AMD EPYC 7601 32-Core testing with a ASRockRack EPYCD8 (P2.10 BIOS) and llvmpipe 126GB on Ubuntu 19.10 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 1911029-HU-EPYC7601E78
EPYC 7601 Extra,
"SVT-AV1 0.7 - Encoder Mode: Enc Mode 0 - Input: 1080p",
Higher Results Are Better
"EPYC 7601",0.042,0.042,0.042,0.042,0.042,0.042
"MKL-DNN DNNL 1.1 - Harness: Convolution Batch conv_all - Data Type: u8s8f32",
Lower Results Are Better
"EPYC 7601",22368,22574.1,22465.1
"MKL-DNN DNNL 1.1 - Harness: Convolution Batch conv_all - Data Type: f32",
Lower Results Are Better
"EPYC 7601",2567.69,2569.68,2561.94
"FFTW 3.3.6 - Build: Float + SSE - Size: 2D FFT Size 4096",
Higher Results Are Better
"EPYC 7601",13384,13516,13368
"MKL-DNN DNNL 1.1 - Harness: Deconvolution Batch deconv_all - Data Type: f32",
Lower Results Are Better
"EPYC 7601",3145.11,3082.94,3159.79
"dav1d 0.5.0 - Video Input: Chimera 1080p 10-bit",
Higher Results Are Better
"EPYC 7601",61.26690073193,61.779748398902,61.599085086917
"FFTW 3.3.6 - Build: Stock - Size: 2D FFT Size 4096",
Higher Results Are Better
"EPYC 7601",4591.2,4603.7,4590.7
"OSPray 1.8.5 - Demo: San Miguel - Renderer: SciVis",
Higher Results Are Better
"EPYC 7601",24.390243902439,24.390243902439,24.390243902439,24.390243902439,24.390243902439,24.390243902439,24.390243902439,24.390243902439,24.390243902439,24.390243902439,24.390243902439,24.390243902439
"OSPray 1.8.5 - Demo: San Miguel - Renderer: Path Tracer",
Higher Results Are Better
"EPYC 7601",1.9569471624266,1.953125,1.9493177387914
"MKL-DNN DNNL 1.1 - Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f32",
Lower Results Are Better
"EPYC 7601",1388.12,1375.7,1385.83
"MKL-DNN DNNL 1.1 - Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32",
Lower Results Are Better
"EPYC 7601",146.013,150.328,146.368
"FFTW 3.3.6 - Build: Float + SSE - Size: 2D FFT Size 2048",
Higher Results Are Better
"EPYC 7601",14869,14828,14930
"OSPray 1.8.5 - Demo: XFrog Forest - Renderer: Path Tracer",
Higher Results Are Better
"EPYC 7601",2.2883295194508,2.2779043280182,2.2727272727273
"MKL-DNN DNNL 1.1 - Harness: IP Batch 1D - Data Type: f32",
Lower Results Are Better
"EPYC 7601",5.20081,5.07518,4.53012,5.15918,5.07227,5.08083,4.57407,4.33914,4.59512,5.00944,5.00112,5.15633,5.03043,5.13427,4.37959
"FFTW 3.3.6 - Build: Stock - Size: 2D FFT Size 2048",
Higher Results Are Better
"EPYC 7601",5092.9,5087.9,5014.9
"OSPray 1.8.5 - Demo: XFrog Forest - Renderer: SciVis",
Higher Results Are Better
"EPYC 7601",4.3478260869565,4.3290043290043,4.3290043290043
"MKL-DNN DNNL 1.1 - Harness: Convolution Batch conv_3d - Data Type: u8s8f32",
Lower Results Are Better
"EPYC 7601",4892.62,4905.26,4900.51
"Embree 3.6.1 - Binary: Pathtracer ISPC - Model: Asian Dragon Obj",
Higher Results Are Better
"EPYC 7601",20.2364,20.0801,20.0681
"MKL-DNN DNNL 1.1 - Harness: IP Batch All - Data Type: f32",
Lower Results Are Better
"EPYC 7601",29.958,28.6286,29.897
"MKL-DNN DNNL 1.1 - Harness: IP Batch All - Data Type: u8s8f32",
Lower Results Are Better
"EPYC 7601",271.987,271.962,273.178
"Embree 3.6.1 - Binary: Pathtracer - Model: Asian Dragon Obj",
Higher Results Are Better
"EPYC 7601",21.579,21.5945,21.6009
"MKL-DNN DNNL 1.1 - Harness: Deconvolution Batch deconv_1d - Data Type: f32",
Lower Results Are Better
"EPYC 7601",6.45668,6.6438,6.18505,6.71389,6.61853,6.51453
"OSPray 1.8.5 - Demo: NASA Streamlines - Renderer: SciVis",
Higher Results Are Better
"EPYC 7601",32.258064516129,32.258064516129,32.258064516129,32.258064516129,32.258064516129,32.258064516129,32.258064516129,32.258064516129,32.258064516129,32.258064516129,32.258064516129,32.258064516129
"MKL-DNN DNNL 1.1 - Harness: Convolution Batch conv_3d - Data Type: f32",
Lower Results Are Better
"EPYC 7601",18.6478,18.5528,18.8124
"MKL-DNN DNNL 1.1 - Harness: Recurrent Neural Network Training - Data Type: f32",
Lower Results Are Better
"EPYC 7601",416.074,410.458,413.175
"OSPray 1.8.5 - Demo: NASA Streamlines - Renderer: Path Tracer",
Higher Results Are Better
"EPYC 7601",6.4102564102564,6.3694267515924,6.3291139240506
"dav1d 0.5.0 - Video Input: Chimera 1080p",
Higher Results Are Better
"EPYC 7601",313.30420059469,315.87910224154,315.75109903934
"Embree 3.6.1 - Binary: Pathtracer ISPC - Model: Crown",
Higher Results Are Better
"EPYC 7601",19.5858,19.5721,19.4743
"Embree 3.6.1 - Binary: Pathtracer - Model: Crown",
Higher Results Are Better
"EPYC 7601",21.0125,20.9501,20.8849
"FFTW 3.3.6 - Build: Float + SSE - Size: 2D FFT Size 1024",
Higher Results Are Better
"EPYC 7601",20946,20795,20925
"Embree 3.6.1 - Binary: Pathtracer ISPC - Model: Asian Dragon",
Higher Results Are Better
"EPYC 7601",22.8521,23.3321,23.1605
"Embree 3.6.1 - Binary: Pathtracer - Model: Asian Dragon",
Higher Results Are Better
"EPYC 7601",24.1937,24.0934,24.0881
"MKL-DNN DNNL 1.1 - Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32",
Lower Results Are Better
"EPYC 7601",1868.75,1867.6,1895.46
"MKL-DNN DNNL 1.1 - Harness: Convolution Batch conv_alexnet - Data Type: u8s8f32",
Lower Results Are Better
"EPYC 7601",2478.47,2490.3,2472.94
"dav1d 0.5.0 - Video Input: Summer Nature 4K",
Higher Results Are Better
"EPYC 7601",152.44872129288,151.34717039977,150.93026084491
"SVT-AV1 0.7 - Encoder Mode: Enc Mode 4 - Input: 1080p",
Higher Results Are Better
"EPYC 7601",4.028,4.155,4.085
"OSPray 1.8.5 - Demo: Magnetic Reconnection - Renderer: SciVis",
Higher Results Are Better
"EPYC 7601",18.181818181818,18.181818181818,18.181818181818,18.181818181818,17.857142857143
"MKL-DNN DNNL 1.1 - Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32",
Lower Results Are Better
"EPYC 7601",3129.92,3147.36,3137.05
"MKL-DNN DNNL 1.1 - Harness: Convolution Batch conv_alexnet - Data Type: f32",
Lower Results Are Better
"EPYC 7601",302.053,303.488,306.211
"FFTW 3.3.6 - Build: Stock - Size: 2D FFT Size 1024",
Higher Results Are Better
"EPYC 7601",5510.1,5532.2,5520.6
"MKL-DNN DNNL 1.1 - Harness: Deconvolution Batch deconv_3d - Data Type: f32",
Lower Results Are Better
"EPYC 7601",8.51352,8.30288,9.23032,8.34072,9.49003,8.31514,9.35739,8.66492,8.517,8.75101,8.54121,8.37366,8.51919,8.82271,9.45499
"MKL-DNN DNNL 1.1 - Harness: IP Batch 1D - Data Type: u8s8f32",
Lower Results Are Better
"EPYC 7601",53.9304,53.2962,53.6026
"Intel Open Image Denoise 1.0.0 - Scene: Memorial",
Higher Results Are Better
"EPYC 7601",7.2048704924529,7.1545599587897,7.1013648823304
"SVT-AV1 0.7 - Encoder Mode: Enc Mode 8 - Input: 1080p",
Higher Results Are Better
"EPYC 7601",39.91,40.12,40.542
"dav1d 0.5.0 - Video Input: Summer Nature 1080p",
Higher Results Are Better
"EPYC 7601",355.24393707483,354.91590986394,358.83301870748
"FFTW 3.3.6 - Build: Float + SSE - Size: 1D FFT Size 4096",
Higher Results Are Better
"EPYC 7601",27661,27803,27763
"FFTW 3.3.6 - Build: Float + SSE - Size: 2D FFT Size 512",
Higher Results Are Better
"EPYC 7601",23559,23543,23627
"FFTW 3.3.6 - Build: Stock - Size: 1D FFT Size 4096",
Higher Results Are Better
"EPYC 7601",6306.6,6327.2,6315.1
"FFTW 3.3.6 - Build: Float + SSE - Size: 1D FFT Size 2048",
Higher Results Are Better
"EPYC 7601",27464,28378,28243
"FFTW 3.3.6 - Build: Stock - Size: 1D FFT Size 2048",
Higher Results Are Better
"EPYC 7601",6399.4,6384.6,6381.7
"FFTW 3.3.6 - Build: Stock - Size: 2D FFT Size 512",
Higher Results Are Better
"EPYC 7601",6088.5,6124.5,6129.2
"FFTW 3.3.6 - Build: Float + SSE - Size: 1D FFT Size 1024",
Higher Results Are Better
"EPYC 7601",27275,27823,28061
"FFTW 3.3.6 - Build: Float + SSE - Size: 2D FFT Size 256",
Higher Results Are Better
"EPYC 7601",23261,23323,23417
"FFTW 3.3.6 - Build: Stock - Size: 1D FFT Size 1024",
Higher Results Are Better
"EPYC 7601",6616,6602.3,6612.3
"FFTW 3.3.6 - Build: Float + SSE - Size: 1D FFT Size 512",
Higher Results Are Better
"EPYC 7601",25509,25804,25676
"FFTW 3.3.6 - Build: Float + SSE - Size: 2D FFT Size 64",
Higher Results Are Better
"EPYC 7601",26119,26183,26072
"FFTW 3.3.6 - Build: Stock - Size: 2D FFT Size 64",
Higher Results Are Better
"EPYC 7601",6447.8,6444.2,6465.4
"FFTW 3.3.6 - Build: Float + SSE - Size: 1D FFT Size 32",
Higher Results Are Better
"EPYC 7601",10632,10624,10652
"FFTW 3.3.6 - Build: Float + SSE - Size: 1D FFT Size 128",
Higher Results Are Better
"EPYC 7601",16418,16438,16346
"FFTW 3.3.6 - Build: Float + SSE - Size: 1D FFT Size 64",
Higher Results Are Better
"EPYC 7601",13431,13235,13358
"FFTW 3.3.6 - Build: Stock - Size: 2D FFT Size 256",
Higher Results Are Better
"EPYC 7601",6139.2,6134.3,6151.8
"FFTW 3.3.6 - Build: Float + SSE - Size: 1D FFT Size 256",
Higher Results Are Better
"EPYC 7601",21049,21169,20711
"FFTW 3.3.6 - Build: Stock - Size: 1D FFT Size 512",
Higher Results Are Better
"EPYC 7601",6478.5,6489.2,6476.2
"FFTW 3.3.6 - Build: Stock - Size: 1D FFT Size 32",
Higher Results Are Better
"EPYC 7601",6314.4,6283,6320.5
"FFTW 3.3.6 - Build: Float + SSE - Size: 2D FFT Size 32",
Higher Results Are Better
"EPYC 7601",26331,26290,26302
"FFTW 3.3.6 - Build: Float + SSE - Size: 2D FFT Size 128",
Higher Results Are Better
"EPYC 7601",25409,25551,25517
"FFTW 3.3.6 - Build: Stock - Size: 1D FFT Size 256",
Higher Results Are Better
"EPYC 7601",6428.5,6419.2,6413.8
"FFTW 3.3.6 - Build: Stock - Size: 2D FFT Size 128",
Higher Results Are Better
"EPYC 7601",6253.8,6381.5,6351.7
"OSPray 1.8.5 - Demo: Magnetic Reconnection - Renderer: Path Tracer",
Higher Results Are Better
"EPYC 7601",
"FFTW 3.3.6 - Build: Stock - Size: 2D FFT Size 32",
Higher Results Are Better
"EPYC 7601",8155.7,8160.1,8159.5
"FFTW 3.3.6 - Build: Stock - Size: 1D FFT Size 128",
Higher Results Are Better
"EPYC 7601",6008,6000.2,6012
"FFTW 3.3.6 - Build: Stock - Size: 1D FFT Size 64",
Higher Results Are Better
"EPYC 7601",5723.1,5727.3,5734.6