NVIDIA Kepler Graphics Cards Lineup Leak To Web
Subject: Graphics Cards | February 6, 2012 - 06:23 PM | Tim Verry
Tagged: nvidia, kepler, graphics, gpu
Although there were quite a few rumors leading up to AMD's Radeon 7000 series launch, the Internet has been very quiet on the greener side of the graphics market. Finally; however, we have some rumors to share with you on the Nvidia front. As always, take these numbers with more than your average grain of salt.
Specifically, EXP Review managed to uncover two charts that supposedly detail specifics about a range of GeForce 600 series Kepler cards from the number of stream processors to the release date. Needless to say, it's a lot of rumored information to take in all at once.
Anyway, without further adieu, let's dive into the two leaked charts.
| Model | Code Name | Die Size | Core Clock (TBD) MHz | Shader Clock (TBD) GHz | Stream Processors | SM Count | ROPs | Memory Clock (effective) GDDR5 | Bus Width | Memory Bus Width |
| GTX690 | GK110x2 | 550mm2 | ~750 | ~1.5 | 2x1024 | 2x32 | 2x56 | 4.5 GHz | 2x448bit | 2x252GB/s |
| GTX680 | GK110 | 550mm2 | ~850 | ~1.7 | 1024 | 32 | 64 | 5.5 GHz | 512bit | 352GB/s |
| GTX670 | GK110 | 550mm2 | ~850 | ~1.7 | 896 | 28 | 56 | 5 GHz | 448bit | 280GB/s |
| GTX660Ti | GK110 | 550mm2 | ~850 | ~1.7 | 768 | 24 | 48 | 5 GHz | 384bit | 240GB/s |
| GTX660 | GK104 | 290mm2 | ~900 | ~1.8 | 512 | 16 | 32 | 5.8 GHz | 256bit | 186GB/s |
| GTX650Ti | GK104 | 290mm2 | ~850 | ~1.7 | 448 | 14 | 28 | 5.5 GHz | 224bit | 154GB/s |
| GTX650 | GK106 | 155mm2 | ~900 | ~1.8 | 256 | 8 | 24 | 5.5 GHz | 192bit | 132GB/s |
| GTX640 | GK106 | 155mm2 | ~850 | ~1.7 | 192 | 6 | 16 | 5.5 GHz | 128bit | 88GB/s |
From the chart above, we can see the entire lineup of Kepler cards from the NVIDIA GTX 640 to the dual GPU GTX 690. The die size in the higher end GeForce cards is approximately 50% larger than that of the AMD Radeon HD 7970, but not much bigger than that of the GTX 580. If only we knew the TDP of these cards! In the next chart, we see alleged performance comparison versus the AMD competition.
| Model | Bus Interface | Frame Buffer | Transistors (Billion) | Price Point | Release Date | Performance Scale |
| GTX690 | PCI-E 3 x16 | 2x1.75 GB | 2x6.4 | $999 | Q3 2012 | |
| GTX680 | PCI-E 3 x16 | 2 GB | 6.4 | $649 | April 2012 | ~45%>HD7970 |
| GTX670 | PCI-E 3 x16 | 1.75 GB | 6.4 | $499 | April 2012 | ~20%>HD7970 |
| GTX660Ti | PCI-E 3 x16 | 1.5 GB | 6.4 | $399 | Q2/Q3 2012 | ~10%>HD7950 |
| GTX660 | PCI-E 3 x16 | 2 GB | 3.4 | $319 | April 2012 | ~GTX580 |
| GTX650Ti | PCI-E 3 x16 | 1.75 GB | 3.4 | $249 | Q2/Q3 2012 | ~GTX570 |
| GTX650 | PCI-E 3 x16 | 1.5 GB | 1.8 | $179 | May 2012 | ~GTX560 |
| GTX640 | PCI-E 3 x16 | 2 GB | 1.8 | $139 | May 2012 | ~GTX550Ti |
If these numbers hold true, NVIDIA will handily beat the current AMD offerings; however, I would wait for reviews to come out before making any purchasing decisions. One interesting aspect is the amount of GDDR5 memory. It seems that NVIDIA is sticking with 2GB frame buffers (or less) per GPU while AMD has really started upping the RAM. It will be interesting to see how this affects gaming in NVIDIA Surround and/or at high resolutions.
What do you guys think about these numbers, do you think Kepler will live up to the alleged performance scale figures?
Podcast #187 - Our thoughts on Ultrabooks, the Radeon HD 7950, ASUS DirectCU GTX cards, and more!
Subject: Editorial, General Tech | February 2, 2012 - 03:11 PM | Ken Addison
Tagged: ssd, sandforce, radeon, podcast, patriot, nvidia, Intel, gtx, arm, amd, 7950
PC Perspective Podcast #187 - 02/02/2012
Join us this week as we talk about our thoughts on Ultrabooks, the Radeon HD 7950, ASUS DirectCU GTX cards, and more!
You can subscribe to us through iTunes and you can still
The URL for the podcast is: http://pcper.com/podcast - Share with your friends!
- iTunes - Subscribe to the podcast directly through the iTunes Store
- RSS - Subscribe through your regular
RSS reader - MP3 - Direct download link to the MP3 file
Hosts: Ryan Shrout, Josh Walrath, Jeremy Hellstrom, and Allyn Malvantano
This Podcast is brought to you by
Program Schedule:
- 0:00:40 Introduction
- 1-888-38-PCPER or podcast@pcper.com
- http://pcper.com/podcast
- http://twitter.com/ryanshrout and http://twitter.com/pcper
- 0:01:20 Ultrabooks: Intel Knows What's Good For You
- 0:08:30 Patriot Pyro and Wildfire SSD Review - IMFT Async vs. Toshiba Toggle-mode Flash
- 0:14:20 AMD Radeon HD 7950 3GB Graphics Card Review
- 0:25:50 This Podcast is brought to you by
MSI Computer , and their all new Sandy Bridge Motherboards!
- 0:26:38 Asus DirectCU II Roundup: ENGTX560, ENGTX570, and ENGTX580 Review
- 0:40:35 Raspberry Pi Linux Computer Will Have Fast GPU For The Price
- 0:44:20 If you thought Intel did well wait until you see ARM
- 0:47:00 AMD 7700 and 7800 Release Dates Leak To Web
- 0:51:20 Live Blog: AMD Financial Analyst Day
- 0:52:20 Hardware / Software Pick of the Week
- Josh: And it is on sale! $770 off!
- Allyn: Corsair Force 3 - very good pricing.
- 1-888-38-PCPER or podcast@pcper.com
- http://pcper.com/podcast
- http://twitter.com/ryanshrout and http://twitter.com/pcper
- Closing
3 NV for DCII
The world of video cards is a much changed place over the past few years. Where once we saw only “sticker versions” of cards mass produced by a handful of manufacturers, we are now seeing some really nice differentiation from the major manufacturers. While the first iterations of these new cards are typically mass produced by NVIDIA or AMD and then distributed to their partners for initial sales, these manufacturers are now more consistently getting their own unique versions out to retail in record time. MSI was one of the first to put out their own unique designs, but now we are seeing Asus becoming much more aggressive with products of their own.
The DirectCU II line is Asus’ response to the growing number of original designs from other manufacturers. The easiest way to categorize these designs is that they straddle nicely the very high end and extreme products like the MSI Lightning series and those of the reference design boards with standard cooling. These are unique designs that integrate features and cooling solutions that are well above that of reference cards.
DirectCU II applies primarily to the cooling solutions on these boards. The copper heatipipes in the DirectCU II cooler are in direct contact with the GPU. These heatpipes then are distributed through two separate aluminum fin arrays, each with their own fan. So each card has either a dual slot or triple slot cooling solution with two 80 mm fans that dynamically adjust to the temperature of the chip. The second part of this is branded “Super Alloy Power” in which Asus has upgraded most of the electrical components on the board to match higher specifications. Hi-C caps, proadlizers, polymer caps, and higher quality chokes round out the upgraded components which should translate into more stable overclocked performance and a longer lifespan.
NVIDIA Updates CUDA: Major Release for Science Research
Subject: General Tech, Graphics Cards | January 29, 2012 - 02:53 AM | Scott Michaud
Tagged: nvidia, gpgpu, CUDA
NVIDIA has traditionally been very interested in acquiring room in the high-performance computing for scientific research market. For a lot of functions, having a fast and highly parallel processor saves time and money compared to having a traditional computer crunch away or having to book time with one of the world’s relatively few supercomputers. Despite the raw performance of a GPU, adequate development tools are required to bring the simulation or calculation into a functional program to execute on said GPU. NVIDIA is said to have had a strong lead with their CUDA platform for quite some time; that lead will likely continue with releases the size of this one.
What does a tuned up GPU purr like? Cuda cuda cuda cuda cuda.
The most recent release, CUDA 4.1, has three main features:
- A visual profiler to point out common mistakes and optimizations and to provide instructions which detail how to alter your code to increase your performance
- A new compiler which is based on the LLVM infrastructure, making good on their promise to open the CUDA platform to other architectures -- both software and hardware
- New image and signal processing functions for their NVIDIA Performance Primitives (NPP) library, relieving developers the need to create their own versions or license a proprietary library
The three features, as NVIDIA describes them in their press release, are listed below.
New Visual Profiler - Easiest path to performance optimization
The new Visual Profiler makes it easy for developers at all experience levels to optimize their code for maximum performance. Featuring automated performance analysis and an expert guidance system that delivers step-by-step optimization suggestions, the Visual Profiler identifies application performance bottlenecks and recommends actions, with links to the optimization guides. Using the new Visual Profiler, performance bottlenecks are easily identified and actionable.
LLVM Compiler - Instant 10 percent increase in application performance
LLVM is a widely-used open-source compiler infrastructure featuring a modular design that makes it easy to add support for new programming languages and processor architectures. Using the new LLVM-based CUDA compiler, developers can achieve up to 10 percent additional performance gains on existing GPU-accelerated applications with a simple recompile. In addition, LLVM's modular design allows third-party software tool developers to provide a custom LLVM solution for non-NVIDIA processor architectures, enabling CUDA applications to run across NVIDIA GPUs, as well as those from other vendors.
New Image, Signal Processing Library Functions - "Drop-in" Acceleration with NPP Library
NVIDIA has doubled the size of its NPP library, with the addition of hundreds of new image and signal processing functions. This enables virtually any developer using image or signal processing algorithms to easily gain the benefit of GPU acceleration, with the simple addition of library calls into their application. The updated NPP library can be used for a wide variety of image and signal processing algorithms, ranging from basic filtering to advanced workflows.
Guess what? Overclocked.
The NVIDIA GTX 580 GPU, based on the GF110 Fermi architecture, is old but it isn't forgotten. Released in November of 2010, NVIDIA had held the single GPU performance grown for more than a year before it was usurped by AMD and the Radeon HD 7970 just this month. Still, the GTX 580 is a solid high-end enthusiast graphics card that has wide spread availability and custom designed, overclocked models from numerous vendors making it a viable option.
Gigabyte sent us this overclocked and custom cooled model quite a while ago but we had simply fallen behind with other reviews until just after CES. In today's market the card has a bit of a different role to fill - it surely won't be able to pass up the new AMD Radeon HD 7970 but can it fight the good fight and keep NVIDIA's current lineup of GPUs more competitive until Kepler finally shows himself?
The Gigabyte GTX 580 1.5GB Super Overclock Card
With the age of the GTX 580 designs, Gigabyte had plenty of time to perfect their PCB and cooler design. This model, the Super Overclock (GV-N580SO-15I), comes in well ahead of the standard reference speeds of the GTX 580 but sticks to the same 1.5 GB frame buffer.
The clock speed is set at 855 MHz core and 1025 MHz memory, compared to the 772 MHz core speed and 1002 MHz clock rate of the reference design. That is a very healthy 10% clock rate difference that should equate to nearly that big of a gap in gaming performance where the GPU is the real bottleneck.
Continue reading our review of the Gigabyte GTX 580 1.5GB Super Overclock graphics card!!
Podcast #185 - ASUS EeePad Transformer Prime, an overclocked XFX HD7970, AMDs Lightning Bolt and more!
Subject: Editorial | January 19, 2012 - 07:05 PM | Ken Addison
Tagged: podcast, Intel, amd, ssd, hdd, nvidia, kepler, GK104, gpu, cpu
PC Perspective Podcast #185 - 01/19/2011
Join us this week as we talk about the ASUS EeePad Transformer Prime, an overclocked XFX HD7970, AMDs Lightning Bolt and more!
You can subscribe to us through iTunes and you can still
The URL for the podcast is: http://pcper.com/podcast - Share with your friends!
- iTunes - Subscribe to the podcast directly through the iTunes Store
- RSS - Subscribe through your regular
RSS reader - MP3 - Direct download link to the MP3 file
Hosts: Ryan Shrout, Josh Walrath, Jeremy Hellstrom, and Allyn Malvantano
This Podcast is brought to you by
Program Schedule:
- 0:00:26 Introduction
- 1-888-38-PCPER or podcast@pcper.com
- http://pcper.com/podcast
- http://twitter.com/ryanshrout and http://twitter.com/pcper
- 0:00:55 SOPA and PIPA Chat...
- 0:06:35 ASUS Eee Pad Transformer Prime Review: Thinner, Faster
- 0:07:23 Acer Extensa 5420 Retrospective Review: How Far Have We Come?
- 0:10:15 Corsair Hydro Series H100 Liquid CPU Cooler Review
- 0:11:10 Video Perspective: Cooler Master Cosmos II Case Review
- 0:14:00 XFX Radeon HD 7970 3GB Black Edition and CrossFire Results
- 0:22:45 AMD and IBM inside the Xbox Next?
- 0:31:30 Lucid Cloud Gaming (VGWare) and XLR8 on Tablets Demo
- 0:44:55 Nvidia May Launch GF104 "Kepler" GPUs Ahead Of Schedule
- 0:51:00 AMD Lightning Bolt Strikes At Intel's Thunderbolt
- 0:53:00 AMD Countering Ultrabooks With Ultrathin Notebooks
- 0:55:00 Random Storage stuff at CES? also this link
- 1:05:00 Alienware X51 Desktop -- Console Sized PC, $700 and up.
- 1:10:30 Email from Tom about 7970 CrossFire
- 1:13:34 Hardware / Software Pick of the Week
- Ryan: Behringer XENYX 802
- Jeremy: EVGA SR-X
- Josh: Decent and cheap for the AMD enthusiast: http://www.newegg.com/Product/Product.aspx?Item=N82E16819103962
- Allyn: Mophie Juice Pack Powerstation, oh, and my current house rep is my anti-pick
- 1-888-38-PCPER or podcast@pcper.com
- http://pcper.com/podcast
- http://twitter.com/ryanshrout and http://twitter.com/pcper
- Closing
FBI Arrests Seven People for Insider Trading of Dell and Nvidia Stock
Subject: General Tech | January 19, 2012 - 06:40 PM | Tim Verry
Tagged: dell, nvidia, insider trading, tech, Law, ethics
There is an important distinction between working within the confines of the law to make the most profit possible and going outside those lines to make a profit while hoping you don't get caught. To drive that point home, the FBI has stated "what distinguishes you from the dozens who have been charged is not that you haven’t been caught; it’s that you haven’t been caught yet." Assistant Director in Charge Janice K. Fedarcyk wrote that when referring to a recent bust of seven individuals accused of insider trading of Dell and Nvidia stock. The arrests, made as part of Operation Perfect Hedge, include seven men who are connected by "friendship or business association."
Thanks to three of the seven men cooperating with the FBI, we know that the men used information about Dell and Nvidia's quarterly earnings prior to any public release of such earnings documents to purchase stock to resell after positive earning documents caused the stock value to increase or to short their stock to avoid losses that would be incurred by lower than expected quarterly earnings causing the stock price to drop. On the Dell side of things, two employees in the know provided quarterly earnings numbers to various hedge funds. The first employee, Sandy Goyal, is charged with providing a hedge fund with Q1 earning results for 2008 in exchange for $175,000. The hedge fund then used that insider information to make $3.8 million dollars. Another (former) Dell employee, Jesse Tortora furnished three hedge funds with quarterly earnings numbers who each then made $4 million on Q1 information and $53 million on the Q2 information, $1 million in profit, and the final Hedge fund avoided #78,000 in losses by selling stock before the inevitable price drop thanks to knowing the negative earning numbers before hand.
Finally, Danny kuo knew someone who worked at Nvidia and provided information to the other members of the insider trading group.
Let this be a lesson to those business folks that slept through ethics classes, stay away from insider trading, especially when you are paid for the information as you are just asking to get caught. (Cue the "Cops" theme song). Normally we don't cover this kind of news; however, I thought it applicable since it involves Dell and Nvidia. Also, speaking of quarterly earnings, Josh will have all the details from today's Intel Earnings Call up soon.
Are AMD's Southern Islands about to be swamped by a Kepler tidal wave?
Subject: Graphics Cards | January 19, 2012 - 03:06 PM | Jeremy Hellstrom
Tagged: nvidia, kepler, GK104, amd
AMD had better hope that SemiAccurate lives up to their tongue in cheek name because according to their sources NVIDIA's new Kepler GPU line is going to better AMD at everything. The variety of negative rumours about Kepler are contradicted by the news that the A2 silicon has arrived, trouble free, at NVIDIA's door step so the early arrival of Kepler in May now seems reasonable.
From the limited description we have it seems that NVIDIA's Achilles Heel, multi-monitor support on a single card is no longer a problem as there are two DVI plugs, a single HDMI, and a DisplayPort connector present. With all of those plugs it seems obvious the card is designed to handle NVIDIA Surround all by its lonesome. From the GDDR5 count the cards bandwidth of 256 bits and total memory size of 2GB is lesser than AMD's 384-bit, 3GB card but from the description at SemiAccurate that is not going to be enough to save AMD's biscuit.
"The short story is that Nvidia (NASDAQ:NVDA) will win this round on just about every metric, some more than others. Look for late March or early April availability, with volumes being the major concern at first. GK104 cards seen by SemiAccurate all look very polished and complete, far from rough prototypes or “Puppies“. A2 silicon is now back from TSMC, and that will likely be the production stepping barring any last second hitches. Currently, there aren’t any."
Here are some more Graphics Card articles from around the web:
- AMD CrossFireX Drivers - Opportunity Lost @ [H]ard|OCP
- Radeon Gallium3D: A Half-Decade Behind Catalyst? @ Phoronix
- Muscle-Flexing: AMD Radeon HD 7970 3GB @ X-bit Labs
- Sapphire Radeon HD 7970 3GB @ Tweaktown
- Workstation Graphics Card Comparison Guide @ TechARP
- Zotac GTX 560 Ti with 448 cores @ Hardwareoverclock
- Zotac GeForce GTX 550 Ti AMP! Edition Review @ Tech-Reviews UK
- Galaxy GTX 560 Ti Super OC White Edition Review @ Hardware Canucks
Nvidia May Launch GK104 "Kepler" GPUs Ahead Of Schedule
Subject: Graphics Cards | January 17, 2012 - 06:08 AM | Tim Verry
Tagged: nvidia, kepler, gtx 680, gf104
AMD's latest graphics cards are already hitting the street, but Nvidia's high end GPUs are nowhere to be seen. Late last year, we heard rumors that Nvidia's GK104 "Kepler" graphics cards may be delayed to an early summer or late spring launch. If the latest rumors reported by Maximum PC hold true; however, Nvidia's Kepler cards' release may be leaning more towards a late spring launch instead of an early summer window.
They report that sources from Chinese website ChipHell.com have indicated that Nvidia's new high end GTX 680 graphics cards may be released as soon as February 2012. This is a relatively big push forwards compared to the previously rumored March or April launch window, and is likely being accelerated in response to AMD's successful launch of their GCN (Graphics Core Next) based Radeon 7970 graphics cards. Although we do not know much about the upcoming cards, the general consensus is that the GTX 680 cards will have 2 GB of video memory on a 256-bit bus. Further, the Kepler cards' core will run at 780 MHz, and the GK104 cards will have TDP (thermal design power) ratings of 225 Watts.
Whether the company will have actual hardware to sell or if it will be more of a "paper launch" remains to be seen. If I had to venture a guess, the cards will likely see limited availability in a late February launch but will not be around in significant quantities until later this year. With the delays caused by manufacturing the 28nm Kepler cards over at TSMC, the company is not likely to have all that big of a stockpile on hand.
CES Day 4 Podcast - 1/11/2012
Subject: Editorial, General Tech | January 12, 2012 - 11:04 AM | Ken Addison
Tagged: ultrabook, razer, podcast, nvidia, Intel, CES, amd
PC Perspective CES 2012 Day 4 - 1/11/12
Join us tonight as we talk about our final full day of CES 2012 - including yet more Ultrabooks, New products from Razer, and our closing thoughts on this years show.
You can subscribe to us through iTunes and you can still
The URL for the podcast is: http://pcper.com/podcast - Share with your friends!
- iTunes - Subscribe to the podcast directly through the iTunes Store
- RSS - Subscribe through your regular
RSS reader - MP3 - Direct download link to the MP3 file
Hosts: Ryan Shrout, Josh Walrath, Matt Smith and Allyn Malventano
PC Perspective's CES 2012 coverage is sponsored by MSI Computer.
Follow all of our coverage of the show at http://pcper.com/ces!
PC Perspective Upcoming Events
|
