year's lineup of chips. With this leaked roadmap, those specs may have to wait until 2027.

While AMD's future plans stop in 2026, and don't show the series of CPUs that will arrive, it's likely to be the follow-up to its Ryzen AI 300 series CPUs. If that's the case, expect a suite of Ryzen AI 400 chips coming early next year.

CPU competition heats up

We've heard about Intel and AMD's chips for a while, but there are others now in the limelight. With rumors of Qualcomm's Snapdragon X2 chips and Nvidia's N1X coming out of the woodwork, Team Blue and Team Red will need to deliver some reputable performance to shine through its rumored competition.

For one, the N1-series chip will reportedly deliver the same performance as an RTX 4070-equipped laptop, but it's also been hinting at RTX 5070 power in recent reports. For gamers and creatives, that's a huge gain in a laptop, especially if it keeps the price down without having a discrete GPU. Plus, it makes for an ultra-portable machine.

While gaming on a Snapdragon X laptop doesn't compete with the likes of an Intel or AMD machine (think Asus ROG Flow Z13), the Snapdragon X2 chip is expected to deliver a major boost in performance, if the rumored 18-core CPU configuration and up to 64GB of LPDDR5X RAM in the X2 Elite is to be believed.

Either way, it's looking to be less than a year until we see what Intel and AMD have up their sleeves with their next-gen CPUs. That's still a while away, though, so for the best laptops around right now, we've got you covered.

More from Tom's Guide

• Nvidia N1X CPU appears in new benchmark — but it doesn't show its true potential
• I played games for 100 hours on Intel Arc B580 — it proved to me that GPUs are Intel's way back from the brink
• Intel’s in trouble — AMD’s Ryzen Z2 Extreme pulls ahead where it counts for handheld gaming

Nvidia's rumored N1X chip has recently shown up on Geekbench, giving use a sneak peek at its specs that appear to match an RTX 5070 desktop GPU. Now, the Arm-based CPU has popped up in yet another benchmark.

The N1X CPU has been spotted in a FurMark benchmaking tool (via VideoCardz), with results showing its GPU capabilities. Codenamed "NVIDIA JMJWOA," it received a OpenGL score of 4,286, but it wasn't at its full GPU usage.

According to the listing, the N1X used 63% of its maximum GPU capabilities, with its result putting it well under an RTX 5060 desktop graphics card. However, seeing as the previous Geekbench benchmark showed it coming with a 20-core CPU and 6,144 CUDA cores, the same number as an RTX 5070 desktop GPU, it's sure to offer far more performance power.

Interestingly, the listing shows that Nvidia, or a manufacturer testing the chip for upcoming PCs, is testing the Arm-based CPU on Windows 11. This simply means that the N1X SoC will run on Windows — as many have expected.

While this benchmark doesn't show just how well the N1X will perform (similar to the Geekbench result), it does give us yet another tease that Nvidia's anticipated N1X may be closer than we think, seeing as benchmarks popping up is a good sign that it's approaching a release.

When will Nvidia's N1X launch?

More from Tom's Guide

• Nvidia wants to make 8GB GPUs great again with AI texture compression — but I'm not convinced
• Nvidia RTX 50 SUPER GPUs may arrive sooner than you think — even before CES 2026
• I just tried Alienware’s new 18-inch RTX 5090 gaming laptop — and it’s ridiculous in the best way possible

Rumors are heating up around Nvidia's N1X Arm-based CPU, and thanks to a leaked benchmark, we may have an idea of its integrated GPU capabilities — and it's hinting at RTX 50-series GPU power.

A Geekbench benchmark listing for the Nvidia N1X CPU has popped up, showing its OpenCL score, which measures GPU performance. Its results show a 46,361 OpenCL score, which puts it around an RTX 2050 GPU. Of course, this isn't ideal, but this early on, it is an early prototype of the chip, so don't expect this result in the final rumored chip.

More impressively, the Geekbench entry shows the specs of the N1X SoC, including a 20-core CPU that's split into two 10-core clusters, and 48 compute units. These are the Streaming Multiprocessors, and Nvidia offers 128 CUDA cores per unit. This translates to 6,144 CUDA cores, which is the same number as an RTX 5070 desktop GPU.

We've heard that the N1X may deliver RTX 4070 power in a previous benchmark leak, but seeing this key spec shows what Nvidia may be aiming to deliver in its rumored Arm-based chip.

What's more, it shows similar specs to the Nvidia GB10 Superchip, seen in DGX Spark-equipped PCs. As rumored, the N1X may use a pared-down version of the GB10, and as that chip combines a Blackwell GPU with a Grace Arm CPU to make a custom SoC, we can expect this chip to offer RTX 50-series performance.

Now, despite the weaker OpenCL result, it still pushes integrated GPU performance in Arm-based chips. Even though it's likely not to be the final results, the N1X chip in this benchmark still pushes past Apple's M3 Max performance, which usually sees a score of 37,500 (as per our sister site, Tom's Hardware).

All in all, with these expected specs now known, it's looking like Nvidia will have a strong chip on its hands, putting Apple's own M-series chips on notice.

When will Nvidia's N1X launch?

More from Tom's Guide

• Nvidia wants to make 8GB GPUs great again with AI texture compression — but I'm not convinced
• macOS Tahoe hands-on preview: Apple Intelligence and Liquid Glass steal the show
• I just tried Alienware’s new 18-inch RTX 5090 gaming laptop — and it’s ridiculous in the best way possible

The anticipated Nvidia N1X CPU is seeing some major roadblocks lately, and now it's reportedly been pushed back until early 2026 — with Microsoft's next Windows OS tipped to be setting it back.

The Windows-on-Arm chip from Nvidia and MediaTek is now expected to launch in 2026, with DigiTimes (via Jukan on X) citing supply chain sources stating it's due to a combination of reasons. This includes delays on Microsoft's next-gen Windows OS (possibly Windows 12), Nvidia's own design changes to the silicon and the market seeing a smaller demand for notebooks.

As per the report, Nvidia and MediaTek's N1X AI PC chip will focus on enterprise-class devices first, shifting to the consumer market once demand picks back up. Along with Microsoft's roadmap for its OS apparently being delayed and Nvidia redesigning the chip, it has led to the Arm-based CPU being pushed to 2026.

Initially, the Nvidia N1X CPU was expected to be announced at Computex 2025, with the report claiming it would feature 180 to 200 TOPS (trillions of operations per second) — a huge jump over Snapdragon X Series chips at 45 TOPS — to drive AI performance further.

However, the chip didn't arrive, and many reports suggested it would arrive by late 2025 instead. Now, a 2026 launch date is looking likely, but earlier reports have indicated it's been pushed back until late 2026 instead.

When is Nvidia N1X expected to launch?

The Nvidia N1X chip's release date appears to be all over the place, but, along with this report, it's looking like industry insiders agree that it will now launch in 2026.

According to SemiAccurate, the Arm-based AI chip has been hit by a few hurdles, with sources suggesting that it would require engineers to make design changes to the silicon.

While this lines up with the DigiTimes report, the tech news site believes the chip has now been delayed until late 2026 instead of the expected first quarter of the year. However, plans can change, of course, so only time will tell until Nvidia and MediaTek announce their Arm-based CPU.

Nvidia's N1X chip is rumored to come in two flavors: the N1X being for desktops, while the N1 chip is for laptops. So far, they look to deliver powerful AI performance along with the same performance as an RTX 4070 GPU for notebooks. It's tipped to use a smaller GB10 Blackwell chip, or a GB206 model, similar to RTX 5060 Ti or RTX 5060.

If Nvidia plans to release the chip around the same time as Microsoft's next-gen Windows OS, this could mean that the chip may launch alongside a possible Windows 12, or something of the like.

Rumors are scattered, but for now, it's looking like we'll have to wait until at least early 2026 to see Nvidia's N1X CPU.

More from Tom's Guide

• The RTX 5090 is the best graphics card I've ever owned — but there's a catch for living room PC gamers
• Nvidia wants to make 8GB GPUs great again with AI texture compression — but I'm not convinced
• I test laptops for a living — and these are my 3 favorite laptops of 2025 (so far)

Nvidia is the undisputed leader of the GPU market (whether you like it or not), with its RTX 50-series graphics cards making waves this year, but Team Green looks to be putting its hat in the ring of another sector, as a new CPU is on the horizon.

Rumors have been swirling of an Nvidia N1X and N1 Arm-based CPUs that would be made for desktops and laptops, respectively. Now, CEO Jensen Huang has confirmed new chips with MediaTek are coming.

Not only does this mean Nvidia will have a stake in PCs in a whole new way, but as reports have pointed out, it could lead to slimmer, more powerful gaming laptops, too. Moreover, it's bringing the fight to Intel Core Ultra Series 3, AMD Ryzen AI 400 and Qualcomm Snapdragon X2 Elite.

Now, there's also a whole new chip in the making, all thanks to a surprise Nvidia and Intel partnership. Team Green is investing $5 billion in Intel to jointly build a new series of Intel x86 RTX chips, bringing Intel CPUs with Nvidia RTX graphics built into them for consumer PC gaming. But, as CEO Jensen Huang made clear, this won't bring Nvidia's own chip plans to a halt.

While Nvidia may have GPU and AI markets in its pocket, its N1X and N1 System on Chips (SoC), and now its Intel x86 RTX silicon, may prove to shake up the competition in AMD, Qualcomm, and Apple's offerings.

It might not be long before the first Nvidia N1X and N1 chips arrive, as recent leaks suggest that Dell and Lenovo could announce N1X-Powered laptops at Computex 2026 (via VideoCardz). With that said, let's dive into what we know so far.

Nvidia NX1 CPU: Release date rumors

Until recently, it was anyone's guess when we'd see Nvidia's N1-series GPUs. Nothing is confirmed, but if recent Dell and Lenovo leaks (via VideoCardz) are accurate, we could see announcements during Computex 2026.

Initially, Nvidia and MediaTek's Arm-based CPU was rumored to be announced at Computex 2025, with the tech giant expected to be gearing up to show off its smaller GB10 Blackwell chip in the Arm SoC coming to laptops. As you can tell, this didn't come to be, as it seems Nvidia wasn't ready to officially announce its chips.

Before the recent Computex 2026 leaks, it was reported that the Nvidia N1X Arm CPU had been delayed until late 2026. As noted by SemiAccurate, Nvidia faced problems that delayed the CPU's arrival in early 2026.

While the report doesn't detail the specific problem with Nvidia's chip, sources say the chip has encountered issues requiring engineers to make design changes to the silicon.

With Nvidia now joining forces with Intel to develop an all-new silicon that brings yet another series of SoCs packing Team Green's GPU power into the Intel x86 RTX silicon, there's a possibility this will alter the tech giant's plans. There's no word on when these chips will launch, so we may get the Nvidia N1X before they're released.

For now, of course, this is all up in the air. At the very least, Huang has confirmed that these PCs-geared CPUs with "powerful AI capabilities" are coming. But with rumors of delays, it's likely to be a while before we see any mention of a new CPU from Nvidia.

Nvidia NX1 CPU: Performance rumors

So, what kind of performance can we expect from Nvidia's N1-series chips? According to leaked benchmarks, we could see some big performance gains in ultraportable laptops.

We've heard that the N1-series chip will be based on a GB10 Superchip, found in Nvidia's announced Project DIGITS AI supercomputer (now known as DGX Spark) for desktops. For the laptop version, which is set to be the N1 SoC, it may be a cut-down version of GB10, with some combination of a Blackwell GPU and a MediaTek CPU.

That said, there's reason to believe it could use a GB206 model. Either way, it's looking to leverage the power of an RTX GPU, with these Blackwell-based GPUs being used in RTX 5060 Ti or RTX 5060 graphics cards.

But the real kicker here is that this N1 chip will reportedly deliver the same performance as an RTX 4070-equipped laptop, but with far better energy efficiency, according to Taiwanese outlet UDN. A CPU that delivers an integrated GPU with that kind of power, along with improvements in power efficiency (possibly longer battery life), is already a good sign that Team Green's chip will be worth waiting for.

But the rumors continue: the N1 chip is expected to draw 65W to match the performance of a 120W RTX 4070 gaming laptop, and another source suggests it would offer a TDP (Thermal Design Power) of 80W to 120W.

Since benchmarking the Intel Core Ultra X9 388H, Team Blue's flagship laptop processor, we've been hugely impressed by the performance and power efficiency gains the chip delivers. Especially when it comes to Intel XeSS 3 and its integrated graphics for gaming — seeing Cyberpunk 2077 at High settings reach 217 frames per second is a highlight. That's a chip Nvidia will have to battle.

According to ComputerBase, Nvidia and MediaTek's chips may only have 8 or 12 CPU cores instead of 20. Benchmark leaks of Nvidia's GB10 Arm superchip (via Notebookcheck) suggest single-core performance reaching 2,960 and multicore at 10,682.

Due to the rumored delay, it's only guesswork if these are the benchmarks (or even specs) that will arrive, as for now, these Geekbench results put it behind Apple's M4 Max chips.

For gaming laptops?

While it's believed the N1X chip is for desktops and the N1 is for laptops, it's looking likely that the latter will be primed for gaming laptops. And reports even suggest the first gamer-focused notebooks that will be getting them.

According to the UDN report, Dell's gaming brand, Alienware, will be among the first to launch new gaming laptops featuring Nvidia and MediaTek CPUs. That means we could see fresh Alienware notebooks that are slimmer and offer better battery life, if rumors about Nvidia and MediaTek's chip are accurate — not unlike the newly designed Alienware 16 Aurora lineup.

If the rumors are accurate, Nvidia's Arm-based SoC is set to bolster ultraportable gaming laptops (and possibly PC gaming handhelds) with improved power efficiency, which should translate to longer battery life in gaming notebooks.

We've seen ARM chips in action before, with Snapdragon X Elite laptops impressing with their long battery life and fast speeds. We've even tested Snapdragon X Elite PCs for gaming, and while impressive, they aren't quite built for demanding titles.

With Nvidia's own chip sporting its GPU tech, however, gaming on machines with this chip could see major performance gains, especially if it uses some form of DLSS 4 and its Multi Frame Generation tech.

But there's already some competition heating up, and that's from two heavy hitters in the laptop market. For one, the AMD Strix Halo APU already delivers close to RTX 4060 desktop GPU power, and Qualcomm's Snapdragon X2 Series chip is set to arrive soon.

Nvidia NX1 CPU: Outlook

It's still early days for the Nvidia N1X and N1 Arm-based CPU. We have an idea of what to expect, especially regarding the power the N1-series chip for laptops may deliver, but all this could change depending on when the chips are released.

We'll be getting another chip from Nvidia, thanks to the partnership with Intel. Will this put a spanner in the works for Nvidia's initial plans? It doesn't look like it, according to Huang during the webcast press conference. Either way, that's great news for us, as it gives us a better chance of seeing an evolution of PC gaming in slimmer, more battery-efficient laptops.

Only time will tell when we see Nvidia's N1X Arm-based CPU arrive and whether it's the consumer CPU we've been expecting. But if it comes from Team Green, we should expect to see a boost in ultraportable laptops, at the very least, along with a touch of AI to improve power-efficiency management.

More from Tom's Guide

• The RTX 5090 is the best graphics card I've ever owned — but there's a catch for living room PC gamers
• Leaked Windows 11 gaming handheld UI hints at long-overdue fix ahead of ROG Xbox Ally launch — and it’s about time
• AMD Ryzen Z2 Extreme benchmarks revealed — here's how the MSI Claw A8 and ROG Xbox Ally X may perform

Nvidia's rumored N1X CPU has been a long time coming, but it may be another while until the Arm-based chip arrives due to major production issues.

The custom Arm CPU is now being pushed back until late 2026, according to tech site SemiAccurate. Sources state that the chip has been hit with problems that require engineers to make design changes to the silicon, with the report calling it another "whoopsie."

The Arm-based laptop chip was initially expected to be revealed back at Computex 2025, but clearly, Nvidia wasn't ready to announce its all-new CPU for gaming laptops, and it won't be for some time, according to the report.

Apparently, this is one of several delays, with Nvidia facing problems that caused a roadblock in the CPU arriving in early 2026. While this was reportedly handled, the new N1X chip is now rumored to be suffering from another hurdle.

Now, Nvidia did officially announce that a new Arm-based CPU is in the works, and would be arriving in a "one-year rhythm." However, with the reported issues, this may not fall in line with CEO Jensen Huang's roadmap.

The report doesn't state the specific problem with the chip, just that it's causing a delay in production. If accurate, it could be another year until we see Nvidia's custom CPU — likely closer to CES 2027.

Time for upgrades?

Recent reports have detailed that Nvidia's Arm-based CPU delivers the same performance as an RTX 4070-equipped laptop, with the benchmarks indicating it could be launching in late 2025 or early 2026. Now, this may not be the case, but it does leave room for possible improvements.

Rumor has it that the Arm-based chip will use a Blackwell-based GPU, with a smaller GB10 Blackwell chip for laptops or a GB206 model as seen in RTX 5060 Ti or RTX 5060 graphics cards. It's also believed to use 65W power to match the performance of a 120W RTX 4070 laptop GPU, which is already impressive, while other leaks suggest the chip would offer a TDP (Thermal Design Power) of 80W to 120W.

This would give gaming laptops more ultraportable designs, with better power efficiency that could translate to improved battery life (something even the best gaming laptops today struggle with). But with this delay, perhaps Nvidia has time to refine its custom CPU, giving it even greater power gains to match current and upcoming chips.

But if the delay is accurate, it also gives time for Nvidia's competition to bolster its offerings. For one, the AMD Strix Halo APU already delivers close to RTX 4060 desktop GPU power, and Qualcomm's Snapdragon X2 Series chip is set to arrive soon.

Only time will tell when we see Nvidia's N1X Arm-based CPU arrive, but in the meantime, we'll be enjoying what its RTX 50-series GPUs have to offer.

More from Tom's Guide

• The RTX 5090 is the best graphics card I've ever owned — but there's a catch for living room PC gamers
• Leaked Windows 11 gaming handheld UI hints at long-overdue fix ahead of ROG Xbox Ally launch — and it’s about time
• AMD Ryzen Z2 Extreme benchmarks revealed — here's how the MSI Claw A8 and ROG Xbox Ally X may perform

The computing conference Computex 2025 is showcasing the future of PCs in Taipei this week with massive announcements from Nvidia, AMD and more.

Intel is among them, and one of the bigger reveals from Team Blue this week was live demos of the upcoming Panther Lake Core Ultra 300 CPUs for laptops.

Our colleagues at Tom's Hardware got to put some eyeballs on the Panther Lake-powered systems while at Computex, and their Intel demos sound promising.

According to them, Intel showcased real-time tests of a Panther Lake CPU as it rendered or powered AI applications. They showed that the "silicon is healthy and on-track for retail availability in early 2026."

The processors were using Cougar Cove P-cores (performance cores) and Darkmont E-cores (efficiency cores) which are being built on Intel's 18A node process, which should deliver better performance than earlier editions.

The new laptop CPUs were described as blend of Intel's power-efficient Lunar Lake chips and the Arrow Lake-H performance CPUs. That sounds like it could be good news for future handheld gaming PCs like the MSI Claw 8 AI+, though we expect to see Panther Lake primarily in laptops.

Intel has also previously teased a "next-gen built-in iGPU" which could also be a major performance booster, but it has not elaborated on that teaser at Computex so far.

Intel has faced severe challenges in the last couple of years, so Panther Lake needs to be a quality chip to right the boat for the struggling chip maker. Especially as companies like AMD are impressing with its Ryzen Z2 Extreme processors, which we saw in a MSI Claw A8 at Computex.

That said, right now Intel has the advantage in battery life efficiency with its Lunar Lake chips and it appears that Panther Lake is on track to keep that lead while providing better performance.

More from Tom's Guide

• Qualcomm slams Intel chips in new Snapdragon ads — and it may have a point
• Intel Core Ultra 200U, 200H, 200HX and 200S PCs coming this month — here's everything we know
• Intel Core Ultra 9 275HX CPU is the best-performing laptop processor according to new benchmarks — but Apple still beats it in this key area