NVIDIA GeForce RTX 3080 Offers Up Incredible Linux GPU Compute Performance
Written by Michael Larabel in Graphics Cards on 6 October 2020 at 11:00 AM EDT. Page 5 of 9. 38 Comments.
Next up was the PlaidML deep learning framework using OpenCL acceleration.
Across the various neural networks tested with PlaidML, the RTX 3080 GPU inference performance was a great deal faster than the TITAN RTX and other Turing GPUs benchmarked.
The RTX 3080 continued leading in performance-per-dollar at $699.
And had the highest peak power consumption, but for this deep learning framework the performance-per-Watt was comparable to the RTX 2080 Ti and TITAN RTX.