NVIDIA CEO Jensen Huang Gives Away 20 "CEO Edition TITAN V" To Titans of the AI Industry

[Raevenlord]

During the Computer Vision and Pattern Recognition conference in Salt Lake City on Wednesday, NVIDIA CEO Jensen Huang pulled a PR stunt eased by NVIDIA's currently entrenched position in the AI-acceleration market. The new TITAN V graphics card being given away under the "CEO Edition" tag feature more than double the amount of HBM memory - up to 32 GB of HBM 2 memory compared to the original Titan V's 12 GB.

This is, essentially, a Titan-branded Quadro GV100 accelerator, made all the more exotic for the limited-edition branding. It features the GV100 graphics processor - a large chip with a die area of 815 mm² and 21,100 million transistors. It features 5120 shading units, 320 texture mapping units and 128 ROPs. Also included are 640 tensor cores which help improve the speed of machine learning applications.



View at TechPowerUp Main Site

 

[FreedomEclipse]

So only two women got to walk away with them out of the 20?? talk about equality. <A U T I S T I C S C R E A M>

 

[dj-electric]

Easily 25K$ a pop

 

[Caring1]

So only two women got to walk away with them out of the 20?? talk about equality. <A U T I S T I C S C R E A M>

If best person for the job means only two female CEOs, then that is equality, discrimination would be giving the women the position due to their gender.

 

[xkm1948]

If best person for the job means only two female CEOs, then that is equality, discrimination would be giving the women the position due to their gender.

Agree woth this.

 

[mouacyk]

So only two women got to walk away with them out of the 20?? talk about equality. <A U T I S T I C S C R E A M>

There's equality in representation and then there's proportionality in presence.

Gotta beat the negative press in some meaningful way.

 

[ensabrenoir]

...so i can safely assume that these will not be on ebay anytime soon.

 

[Vya Domus]

If best person for the job means only two female CEOs, then that is equality, discrimination would be giving the women the position due to their gender.

Too bad feminists only care about equality of outcome and nothing else.

 

[kastriot]

Lucky bastards!

 

[jabbadap]

Lucky bastards!

It's actually win-win for both, those are some pro DL/ML/AI people and nvidia can assure them continuing to use their proprietary cuda. Nvidia actually did the same too with tesla V100 pcie ~year ago.

 

[IceShroom]

Does Titan V support Pro Driver? Or still use GeForce Driver?

 

[jabbadap]

Does Titan V support Pro Driver? Or still use GeForce Driver?

Geforce driver. Probably does not matter anyway, these things are usually bought and used for cuda computing, not so much on pro rendering tasks. Though nvidia did enabled couple of pro driver rendering paths, after AMD marketed Vega FE beating Titan on some specviewperf benches.

 

[IceShroom]

Geforce driver. Probably does not matter anyway, these things are usually bought and used for cuda computing, not so much on pro rendering tasks. Though nvidia did enabled couple of pro driver rendering paths, after AMD marketed Vega FE beating Titan on some specviewperf benches.

So does mean 1/64 FP16 rating like all GeForce card?

 

[jabbadap]

So does mean 1/64 FP16 rating like all GeForce card?

No that has nothing to do with drivers. Titan V has full compute of everything: FP16:FP32 = 2:1, FP64:FP32 = 1:2, int8:FP32 = 4:1. And some ~125 TFlops of fp16 on certain deep learning matrix calculations.

 

[IceShroom]

No that has nothing to do with drivers. Titan V has full compute of everything: FP16:FP32 = 2:1, FP64:FP32 = 1:2, int8:FP32 = 4:1. And some ~125 TFlops of fp16 on certain deep learning matrix calculations.

Hmm. Previous Titans had crippled compute performance than their Quardo counterparts. Thats why asking.

 

[jabbadap]

Hmm. Previous Titans had crippled compute performance than their Quardo counterparts. Thats why asking.

Can't remember any same chip quadro and titans, which would have different compute capabilities.

 

[IceShroom]

Can't remember any same chip quadro and titans, which would have different compute capabilities.

Quardo P6000 and TITAN XP and Xp both use the same GP102.
https://www.anandtech.com/show/11253/nvidia-announces-nvidia-titan-xp-full-gp102-for-1200
https://www.anandtech.com/show/10516/nvidia-announces-quadro-pascal-family-quadro-p6000-p5000

 

[jabbadap]

Quardo P6000 and TITAN XP and Xp both use the same GP102.
https://www.anandtech.com/show/11253/nvidia-announces-nvidia-titan-xp-full-gp102-for-1200
https://www.anandtech.com/show/10516/nvidia-announces-quadro-pascal-family-quadro-p6000-p5000

And both have the same compute capabilities: fp16:fp32 = 1:64, FP64:fp32 = 1:32, int8:fp32 = 4:1. All that gp102 gpu have in hardware.

 

[bogami]

Cheap ad for giant. and when I see the pictures I wonder whether they are exploiting children as the main developmental force, etc. this pig is so thick .... ,and at all what will you do with one. free promotion !

 

[IceShroom]

And both have the same compute capabilities: fp16:fp32 = 1:64, FP64:fp32 = 1:32, int8:fp32 = 4:1. All that gp102 gpu have in hardware.

Didnt notice pro level Quardo has such low FP16 rate.

 

[AnarchoPrimitiv]

Yes, because the "Titans of the AI Industry" are the ones that really need free merchandise

 

[Fluffmeister]

Nvidia release Titan V *whine*. Nvidia give a few Titan V to true cutting edge (not blockchain) pioneers *whine*.

 

[TheGuruStud]

This guy won't be happy till he's in place of the Lincoln memorial.

 

[R-T-B]

There's equality in representation and then there's proportionality in presence.

This. So many people fail to grasp there are two kinds of ways to represent equality. So you are both right.

This guy won't be happy till he's in place of the Lincoln memorial.

I'm sure that was his goal in giving away free stuff...

Publicity stunt it may be, but it certainly can't be construed as greedy no matter how far you twist it.

 

[Prima.Vera]

So a ~100,000 bucks PR stunt??