AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (3801 BIOS) and NVIDIA GeForce RTX 3090 24GB on Ubuntu 21.10 via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2109061-TJ-RTX30COMP69
RTX 30 Compute September
AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (3801 BIOS) and NVIDIA GeForce RTX 3090 24GB on Ubuntu 21.10 via the Phoronix Test Suite.
RTX 3070:
Processor: AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (3801 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 1000GB Western Digital WDS100T1X0E-00AFY0 + 1000GB Western Digital WD_BLACK SN850 1TB + 2000GB, Graphics: NVIDIA GeForce RTX 3070 8GB, Audio: NVIDIA GA104 HD Audio, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel I211
OS: Ubuntu 21.10, Kernel: 5.13.0-14-generic (x86_64), Desktop: GNOME Shell 40.2, Display Server: X Server 1.20.11, Display Driver: NVIDIA 470.63.01, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 11.4.112, Vulkan: 1.2.175, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160
RTX 3090:
Processor: AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (3801 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 1000GB Western Digital WDS100T1X0E-00AFY0 + 1000GB Western Digital WD_BLACK SN850 1TB + 2000GB, Graphics: NVIDIA GeForce RTX 3090 24GB, Audio: NVIDIA GA102 HD Audio, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel I211
OS: Ubuntu 21.10, Kernel: 5.13.0-14-generic (x86_64), Desktop: GNOME Shell 40.2, Display Server: X Server 1.20.11, Display Driver: NVIDIA 470.63.01, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 11.4.112, Vulkan: 1.2.175, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160
GeForce RTX 3090:
Processor: AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (3801 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 1000GB Western Digital WDS100T1X0E-00AFY0 + 1000GB Western Digital WD_BLACK SN850 1TB + 2000GB, Graphics: NVIDIA GeForce RTX 3090 24GB, Audio: NVIDIA GA102 HD Audio, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel I211
OS: Ubuntu 21.10, Kernel: 5.13.0-14-generic (x86_64), Desktop: GNOME Shell 40.2, Display Server: X Server 1.20.11, Display Driver: NVIDIA 470.63.01, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 11.4.112, Vulkan: 1.2.175, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: GEMM SGEMM_N
GFLOPS > Higher Is Better
RTX 3070 ......... 3732.87 |=======================
RTX 3090 ......... 8174.54 |===================================================
GeForce RTX 3090 . 8105.73 |===================================================
cl-mem 2017-01-13
Benchmark: Read
GB/s > Higher Is Better
RTX 3070 . 393.3 |=============================
RTX 3090 . 826.6 |=============================================================
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: FFT SP
GFLOPS > Higher Is Better
RTX 3070 ......... 1134.82 |=========================
RTX 3090 ......... 2349.19 |===================================================
GeForce RTX 3090 . 2348.03 |===================================================
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: S3D
GFLOPS > Higher Is Better
RTX 3070 ......... 219.20 |===========================
RTX 3090 ......... 427.94 |====================================================
GeForce RTX 3090 . 427.08 |====================================================
cl-mem 2017-01-13
Benchmark: Write
GB/s > Higher Is Better
RTX 3070 . 382.0 |===============================
RTX 3090 . 745.0 |=============================================================
VkResample 1.0
Upscale: 2x - Precision: Single
ms < Lower Is Better
RTX 3070 . 17.529 |============================================================
RTX 3090 . 9.254 |================================
vkpeak 20210424
fp32-scalar
GFLOPS > Higher Is Better
GeForce RTX 3090 . 20871.36 |=================================================
RTX 3070 ......... 11589.59 |===========================
RTX 3090 ......... 21074.89 |==================================================
GeForce RTX 3090 . 21146.01 |==================================================
ViennaCL 1.7.1
Test: OpenCL BLAS - dAXPY
GB/s > Higher Is Better
RTX 3070 . 396 |===================================
RTX 3090 . 721 |===============================================================
vkpeak 20210424
fp16-vec4
GFLOPS > Higher Is Better
GeForce RTX 3090 . 40423.41 |=================================================
RTX 3070 ......... 22503.16 |============================
RTX 3090 ......... 40868.96 |==================================================
GeForce RTX 3090 . 40869.19 |==================================================
vkpeak 20210424
fp32-vec4
GFLOPS > Higher Is Better
GeForce RTX 3090 . 27390.08 |=================================================
RTX 3070 ......... 15319.31 |============================
RTX 3090 ......... 27597.60 |==================================================
GeForce RTX 3090 . 27807.75 |==================================================
vkpeak 20210424
fp64-scalar
GFLOPS > Higher Is Better
GeForce RTX 3090 . 651.31 |===================================================
RTX 3070 ......... 363.75 |=============================
RTX 3090 ......... 655.42 |====================================================
GeForce RTX 3090 . 658.56 |====================================================
vkpeak 20210424
int32-scalar
GIOPS > Higher Is Better
GeForce RTX 3090 . 20687.55 |=================================================
RTX 3070 ......... 11571.35 |============================
RTX 3090 ......... 20836.32 |==================================================
GeForce RTX 3090 . 20925.87 |==================================================
vkpeak 20210424
fp16-scalar
GFLOPS > Higher Is Better
GeForce RTX 3090 . 20714.14 |=================================================
RTX 3070 ......... 11590.48 |============================
RTX 3090 ......... 20954.00 |==================================================
GeForce RTX 3090 . 20953.65 |==================================================
VkResample 1.0
Upscale: 2x - Precision: Double
ms < Lower Is Better
RTX 3070 . 216.41 |============================================================
RTX 3090 . 119.82 |=================================
vkpeak 20210424
int32-vec4
GIOPS > Higher Is Better
GeForce RTX 3090 . 20597.19 |==================================================
RTX 3070 ......... 11521.48 |============================
RTX 3090 ......... 20743.80 |==================================================
GeForce RTX 3090 . 20746.67 |==================================================
vkpeak 20210424
int16-scalar
GIOPS > Higher Is Better
GeForce RTX 3090 . 13655.97 |==================================================
RTX 3070 ......... 7644.32 |============================
RTX 3090 ......... 13646.56 |==================================================
GeForce RTX 3090 . 13752.19 |==================================================
vkpeak 20210424
fp64-vec4
GFLOPS > Higher Is Better
RTX 3070 ......... 364.29 |=============================
RTX 3090 ......... 655.31 |====================================================
GeForce RTX 3090 . 651.20 |====================================================
ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMM-TN
GFLOPs/s > Higher Is Better
RTX 3070 . 337 |====================================
RTX 3090 . 598 |===============================================================
ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMM-NT
GFLOPs/s > Higher Is Better
RTX 3070 . 342 |====================================
RTX 3090 . 602 |===============================================================
ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMM-NN
GFLOPs/s > Higher Is Better
RTX 3070 . 341 |====================================
RTX 3090 . 600 |===============================================================
Blender 2.92
Blend File: Fishy Cat - Compute: NVIDIA OptiX
Seconds < Lower Is Better
RTX 3070 . 33.34 |=============================================================
RTX 3090 . 18.97 |===================================
LuxCoreRender 2.5
Scene: LuxCore Benchmark - Acceleration: GPU
M samples/sec > Higher Is Better
RTX 3070 . 6.22 |===================================
RTX 3090 . 10.92 |=============================================================
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: MD5 Hash
GHash/s > Higher Is Better
RTX 3070 ......... 25.41 |==============================
RTX 3090 ......... 44.43 |=====================================================
GeForce RTX 3090 . 44.19 |=====================================================
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: Max SP Flops
GFLOPS > Higher Is Better
RTX 3070 . 23102.9 |==================================
RTX 3090 . 40344.1 |===========================================================
Hashcat 6.1.1
Benchmark: SHA1
H/s > Higher Is Better
RTX 3070 ......... 13194833333 |===========================
RTX 3090 ......... 22853700000 |===============================================
GeForce RTX 3090 . 22798500000 |===============================================
LuxCoreRender 2.5
Scene: DLSC - Acceleration: GPU
M samples/sec > Higher Is Better
RTX 3070 . 6.68 |===================================
RTX 3090 . 11.53 |=============================================================
Hashcat 6.1.1
Benchmark: SHA-512
H/s > Higher Is Better
RTX 3070 ......... 1673633333 |============================
RTX 3090 ......... 2884600000 |================================================
GeForce RTX 3090 . 2881300000 |================================================
Hashcat 6.1.1
Benchmark: MD5
H/s > Higher Is Better
RTX 3070 ......... 38925166667 |===========================
RTX 3090 ......... 66739300000 |===============================================
GeForce RTX 3090 . 66589100000 |===============================================
Hashcat 6.1.1
Benchmark: TrueCrypt RIPEMD160 + XTS
H/s > Higher Is Better
RTX 3070 ......... 503167 |==============================
RTX 3090 ......... 857500 |====================================================
GeForce RTX 3090 . 859400 |====================================================
LuxCoreRender 2.5
Scene: Danish Mood - Acceleration: GPU
M samples/sec > Higher Is Better
RTX 3070 . 5.06 |====================================
RTX 3090 . 8.64 |==============================================================
Hashcat 6.1.1
Benchmark: 7-Zip
H/s > Higher Is Better
RTX 3070 ......... 689667 |==============================
RTX 3090 ......... 1154600 |===================================================
GeForce RTX 3090 . 1161100 |===================================================
vkpeak 20210424
int16-vec4
GIOPS > Higher Is Better
GeForce RTX 3090 . 16887.92 |==================================================
RTX 3070 ......... 10106.83 |==============================
RTX 3090 ......... 16889.26 |==================================================
GeForce RTX 3090 . 17006.61 |==================================================
ViennaCL 1.7.1
Test: OpenCL BLAS - dDOT
GB/s > Higher Is Better
RTX 3070 . 395 |======================================
RTX 3090 . 663 |===============================================================
Blender 2.92
Blend File: BMW27 - Compute: NVIDIA OptiX
Seconds < Lower Is Better
RTX 3070 . 15.82 |=============================================================
RTX 3090 . 9.52 |=====================================
RealSR-NCNN 20200818
Scale: 4x - TAA: Yes
Seconds < Lower Is Better
RTX 3070 ......... 46.79 |=====================================================
RTX 3090 ......... 28.42 |================================
GeForce RTX 3090 . 28.42 |================================
OctaneBench 2020.1
Total Score
Score > Higher Is Better
RTX 3070 . 415.20 |====================================
RTX 3090 . 683.60 |============================================================
FinanceBench 2016-07-25
Benchmark: Black-Scholes OpenCL
ms < Lower Is Better
RTX 3070 . 10.457 |============================================================
RTX 3090 . 6.357 |====================================
ViennaCL 1.7.1
Test: OpenCL BLAS - dCOPY
GB/s > Higher Is Better
RTX 3070 . 370 |======================================
RTX 3090 . 607 |===============================================================
RedShift Demo 3.0
Seconds < Lower Is Better
RTX 3070 . 226 |===============================================================
RTX 3090 . 139 |=======================================
IndigoBench 4.4
Acceleration: OpenCL GPU - Scene: Bedroom
M samples/s > Higher Is Better
RTX 3070 . 12.94 |======================================
RTX 3090 . 20.97 |=============================================================
Rodinia 3.1
Test: OpenCL Particle Filter
Seconds < Lower Is Better
RTX 3070 . 5.920 |=============================================================
RTX 3090 . 3.680 |======================================
Blender 2.92
Blend File: Classroom - Compute: NVIDIA OptiX
Seconds < Lower Is Better
RTX 3070 . 45.77 |=============================================================
RTX 3090 . 28.66 |======================================
Blender 2.92
Blend File: BMW27 - Compute: CUDA
Seconds < Lower Is Better
RTX 3070 . 29.00 |=============================================================
RTX 3090 . 18.32 |=======================================
Blender 2.92
Blend File: Fishy Cat - Compute: CUDA
Seconds < Lower Is Better
RTX 3070 . 54.69 |=============================================================
RTX 3090 . 34.65 |=======================================
LuxCoreRender 2.5
Scene: Rainbow Colors and Prism - Acceleration: GPU
M samples/sec > Higher Is Better
RTX 3070 . 21.27 |========================================
RTX 3090 . 32.09 |=============================================================
LuxCoreRender 2.5
Scene: Orange Juice - Acceleration: GPU
M samples/sec > Higher Is Better
RTX 3070 . 6.99 |=========================================
RTX 3090 . 10.42 |=============================================================
Blender 2.92
Blend File: Classroom - Compute: CUDA
Seconds < Lower Is Better
RTX 3070 . 76.06 |=============================================================
RTX 3090 . 51.18 |=========================================
IndigoBench 4.4
Acceleration: OpenCL GPU - Scene: Supercar
M samples/s > Higher Is Better
RTX 3070 . 37.32 |===========================================
RTX 3090 . 53.36 |=============================================================
LeelaChessZero 0.28
Backend: OpenCL
Nodes Per Second > Higher Is Better
RTX 3070 . 37790 |===========================================
RTX 3090 . 53653 |=============================================================
VkFFT 1.1.1
Benchmark Score > Higher Is Better
RTX 3070 ......... 31554 |======================================
RTX 3090 ......... 43295 |===================================================
GeForce RTX 3090 . 44567 |=====================================================
ArrayFire 3.7
Test: Conjugate Gradient OpenCL
ms < Lower Is Better
RTX 3070 . 2.089 |=============================================================
RTX 3090 . 1.480 |===========================================
RealSR-NCNN 20200818
Scale: 4x - TAA: No
Seconds < Lower Is Better
RTX 3070 ......... 7.856 |=====================================================
RTX 3090 ......... 5.566 |======================================
GeForce RTX 3090 . 5.595 |======================================
ViennaCL 1.7.1
Test: OpenCL BLAS - sAXPY
GB/s > Higher Is Better
RTX 3070 . 359 |=============================================
RTX 3090 . 501 |===============================================================
Blender 2.92
Blend File: Barbershop - Compute: CUDA
Seconds < Lower Is Better
RTX 3070 . 507.39 |============================================================
RTX 3090 . 373.32 |============================================
FAHBench 2.3.2
Ns Per Day > Higher Is Better
RTX 3070 . 265.51 |==============================================
RTX 3090 . 344.02 |============================================================
Waifu2x-NCNN Vulkan 20200818
Scale: 2x - Denoise: 3 - TAA: Yes
Seconds < Lower Is Better
RTX 3070 ......... 4.112 |=====================================================
RTX 3090 ......... 3.276 |==========================================
GeForce RTX 3090 . 3.275 |==========================================
ViennaCL 1.7.1
Test: OpenCL BLAS - sCOPY
GB/s > Higher Is Better
RTX 3070 . 291 |==================================================
RTX 3090 . 364 |===============================================================
NCNN 20210720
Target: Vulkan GPU - Model: vgg16
ms < Lower Is Better
RTX 3090 . 4.44 |===================================================
RTX 3070 . 5.43 |==============================================================
RTX 3090 . 4.35 |==================================================
cl-mem 2017-01-13
Benchmark: Copy
GB/s > Higher Is Better
RTX 3070 . 296.8 |==================================================
RTX 3090 . 364.3 |=============================================================
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: Reduction
GB/s > Higher Is Better
RTX 3070 ......... 326.57 |===========================================
RTX 3090 ......... 391.21 |====================================================
GeForce RTX 3090 . 391.09 |====================================================
NCNN 20210720
Target: Vulkan GPU - Model: yolov4-tiny
ms < Lower Is Better
RTX 3090 . 7.66 |==============================================================
RTX 3070 . 6.57 |=====================================================
RTX 3090 . 7.20 |==========================================================
ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMV-T
GB/s > Higher Is Better
RTX 3070 . 333 |========================================================
RTX 3090 . 376 |===============================================================
ViennaCL 1.7.1
Test: CPU BLAS - dDOT
GB/s > Higher Is Better
RTX 3070 . 48.6 |==============================================================
RTX 3090 . 43.2 |=======================================================
ViennaCL 1.7.1
Test: CPU BLAS - sAXPY
GB/s > Higher Is Better
RTX 3070 . 100.8 |=============================================================
RTX 3090 . 89.6 |======================================================
NCNN 20210720
Target: Vulkan GPU - Model: blazeface
ms < Lower Is Better
RTX 3090 . 0.96 |========================================================
RTX 3070 . 0.98 |=========================================================
RTX 3090 . 1.07 |==============================================================
ViennaCL 1.7.1
Test: CPU BLAS - sDOT
GB/s > Higher Is Better
RTX 3070 . 151 |===============================================================
RTX 3090 . 136 |=========================================================
NCNN 20210720
Target: Vulkan GPU - Model: resnet18
ms < Lower Is Better
RTX 3090 . 1.81 |==============================================================
RTX 3070 . 1.68 |==========================================================
RTX 3090 . 1.66 |=========================================================
ViennaCL 1.7.1
Test: OpenCL BLAS - sDOT
GB/s > Higher Is Better
RTX 3070 . 322 |==========================================================
RTX 3090 . 351 |===============================================================
ViennaCL 1.7.1
Test: CPU BLAS - dCOPY
GB/s > Higher Is Better
RTX 3070 . 24.6 |==============================================================
RTX 3090 . 22.6 |=========================================================
NCNN 20210720
Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3
ms < Lower Is Better
RTX 3090 . 2.21 |===========================================================
RTX 3070 . 2.14 |=========================================================
RTX 3090 . 2.31 |==============================================================
ViennaCL 1.7.1
Test: CPU BLAS - dGEMV-T
GB/s > Higher Is Better
RTX 3070 . 89.8 |==============================================================
RTX 3090 . 83.8 |==========================================================
NCNN 20210720
Target: Vulkan GPU - Model: googlenet
ms < Lower Is Better
RTX 3090 . 4.38 |==============================================================
RTX 3070 . 4.13 |==========================================================
RTX 3090 . 4.41 |==============================================================
ViennaCL 1.7.1
Test: CPU BLAS - dGEMV-N
GB/s > Higher Is Better
RTX 3070 . 82.8 |==============================================================
RTX 3090 . 77.6 |==========================================================
NCNN 20210720
Target: Vulkan GPU - Model: mnasnet
ms < Lower Is Better
RTX 3090 . 2.09 |=============================================================
RTX 3070 . 2.01 |==========================================================
RTX 3090 . 2.14 |==============================================================
ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMV-N
GB/s > Higher Is Better
RTX 3070 . 222 |===========================================================
RTX 3090 . 236 |===============================================================
NCNN 20210720
Target: Vulkan GPU - Model: regnety_400m
ms < Lower Is Better
RTX 3090 . 2.57 |==============================================================
RTX 3070 . 2.42 |==========================================================
RTX 3090 . 2.52 |=============================================================
ViennaCL 1.7.1
Test: CPU BLAS - sCOPY
GB/s > Higher Is Better
RTX 3070 . 66.9 |==============================================================
RTX 3090 . 63.0 |==========================================================
NCNN 20210720
Target: Vulkan GPU - Model: efficientnet-b0
ms < Lower Is Better
RTX 3090 . 3.35 |==============================================================
RTX 3070 . 3.17 |===========================================================
RTX 3090 . 3.34 |==============================================================
ViennaCL 1.7.1
Test: CPU BLAS - dAXPY
GB/s > Higher Is Better
RTX 3070 . 36.5 |==============================================================
RTX 3090 . 34.8 |===========================================================
NAMD CUDA 2.14
ATPase Simulation - 327,506 Atoms
days/ns < Lower Is Better
RTX 3070 . 0.12555 |========================================================
RTX 3090 . 0.13155 |===========================================================
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: Texture Read Bandwidth
GB/s > Higher Is Better
RTX 3070 . 2130.04 |========================================================
RTX 3090 . 2228.01 |===========================================================
ViennaCL 1.7.1
Test: CPU BLAS - dGEMM-TT
GFLOPs/s > Higher Is Better
RTX 3070 . 56.4 |==============================================================
RTX 3090 . 54.0 |===========================================================
NCNN 20210720
Target: Vulkan GPU - Model: mobilenet
ms < Lower Is Better
RTX 3090 . 4.48 |==============================================================
RTX 3070 . 4.31 |============================================================
RTX 3090 . 4.47 |==============================================================
ViennaCL 1.7.1
Test: CPU BLAS - dGEMM-TN
GFLOPs/s > Higher Is Better
RTX 3070 . 58.2 |==============================================================
RTX 3090 . 56.1 |============================================================
NCNN 20210720
Target: Vulkan GPU - Model: alexnet
ms < Lower Is Better
RTX 3090 . 1.88 |============================================================
RTX 3070 . 1.90 |============================================================
RTX 3090 . 1.95 |==============================================================
NCNN 20210720
Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2
ms < Lower Is Better
RTX 3070 . 1.91 |============================================================
RTX 3090 . 1.98 |==============================================================
NCNN 20210720
Target: Vulkan GPU - Model: shufflenet-v2
ms < Lower Is Better
RTX 3070 . 1.71 |============================================================
RTX 3090 . 1.77 |==============================================================
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: Triad
GB/s > Higher Is Better
RTX 3070 ......... 24.67 |===================================================
RTX 3090 ......... 25.53 |=====================================================
GeForce RTX 3090 . 25.45 |=====================================================
ViennaCL 1.7.1
Test: CPU BLAS - dGEMM-NT
GFLOPs/s > Higher Is Better
RTX 3070 . 53.6 |==============================================================
RTX 3090 . 52.0 |============================================================
NCNN 20210720
Target: Vulkan GPU - Model: resnet50
ms < Lower Is Better
RTX 3090 . 3.74 |==============================================================
RTX 3070 . 3.74 |==============================================================
RTX 3090 . 3.68 |=============================================================
NCNN 20210720
Target: Vulkan GPU - Model: squeezenet_ssd
ms < Lower Is Better
RTX 3070 . 4.43 |==============================================================
RTX 3090 . 4.37 |=============================================================
ViennaCL 1.7.1
Test: CPU BLAS - dGEMM-NN
GFLOPs/s > Higher Is Better
RTX 3070 . 53.1 |==============================================================
RTX 3090 . 52.4 |=============================================================
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: Bus Speed Readback
GB/s > Higher Is Better
RTX 3070 . 27.11 |=============================================================
RTX 3090 . 27.08 |=============================================================
SHOC Scalable HeterOgeneous Computing 2020-04-17
Target: OpenCL - Benchmark: Bus Speed Download
GB/s > Higher Is Better
RTX 3070 . 26.28 |=============================================================
RTX 3090 . 26.26 |=============================================================
ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMM-TT
GFLOPs/s > Higher Is Better
RTX 3090 . 601 |===============================================================
PlaidML
FP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCL
Examples Per Second > Higher Is Better
PlaidML
FP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCL
Examples Per Second > Higher Is Better
PlaidML
FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL
Examples Per Second > Higher Is Better
PlaidML
FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL
Examples Per Second > Higher Is Better
PlaidML
FP16: No - Mode: Training - Network: Mobilenet - Device: OpenCL
Examples Per Second > Higher Is Better
Caffe 2020-02-13
Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 1000
Milli-Seconds < Lower Is Better
Caffe 2020-02-13
Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 200
Milli-Seconds < Lower Is Better
Caffe 2020-02-13
Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 100
Milli-Seconds < Lower Is Better
Caffe 2020-02-13
Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 1000
Milli-Seconds < Lower Is Better
Caffe 2020-02-13
Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 200
Milli-Seconds < Lower Is Better
Caffe 2020-02-13
Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 100
Milli-Seconds < Lower Is Better
GROMACS 2021.2
Implementation: NVIDIA CUDA GPU - Input: water_GMX50_bare
Ns Per Day > Higher Is Better
Betsy GPU Compressor 1.1 Beta
Codec: ETC2 RGB - Quality: Highest
Seconds < Lower Is Better
Betsy GPU Compressor 1.1 Beta
Codec: ETC1 - Quality: Highest
Seconds < Lower Is Better
Libplacebo 2.72.2
FPS > Higher Is Better
Mixbench 2020-06-23
Backend: NVIDIA CUDA - Benchmark: Single Precision
Mixbench 2020-06-23
Backend: NVIDIA CUDA - Benchmark: Double Precision
Mixbench 2020-06-23
Backend: NVIDIA CUDA - Benchmark: Half Precision
Mixbench 2020-06-23
Backend: OpenCL - Benchmark: Single Precision
Mixbench 2020-06-23
Backend: OpenCL - Benchmark: Double Precision
Mixbench 2020-06-23
Backend: NVIDIA CUDA - Benchmark: Integer
Mixbench 2020-06-23
Backend: OpenCL - Benchmark: Integer
Waifu2x-NCNN Vulkan 20200818
Scale: 2x - Denoise: 3 - TAA: No
Seconds < Lower Is Better