NVIDIA GPU Compute
Some benchmarks by Michael Larabel.
HTML result view exported from: https://openbenchmarking.org/result/2009116-FI-GPUCOMPUT55&grs&rdt.
Darktable
Test: Boat - Acceleration: OpenCL
Mixbench
Backend: NVIDIA CUDA - Benchmark: Half Precision
clpeak
OpenCL Test: Double-Precision Double
MandelGPU
OpenCL Device: GPU
Mixbench
Backend: OpenCL - Benchmark: Single Precision
clpeak
OpenCL Test: Single-Precision Float
PlaidML
FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL
cl-mem
Benchmark: Read
Mixbench
Backend: OpenCL - Benchmark: Double Precision
clpeak
OpenCL Test: Global Memory Bandwidth
Mixbench
Backend: NVIDIA CUDA - Benchmark: Single Precision
OctaneBench
Total Score
cl-mem
Benchmark: Write
PlaidML
FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL
LuxCoreRender OpenCL
Scene: LuxCore Benchmark
LuxCoreRender OpenCL
Scene: DLSC
PlaidML
FP16: No - Mode: Inference - Network: Inception V3 - Device: OpenCL
LuxCoreRender OpenCL
Scene: Food
PlaidML
FP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCL
Rodinia
Test: OpenCL Particle Filter
FAHBench
PlaidML
FP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCL
ArrayFire
Test: Conjugate Gradient OpenCL
NeatBench
Acceleration: GPU
ArrayFire
Test: BLAS OpenCL
Mixbench
Backend: NVIDIA CUDA - Benchmark: Double Precision
Darktable
Test: Server Room - Acceleration: OpenCL
LuxCoreRender OpenCL
Scene: Rainbow Colors and Prism
Meta Performance Per Watts
Performance Per Watts
cl-mem
Benchmark: Copy
PlaidML
FP16: No - Mode: Inference - Network: VGG19 - Device: OpenCL
GROMACS
Water Benchmark
PlaidML
FP16: No - Mode: Training - Network: Mobilenet - Device: OpenCL
NAMD CUDA
ATPase Simulation - 327,506 Atoms
Mixbench
Backend: OpenCL - Benchmark: Integer
PlaidML
FP16: No - Mode: Inference - Network: VGG16 - Device: OpenCL
FinanceBench
Benchmark: Monte-Carlo OpenCL
clpeak
OpenCL Test: Integer Compute INT
Mixbench
Backend: NVIDIA CUDA - Benchmark: Integer
LeelaChessZero
Backend: OpenCL
Darktable
Test: Server Rack - Acceleration: OpenCL
Darktable
Test: Masskrug - Acceleration: OpenCL
Rodinia
Test: OpenCL Myocyte
Rodinia
Test: NVIDIA CUDA GPU Myocyte
ViennaCL
OpenCL LU Factorization
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: NVIDIA CUDA - Kernel: Euclidean Cluster
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: OpenCL - Kernel: Euclidean Cluster
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: OpenCL - Kernel: NDT Mapping
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: OpenCL - Kernel: Points2Image
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: NVIDIA CUDA - Kernel: NDT Mapping
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: NVIDIA CUDA - Kernel: Points2Image
Darktable
GPU Power Consumption Monitor
Darktable
GPU Power Consumption Monitor
Darktable
GPU Power Consumption Monitor
Darktable
GPU Power Consumption Monitor
Darktable
GPU Temperature Monitor
Darktable
GPU Temperature Monitor
Darktable
GPU Temperature Monitor
Darktable
GPU Temperature Monitor
GPU Temperature Monitor
Phoronix Test Suite System Monitoring
GPU Power Consumption Monitor
Phoronix Test Suite System Monitoring
ViennaCL
GPU Temperature Monitor
ViennaCL
GPU Power Consumption Monitor
ViennaCL
OpenCL LU Factorization
MandelGPU
GPU Temperature Monitor
MandelGPU
GPU Power Consumption Monitor
MandelGPU
OpenCL Device: GPU
cl-mem
GPU Temperature Monitor
cl-mem
GPU Power Consumption Monitor
cl-mem
Benchmark: Copy
cl-mem
GPU Temperature Monitor
cl-mem
GPU Power Consumption Monitor
cl-mem
Benchmark: Write
cl-mem
GPU Temperature Monitor
cl-mem
GPU Power Consumption Monitor
cl-mem
Benchmark: Read
Darktable
GPU Temperature Monitor
Darktable
GPU Power Consumption Monitor
Darktable
GPU Temperature Monitor
Darktable
GPU Power Consumption Monitor
Darktable
GPU Temperature Monitor
Darktable
GPU Power Consumption Monitor
Darktable
GPU Temperature Monitor
Darktable
GPU Power Consumption Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Temperature Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Power Consumption Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: NVIDIA CUDA - Kernel: Points2Image
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Temperature Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Power Consumption Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: NVIDIA CUDA - Kernel: NDT Mapping
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Temperature Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Power Consumption Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: NVIDIA CUDA - Kernel: Euclidean Cluster
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Temperature Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Power Consumption Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: OpenCL - Kernel: Points2Image
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Temperature Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Power Consumption Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: OpenCL - Kernel: NDT Mapping
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Temperature Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
GPU Power Consumption Monitor
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: OpenCL - Kernel: Euclidean Cluster
LeelaChessZero
GPU Temperature Monitor
LeelaChessZero
GPU Power Consumption Monitor
LeelaChessZero
Backend: OpenCL
PlaidML
GPU Temperature Monitor
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Inference - Network: VGG19 - Device: OpenCL
PlaidML
GPU Temperature Monitor
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Inference - Network: VGG16 - Device: OpenCL
PlaidML
GPU Temperature Monitor
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL
PlaidML
GPU Temperature Monitor
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCL
PlaidML
GPU Temperature Monitor
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Inference - Network: Inception V3 - Device: OpenCL
PlaidML
GPU Temperature Monitor
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL
PlaidML
GPU Temperature Monitor
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: No - Mode: Training - Network: Mobilenet - Device: OpenCL
PlaidML
GPU Temperature Monitor
PlaidML
GPU Power Consumption Monitor
PlaidML
FP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCL
FinanceBench
GPU Temperature Monitor
FinanceBench
GPU Power Consumption Monitor
FinanceBench
GPU Temperature Monitor
FinanceBench
GPU Power Consumption Monitor
FinanceBench
Benchmark: Black-Scholes OpenCL
NeatBench
GPU Temperature Monitor
NeatBench
GPU Power Consumption Monitor
NeatBench
Acceleration: GPU
clpeak
GPU Temperature Monitor
clpeak
GPU Power Consumption Monitor
clpeak
OpenCL Test: Integer Compute INT
clpeak
GPU Temperature Monitor
clpeak
GPU Power Consumption Monitor
clpeak
OpenCL Test: Double-Precision Double
clpeak
GPU Temperature Monitor
clpeak
GPU Power Consumption Monitor
clpeak
OpenCL Test: Single-Precision Float
clpeak
GPU Temperature Monitor
clpeak
GPU Power Consumption Monitor
clpeak
OpenCL Test: Global Memory Bandwidth
ArrayFire
GPU Temperature Monitor
ArrayFire
GPU Power Consumption Monitor
ArrayFire
GPU Temperature Monitor
ArrayFire
GPU Power Consumption Monitor
ArrayFire
Test: BLAS OpenCL
Rodinia
GPU Temperature Monitor
Rodinia
GPU Power Consumption Monitor
Rodinia
GPU Temperature Monitor
Rodinia
GPU Power Consumption Monitor
Rodinia
GPU Temperature Monitor
Rodinia
GPU Power Consumption Monitor
LuxCoreRender OpenCL
GPU Temperature Monitor
LuxCoreRender OpenCL
GPU Power Consumption Monitor
LuxCoreRender OpenCL
Scene: Food
LuxCoreRender OpenCL
GPU Temperature Monitor
LuxCoreRender OpenCL
GPU Power Consumption Monitor
LuxCoreRender OpenCL
Scene: Rainbow Colors and Prism
LuxCoreRender OpenCL
GPU Temperature Monitor
LuxCoreRender OpenCL
GPU Power Consumption Monitor
LuxCoreRender OpenCL
Scene: DLSC
LuxCoreRender OpenCL
GPU Temperature Monitor
LuxCoreRender OpenCL
GPU Power Consumption Monitor
LuxCoreRender OpenCL
Scene: LuxCore Benchmark
OctaneBench
GPU Temperature Monitor
OctaneBench
GPU Power Consumption Monitor
OctaneBench
Total Score
Mixbench
GPU Temperature Monitor
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: NVIDIA CUDA - Benchmark: Integer
Mixbench
GPU Temperature Monitor
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: NVIDIA CUDA - Benchmark: Half Precision
Mixbench
GPU Temperature Monitor
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: NVIDIA CUDA - Benchmark: Double Precision
Mixbench
GPU Temperature Monitor
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: NVIDIA CUDA - Benchmark: Single Precision
Mixbench
GPU Temperature Monitor
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: OpenCL - Benchmark: Integer
Mixbench
GPU Temperature Monitor
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: OpenCL - Benchmark: Double Precision
Mixbench
GPU Temperature Monitor
Mixbench
GPU Power Consumption Monitor
Mixbench
Backend: OpenCL - Benchmark: Single Precision
NAMD CUDA
GPU Temperature Monitor
NAMD CUDA
GPU Power Consumption Monitor
GROMACS
GPU Temperature Monitor
GROMACS
GPU Power Consumption Monitor
GROMACS
Water Benchmark
FAHBench
GPU Temperature Monitor
FAHBench
GPU Power Consumption Monitor
FAHBench
Phoronix Test Suite v10.8.5