NVIDIA Linux OpenCL CUDA RTX SUPER Compute

NVIDIA GeForce RTX SUPER Linux OpenCL/CUDA GPU compute benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/1909299-AS-RTXSUPERC34&grr.

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Max SP Flops

OctaneBench

Total Score

LuxMark

OpenCL Device: GPU - Scene: Microphone

LuxMark

OpenCL Device: GPU - Scene: Luxball HDR

LuxMark

OpenCL Device: GPU - Scene: Hotel

FAHBench

PlaidML

FP16: Yes - Mode: Inference - Network: NASNer Large - Device: OpenCL

PlaidML

FP16: Yes - Mode: Inference - Network: VGG19 - Device: OpenCL

PlaidML

FP16: Yes - Mode: Inference - Network: DenseNet 201 - Device: OpenCL

Rodinia

Test: OpenCL Myocyte

PlaidML

FP16: Yes - Mode: Inference - Network: VGG16 - Device: OpenCL

clpeak

OpenCL Test: Double-Precision Double

PlaidML

FP16: No - Mode: Inference - Network: VGG19 - Device: OpenCL

PlaidML

FP16: Yes - Mode: Inference - Network: Inception V3 - Device: OpenCL

LeelaChessZero

Backend: OpenCL

PlaidML

FP16: No - Mode: Inference - Network: VGG16 - Device: OpenCL

PlaidML

FP16: No - Mode: Inference - Network: Inception V3 - Device: OpenCL

PlaidML

FP16: Yes - Mode: Inference - Network: ResNet 50 - Device: OpenCL

NAMD CUDA

ATPase Simulation - 327,506 Atoms

PlaidML

FP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCL

PlaidML

FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL

Rodinia

Test: OpenCL Particle Filter

cl-mem

Benchmark: Read

JuliaGPU

OpenCL Device: GPU

PlaidML

FP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCL

cl-mem

Benchmark: Copy

cl-mem

Benchmark: Write

PlaidML

FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL

Darktable

Test: Masskrug - Acceleration: OpenCL

clpeak

OpenCL Test: Integer Compute INT

Darktable

Test: Boat - Acceleration: OpenCL

clpeak

OpenCL Test: Single-Precision Float

Darktable

Test: Server Room - Acceleration: OpenCL

ViennaCL

OpenCL LU Factorization

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: FFT SP

clpeak

OpenCL Test: Global Memory Bandwidth

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: MD5 Hash

System Power Consumption Monitor

Phoronix Test Suite System Monitoring

GPU Temperature Monitor

Phoronix Test Suite System Monitoring

OctaneBench

System Power Consumption Monitor

OctaneBench

Total Score

LeelaChessZero

System Power Consumption Monitor

LeelaChessZero

Backend: OpenCL

PlaidML

System Power Consumption Monitor

PlaidML

FP16: No - Mode: Inference - Network: VGG19 - Device: OpenCL

PlaidML

System Power Consumption Monitor

PlaidML

FP16: No - Mode: Inference - Network: VGG16 - Device: OpenCL

PlaidML

System Power Consumption Monitor

PlaidML

FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL

PlaidML

System Power Consumption Monitor

PlaidML

FP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCL

PlaidML

System Power Consumption Monitor

PlaidML

FP16: No - Mode: Inference - Network: Inception V3 - Device: OpenCL

PlaidML

FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL