NVIDIA Linux GPU Performance Comptue RTX 30 RTX 20

Benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2012250-HA-NVIDIAGPU61&grw&sro&rro.

Betsy GPU Compressor

Codec: ETC1 - Quality: Highest

Betsy GPU Compressor

Codec: ETC2 RGB - Quality: Highest

PlaidML

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

LuxCoreRender OpenCL

Scene: DLSC

PlaidML

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

PlaidML

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

LuxCoreRender OpenCL

Scene: Food

LuxCoreRender OpenCL

Scene: Rainbow Colors and Prism

PlaidML

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

PlaidML

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

PlaidML

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

PlaidML

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

PlaidML

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

PlaidML

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

PlaidML

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

PlaidML

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

LeelaChessZero

Backend: OpenCL

NCNN

Target: Vulkan GPU - Model: mobilenet

NCNN

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

NCNN

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

NCNN

Target: Vulkan GPU - Model: shufflenet-v2

LuxCoreRender OpenCL

Scene: LuxCore Benchmark

NCNN

Target: Vulkan GPU - Model: mnasnet

NCNN

Target: Vulkan GPU - Model: efficientnet-b0

NCNN

Target: Vulkan GPU - Model: blazeface

NCNN

Target: Vulkan GPU - Model: googlenet

NCNN

Target: Vulkan GPU - Model: vgg16

NCNN

Target: Vulkan GPU - Model: resnet18

NCNN

Target: Vulkan GPU - Model: alexnet

NCNN

Target: Vulkan GPU - Model: resnet50

NCNN

Target: Vulkan GPU - Model: yolov4-tiny

NCNN

Target: Vulkan GPU - Model: squeezenet_ssd

NCNN

Target: Vulkan GPU - Model: regnety_400m

Rodinia

Test: OpenCL Particle Filter

ArrayFire

Test: Conjugate Gradient OpenCL

Blender

Blend File: BMW27 - Compute: NVIDIA OptiX

Blender

Blend File: Classroom - Compute: NVIDIA OptiX

Blender

Blend File: Fishy Cat - Compute: NVIDIA OptiX

Blender

Blend File: Barbershop - Compute: NVIDIA OptiX

Blender

Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX

FAHBench

Hashcat

Benchmark: MD5

Hashcat

Benchmark: SHA1

Hashcat

Benchmark: 7-Zip

Hashcat

Benchmark: SHA-512

Hashcat

Benchmark: TrueCrypt RIPEMD160 + XTS

OctaneBench

Total Score

RedShift Demo

FinanceBench

Benchmark: Black-Scholes OpenCL

cl-mem

Benchmark: Copy

cl-mem

Benchmark: Read

cl-mem

Benchmark: Write

clpeak

OpenCL Test: Integer Compute INT

clpeak

OpenCL Test: Single-Precision Float

clpeak

OpenCL Test: Double-Precision Double

clpeak

OpenCL Test: Global Memory Bandwidth

MandelGPU

OpenCL Device: GPU

ViennaCL

OpenCL LU Factorization

RealSR-NCNN

Scale: 4x - TAA: No

RealSR-NCNN

Scale: 4x - TAA: Yes

VkFFT

VkResample

Upscale: 2x - Precision: Double

VkResample

Upscale: 2x - Precision: Single

Waifu2x-NCNN Vulkan

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

Phoronix Test Suite v10.8.5