[Solved] NVidia Kepler: News and Rumors, Releases and Reviews, Technical Data..

http://lenzfire.com/2011/08/nvidia [...] ate-72105/
Nvidia Confirms Kepler Release Date

Ken Brown’s Speech

A Nvidia employee spoke about the Kepler delivery this year, but this was apparently misunderstood by many. That was Ken Brown, PR manager at Nvidia, who has made a statement on the launch date of Kepler.





According to Brown, Nvidia have received “Early Silicon” Kepler-based GPUs this year from TSMC. That does not change the fact that Kepler will not go into mass production until 2012 and accordingly, it will only be available in stores next year. It was misinterpreted by the media. He meant that Nvidia will have samples of Kepler by the end of this year but that does not mean products will come out by the end of this year.
Rumors on Features of Nvidia GeForce 600 series Kepler GPUs

Brown did not mention anything about the rumor which says, Kepler will be coming with native multi-monitor technology vs Eyefinity, which may be launched as a new version of its current Nvidia Surround technology, which currently requires using 2 GeForce video card to display games, videos and applications to 3 monitors. The new incarnation of Kepler Surround technology will use 3 to 5 monitors without requiring a 2nd board and eliminate the need for third-party chips to achieve this function. It should be remembered that manufacturers like Galaxy/KFA2 offer solutions based on Nvidia GPUs with support for 3, 4 and 5 monitors, but this is possible with ViewXpand VMM 1400/1402 chip, which is not a native capacity of the current Nvidia GPUs .
Conclusions on Release Date

Nvidia made this statement to avoid any false expectations or bad names. So the first GPUs based on Kepler will begin mass production in early 2012. It could be an indication that we can see the first GeForce 600 Series GPUs from February to June next year, although again Nvidia has not given a definite date. To further specify the release date, currently it may be scheduled for the presentation on CeBIT trade fair.

http://www.xbitlabs.com/news/graph [...] _2011.html
Kepler is Nvidia's next-generation graphics processor architecture that is projected to bring considerable performance improvements and will likely make the GPU more flexible in terms of programmability, which will speed up development of applications that take advantage of GPGPU (general purpose processing on GPU) technologies. Some of the technologies that Nvidia promised to introduce in Kepler and Maxwell (the architecture that will succeed Kepler) include virtual memory space (which will allow CPUs and GPUs to use the "unified" virtual memory), pre-emption, enhance the ability of GPU to autonomously process the data without the help of CPU and so on.

The new chip is projected to be made using 28nm process technology. Many believe that Taiwan Semiconductor Manufacturing Company, which makes chips for Nvidia, AMD and many others, will not be able to supply enough 28nm products this calendar year.

Nvidia's Kepler family of products, which will likely get GeForce 600-series name in the consumer market segment, will not only power Nvidia's mid-term future products, but will also help Nvidia to boost sales of its desktop discrete graphics cards. In Q2 2011 shipments of discrete graphics boards for desktops were down 15%, according to some analysts.



http://www.nvidia.co.uk/page/channel_partners.html

Kepler promises to double the computing power of double-precision GPGPU compared to current Fermi, and to provide a performance / watt 50% higher than Fermi, Nvidia has not revealed details on its new architecture, but it is expected that as the Current Fermi (GeForce GTX 400 and 500 Series) is used in at least 1 or 2 families of products (Geforce GTX 600 and 700 Series) that will be available from late 2011(Release Date info Updated) until late 2013 when Maxwell may make his appearance.
Returning to Kepler, Nvidia has completed the design (known as Taped-Out), so no new features added with the exception of the base architecture optimizations.
It is expected that within a few months after Nvidia reveals some details on this new architecture.

http://news.softpedia.com/news/Nvi [...] 0284.shtml

Even though AMD will most certainly be the first graphics card maker to release a 28nm GPU, Nvidia hasn’t abandoned the fight and has recently disclosed that Kepler is on track and that they have already entered into the possession of the first 28nm chip samples.

“We are on track with our Kepler roadmap. We have 28nm silicon in house now,” said Igor Stanek, Senior Product PR Manager at Nvidia during an interview with the Fudzilla website.

“Our transition to 28nm is going better than 40nm, and yields are better than our original plan,” concluded the company’s rep without going into any details regarding Kepler’s roadmap.

However, over the course of the last few months, multiple Nvidia representatives have said that the first Kepler-based graphics cards aren’t expected to arrive until Q2 of 2012, so this is probably the timeline that Stanek is referring to.

Before Kepler arrives, the GPU maker is expected to release a series of die shrinks of Fermi built using TSMC's 28nm fabrication process.

The first chips based on the Kepler architecture were taped out by Nvidia at the beginning of September.

Kepler is the code name used by Nvidia to refer to its next-generation graphics processing unit architecture, which, just like AMD's Radeon HD 7000 GPUs, will be manufactured using TSMC's 28nm fabrication process.

The new graphics core is expected to be more flexible in terms of programmability than the current Fermi architecture.

In the second half of 2010, Nvidia promised that Kepler, and its successor Maxwell, will include virtual memory space (allowing both the CPU and the GPU to use a unified virtual memory) and pre-emption support, as well as a series of other technologies meant to improve the GPU's ability to process data without the help of the system's processor.

According to previous Nvidia estimates, these changes, combined with the new manufacturing process, should deliver 3 to 4 times the performance per Watt of the Fermi architecture in double-precision 64-bit floating point operations.

Oak Ridge National Library this week said that it would upgrade its Jaguar supercomputer, which was once the most powerful system in the world, and create petascale Titan supercomputer. Powered by sixteen-core AMD Opteron "Interlagos" 6200 central processing units and Nvidia Tesla M2090 as well as Kepler next-gen compute accelerators, Titan will deliver a peak speed between 10 PetaFLOPS and 20 PetaFLOPS.

"All of these areas of science will benefit from Titan's enormous increase in computational power. Titan will allow for significantly greater realism in models and simulations and the resulting scientific breakthroughs and technological innovations will provide the return on this national investment. Discoveries that take weeks even on a system as powerful as Jaguar might take days on Titan," said Thom Mason, the director of ORNL.

The initial system upgrade will replace all of the processors with the latest AMD Opteron processors code named "Interlagos", taking the system from two six-core processors per node to one 16-core processor per node, and facilitate the addition of 960 Nvidia Tesla 2090 "Fermi" compute accelerators. In the second phase, to be performed in the second half of 2012, between 7000 and 18 000 of Nvidia's Tesla 3000-series "Kepler" compute accelerators GPUs will be added to the system, increasing the peak performance to between 10PFLOPS and 20PFLOPS, depending on architectural options executed. In total, the system will include 299 008 x86 cores (18688 CPUs) and 600TB of memory.



As one of the first deployments of the latest generation of AMD Opteron processors, the upgrade delivers one-third more cores in the same physical space compared with Jaguar while also doubling the memory and outfitting the system with Cray's powerful "Gemini" network to improve performance on scientific applications.

"But the game-changing new technology will be the inclusion of advanced GPUs. These Tesla GPUs are able to perform many more calculations for the same amount of power as conventional microprocessors and will work hand-in-hand with CPUs to deliver new levels of energy-efficient application acceleration," added Mr. Mason.

ORNL's Titan will be based on the Cray XK6 systems that pair AMD's next-generation Opteron processors with up to 16 cores code-named Interlagos powered by the Bulldozer micro-architecture (up to 96 per cabinet) and Nvidia Tesla compute cards (up to 96 per cabinet) to create a true hybrid supercomputer with intra-node flexibility.

"Cray has been fortunate to share in a great partnership with the Oak Ridge National Laboratory and the Department of Energy's Office of Science, and we are excited that our relationship will continue with Titan. Scientists at Oak Ridge and across the DOE have been researching some of the world's most pressing energy and environmental challenges. As a company, it's an honor to know that this vitally important work has been, and will now continue to be, done on supercomputers built by Cray. It's a partnership we are very proud of," said Peter Ungaro, president and chief executive officer of Cray.

The system will be used for research in a broad range of fields, including material science, energy technology, medical research, geoscience, and others.

Sorry to tell you but it is supposed to read Kepler

The title of this topic has been edited by Aford10

http://news.softpedia.com/news/Nvi [...] 0902.shtml

Many suppositions regarding Nvidia’s next-generation Kepler GPUs have come our way during this last week, but now an even more startling rumor has started circulating on the Web.

According to this, Nvidia’s upcoming solutions will be released into the GTX 700-series and, the top part, named the GTX 780, will be able to deliver a twofold increase in performance over the current GTX 580.

At the heart of these rumors, stands a slide that was recently published on a Chinese forum, which pretty much suggests that the GTX 780 will blow all its competitors out of the water.

It goes without saying that we doubt the authenticity of this slide as the performance increase seems too high from one generation to the other, but if indeed true than Nvidia has a winner on its hands.

I can delete your account if you wish.

any ideas about price of GK104?
I'm eager to grab it whn it releases
(given) it'll be replacing gtx 560ti nd packing 768 SPs woof!

The "stapler" movie was Office Space

And get deleted again ? Thats even less respectfull i think.

Also, i do appreciate your advice and i have nothing against you but you yourself use words sometimes which are much worse, so . . . ( only you don't use them against a mod, i don't care about status or power though )

Anyway, can we please close this subject now, i really hate hijacking Mal's thread this way, it was never my intention that it went this way.

Maybe better close it and make it a new thread, this is all bull.

SORRY MAL, i really am !

I have done mate and will not deny it one bit.
People with power have power it is as simple as that, i learned this in the forces

http://www.tomshardware.com/forum/ [...] discussion
HD 7000 series thread

this GTX 7 series and basically designating the GTX 6 series OEM has really thrown me for a loop..
I do not know what to believe..

anyone remember the 10 day older kepler rumours? i will give a short summary

1. gk106 and gk107. low end.
2.gk104 to be released. successor to gtx560
3.gk102= 2xgk104. high-ish end. release in q3 2012.
4. no gk100
5. gk112. top of the block. said to be about 2x590. release q12013.

GF110 (GeForce GTX 570/580/590)
3 billion transistors on 530mm ² die area in 40nm 3 billion transistors on a 530mm ² area in the 40nm
Fermi architecture with 4 grid with 4 grid architecture Fermi Engines Engines
Engines with 16 Polymorph 16 tesselations units 16 Polymorph Engines tesselations with 16 units
512 (1D) shader units 512 (1D) shader units
64 texture units (TMUs) 64 texture units (TMUs)
48 raster operation units (ROPs) 48 raster operation units (ROPs)
384-bit DDR memory interface (up to GDDR5) 384-bit DDR memory interface (up to GDDR5)
maximum clock chip 772 MHz Shader 1544 MHz, 2000 MHz memory maximum clock chip 772 MHz Shader 1544 MHz, 2000 MHz memory
maximum computing performance: 1.58 TFLOPS ( SP) maximum computing performance: 1.58 TFLOPS (SP)
Maximum Memory Bandwidth: 192 GB / sec Maximum Memory Bandwidth: 192 GB / sec GK100 GK100
Kepler architecture with completely changed shader units (no Hotclocks more, ie Shader Clock = Chip clock) Kepler architecture with completely changed shader units (no more Hotclocks, ie Shader Clock = Chip clock)
estimated ~ 1500 (1D) shader units estimated ~ 1500 (1D) shader units
estimated 512-bit DDR memory interface (up to GDDR5) estimated 512-bit DDR memory interface ( up to GDDR5)
Performance: still too uncertain for a forecast, but certainly about half the level of the GeForce GTX 590 Performance: still too uncertain for a forecast, but certainly about the level of half 590 GeForce GTX
launch: second quarter 2012 launch: second quarter of 2012
GF114 (GeForce GTX 560/Ti) GF114 (GeForce GTX 560/Ti)
1.95 billion transistors on a 358mm ² die area in 1.95 billion transistors on a 40nm 358mm ² area in the 40nm
architecture Fermi architecture with two raster engines Fermi with two scanning engines
8 Polymorph engines with 8 tesselations units 8 Polymorph engines with 8 units tesselations
384 (1D) shader units 384 (1D) shader units
64 texture units (TMUs) 64 texture units (TMUs)
32 raster operation units (ROPs) 32 raster operation units (ROPs)
256-bit DDR memory interface (up to GDDR5) 256-bit DDR memory interface (up to GDDR5)
maximum clock chip 822 MHz, Shader 1644 MHz, Memory 2000 MHz maximum clock chip 822 MHz, Shader 1644 MHz, 2000 MHz memory
maximum computing performance: 1.26 TFLOPS (SP) maximum computing performance: 1.26 TFLOPS (SP)
Maximum Memory Bandwidth: 128 GB / sec Maximum Memory Bandwidth: 128 GB / sec GK104 GK104
supposedly slightly smaller die size than the GF114 The apparently slightly lower than the size GF114
Kepler architecture with completely changed shader units (no Hotclocks more, ie Shader Clock = Chip clock) Kepler architecture with completely changed shader units (no more Hotclocks, ie Shader Clock = Chip clock)
estimated ~ 1000 (1D) shader units estimated ~ 1000 (1D) shader units
memory interface unclear, 256 to 384-bit DDR possible Unclear memory interface, 256 to 384 are possible bit DDR
clock goal rumored clearly above 1000 MHz target clock: 1000 MHz rumored clear about
maximum computing power: rumors "well over 2 TFLOPs" maximum computing performance: rumors "well over 2 TFLOPs"
Performance: allegedly a bit faster than the GeForce GTX 580 Performance: allegedly a bit faster than the GeForce GTX 580
Launch: early / mid first quarter 2012 launch: early / mid first quarter of 2012 http://translate.googleusercontent [...] PQECt-Wdhg Hmmmm ....

another technical info about nvidia kepler particularly GK104:

http://www.techpowerup.com/157039/ [...] locks.html

Something seems fishy with No high end till 2013 for Kepler. I don't fully believe that. I figure by SI's release Nvidia will be releasing more info.

March 2012, all I can say.

You are way behind. The latest 290 beta drivers are the best for me, and are particularly focused on Skyrim enhancements and performance.

To keep in line with the discussion, of course I have already carved out some Best Buy credit to purchase the top of the line Kepler as soon as it becomes available. I don't plan on needing an upgrade anytime soon with a GTX 580, so I'm happy to wait a few months if necessary. I don't tend to think it will be that long though.

I don't and the only issue I've noticed is a few crashes from F@H.

I think they're good drivers all the same. I am a big fan of Guru3d who usually have a lively discussion thread about the new drivers when they come out:
http://forums.guru3d.com/showthread.php?p=4176933

If you want to get a real kick, compare the discussion about Nvidia drivers with the discussion thread for the AMD drivers.
http://forums.guru3d.com/showthread.php?p=4192849

Actually I'm on 290.36 and apart from the F@H crashes (which may or not be down to v.7) everything seems to be tickty boo.

was a big fan of guru3d when i had a 9600gt.
when i got a gtx580, guru3d just became inconsequential.

Actually this last time, when updating to the 290.36's I was forced to reinsall with the clean install option. I also use something called Detonator R.I.P. to clean everything out. My Skyrim settings didn't show up properly in the Control Panel, but the clean install fixed everything right up.
http://downloads.guru3d.com/download.php?det=603