[WCCF] NVIDIA Ampere GA100 GPU Rumored Specifications Detailed – 8192 CUDA Cores, Up To 48 GB HBM2e Memory, Up To 2.2 GHz Clocks & 300W TDP
According to the forum member, the flagship Ampere GPU would be the GA100 and as expected, the full configuration would feature 128 streaming multi-processor units or 8192 CUDA cores. Utilizing the new process and GPU architecture, the chip is rumored to feature a maximum boost clock of up to 2.2 GHz on the GPU core. Based on the number of cores and the boost clock of the GA100 GPU, we are looking at a massive 36 TFLOPs of FP32 Compute performance.
These rumored specs for this compute card is insane. The GTX 2080Ti is only 12 TFLOP and GV100 is like 16.6 TFLOPs. if GA100 pushes 36TFLOS, that's insane. Previous rumors about the 8000+ cuda core compute card being at the low 1000 Mhz range. If this thing can clock to over 2.2Ghz, that means we can expect insane clock speeds on the cut down GA102 cards. 2.2Ghz? 2.3Ghz?
...could be the full-fat enterprise 'QuadroRTX or Tesla V100' successors, with a focus on AI / DL and for a gazillion $s...3080 Ti would be 'cut down' version, though when starting with over 8000 cuda cores, cutting is not so hard