NVIDIA Linux GPU Performance Comptue RTX 3000 Series

Benchmarks by Michael Larabel.

HTML result view exported from: https://openbenchmarking.org/result/2012252-HA-NVIDIAGPU08&sor&grr.

Blender

Blend File: Barbershop - Compute: NVIDIA OptiX

VkFFT

LeelaChessZero

Backend: OpenCL

RedShift Demo

LuxCoreRender OpenCL

Scene: Food

LuxCoreRender OpenCL

Scene: LuxCore Benchmark

OctaneBench

Total Score

LuxCoreRender OpenCL

Scene: DLSC

Blender

Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX

PlaidML

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

PlaidML

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

Blender

Blend File: Classroom - Compute: NVIDIA OptiX

FAHBench

NCNN

Target: Vulkan GPU - Model: regnety_400m

NCNN

Target: Vulkan GPU - Model: squeezenet_ssd

NCNN

Target: Vulkan GPU - Model: yolov4-tiny

NCNN

Target: Vulkan GPU - Model: resnet50

NCNN

Target: Vulkan GPU - Model: alexnet

NCNN

Target: Vulkan GPU - Model: resnet18

NCNN

Target: Vulkan GPU - Model: vgg16

NCNN

Target: Vulkan GPU - Model: googlenet

NCNN

Target: Vulkan GPU - Model: blazeface

NCNN

Target: Vulkan GPU - Model: efficientnet-b0

NCNN

Target: Vulkan GPU - Model: mnasnet

NCNN

Target: Vulkan GPU - Model: shufflenet-v2

NCNN

Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3

NCNN

Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2

NCNN

Target: Vulkan GPU - Model: mobilenet

PlaidML

FP16: No - Mode: Inference - Network: NASNer Large - Device: OpenCL

RealSR-NCNN

Scale: 4x - TAA: Yes

Blender

Blend File: Fishy Cat - Compute: NVIDIA OptiX

Blender

Blend File: BMW27 - Compute: NVIDIA OptiX

VkResample

Upscale: 2x - Precision: Double

PlaidML

FP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCL

clpeak

OpenCL Test: Double-Precision Double

LuxCoreRender OpenCL

Scene: Rainbow Colors and Prism

Betsy GPU Compressor

Codec: ETC2 RGB - Quality: Highest

PlaidML

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

Betsy GPU Compressor

Codec: ETC1 - Quality: Highest

PlaidML

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

PlaidML

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

VkResample

Upscale: 2x - Precision: Single

PlaidML

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

RealSR-NCNN

Scale: 4x - TAA: No

PlaidML

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

Rodinia

Test: OpenCL Particle Filter

Hashcat

Benchmark: TrueCrypt RIPEMD160 + XTS

Hashcat

Benchmark: SHA-512

Hashcat

Benchmark: SHA1

Hashcat

Benchmark: MD5

PlaidML

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

PlaidML

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

clpeak

OpenCL Test: Integer Compute INT

Waifu2x-NCNN Vulkan

Scale: 2x - Denoise: 3 - TAA: Yes

cl-mem

Benchmark: Copy

cl-mem

Benchmark: Write

cl-mem

Benchmark: Read

clpeak

OpenCL Test: Single-Precision Float

MandelGPU

OpenCL Device: GPU

ArrayFire

Test: Conjugate Gradient OpenCL

Hashcat

Benchmark: 7-Zip

ViennaCL

OpenCL LU Factorization

clpeak

OpenCL Test: Global Memory Bandwidth

FinanceBench

Benchmark: Black-Scholes OpenCL

Phoronix Test Suite v10.8.4