Jules Urbach, CEO of cloud graphics company OTOY, has shared one of the earliest benchmarks for NVIDIA’s Ampere-based A100 Tensor Core GPU. This shouldn’t be much of a surprise (gee whiz, newer hardware is faster!), but green team’s first 7 nm product managed to beat all contenders on OctaneBench, the official benchmark for OctaneRender (the world’s “first and fastest” GPU render engine).
You can check out the full results in the tweet below, but Ampere, as Urbach points out, appears to be around 43 percent faster than Turing. That’s not too shabby, especially when you consider the fact that RTX was off!
A record breaking week🚀
— Jules Urbach (@JulesUrbach) July 23, 2020
The @NVIDIA A100 has now become the fastest GPU ever recorded on #OctaneBench: 446 OB4*#Ampere appears to be ~43% faster than #Turing in #OctaneRender – even w/ #RTX off!
(*standard Linux OB4 benchmark, RTX off, recompiled for CUDA11, ref. 980=102 OB) pic.twitter.com/WlpG7ZUq2R