mnn ncnn zen 1 epyc

AMD EPYC 7551 32-Core testing with a GIGABYTE MZ31-AR0-00 v01010101 (F10 BIOS) and ASPEED on Debian 11 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 2208156-NE-MNNNCNNZE48
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

HPC - High Performance Computing 2 Tests
Machine Learning 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
A
August 15 2022
  19 Hours, 22 Minutes
B
August 15 2022
  21 Hours, 6 Minutes
C
August 15 2022
  21 Hours, 3 Minutes
Invert Hiding All Results Option
  20 Hours, 30 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

mnn ncnn zen 1 epyc, "Mobile Neural Network 2.0 - Model: mobilenetV3", Lower Results Are Better "A",5.487,5.061,5.397,5.228,5.232 "B",6.518,4.838,5.069,5.295,5.564,4.828,5.195,5.164,6.708 "C",5.05,7.015,6.664,5.031,6.766,6.398,6.973,5.551,5.183 "Mobile Neural Network 2.0 - Model: squeezenetv1.1", Lower Results Are Better "A",9.702,9.188,8.972,8.731,9.194 "B",9.021,8.919,7.971,9.351,9.027,8.231,8.932,8.662,10.966 "C",8.365,11.061,9.095,8.452,9.87,9.002,9.511,11.441,8.846 "Mobile Neural Network 2.0 - Model: resnet-v2-50", Lower Results Are Better "A",55.705,51.374,49.082,53.723,52.018 "B",48.955,48.753,53.191,52.002,51.686,53.404,51.095,52.469,75.81 "C",53.071,67.791,52.476,51.598,71.772,61.299,54.563,53.828,54.033 "Mobile Neural Network 2.0 - Model: SqueezeNetV1.0", Lower Results Are Better "A",15.277,13.964,13.989,13.829,13.39 "B",13.707,13.634,13.62,13.6,14.188,12.955,13.901,13.514,16.73 "C",14.174,16.407,13.786,13.197,15.395,15.48,13.61,14.264,14.112 "Mobile Neural Network 2.0 - Model: MobileNetV2_224", Lower Results Are Better "A",9.618,10.746,10.265,9.669,9.942 "B",8.6,8.464,8.457,8.596,8.945,8.253,8.753,8.707,8.582 "C",9.237,10.367,10.329,9.621,9.607,9.457,11.052,9.986,9.426 "Mobile Neural Network 2.0 - Model: mobilenet-v1-1.0", Lower Results Are Better "A",9.729,7.361,7.624,7.202,9.789 "B",7.334,7.715,7.745,7.797,7.951,7.677,7.345,7.603,11.903 "C",6.902,11.111,7.89,7.552,10.6,8.279,8.055,11.658,6.996 "Mobile Neural Network 2.0 - Model: inception-v3", Lower Results Are Better "A",67.662,56.381,59.029,59.199,55.673 "B",54.737,54.458,55.088,55.633,53.299,53.945,53.682,55.131,84.652 "C",57.423,81.805,55.265,52.575,64.9,57.532,53.529,58.481,58.089 "NCNN 20220729 - Target: CPU - Model: mobilenet", Lower Results Are Better "A",41.59,44.82,53.57,36.88,42.21,53.44,44.34,37.89,50.22 "B",41.08,46.87,38.73,40.68,41.36,46.37,43.48,37.77,37.35 "C",54.83,62.13,38.52,36.26,37.1,37.95,64.21,41.54,40.41 "NCNN 20220729 - Target: CPU-v2-v2 - Model: mobilenet-v2", Lower Results Are Better "A",22.27,23.22,20.41,23.77,22.63,24.94,20.83,31.17,27.09 "B",21.5,32.04,35.79,19.56,19.98,20.82,18.82,30.06,23.14 "C",30.66,29.88,23.65,39.21,20.98,24.45,32.22,21.48,20.95 "NCNN 20220729 - Target: CPU-v3-v3 - Model: mobilenet-v3", Lower Results Are Better "A",21.5,27.5,19.71,19.39,20.52,20,19.21,20.18,26.42 "B",29.78,23.72,23.46,18.83,36.15,21.1,19.04,18.42,21.05 "C",19.69,22.21,20.85,19.46,21.3,20.62,22.85,21.85,32.33 "NCNN 20220729 - Target: CPU - Model: shufflenet-v2", Lower Results Are Better "A",25.03,24.07,61.5,22.53,23.16,23.02,22.93,35.3,24.91 "B",22.2,26.47,23.85,22.33,22.61,32.53,30.84,34.99,23.65 "C",22.41,32.6,30.72,23.7,24.22,23.98,25.13,37.5,23.38 "NCNN 20220729 - Target: CPU - Model: mnasnet", Lower Results Are Better "A",20.8,20.44,21.14,19.58,19.59,22.43,19.25,35.12,23.02 "B",19.25,30.93,21.4,29.79,18.26,18.29,18.25,17.56,24.89 "C",19.1,29.1,26.13,34.83,19.07,19.91,24.83,28.02,35.61 "NCNN 20220729 - Target: CPU - Model: efficientnet-b0", Lower Results Are Better "A",50.24,31.27,31.58,29.79,36.33,31.97,29.03,29.06,29.48 "B",27.93,45.57,49.55,33.54,35.42,38.84,25.61,24.73,30.66 "C",28.34,42.75,27.98,31.88,27.76,27.8,35.34,27.67,36.65 "NCNN 20220729 - Target: CPU - Model: blazeface", Lower Results Are Better "A",10.79,12.61,12.89,10.61,10.83,12.16,17.02,13.49,10.08 "B",10.79,11.25,33.16,15.49,10.7,9.93,10.7,14.58,13.59 "C",10.78,11.6,11.68,11.81,11.26,11.66,11.17,11.69,10.78 "NCNN 20220729 - Target: CPU - Model: googlenet", Lower Results Are Better "A",60.74,49.65,75.83,50.95,46.03,79.61,48.71,53.06,70.08 "B",75.32,65.7,68.7,44.18,44.6,48.26,47.29,41.09,49.2 "C",64.16,45.92,51.03,50.33,41.6,62.55,71.19,48.39,43.97 "NCNN 20220729 - Target: CPU - Model: vgg16", Lower Results Are Better "A",108.43,105.03,91.07,88.24,120.13,77.87,76.46,87.98,95.06 "B",136.89,119.69,127.29,80.78,79.46,88.39,79.79,85.19,93.39 "C",77.64,91.24,110.46,79.61,78,78.68,131.78,106.95,79.93 "NCNN 20220729 - Target: CPU - Model: resnet18", Lower Results Are Better "A",57.85,39.38,35.99,38.74,63.21,38.62,30.62,36,68.35 "B",52.27,74.85,65.78,32.48,30.55,47.67,37.28,31.2,38.76 "C",31.23,36.39,57.47,30.44,43.76,40.54,50.09,31.63,41.99 "NCNN 20220729 - Target: CPU - Model: alexnet", Lower Results Are Better "A",58.55,54.03,34.15,52.91,49.34,25.2,32.97,32.33,32.76 "B",48.93,63.12,64.28,31.48,48.86,29.5,29.5,26.1,26.71 "C",23.61,24.81,51.95,24.37,38.43,34.66,77.15,62.93,44.36 "NCNN 20220729 - Target: CPU - Model: resnet50", Lower Results Are Better "A",71.51,64.06,70.44,53.69,96.39,57.38,68.74,71.59,88.28 "B",92.75,77.45,76.44,49.99,79.65,74.93,93.08,75.27,78.91 "C",61.44,52.64,65.27,54.01,64.14,64.25,96.25,69.41,61.25 "NCNN 20220729 - Target: CPU - Model: yolov4-tiny", Lower Results Are Better "A",55.35,72.78,62.4,61.49,64.08,76.67,50.65,52.88,73.2 "B",57.87,53.38,54.51,56.54,49.95,58.25,58.9,51.58,58.37 "C",55.37,55.12,61.89,55.48,62.87,52.26,75.82,51.2,70.5 "NCNN 20220729 - Target: CPU - Model: squeezenet_ssd", Lower Results Are Better "A",52,65.23,49.04,81.63,57.72,54.55,55.51,76.89,55.38 "B",51.72,75.32,65.45,47.51,46.65,61.11,46.88,58.18,58.41 "C",59.04,46.92,62.01,56.14,55.78,64.82,53.19,71.26 "NCNN 20220729 - Target: CPU - Model: regnety_400m", Lower Results Are Better "A",89.95,92.56,123.13,136.84,90.94,88.19,101.61,120.63,86.82 "B",85.26,106.23,96.03,85.14,83.72,85.73,92.36,97.87,140.08 "C",91.84,126.66,86.73,85.58,93.14,106.16,96.39,103.07,117.14 "NCNN 20220729 - Target: CPU - Model: vision_transformer", Lower Results Are Better "A",345.93,358.92,348.25,358.74,354.96,367.52,340.19,352.36,356.25 "B",352.85,348.51,353.65,350.79,343.52,354.05,354.38,345.29,368.17 "C",353.41,348.98,355.48,352.29,346.64,345.1,346.99,347.34,359.02 "NCNN 20220729 - Target: CPU - Model: FastestDet", Lower Results Are Better "A",30.9,37.72,33.03,34.08,30.44,27.59,26.79,27.48,42.52 "B",29.32,27.85,42.61,27.37,45.67,38.88,27.27,27.46,28.28 "C",28,34.01,29.15,27.63,37.78,27.97,27.92,33.7,27.86