Author:
Manufacturer: NVIDIA

GK110 Makes Its Way to Gamers

Our NVIDIA GeForce GTX TITAN Coverage Schedule:

Back in May of 2012 NVIDIA released information on GK110, a new GPU that the company was targeting towards HPC (high performance computing) and the GPGPU markets that are eager for more processing power.  Almost immediately the questions began on when we might see the GK110 part make its way to consumers and gamers in addition to finding a home in supercomputers like Cray's Titan system capable of 17.59 Petaflops/s. 

 

Video Loading...

Watch this same video on our YouTube channel

02.jpg

Nine months later we finally have an answer - the GeForce GTX TITAN is a consumer graphics card built around the GK110 GPU.  Comprised of 2,688 CUDA cores, 7.1 billion transistors and with a die size of 551 mm^2, the GTX TITAN is a big step forward (both in performance and physical size).

specs3.jpg

From a pure specifications standpoint the GeForce GTX TITAN based on GK110 is a powerhouse.  While the full GPU sports a total of 15 SMX units, TITAN will have 14 of them enabled for a total of 2688 shaders and 224 texture units.  Clock speeds on TITAN are a bit lower than on GK104 with a base clock rate of 836 MHz and a Boost Clock of 876 MHz.  As we will show you later in this article though the GPU Boost technology has been updated and changed quite a bit from what we first saw with the GTX 680.

The bump in the memory bus width is also key, being able to feed that many CUDA cores definitely required a boost from 256-bit to 384-bit, a 50% increase.  Even better, the memory bus is still running at 6.0 GHz resulting in total memory bandwdith of 288.4 GB/s

Continue reading our preview of the brand new NVIDIA GeForce GTX TITAN graphics card!!

Podcast #235 - AMD Hardware in the PS4, a GK110 NVIDIA product, Corsair 200R case and more!

Subject: General Tech | January 24, 2013 - 03:31 PM |
Tagged: video, titan, ps4, podcast, nvidia, kavari, Kabini, H80i, gk110, GCN, corsair, APU, amd, 200r

PC Perspective Podcast #235 - 01/24/2013

Join us this week as we discuss potential AMD Hardware in the PS4, a GK110 NVIDIA product, Corsair 200R case and more!

You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE.

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, Jeremy Hellstrom, Josh Walrath and Allyn Malventano

This Podcast is brought to you by MSI!

Program length: 1:16:39

Podcast topics of discussion:

  1. Week in Reviews:
    1. 0:01:25 Corsair Hydro Series H80i Taken Apart
    2. 0:03:45 Video: Corsair Carbide 200R Overview
    3. 0:10:30 Silverstone Zeus 1350 watt power supply
  2. 0:13:35 This Podcast is brought to you by MSI!
  3. News items of interest:
    1. 0:15:15 Corsair Launches 140mm and 280mm coolers
    2. 0:17:50 Catalyst 13.1 drivers released
    3. 0:19:30 SimCity Beta Weekend
    4. 0:25:00 NVIDIA GK110 Titan Rumored for February
    5. 0:30:50 Intel to exit motherboard business after Haswell
    6. 0:38:00 But ASUS says its okay...
    7. 0:41:30 PS4 Hardware discussion
  4. Closing:
    1. Hardware / Software Pick of the Week
      1. Ryan: Pegasus R4 with Thunderbolt
      2. Jeremy: Not since the Sumosac has there been something more sure to get you the ladies!
      3. Josh: Just built a machine with one of these
      4. Allyn: Zip Snip ($20 at Lowes)
  1. 1-888-38-PCPER or podcast@pcper.com
  2. http://pcper.com/podcast
  3. http://twitter.com/ryanshrout and http://twitter.com/pcper
  4. Closing/outro

Be sure to subscribe to the PC Perspective YouTube channel!!

 

Rumor: NVIDIA GK110 based GeForce GPU 'Titan' to be released late February

Subject: Graphics Cards | January 22, 2013 - 02:44 PM |
Tagged: nvidia, geforce, gk110, titan, rumor

A combination of rumors and news pieces found online and in some recent conversations with partners indicates that February will see the release of a new super-high-end graphics card from NVIDIA based on the GK110 GPU.  Apparently using the name "Titan" based on a report from Sweclockers.com, this new single GPU card will feature 2688 CUDA cores, compared to the 1536 in the GeForce GTX 680. 

gk110.jpg

If true, the name Titan likely refers to the Cray super computer of the same name built using GK110 Kepler Tesla cards.  Sweclockers.com's sources are quoted with the clocks of this new super-GPU as well: 732 MHz core clock and 5.2 GHz GDDR5 memory clock.  While those numbers are low compared to the 1000+ MHz speeds of the GK104 parts out today, this GPU would have 75% more compute units and presumably additional memory capacity as well.  The memory bus width of 384-bits is a 50% increase as well which would indicate another big jump in performance over current cards.  The CUDA core count of 2688 is actually indicative of a GK110 GPU with a single SMX disabled as well.

teslagpu2.jpg

The NVIDIA Titan card will apparently be the replacement for the GeForce GTX 690, a dual-GK104 card launched in May of last year.  The performance estimate for the Titan is approximately 85% of that GTX 690 and if the rumors are right it would see an $899 price tag.

Based on other conversations I have had recently you should only expect those same partners that were able to sell the GTX 690 to stock this new GK110-based part.  There won't be any modifications and you will see very little differentiation between vendors branding on it.  If dates are to be believed, we are hearing that a Feb 25th (or at least that week) launch is the current target.

NVIDIA Launches Maximus 2.0, Combining Kepler and Tesla

Subject: General Tech, Graphics Cards | August 10, 2012 - 05:34 AM |
Tagged: tesla, quadro, nvidia, maximus, kepler, gk110

At SIGGRAPH 2012 NVIDIA announced a refresh of its Maximus workstation platform technology. Maximus is a technology aimed at professionals that work with simulations or content creation and editing. The updated platform features a Tesla K20 accelerator card as well as a Kepler-based NVIDIA Quadro K5000 graphics card. The K5000 in particular has 4GB of GDDR5 memory on a 256-bit bus and 1536 CUDA cores. NVIDIA states that the Quadro graphics card has 2.1 Teraflops of single precision compute power and draws 122 watts.

The K20 on the other hand features a GK110 Kepler GPU with Dynamic Parallelism and Hyper Q features that reportedly enable more than 1 Teraflop of peak double precision performance. Unfortunately, we do not know much more than that on the new K20 Tesla card as the exact specifications are still listed as “to be announced.” It is slated for a Q4 2012 release.

displaymedia.jpg

The Quadro K5000 workstation GPU

Beyond the hardware itself, the company’s Maximus platform has received software support from several high-profile software companies and system integrators. Some of the companies that certify and support Maximus are Adobe, Autodesk, Mathworks, and Paradigm among others. Dell, Fujitsu, HP, Lenovo, and Supermicro are OEMs that support the hardware and manufacture Maximus-powered workstations.

NVIDIA Tesla K20 GK110 GPU.jpg

The Tesla K20 accelerator card.

The second-generation Maximus technology will be available in desktop workstations as early as December 2012. Further, the NVIDIA Quadro K5000 will be available for purchase as a separate discrete card in October 2012 for $2,249 (MSRP). The Tesla K20 will (for now) only be available integrated in a workstation, but NVIDIA lists the MSRP at $3,199.

More information on the NVIDIA Maximus refresh can be found in the company’s press release.

displaymedia (1).jpg

Source: NVIDIA

NVIDIA's big chip, the GK110

Subject: General Tech | May 25, 2012 - 11:28 AM |
Tagged: nvidia, kepler, GTC 2012, gk110

We at PC Perspective were not the only ones who became a wee bit excited when we had the news from NVIDIA about what the GK110 Kepler chip is going to be capable of.  The chip will be powering professional HPC systems with the Telsa K20 board which will deliver over a teraflop of double precision processing power.  That precision is not so important to the proper rendering of fluid dynamics in the underground water of Crysis 2 but for scientists trying to model the real world it is double what they say from the previous generation of Fermi based Tesla boards.  Check out The Tech Report as they delve into how NVIDIA tweaked their new architecture to deal with new choke points and the compute enhancements they've added.

TR_mandelbrot.jpg

"At its 2012 GPU Technology Conference, Nvidia revealed plenty of details about the biggest GPU of its Kepler generation. Here's what you need to know."

Here is some more Tech News from around the web:

Tech Talk

 

Podcast #202 - GTX 670, NVIDIA's GK110 Tesla card, our AMD Trinity Mobile review and more!

Subject: General Tech | May 17, 2012 - 03:16 PM |
Tagged: trinity, tesla, podcast, nvidia, kepler, gtx670, GTC 2012, gk110, GK104, dv nation, a10

PC Perspective Podcast #202 - 05/17/2012

Join us this week as we talk about the GTX 670, NVIDIA's GK110 Tesla card, our AMD Trinity Mobile review and more!

If you want even more PC Perspective this, check out our "aftershow" event as well.  Event might be an over-statement though...

You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE.

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: Jeremy Hellstrom, Josh Walrath, and Allyn Malvantano

Program length: 1:05:16

Program Schedule:

  1. 0:00:21 Introduction
  2. 1-888-38-PCPER or podcast@pcper.com
  3. http://pcper.com/podcast
  4. http://twitter.com/ryanshrout and http://twitter.com/pcper
  5. 0:01:15 NVIDIA GeForce GTX 670 2GB Graphics Card Review - Kepler for $399
    1. GeForce GTX 670 vs GTX 570 Performance Update
    2. The GTX 670 and the Case of the Missing (and Returning) 4-Way SLI Support
  6. 0:11:20 Graphics Card (GPU) Stock Check - May 10th, 2012
    1. Hard to make a profit when no one can find Kepler cards for sale, NVIDIA
  7. 0:14:25 NVIDIA Reveals GK110 GPU - Kepler at 7.1B Transistors, 15 SMX Units
  8. 0:20:20 Lenovo IdeaCentre Q180: Atom's Wake
  9. 0:24:30 AMD A10-4600M Trinity For Mobile Review: Trying To Cut The Ivy
  10. 0:33:40 Just Delivered: DV Nation RAMRod PC - Sandy Bridge-E, 64GB DDR3, 480GB RevoDrive 3 X2
  11. 0:35:42 Plug and Pray PCIe SSD that you can upgrade; OWC's Mercury Accelsior
  12. 0:40:40 GTC 2012: NVIDIA Announces GeForce GRID Cloud Gaming Platform
    1. NVIDIA Pioneers New Standard for High Performance Computing with Tesla GPUs
    2. NVIDIA Introduces World's First Virtualized GPU, Accelerating Graphics for Cloud Computing
  13. 0:53:00 ZOTAC announces ZOTAC GeForce GT 630, GT 620 and GT 610 series
  14. 0:55:00 Hardware / Software Pick of the Week
    1. Jeremy: Only to be used for evil
    2. Josh: Since NV doesn't have an answer yet at this price range...
    3. Allyn: If you need your files secure - without the destruction
  15. 1-888-38-PCPER or podcast@pcper.com
  16. http://pcper.com/podcast   
  17. http://twitter.com/ryanshrout and http://twitter.com/pcper
  18. Closing

Author:
Manufacturer: NVIDIA

GK110 Specifications

When the Fermi architecture was first discussed in September of 2009 at the NVIDIA GPU Technology Conference it marked an interesting turn for the company. Not only was NVIDIA releasing details about a GPU that wasn’t going to be available to consumers for another six months, but also that NVIDIA was building GPUs not strictly for gaming anymore – HPC and GPGPU were a defining target of all the company’s resources going forward.

Kepler on the other hand seemed to go back in the other direction with a consumer graphics release in March of this year without discussion of the Tesla / Quadro side of the picture. While the company liked to tout that Kepler was built for gamers I think you’ll find that with the information NVIDIA released today, Kepler was still very much designed to be an HPC powerhouse. More than likely NVIDIA’s release schedules were altered by the very successful launch of AMD’s Tahiti graphics cards under the HD 7900 brand. As a result, gamers got access to GK104 before NVIDIA’s flagship professional conference and the announcement of GK110 – a 7.1 billion transistor GPU aimed squarely at parallel computing workloads.

Kepler GK110

With the Fermi design NVIDIA took a gamble and changed directions with its GPU design betting that it could develop a microprocessor that was primarily intended for the professional markets while still appealing to the gaming markets that have sustained it for the majority of the company’s existence. While the GTX 480 flagship consumer card and the GTX 580 to some degree had overheating and efficiency drawbacks for gaming workloads compared to AMD GPUs, the GTX 680 based on Kepler GK104 has improved on them greatly. NVIDIA has still designed Kepler for high-performance computing though with a focus this time on power efficiency as well as performance though we haven’t seen the true king of this product line until today.

dieshot.jpg

GK110 Die Shot

Built on the 28nm process technology from TSMC, GK110 is an absolutely MASSIVE chip built on 7.1 billion transistors and though NVIDIA hasn’t given us a die size, it is likely coming close the reticle limit of 550 square millimeters. NVIDIA is proud to call this chip the most ‘architecturally complex’ microprocessor ever built and while impressive, it means there is potential for some issues when it comes to producing a chip of this size. This GPU will be able to offer more than 1 TFlop of double precision computing power with greater than 80% efficiency and 3x the performance per watt of Fermi designs.

02.jpg

Continue reading our overview of the newly announced NVIDIA Kepler GK110 GPU!

Live: First NVIDIA Kepler GK110 GPU Details

Subject: Graphics Cards | May 16, 2012 - 05:14 PM |
Tagged: nvidia, kepler, GTC 2012, gk110

We are posting live from the "Inside Kepler" talk at NVIDIA's GPU Technology Conference with details on the new GK110 GPU.  Here is what we know so far:

  • 7.1 billion transistors
  • 15 SMX (modified) units
  • 2880 available CUDA cores
  • Greater than 1 TFLOP FP64 (double precision) compute
  • 384-bit GDDR5 memory bus

01.jpg

The block diagram for the GK110 GPU

We will update this post with more photos and information as we have it!

02.jpg

Diagram of the updated SMX for GK110

Don't expect to see this GPU until at least  Q4 of this year.

Check out our full Kepler GK110 deep-dive article we just posted!  

Rumored NVIDIA GK110 Chip Coming This Summer

Subject: Graphics Cards | March 22, 2012 - 02:19 AM |
Tagged: nvidia, gpu, gk110, GK104

Update: The GK104 GeForce GTX 680 has now been released and we have our review published.

There have been some rumors going around the web about an alleged NVIDIA GK110 chip that is to be the successor to the upcoming GK104, or GTX 600 series, graphics cards. Videocardz is the latest site to pick up on the rumors, and they were able to translate the German blog from which the GK110 chip rumors originated. The NVIDIA GTX 600 series is not yet out, but rumored specs put the top end GTX 680 at 1536 CUDA cores, 2 GB GDDR5 memory, 3.54 billion transistors and a die size of 294mm^2. Clockspeed wise, the GTX 680 will run at a 1.006 GHz core clock, 2.012 GHz shader clock, and effective memory clock of 6.008 GHz.

NVIDIA-logo.gif

The rumored GK110 chip, on the other hand, will have 6 billion transistors, 2,304 CUDA cores, and a die size of 550mm^2. That die size figure is where many of the doubts about the rumors come from. Many people have expressed their doubts that NVIDIA would ever create a GPU using such a large die as it would be very cost prohibitive to manufacture. Further, at 550mm^2, that would mean a 87% jump versus the GTX 680, which to many seems too large to consider realistic. Even AMD's 7970 Tahiti XT part only has a die size of 365mm^2. While the other reported figures seem to match up to what's realistically possible, the jump in die size is taken by many to indicate the rumors are false.

  AMD HD 7970 Tahiti XT NVIDIA GTX 680 GK104 NVIDIA GK110
Die Size 365mm^2 294mm^2 550mm^2
Transistors 4.3 billion 3.54 billion 6 billion
GCN/CUDA Cores 2048 1536 2304
Single Precision Computing Power 3.79 TFlops 3.09 TFlops 4.5 TFlops
Gaming Performance (relative) 85% 100% 150%
Power Consumption 211 Watts 185 Watts 250-300 Watts
Launch Price $549 $549 $650
Release Date 12-22-2011 (paper) 3-22-2012 August 2012

Rumored GK110 Specs

The GK104 is likely going to be released today (be sure to stay tuned to PC Per Live for more information on the NVIDIA event today), and rumors suggest that the GK110 will be released sometime between August 2012 and January 2012. Performance is said to be 1.41 Teraflops more than the GTX 680 based on the GK104 chip.

It could be possible but it would be questionable for NVIDIA to release another big chip within five months of GK104's release. Videocardz suggests that it may in face be a "GTX 685" type of card, but it is bigger and uses more power than the GTX 680 so I honestly don't see NVIDIA doing such a thing. If anything, a refresh should use less power or provide the same performance at a lower price thanks to manufacturing advances. Should the 550mm^2 die hold true, it's not looking likely even for a GTX 700 series card but who knows what NVIDIA has up their sleeves. For now, I am going to focus on the GTX 680 release (heh).

Source: Videocardz