nn

AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (3501 BIOS) and AMD Radeon RX 6800/6800 XT / 6900 16GB on Ubuntu 21.04 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 2106189-PTS-NN67322042
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
Show Result Confidence Charts

Limit displaying results to tests within:

HPC - High Performance Computing 3 Tests
Machine Learning 3 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
Performance Per
Dollar
Date
Run
  Test
  Duration
1
June 18 2021
  44 Minutes
1a
June 18 2021
  39 Minutes
2
June 18 2021
  1 Hour, 23 Minutes
3
June 18 2021
  2 Hours, 5 Minutes
Invert Hiding All Results Option
  1 Hour, 13 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):


nnProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen Resolution11a23AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads)ASUS ROG CROSSHAIR VIII HERO (3501 BIOS)AMD Starship/Matisse16GB1000GB Sabrent Rocket 4.0 Plus + 2000GBAMD Radeon RX 6800/6800 XT / 6900 16GB (2475/1000MHz)AMD Navi 21 HDMI AudioASUS VP28URealtek RTL8125 2.5GbE + Intel I211Ubuntu 21.045.13.0-051300rc6daily20210617-generic (x86_64) 20210616GNOME Shell 3.38.4X Server 1.20.11 + Wayland4.6 Mesa 21.2.0-devel (git-849ab4e 2021-06-17 hirsute-oibaf-ppa) (LLVM 12.0.0)1.2.180GCC 10.3.0 + CUDA 11.3ext43840x2160OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

nntnn: CPU - DenseNetmnn: inception-v3mnn: mobilenet-v1-1.0mnn: MobileNetV2_224mnn: SqueezeNetV1.0mnn: resnet-v2-50mnn: squeezenetv1.1mnn: mobilenetV3ncnn: CPU - regnety_400mncnn: CPU - squeezenet_ssdncnn: CPU - yolov4-tinyncnn: CPU - resnet50ncnn: CPU - alexnetncnn: CPU - resnet18ncnn: CPU - vgg16ncnn: CPU - googlenetncnn: CPU - blazefacencnn: CPU - efficientnet-b0ncnn: CPU - mnasnetncnn: CPU - shufflenet-v2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU - mobilenettnn: CPU - MobileNet v2tnn: CPU - SqueezeNet v1.1tnn: CPU - SqueezeNet v211a232510.76226.1034.3843.2775.05727.9443.7432.129226.928212.47650.9309.4214.8021.4822.8011.0713.7455.0712.381.715.103.674.003.914.1012.142504.44226.1694.3523.2465.12127.4033.7362.1169.5014.7822.0022.5511.2013.5255.2612.451.795.283.824.104.004.1612.12226.523212.73851.1132505.20425.6164.3253.2115.02327.0103.6192.0599.3414.7521.9122.6911.1313.6155.9812.511.705.133.794.053.964.1312.05224.051211.08650.718OpenBenchmarking.org

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: DenseNet1325001000150020002500SE +/- 3.60, N = 3SE +/- 1.51, N = 3SE +/- 3.49, N = 32510.762505.202504.44MIN: 2446.86 / MAX: 2649.04MIN: 2430 / MAX: 2579.39MIN: 2433.14 / MAX: 2583.961. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: inception-v3213612182430SE +/- 0.24, N = 3SE +/- 0.23, N = 3SE +/- 0.13, N = 726.1726.1025.62MIN: 24.88 / MAX: 34.4MIN: 24.2 / MAX: 62.16MIN: 24.49 / MAX: 42.771. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenet-v1-1.01230.98641.97282.95923.94564.932SE +/- 0.070, N = 3SE +/- 0.051, N = 3SE +/- 0.022, N = 74.3844.3524.325MIN: 4.07 / MAX: 13.17MIN: 4.08 / MAX: 11.71MIN: 4.09 / MAX: 11.491. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: MobileNetV2_2241230.73731.47462.21192.94923.6865SE +/- 0.035, N = 3SE +/- 0.041, N = 3SE +/- 0.032, N = 73.2773.2463.211MIN: 3.07 / MAX: 11.36MIN: 3.07 / MAX: 10.44MIN: 2.97 / MAX: 111. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: SqueezeNetV1.02131.15222.30443.45664.60885.761SE +/- 0.050, N = 3SE +/- 0.054, N = 3SE +/- 0.049, N = 75.1215.0575.023MIN: 4.85 / MAX: 13.16MIN: 4.64 / MAX: 33.19MIN: 4.59 / MAX: 14.331. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: resnet-v2-50123714212835SE +/- 0.51, N = 3SE +/- 0.15, N = 3SE +/- 0.04, N = 727.9427.4027.01MIN: 25.6 / MAX: 40.37MIN: 26.03 / MAX: 50.79MIN: 25.86 / MAX: 50.681. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: squeezenetv1.11230.84221.68442.52663.36884.211SE +/- 0.068, N = 3SE +/- 0.012, N = 3SE +/- 0.058, N = 73.7433.7363.619MIN: 3.36 / MAX: 12.79MIN: 3.56 / MAX: 11.48MIN: 3.25 / MAX: 12.541. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenetV31230.4790.9581.4371.9162.395SE +/- 0.027, N = 3SE +/- 0.009, N = 3SE +/- 0.019, N = 72.1292.1162.059MIN: 1.91 / MAX: 11.33MIN: 2.01 / MAX: 10.13MIN: 1.88 / MAX: 10.151. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: regnety_400m21a33691215SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 39.509.429.34MIN: 9.03 / MAX: 17.41MIN: 9.01 / MAX: 26.23MIN: 8.86 / MAX: 17.281. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: squeezenet_ssd1a2348121620SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 314.8014.7814.75MIN: 13.52 / MAX: 64.82MIN: 13.64 / MAX: 45.78MIN: 13.81 / MAX: 23.131. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: yolov4-tiny231a510152025SE +/- 0.18, N = 3SE +/- 0.11, N = 3SE +/- 0.28, N = 322.0021.9121.48MIN: 20.67 / MAX: 30.95MIN: 19.58 / MAX: 57.14MIN: 19.65 / MAX: 48.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: resnet501a32510152025SE +/- 0.36, N = 3SE +/- 0.18, N = 3SE +/- 0.28, N = 322.8022.6922.55MIN: 20.6 / MAX: 104.7MIN: 21.25 / MAX: 31.38MIN: 20.93 / MAX: 31.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: alexnet231a3691215SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 311.2011.1311.07MIN: 10.37 / MAX: 19.33MIN: 10.19 / MAX: 19.44MIN: 10.16 / MAX: 19.671. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: resnet181a3248121620SE +/- 0.16, N = 3SE +/- 0.16, N = 3SE +/- 0.04, N = 313.7413.6113.52MIN: 12.93 / MAX: 22.43MIN: 12.65 / MAX: 21.63MIN: 12.87 / MAX: 22.241. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: vgg16321a1326395265SE +/- 0.90, N = 3SE +/- 0.21, N = 3SE +/- 0.20, N = 355.9855.2655.07MIN: 51.31 / MAX: 862.27MIN: 52.4 / MAX: 85.07MIN: 52.29 / MAX: 67.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: googlenet321a3691215SE +/- 0.17, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 312.5112.4512.38MIN: 11.61 / MAX: 32.56MIN: 11.62 / MAX: 30.43MIN: 11.66 / MAX: 20.81. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: blazeface21a30.40280.80561.20841.61122.014SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.791.711.70MIN: 1.65 / MAX: 9.61MIN: 1.62 / MAX: 8.08MIN: 1.6 / MAX: 9.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: efficientnet-b0231a1.1882.3763.5644.7525.94SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 35.285.135.10MIN: 4.9 / MAX: 13.73MIN: 4.8 / MAX: 38.8MIN: 4.83 / MAX: 13.321. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: mnasnet231a0.85951.7192.57853.4384.2975SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.02, N = 33.823.793.67MIN: 3.57 / MAX: 11.88MIN: 3.51 / MAX: 12.14MIN: 3.5 / MAX: 11.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: shufflenet-v2231a0.92251.8452.76753.694.6125SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 34.104.054.00MIN: 3.89 / MAX: 11.86MIN: 3.86 / MAX: 11.94MIN: 3.74 / MAX: 11.711. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU-v3-v3 - Model: mobilenet-v3231a0.91.82.73.64.5SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 34.003.963.91MIN: 3.76 / MAX: 12.05MIN: 3.75 / MAX: 12.14MIN: 3.73 / MAX: 11.81. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU-v2-v2 - Model: mobilenet-v2231a0.9361.8722.8083.7444.68SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 34.164.134.10MIN: 3.87 / MAX: 12.37MIN: 3.88 / MAX: 12.67MIN: 3.86 / MAX: 12.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210525Target: CPU - Model: mobilenet1a233691215SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 312.1412.1212.05MIN: 11.26 / MAX: 38.92MIN: 11.38 / MAX: 34.52MIN: 11.3 / MAX: 29.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v212350100150200250SE +/- 0.40, N = 3SE +/- 2.57, N = 3SE +/- 1.38, N = 3226.93226.52224.05MIN: 222.06 / MAX: 239.49MIN: 218.62 / MAX: 239.35MIN: 218.93 / MAX: 236.841. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.121350100150200250SE +/- 3.04, N = 3SE +/- 0.82, N = 3SE +/- 0.10, N = 3212.74212.48211.09MIN: 206.57 / MAX: 219.72MIN: 210.54 / MAX: 219.75MIN: 210.04 / MAX: 211.621. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v22131224364860SE +/- 0.11, N = 3SE +/- 0.28, N = 3SE +/- 0.25, N = 351.1150.9350.72MIN: 50.74 / MAX: 51.46MIN: 50.31 / MAX: 51.63MIN: 50.05 / MAX: 51.131. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl