|Image source anandtech|
Does anybody know the snappiest method to wind up an AI specialist? We aren’t especially keen on making Skynet- – or more probable preparing PCs to show more important promotions – however we wouldn’t state no to the restricted release Titan V GPUs that Nvidia’s CEO Jensen Huang gave away at the Computer Vision and Pattern Recognition meeting in Salt Lake City.
The new Titan V CEO Edition depends on the Volta stage and highlights the Tensor Cores that Nvidia made particularly for AI investigate. You can most likely think about how it got the CEO Edition moniker, yet on the off chance that it isn’t self-evident, this is on the grounds that Huang marked every one of the GPUs he gave away at the Utah meeting. In any case, a mark is the main thing isolating the Titan V CEO Edition from its unsigned partners – it gloats different changes too.
The most clear contrast between the Titan V CEO Edition and the plain-ol’ Titan V is this new version highlights 32GB of memory; its forerunner has 12GB. AnandTech discovered that NVIDIA likewise included with the new unit 128 render, yield units (ROPs) contrasted with the Titan V’s 96, enhanced the memory transport width to 4,096 bits (from 3,072), and expanded tensor execution to 125 TFLOPS from 110 TFLOPS.
Nvidia said it gave away 20 of the Titan V CEO Edition GPUs at the Computer Vision and Pattern Recognition meeting. The organization hasn’t formally reported the Titan V CEO Edition for the retail showcase, however, we likewise question it would make 20 units of another Titan V just so Huang could give them away amid an AI meeting. We speculate the organization will uncover an also spiced item at some point later on.
For the time being, however, it appears like Huang just needed to charm the analysts at this meeting. He additionally endeavored to prevail upon the group with words:
“There’s all kinds of research being done here. As someone who benefits from your work, as a person who is going to enjoy the incredible research you guys do — solving some of the world’s grand challenges — and to be able to witness artificial intelligence happen in my lifetime, I want to thank all of you guys for that,” Huang said. “You guys bring me so much joy.”
|NVIDIA Compute Accelerator Specification Comparison|
|Titan V||Tesla V100
|Memory Clock||1.7Gbps HBM2?||1.7Gbps HBM2||1.75Gbps HBM2||11.4Gbps GDDR5X|
|Memory Bus Width||4096-bit||3072-bit||4096-bit||384-bit|
|Single Precision||13.8 TFLOPS||13.8 TFLOPS||14 TFLOPS||12.1 TFLOPS|
|Double Precision||6.9 TFLOPS
|125 TFLOPS||110 TFLOPS||112 TFLOPS||N/A|
|Manufacturing Process||TSMC 12nm FFN||TSMC 12nm FFN||TSMC 12nm FFN||TSMC 16nm FinFET|