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

Written by Michael Larabel in Graphics Cards on 21 February 2023 at 01:00 PM EST. Page 7 of 8. 31 Comments.
NAMD CUDA benchmark with settings of ATPase Simulation, 327,506 Atoms. RTX 4090 was the fastest.
SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: Texture Read Bandwidth. RTX 4090 was the fastest.
SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: GEMM SGEMM_N. RTX 4090 was the fastest.
SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: MD5 Hash. RTX 4090 was the fastest.
SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: Reduction. RTX 4090 was the fastest.

It would have been more interesting if the AMD ROCm stack was playing well for this round of testing with AMD Radeon comparison points.

FluidX3D benchmark with settings of Test: FP32-FP16C. RTX 4090 was the fastest.

With FluidX3D the RTX 4090 performance was terrific though strangely the RTX 4080 was performing similar to the RTX 3080 in this particular benchmark.


Related Articles
AMD Radeon RX 7900 GRE Linux Performance
Intel Arc Graphics A580 / A750 / A770 Linux Performance For Early 2024
35-Way Linux GPU Graphics Comparison, Initial NVIDIA RTX 40 SUPER Linux Benchmarks
AMD Radeon RX 7600 XT Linux Performance
AMD Announces The Radeon RX 7600 XT For 1080p~1440p Gaming At $329
Intel Integrated Graphics Performance From Gen9 To Meteor Lake Arc Graphics