NVIDIA GPU Compute Benchmarks
Benchmarks for a future article.
HTML result view exported from: https://openbenchmarking.org/result/2106089-IB-NVIDIACOM84&sgm=1&hgv=RTX+3070+Ti%2CRTX+3080+Ti&grw&sor.
Betsy GPU Compressor
Codec: ETC1 - Quality: Highest
Betsy GPU Compressor
Codec: ETC2 RGB - Quality: Highest
PlaidML
FP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCL
PlaidML
FP16: No - Mode: Inference - Network: VGG16 - Device: OpenCL
PlaidML
FP16: No - Mode: Inference - Network: VGG19 - Device: OpenCL
LeelaChessZero
Backend: OpenCL
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: Texture Read Bandwidth
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: FFT SP
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: GEMM SGEMM_N
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: MD5 Hash
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: S3D
ArrayFire
Test: BLAS OpenCL
ArrayFire
Test: Conjugate Gradient OpenCL
Chaos Group V-RAY
Mode: NVIDIA CUDA GPU
Chaos Group V-RAY
Mode: NVIDIA RTX GPU
Blender
Blend File: BMW27 - Compute: CUDA
Blender
Blend File: BMW27 - Compute: NVIDIA OptiX
Blender
Blend File: Classroom - Compute: CUDA
Blender
Blend File: Classroom - Compute: NVIDIA OptiX
Blender
Blend File: Fishy Cat - Compute: CUDA
Blender
Blend File: Fishy Cat - Compute: NVIDIA OptiX
Blender
Blend File: Pabellon Barcelona - Compute: CUDA
Blender
Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX
Blender
Blend File: Barbershop - Compute: CUDA
Blender
Blend File: Barbershop - Compute: NVIDIA OptiX
IndigoBench
Acceleration: OpenCL GPU - Scene: Supercar
IndigoBench
Acceleration: OpenCL GPU - Scene: Bedroom
LuxCoreRender
Scene: DLSC - Acceleration: GPU
LuxCoreRender
Scene: Rainbow Colors and Prism - Acceleration: GPU
LuxCoreRender
Scene: LuxCore Benchmark - Acceleration: GPU
LuxCoreRender
Scene: Orange Juice - Acceleration: GPU
LuxCoreRender
Scene: Danish Mood - Acceleration: GPU
FAHBench
Hashcat
Benchmark: MD5
Hashcat
Benchmark: SHA1
Hashcat
Benchmark: SHA-512
Hashcat
Benchmark: 7-Zip
Hashcat
Benchmark: TrueCrypt RIPEMD160 + XTS
Mixbench
Backend: NVIDIA CUDA - Benchmark: Single Precision
Mixbench
Backend: NVIDIA CUDA - Benchmark: Double Precision
Mixbench
Backend: NVIDIA CUDA - Benchmark: Half Precision
Mixbench
Backend: NVIDIA CUDA - Benchmark: Integer
NAMD CUDA
ATPase Simulation - 327,506 Atoms
OctaneBench
Total Score
RedShift Demo
cl-mem
Benchmark: Read
cl-mem
Benchmark: Write
cl-mem
Benchmark: Copy
clpeak
OpenCL Test: Global Memory Bandwidth
clpeak
OpenCL Test: Single-Precision Float
clpeak
OpenCL Test: Double-Precision Double
clpeak
OpenCL Test: Integer Compute INT
ViennaCL
Test: OpenCL BLAS - sCOPY
ViennaCL
Test: OpenCL BLAS - sAXPY
ViennaCL
Test: OpenCL BLAS - dCOPY
ViennaCL
Test: OpenCL BLAS - dAXPY
ViennaCL
Test: OpenCL BLAS - dDOT
ViennaCL
Test: OpenCL BLAS - dGEMM-NN
ViennaCL
Test: OpenCL BLAS - dGEMM-NT
ViennaCL
Test: OpenCL BLAS - dGEMM-TN
ViennaCL
Test: OpenCL BLAS - sDOT
ViennaCL
Test: OpenCL BLAS - dGEMV-N
ViennaCL
Test: OpenCL BLAS - dGEMV-T
ViennaCL
Test: OpenCL BLAS - dGEMM-TT
RealSR-NCNN
Scale: 4x - TAA: Yes
RealSR-NCNN
Scale: 4x - TAA: No
vkpeak
fp32-scalar
vkpeak
fp32-vec4
vkpeak
fp16-scalar
vkpeak
fp16-vec4
vkpeak
fp64-scalar
vkpeak
fp64-vec4
vkpeak
int32-scalar
vkpeak
int32-vec4
vkpeak
int16-scalar
vkpeak
int16-vec4
VkResample
Upscale: 2x - Precision: Single
Waifu2x-NCNN Vulkan
Scale: 2x - Denoise: 3 - TAA: Yes
Hashcat
Benchmark: MD5
Hashcat
GPU Power Consumption Monitor
Hashcat
Benchmark: SHA1
Hashcat
GPU Power Consumption Monitor
Hashcat
Benchmark: SHA-512
Hashcat
GPU Power Consumption Monitor
Hashcat
Benchmark: 7-Zip
Hashcat
GPU Power Consumption Monitor
Hashcat
Benchmark: TrueCrypt RIPEMD160 + XTS
Hashcat
GPU Power Consumption Monitor
FAHBench
FAHBench
GPU Power Consumption Monitor
NAMD CUDA
GPU Power Consumption Monitor
Mixbench
Backend: NVIDIA CUDA - Benchmark: Single Precision
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: NVIDIA CUDA - Benchmark: Double Precision
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: NVIDIA CUDA - Benchmark: Half Precision
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: NVIDIA CUDA - Benchmark: Integer
Mixbench
GPU Power Consumption Monitor
OctaneBench
Total Score
OctaneBench
GPU Power Consumption Monitor
LuxCoreRender
Scene: DLSC - Acceleration: GPU
LuxCoreRender
GPU Power Consumption Monitor
LuxCoreRender
Scene: Rainbow Colors and Prism - Acceleration: GPU
LuxCoreRender
GPU Power Consumption Monitor
LuxCoreRender
Scene: LuxCore Benchmark - Acceleration: GPU
LuxCoreRender
GPU Power Consumption Monitor
LuxCoreRender
Scene: Orange Juice - Acceleration: GPU
LuxCoreRender
GPU Power Consumption Monitor
LuxCoreRender
Scene: Danish Mood - Acceleration: GPU
LuxCoreRender
GPU Power Consumption Monitor
ArrayFire
Test: BLAS OpenCL
ArrayFire
GPU Power Consumption Monitor
ArrayFire
GPU Power Consumption Monitor
clpeak
OpenCL Test: Global Memory Bandwidth
clpeak
GPU Power Consumption Monitor
clpeak
OpenCL Test: Single-Precision Float
clpeak
GPU Power Consumption Monitor
clpeak
OpenCL Test: Double-Precision Double
clpeak
GPU Power Consumption Monitor
clpeak
OpenCL Test: Integer Compute INT
clpeak
GPU Power Consumption Monitor
vkpeak
int16-vec4
vkpeak
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCL
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Inference - Network: VGG16 - Device: OpenCL
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Inference - Network: VGG19 - Device: OpenCL
PlaidML
GPU Power Consumption Monitor
LeelaChessZero
Backend: OpenCL
LeelaChessZero
GPU Power Consumption Monitor
cl-mem
Benchmark: Read
cl-mem
GPU Power Consumption Monitor
cl-mem
Benchmark: Write
cl-mem
GPU Power Consumption Monitor
cl-mem
Benchmark: Copy
cl-mem
GPU Power Consumption Monitor
ViennaCL
Test: OpenCL BLAS - dGEMM-TN
ViennaCL
GPU Power Consumption Monitor
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: Texture Read Bandwidth
SHOC Scalable HeterOgeneous Computing
GPU Power Consumption Monitor
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: FFT SP
SHOC Scalable HeterOgeneous Computing
GPU Power Consumption Monitor
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: GEMM SGEMM_N
SHOC Scalable HeterOgeneous Computing
GPU Power Consumption Monitor
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: MD5 Hash
SHOC Scalable HeterOgeneous Computing
GPU Power Consumption Monitor
SHOC Scalable HeterOgeneous Computing
Target: OpenCL - Benchmark: S3D
SHOC Scalable HeterOgeneous Computing
GPU Power Consumption Monitor
IndigoBench
Acceleration: OpenCL GPU - Scene: Supercar
IndigoBench
GPU Power Consumption Monitor
IndigoBench
Acceleration: OpenCL GPU - Scene: Bedroom
IndigoBench
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Blender
GPU Power Consumption Monitor
Chaos Group V-RAY
Mode: NVIDIA CUDA GPU
Chaos Group V-RAY
GPU Power Consumption Monitor
Chaos Group V-RAY
Mode: NVIDIA RTX GPU
Chaos Group V-RAY
GPU Power Consumption Monitor
VkResample
GPU Power Consumption Monitor
RealSR-NCNN
GPU Power Consumption Monitor
RealSR-NCNN
GPU Power Consumption Monitor
Waifu2x-NCNN Vulkan
GPU Power Consumption Monitor
Betsy GPU Compressor
GPU Power Consumption Monitor
Betsy GPU Compressor
GPU Power Consumption Monitor
RedShift Demo
GPU Power Consumption Monitor
ViennaCL
Test: OpenCL BLAS - dGEMM-TT
ViennaCL
Test: OpenCL BLAS - dGEMM-NT
vkpeak
fp64-vec4
ViennaCL
Test: OpenCL BLAS - dDOT
Geometric Mean Of All Test Results
Result Composite - NVIDIA GPU Compute Benchmarks
Phoronix Test Suite v10.8.5