Tuesday, July 23rd 2013

NVIDIA Unveils New Flagship GPU for Visual Computing

NVIDIA today unveiled the visual computing industry's new flagship technology - the NVIDIA Quadro K6000 GPU, the fastest and most capable GPU ever built. NVIDIA today also launched a new line of professional graphics GPUs for mobile workstations, delivering the highest levels of performance and graphics memory ever available on mobile platforms.

The Quadro K6000 GPU delivers five-times higher compute performance and nearly double the graphics capability of its predecessor, the NVIDIA Quadro 6000 GPU, and features the world's largest and fastest graphics memory.

Combining breakthrough performance and advanced capabilities in a power-efficient design, the Quadro K6000 GPU enables leading organizations such as Pixar, Nissan, Apache Corporation and the Weather Channel's WSI division to tackle visualization and analysis workloads of unprecedented size and scope.
  • Animation and Visual Effects - Pixar
  • "The Kepler features are key to our next generation of real-time lighting and geometry handling. We were thrilled to get an early look at the K6000. The added memory and other features allow our artists to see much more of the final scene in a real-time, interactive form, and allow many more artistic iterations."
    Guido Quaroni, Pixar vice president of Software R&D
  • Product Styling - Nissan
  • "With Quadro K6000's 12 GB of memory, I am now able to load nearly complete vehicle models into RTT Deltagen and have stunning photorealism almost instantly. Instead of spending significant time simplifying the models to fit into previous hardware, we can now spend more time reviewing and iterating designs up front which helps avoid costly changes to tooling."
    Dennis Malone, associate engineer, Nissan North America
  • Energy Exploration - Apache
  • "Compared to the Quadro K5000, the Quadro K6000 tripled the performance when running jobs on Terraspark's InsightEarth application. With jobs running in mere minutes, we can run more simulations and get better insight into where to drill. In this business, drilling in the wrong place is a multi-million dollar mistake, and the Quadro K6000 gives us the edge to make better decisions."
    Klaas Koster, manager, seismic interpretation, Apache Corporation
Unprecedented Performance
The Quadro K6000 GPU is based on the NVIDIA Kepler architecture - the world's fastest, most efficient GPU architecture. Key performance features and capabilities include:
  • 12 GB ultra-fast GDDR5 graphics memory lets designers and animators model and render characters and scenes at unprecedented scale, complexity and richness
  • 2,880 streaming multiprocessor (SMX) cores deliver faster visualization and compute horsepower than previous-generation products
  • Supports four simultaneous displays and up to 4k resolution with DisplayPort 1.2
  • Ultra-low latency video I/O and support for large-scale visualizations
"The NVIDIA Quadro K6000 GPU is the highest performance, most capable GPU ever created for the professional graphics market," said Ed Ellett, senior vice president, Professional Solutions Group at NVIDIA. "It will significantly change the game for animators, digital designers and engineers, enabling them to make the impossible possible."

New Mobile Workstation GPUs
NVIDIA today also revealed a new flagship professional graphics GPU for workstation notebooks, the NVIDIA Quadro K5100M GPU. Delivering the highest levels of performance and graphics memory available on notebook platforms, the Quadro K5100M anchors a new line of workstation notebook graphics that includes the Quadro K4100M, K3100M, K2100M, K1100M, K610M, and K510M GPUs.

Quadro GPUs are designed, built and tested by NVIDIA to provide the superb reliability, compatibility and dependability that professionals require. They are certified and recommended by more than 150 leading software application providers worldwide.

Availability
The NVIDIA Quadro K6000 will be available beginning this fall from Dell, HP, Lenovo and other major workstation providers; from systems integrators, including BOXX Technologies and Supermicro; and from authorized distribution partners, including PNY Technologies in North America and Europe, ELSA and Ryoyo in Japan, and Leadtek in Asia Pacific.

The new Quadro mobile workstation graphics product line will also be available beginning this fall from major mobile workstation OEMs.
Add your own comment

22 Comments on NVIDIA Unveils New Flagship GPU for Visual Computing

#1
cheesy999
by: Cristian_25H
12 GB of memory
:eek:

that graphics card has 3x the ram my system has.
Posted on Reply
#2
Xero717
2880 SMX Cores.... Titan Ultra incoming Q1 2014? Looks like Nvidia has been hoarding their best binned GK110s. :shadedshu
Posted on Reply
#3
v12dock
I just installed a computer at work today with a K5000, 128GB ram and duel Xeon E5-2687W. I dont think the K5000 cuts it
Posted on Reply
#5
Prima.Vera
Better, maybe, than AMD's FirePro W9000 ?
Posted on Reply
#6
Recus
by: Prima.Vera
Better, maybe, than AMD's FirePro W9000 ?
W9000 will be eaten without any milk.
Posted on Reply
#7
Prima.Vera
by: Recus
W9000 will be eaten without any milk.
Benches or it didn't happened! ;)
Posted on Reply
#8
fusionblu
Looks like a fully unlocked Titan (or Titan Ultra) with double the amount of RAM to me. I can see this helping animation companies and game developers a lot, and I'm sure more than a few enthusiasts wouldn't mind "borrowing" one of these. :laugh:

I will still keep with my Titan though as it's more than enough for the gaming I do. :nutkick:
Posted on Reply
#9
Xzibit
by: Prima.Vera
Better, maybe, than AMD's FirePro W9000 ?
Yes.

Although its not competing with the W9000. The W9000 competes with the K5000

The K6000 is Nvidias response to AMD FirePro S10000 which has been out for 7-8months. SP and DP are almost similar with the S10000 slightly ahead.

You can only drive 4 monitors with the K6000 as oppose to 6 with the S10000 but you get 12GB of memory instead of 6GB (3GBx2)

S10000 sells for as low as $3,000 (MSRP $3599) and I don't think the K6000 will compete with that price range. Expect it to be from $3,599 - $4,599.
Posted on Reply
#10
amdftw
by: Xero717
2880 SMX Cores.... Titan Ultra incoming Q1 2014? Looks like Nvidia has been hoarding their best binned GK110s. :shadedshu
Whatever, it will be old shitty kepler which does not support Dx11.1 fully...
Nv in trouble TSMC cant make 20nm chips before 2014 Q2-Q3, and the games just coming.
Posted on Reply
#12
Fluffmeister
by: amdftw
Whatever, it will be old shitty kepler which does not support Dx11.1 fully...
Nv in trouble TSMC cant make 20nm chips before 2014 Q2-Q3, and the games just coming.
Luckily the bits they don't support have no relevance to gaming. Good effort though, AMD would pay you if they could afford it.
Posted on Reply
#13
tigger
I'm the only one
Now that is Fap worthy.
Posted on Reply
#14
HumanSmoke
by: Xzibit
The K6000 is Nvidias response to AMD FirePro S10000 which has been out for 7-8months. SP and DP are almost similar with the S10000 slightly ahead.
Yes and No.
S10000's theoretical SP32 is 5.91 TFlops/sec against the K6000's 5.2 TFlops/sec. The S10000 is limited to 1:4 double precision (1.48 TFlops), while the K6000 is at 1:3 rate (1.73 TFlops)....even though some sites are quoting ~1.4 TF. Quite how 5.2 single precision divided by three equates to 1.4 IDK.

Obviously, the actual performance is dependant upon the software suites used, and how closely tailored they are to the relative architectures.
Posted on Reply
#15
Fluffmeister
The age-old battle of two vs one. :roll:

Prima.Vera thanks me.
Posted on Reply
#16
NeoXF
by: Xero717
2880 SMX Cores.... Titan Ultra incoming Q1 2014? Looks like Nvidia has been hoarding their best binned GK110s. :shadedshu
And clock it at what? 750MHz? I don't see the point, Kepler arch (like almost anything else) gives diminishing returns, GTX 780 can clock better than TITANs, and for the most part, will be faster max OC vs max OC, than them. TITAN Ultra or whatever won't do much to change that.


God damn it, where's AMD's new GPUs.
Posted on Reply
#17
Xaser04
by: Xero717
2880 SMX Cores.... Titan Ultra incoming Q1 2014? Looks like Nvidia has been hoarding their best binned GK110s. :shadedshu
A Titan Ultra is unlikley simply because Nvidia would want to use all of the fully functional binned cores for these significantly more profitable Quadro cards.
Posted on Reply
#18
Fluffmeister
Well this is already clocked at 900Mhz within a 225W power envelope, so there is plenty of performance left on the table should nV feel the need to respond once AMD finally launch something new.
Posted on Reply
#19
Xzibit
by: Fluffmeister
Well this is already clocked at 900Mhz within a 225W power envelope, so there is plenty of performance left on the table should nV feel the need to respond once AMD finally launch something new.
I don't think AMD needs to launch something new. The K5000 was released in Aug 2012 when AMD also released their FirePro Wx000 Series line of cards. Like I mentioned before the AMD FirePro S10000 was available November 2012. The Quadro K6000 has a release window of Sep-Nov 2013. That's 1 full year of no competition. The Quadro K5000 (GK104) isn't much of a competitor when it comes to DP.

So the need to finally release something new has been there for Nvidia for almost a year.
Posted on Reply
#20
Fluffmeister
My comment was more regarding the need/potential of a Titan Ultra coming to the consumer market, although it's funny you say AMD have no competition in a market nVidia dominate.

I'm sure nVidia's established client base welcome this card coming to the market.
Posted on Reply
#22
HumanSmoke
by: Xzibit
I don't think AMD needs to launch something new. The K5000 was released in Aug 2012 when AMD also released their FirePro Wx000 Series line of cards. Like I mentioned before the AMD FirePro S10000 was available November 2012. The Quadro K6000 has a release window of Sep-Nov 2013. That's 1 full year of no competition. The Quadro K5000 (GK104) isn't much of a competitor when it comes to DP.

So the need to finally release something new has been there for Nvidia for almost a year.
There's always a "need" to release new product regardless of vendor. I think you'll find that a complete lack of professional drivers makes AMD a non-starter for many design houses.
With Nvidia owning 80% of the of the pro graphics market based upon Fermi (or older) architectures holding their own against the latest cards (see Fluffmeister's Hot Hardware review link- or the TH review for instance), it could be argued that maybe other vendors should be just as -if not more concerned.

Bear in mind that:
1. Actual performance doesn't reflect theoretical performance ( note Fluffmeister's Hot Hardware FirePro review, as well as Tom's review). Theoretically, castrated Fermi based cards shouldn't be able to live with fully enabled Tahiti. Actuality is somewhat different.
2. The pro market numbers for both Nvidia and AMD will likely decrease slightly as Intel pushes Xeon Phi into the HPC arena. On that note, the GK110's closest competitor in theoretical performance for a single GPU is the Xeon Phi at around 85% of its rated theoretical performance....actual performance seems to be another matter entirely...

(Courtesy of HPC Magazine)
Posted on Reply
Add your own comment