By using this site, you agree to our Privacy Policy and our Terms of Use. Close
Netyaroze said:
Slimebeast said:
Netyaroze said:
Volta will have like 6-8 Tflops ?

Nvidia said some time ago that in 2019 they will have 40 Tflop in a 100 Watt GPU. Thats like 3 shrinks away. Volta: 20nm xx: 16nm xxx: 10nm

That 1 TB bandwith might soon be necessary and if stacked ram becomes reality we can look forward to a smaller quiter cheaper PS4.

2020-2025 Avatar realtime will hopefully become reality.

How does that projected number, 40TFlop, correspond to the manufactoring process of 10 nm? I have forgotten the maths behind die shrinks (the correlation between process shrink to amount of transistors and power consumption). Could a shrink from 28nm (today) to 10 nm really allow an increase of around 20 x the processing power that we today get from a 100W GPU?

 

Its roughly 8-9 times more transistors from 28nm to 10nm. So you could get a 8-9 times increase in raw gpu performance just by putting 8-9 28nm Gpus on the wafer, theoretically. But there is no math for that anymore the development isn't as straightforward as it once was.

It also depends on architecture. There are many transistors who have not really something to do with flop performance this comes from the cores. And I doubt Nvidia needs 9 times more of everything. So they probably target to put a higher percentage of transistors to use for flop performance than in current Gpus. Its possible to have a 20 times increase tflop performance from 28nm to 10nm it all depends on the architecture the wafer quality etc. Could be anywhere really without inside knowledge its impossible to say. Nvidia always reached roughly their goals. So a 30-40 Tflop 100-120 Watt Gpu from 2019-2022 can be expected.        

 

 

 

So 8-9 times by going from 28nm to 10nm, but what about from 28nm to 20nm which is around the corner? What sort of increase does that give in raw performance (ignoring any architectural advances)?