Stefan Mileschin

21 billion transistors 12nm

And it happened, just as Fudzilla reported, Nvidia just announced Tesla V100.

This is the 21 billion transistor core manufactured in TSMCs 12 FFN nanometer process. This is a huge chip, we are talking about 815 mm2. Jensen Huang the CEO of Nvidia compared the die size with the size of an apple watch.

Nvidia spent 3 billion of their R&D budget to get to this point, and this is the first of many Volta GPUs and processors. Volta has 5120 Cud cores and can perform 7.5 FP64 TFLOPS or 15 FP32 TFLOPS.

Nvidia has a special part of silicon, a die dedication to a Tensor core. There are 120 Tensor TFLOPS units. The chip is made to gloat in artificial intelligence and deep learning.

The Tesla V100 has 20 MB SM RF with 16GB Cache. The memory of choice is the 16GB HBM2 at 900 GB/s and support for 300 GB/s NVlink.

Pascal P100 had 7.1 Billion transistors and came with 3584 CUDA processors. You can tell that the transistor number can be attributed to this new functional units called Tensor units that will increase performance in deep learning and AI applications.

We will get back with a bit more details after the Q&A that is happening in the next few hours.

http://fudzilla.com/news/graphics/43...sla-v100-volta