The PlayStation 4 is giving "Intel Inside" a whole new meaning.
When the term "discrete graphics" comes to mind, many immediately think of Nvidia and ATI. The two rivals represent the lions share in the discrete graphics market. However, the pond may need to make room for another big fish.
Intel, a company known more for its CPU's and motherboards, is now making a push into the discrete graphics market. According to The Inquirer, Sony has commissioned the California-based chip maker to build the graphics muscle in its next gaming console. While the PlayStation 4 may be several years off, and is (probably) powered by a Cell processor refresh, Sony is certainly shaking things up.
Anyone who follows hardware already knows the bitter rivalry that exists between Intel and Nvidia. The former believes that a harmony of CPU and GPU, with an emphasis on the CPU, is appropriate for personal computers. Nvidia, whose bread and butter is computer graphics, is pushing a "GPGPU" philosophy. This General Purpose Graphics Processing Unit, in the simplest terms, would be something like current video cards, but used for all computing functions inside of ones PC.
Nvidia has now been blacklisted in the console realm. It broke into the market when Microsoft commissioned them for the original Xbox GPU, and is the company behind the graphics power in the PlayStation 3. With the Xbox 360 and Nintendo Wii powered by ATI, and both Microsoft and Nintendo's next offering rumored to be ATI-powered as well, this newest development with Intel leads one to believe that Nvidia is no longer welcome by the three console makers.
Intel also offers stability by means of its very deep coffers, something Nvidia does not. "The nice Sony engineering lady at CES told us that Intel essentially bought the win," says Charlie Demerjiana of the Inquirer. "...theoretically good architecture, no imminent threats of going bust, and not being hated by Sony all contributed too. With a couple of [variables] satisfied, the PS4 GPU belongs to Intel. No word if this is going to be the entire architecture, CPU as well, or not. That, from what we are told, is not final yet. Perhaps the most important aspect of this deal is Intel showing its GPU know-how. With Larrabee in the back of many enthusiasts minds, a PS4 with Intel graphics that isn't a technological flop can open a window in the PC discrete graphics market for Intel.
Demerjiana also said to expect the Wii2 and Xbox3 (or whatever they end up being named), in or around 2012. If that's true, expect the PS4 at the same time. It may not be that coveted ten year life cycle Sony is shooting for, but in the interest of sales and competition, shaving a few years off wouldn't hurt.
Has Nvidia been given the boot out of the console market? Sure looks like it. However, where one opportunity closes, another begins. If the Microsoft phone rumors are true, Nvidia's Tegra platform may have some serious muscle behind it.
Are you retarded?
The PS1's life was 11 years, with games being published until '05. The PS2 is still kicking 8 years from it's launch.
Poor ppl who buy a ps4 Intel has been able to deliver gpu that are nice looking on paper but none that work out in tests very well.
http://en.wikipedia.org/wiki/Larrabee_(GPU)
According to this article, Larrabee is expected to have nearly 2 teraflops of computing power (a lil more raw computing power than a single 4870X2, lets w8 until the first benchmarks ...)
I remember there being bad blood between Microsoft and nVidia with the Xbox 1. Since Microsoft used basically off the shelf graphics and didn't own the silicon microsoft couldn't control the pricing of the Xbox 1. I imagine that Sony is having more issues with the performance end of what they got from nVidia.
umm... u meant that as PS4, right? we already have PS3
Intel may not be able (yet) to make a "video card" but they can make chips that will have more (perhaps) raw processing power then the competition, which sony can use.
We all forget the turn arounds intel have - Pentium D to Core 2 Duo, jumps with 65nm vs 45nm core 2's, SSD's (first attempt dominates) etc, and do we all forget who owns most of the "graphics" market? INTEL - billions of dells etc use Intel Integrated video making up more then 50% of the market.
Out of curiosity whats the fastest card Intel has out right now? How does it compare to Nvidia/ATI?
Intel doesn't have a single card out. Just a bunch of integrated solutions. None of which compete with any serious solutions, but none of them were meant to.
Although I found something rather interesting when I did a bit of research into the matter:
Larrabee's philosophy of using many small, simple cores is similar to the ideas behind the Cell processor. There are some further commonalities, such as the use of a high-bandwidth ring bus to communicate between cores. However, there are many significant differences in implementation which should make programming Larrabee simpler.
* The Cell processor includes one main processor which controls many smaller processors. Additionally, the main processor can run an operating system. In contrast, all of Larrabee's cores are the same, and the Larrabee is not expected to run an OS.
* Each compute core in the Cell (SPE) has a local store, for which explicit (DMA) operations are used for all accesses to DRAM. Ordinary reads/writes to DRAM are not allowed. In Larrabee, all on-chip and off-chip memories are under automatically-managed coherent cache hierarchy, so that its cores virtually share a uniform memory space through standard load/store instructions.
* Because of the cache coherency noted above, each program running in Larrabee has virtually a large linear memory just as in traditional general-purpose CPU; whereas an application for Cell should be programmed taking into consideration limited memory footprint of the local store associated with each SPE (for details see this article) but with theoretically higher bandwidth.
* Cell uses DMA for data transfer to/from on-chip local memories, which has a merit in flexibility and throughput; whereas Larrabee uses special instructions for cache manipulation (notably cache eviction hints and pre-fetch instructions), which has a merit in that it can maintain cache coherence (hence the standard memory hierarchy) while boosting performance for e.g. rendering pipelines and other stream-like computation.
* Each compute core in the Cell runs only one thread at a time, in-order. A core in Larrabee runs up to four threads. Larrabee's hyperthreading helps hide latencies and compensates for lack of out-of-order execution.
http://en.wikipedia.org/wiki/Larrabee_(GPU)
I'd like to have faith in Intel, considering their great success with their SSDs on their first try. But that was a relatively new technology, that doesn't have a large market for it. So I can only hope it does well.