NVIDIA GeForce RTX 4090/4080 Linux Compute CUDA & OpenCL Benchmarks, Blender Performance

[phoronix]

Phoronix: NVIDIA GeForce RTX 4090/4090 Linux Compute CUDA & OpenCL Benchmarks, Blender Performance

Last week I published a number of Linux gaming benchmarks for the GeForce RTX 4080 and RTX 4090 high-end graphics cards now that they finally arrived for my Linux testing on Phoronix. For those more interested in creator workloads and GPU OpenCL and CUDA compute performance for these high-end consumer Ada Lovelace graphics cards, this article is for you with an initial look at the compute performance across a wide range of workloads from Blender OptiX and CUDA rendering to common OpenCL GPU benchmarks.

https://www.phoronix.com/review/nvidia-rtx4080-rtx4090-compute

[zexelon]

This is what I have been waiting for and literally why I pay money to Phoronix to support what Michael does!

Very well done, thank you!

Unfortunately AMD could not even show up, which is to bad. ROCm has seen a LOT of effort in the last 8 months or so... but its still years behind CUDA. AMD strategy seriously dropped the ball on this front and are not appearing to be able to pick it back up.

Nvidia clearly invested the majority of their Ada dev budget into compute and it is awesomely evident here. Kudos to team green for utterly dominating here! Guess my future servers are going to have to be liquid cooled after all to keep thermals in line...

[luben]

Is the title wrong? Shouldn't it be "4090/4080"?

[Mahboi]

Crushing, as expected.
Nvidia has been extremely sturdy in their power growth since the generally poor 2000s. RT has worked great for them too.

AMD still has leagues of problems to clean up...I like them, but this is a hard fight for them.

[Michael]

Is the title wrong? Shouldn't it be "4090/4080"?

Yep fixed, thanks. overlooked it.

[piotrj3]

.... Nvidia literally got twice as fast in Blender, does it consuming 100W less then previous flagship. Literally ~~3x better power efficiency... It is so crushing.

And of course supreme linux support AMD ... is disqualified for failing to run workloads on linux.

[Lycanthropist]

I would like to see a comparison between nVidia and AMD using Vulkan Compute Workloads.

[tildearrow]

.... Nvidia literally got twice as fast in Blender, does it consuming 100W less then previous flagship. Literally ~~3x better power efficiency...

...because Blender has first-class NVIDIA support, whereas the AMD support is still new.

[Volta]

X Server? This nvidia trash doesn't run on Wayland, yet?

[mbriar]

I would like to see a comparison between nVidia and AMD using Vulkan Compute Workloads.

Like which ones that are actually relevant?