AMD’s next huge GPU, A 7nm GPU, but Vega 20 GPU Is probably a Tesla V100 Rival, Not GTX 1180

AMD’s upcoming GPUs, the new 7 nm Vega 20 expected to hit the 20 TFLOP barrier, the downside is huge power consumption figure of 400 W, reports say.7 nm Vega 20 GPUs, the next-gen Radeon Instinct video cards are expected to appear in the news very soon. AMD already provided a sneak peek at the upcoming 7 nm server parts, and based on what they unveiled, ASCII.jp concluded that the solutions based on the Vega 20 can break hit 20 TFLOPs of processing power. ASCII.jp’s estimations put the Vega 20 die size at 360 square millimeters, which is a solid 70 percent reduction in total die size when compared to the Vega 10 chip. 7 nm Vega 20 GPUs are probably solutions that would take on the Tesla V100 from Nvidia and not the long expected 1180 gpus. If the estimations and reports are correct, AMD 7nm Vega 20 GPU Estimated To Feature 20 TFLOPs of Compute  A HPC Powerhouse in The Making, 32 GB of HBM2 VRAM, Up To 400W TDP. And this is not even rational/ normal for gaming platform.

Radeon instinct mi25 credit AMD

Radeon instinct mi25 credit AMD

The clock rate on 7nm LPP is a big trick, you either get 55% power savings from the same frequency as 14nm or you can get 40% faster clock rates from the same power as 14nm. Here’s the thing though, if you were to aim for 40% faster clocks at the same power package, you’d also have to feed power to the extra HBM2 VRAM (4 Stacks vs 2 Stacks), the new DNN instruction sets.

Source(s)

Wccftech

ASCII.jp (translated)

Tags: Amd, AMD VEGA, amd vega 20, gtx, GTX 1180, new amd gpu 2018, Next Gen Nvidia GPUs GTX 1180, NVDA, Nvidia, Gaming, technology, video games

Mohsen Daemi
Latest posts by Mohsen Daemi (see all)