Nvidia’s 7nm GPU blows every other card out of the water in Octane Render benchmarks
Alright gamers, we finally have something concrete on the benchmarks front of Nvidia’s latest Ampere graphics card series. Jules Urbach has tweeted Nvidia A100 Ampere benchmarks.
As you already know that Nvidia A100 gpu architecture was announced earlier this year with heavy emphasis on machine learning and AI usability of Nvidia Ampere series.
Consumer graphics card are also going to be based on this architecture as well and with a benchmarking score like this, we can really keep our hopes high when the said series releases.
Jules Urbach had tweeted on 24 July 2020, 5:18 AM that they have found Nvidia A100 to be 43% faster than Nvidia Turing in their Octane Render synthetic benchmarking software.
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
RTX was turned off because with normal gaming, encoding/transcoding, RTX will be off anyway for most gamers and developers.
The introduced Nvidia A100 was a 7nm gpu and it is still being speculated whether Nvidia RTX 3000 series is going to be on 7nm or Samsung’s 8nm fabrication process.
Another change that we saw is regarding the upgraded version of PCI express, i.e. 4.0. The cards are said to be backwards compatible with PCIe 3.0 as well.
The benchmarked card had 6912 CUDA cores and 40 GB of HBM2 memory. In comparison to that, Nvidia RTX Titan has 4608 CUDA cores and 72 RT cores.
We were thinking that Ampere architecture will be around 20-30% faster than Turing but it has successfully proved that it is 40+% faster than Turing.
Let’s hope that the same is true with Nvidia RTX 3000 series in comparison to RTX 2000 series.
So…what do you think? How much faster RTX 3000 gpus are going to be? Let me know in the comment section below.