3900xt-november

AMD Ryzen 9 3900XT 12-Core testing with a MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.B3 BIOS) and AMD Radeon RX 56/64 8GB on Ubuntu 22.04 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 2211180-SYST-3900XTN38
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:

AV1 2 Tests
C/C++ Compiler Tests 4 Tests
CPU Massive 6 Tests
Creator Workloads 8 Tests
Cryptocurrency Benchmarks, CPU Mining Tests 2 Tests
Cryptography 3 Tests
Encoding 4 Tests
HPC - High Performance Computing 8 Tests
Imaging 3 Tests
Machine Learning 4 Tests
Multi-Core 6 Tests
OpenMPI Tests 3 Tests
Python Tests 6 Tests
Server CPU Tests 3 Tests
Single-Threaded 2 Tests
Video Encoding 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
a
November 17 2022
  5 Hours, 24 Minutes
aa
November 17 2022
  5 Hours, 21 Minutes
b
November 17 2022
  13 Hours, 42 Minutes
Invert Hiding All Results Option
  8 Hours, 9 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):

a

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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 Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701021
Graphics Notes: BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D0500100-102
Python Notes: Python 3.10.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + 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 Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

aa

b

Processor: AMD Ryzen 9 3900XT 12-Core @ 3.80GHz (12 Cores / 24 Threads), Motherboard: MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.B3 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 500GB Seagate FireCuda 520 SSD ZP500GM30002, Graphics: AMD Radeon RX 56/64 8GB (1630/945MHz), Audio: AMD Vega 10 HDMI Audio, Monitor: ASUS MG28U, Network: Realtek Device 2600 + Realtek Killer E3000 2.5GbE + Intel Wi-Fi 6 AX200

OS: Ubuntu 22.04, Kernel: 5.15.0-47-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.42), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160

3900xt-november OpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 3900XT 12-Core @ 3.80GHz (12 Cores / 24 Threads)MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.B3 BIOS)AMD Starship/Matisse16GB500GB Seagate FireCuda 520 SSD ZP500GM30002AMD Radeon RX 56/64 8GB (1630/945MHz)AMD Vega 10 HDMI AudioASUS MG28URealtek Device 2600 + Realtek Killer E3000 2.5GbE + Intel Wi-Fi 6 AX200Ubuntu 22.045.15.0-47-generic (x86_64)GNOME Shell 42.2X Server + Wayland4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.42)1.3.204GCC 11.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen Resolution3900xt-november BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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 - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701021- BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D0500100-102- Python 3.10.4- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + 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 Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

aaabResult OverviewPhoronix Test Suite100%106%112%118%spaCyoneDNNStress-NGnekRSEnCodecXmrigJPEG XL Decoding libjxlnginxQuadRayJPEG XL libjxlSMHasherAOM AV1FLAC Audio EncodingCpuminer-Optlibavif avifencY-CruncherOpenRadiossminiBUDENeural Magic DeepSparseOpenFOAMFFmpegLibplaceboTensorFlow

3900xt-november stress-ng: IO_uringspacy: en_core_web_trfstress-ng: Glibc C String Functionscpuminer-opt: Garlicoinstress-ng: Context Switchingstress-ng: CPU Stressstress-ng: Glibc Qsort Data Sortingsmhasher: t1ha2_atoncestress-ng: CPU Cachequadray: 1 - 4Kstress-ng: Futexstress-ng: Socket Activitysmhasher: FarmHash32 x86_64 AVXstress-ng: SENDFILEsmhasher: t1ha0_aes_avx2 x86_64smhasher: FarmHash128nginx: 100stress-ng: Memory Copyingstress-ng: Matrix Mathaom-av1: Speed 6 Realtime - Bosphorus 1080pstress-ng: Vector Mathsmhasher: fasthash32smhasher: MeowHash x86_64 AES-NIstress-ng: System V Message Passingjpegxl: JPEG - 80stress-ng: MEMFDencodec: 1.5 kbpsnekrs: TurboPipe Periodicjpegxl-decode: 1libplacebo: polar_nocomputeopenradioss: INIVOL and Fluid Structure Interaction Drop Containerxmrig: Monero - 1Mcpuminer-opt: Skeincoinencodec: 6 kbpsencodec: 3 kbpsquadray: 1 - 1080pquadray: 3 - 4Ksmhasher: SHA3-256cpuminer-opt: LBC, LBRY Creditssmhasher: wyhashjpegxl: JPEG - 90deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamspacy: en_core_web_lgavifenc: 2nginx: 500aom-av1: Speed 10 Realtime - Bosphorus 1080pnginx: 1000onednn: Recurrent Neural Network Training - f32 - CPUnginx: 200cpuminer-opt: Blake-2 Sxmrig: Wownero - 1Mtensorflow: CPU - 512 - GoogLeNetlibplacebo: av1_grain_lapstress-ng: Forkinglibplacebo: hdr_lutaom-av1: Speed 6 Two-Pass - Bosphorus 1080paom-av1: Speed 10 Realtime - Bosphorus 4Kquadray: 2 - 4Kaom-av1: Speed 9 Realtime - Bosphorus 1080pjpegxl-decode: Allaom-av1: Speed 8 Realtime - Bosphorus 4Kdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamcpuminer-opt: Myriad-Groestlonednn: IP Shapes 1D - f32 - CPUsmhasher: Spooky32stress-ng: Mallocaom-av1: Speed 8 Realtime - Bosphorus 1080pdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamquadray: 3 - 1080pavifenc: 0libplacebo: deband_heavydeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamy-cruncher: 1Bavifenc: 6cpuminer-opt: Ringcoindeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamquadray: 2 - 1080popenfoam: drivaerFastback, Small Mesh Size - Execution Timeonednn: Recurrent Neural Network Inference - f32 - CPUdeepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streamstress-ng: Mutexdeepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streamencodec: 24 kbpsffmpeg: libx264 - Platformffmpeg: libx264 - Platformffmpeg: libx265 - Platformffmpeg: libx265 - Platformquadray: 5 - 1080paom-av1: Speed 4 Two-Pass - Bosphorus 1080popenradioss: Rubber O-Ring Seal Installationopenradioss: Bumper Beamonednn: Deconvolution Batch shapes_3d - f32 - CPUstress-ng: Atomicencode-flac: WAV To FLACstress-ng: MMAPminibude: OpenMP - BM1minibude: OpenMP - BM1openfoam: drivaerFastback, Small Mesh Size - Mesh Timetensorflow: CPU - 64 - ResNet-50avifenc: 6, Losslessopenradioss: Bird Strike on Windshieldjpegxl: PNG - 90ffmpeg: libx265 - Livejpegxl: PNG - 80ffmpeg: libx265 - Video On Demandaom-av1: Speed 4 Two-Pass - Bosphorus 4Kffmpeg: libx265 - Liveffmpeg: libx265 - Video On Demandcpuminer-opt: x25xcpuminer-opt: Deepcointensorflow: CPU - 16 - ResNet-50stress-ng: NUMAavifenc: 10, Losslesscpuminer-opt: Magiaom-av1: Speed 6 Realtime - Bosphorus 4Klibplacebo: hdr_peakdetecttensorflow: CPU - 512 - AlexNetaom-av1: Speed 9 Realtime - Bosphorus 4Kdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamstress-ng: Cryptotensorflow: CPU - 32 - GoogLeNettensorflow: CPU - 32 - ResNet-50openradioss: Cell Phone Drop Testdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamffmpeg: libx265 - Uploadffmpeg: libx265 - Uploadaom-av1: Speed 6 Two-Pass - Bosphorus 4Kcpuminer-opt: scrypttensorflow: CPU - 16 - AlexNetonednn: IP Shapes 3D - f32 - CPUffmpeg: libx264 - Uploaddeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamffmpeg: libx264 - Uploadtensorflow: CPU - 32 - AlexNettensorflow: CPU - 16 - GoogLeNetdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamffmpeg: libx264 - Video On Demandtensorflow: CPU - 64 - GoogLeNetdeepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Streamffmpeg: libx264 - Video On Demandffmpeg: libx264 - Livedeepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Streamcpuminer-opt: Quad SHA-256, Pyriteffmpeg: libx264 - Livedeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamtensorflow: CPU - 256 - AlexNetdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamonednn: Convolution Batch Shapes Auto - f32 - CPUstress-ng: Semaphoresminibude: OpenMP - BM2minibude: OpenMP - BM2tensorflow: CPU - 256 - GoogLeNety-cruncher: 500Mtensorflow: CPU - 64 - AlexNetcpuminer-opt: Triple SHA-256, Onecoindeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamaom-av1: Speed 0 Two-Pass - Bosphorus 1080paom-av1: Speed 0 Two-Pass - Bosphorus 4Kquadray: 5 - 4Kjpegxl: JPEG - 100jpegxl: PNG - 100onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUsmhasher: MeowHash x86_64 AES-NIsmhasher: t1ha0_aes_avx2 x86_64smhasher: FarmHash32 x86_64 AVXsmhasher: t1ha2_atoncesmhasher: FarmHash128smhasher: fasthash32smhasher: Spooky32smhasher: SHA3-256smhasher: wyhashaaab5126.998322048965.251939.654076744.5232900.39180.7716686.57158.8910.212600010.058595.229093.03215308.1466796.4615884.2487222.683599.4261078.0645.290999.656658.9237183.287925418.158.68747.3145.143155990000048.69977.72633.977982.610688047.32647.23139.482.46149.083126023915.758.6248.486920.62081219663.12175259.81109.5265135.634075.5783513.1452468010466.628.792073.6940818.922797.2341.5747.552.89108.47146.6334.46617.6457211304.6902614909.0213779284.6285.229.71339.71130.357476.59611.84049.774841.5246.0492705.27180.605233.214411.12363.216572503.12113.46036319425.8652.847654.416155.1148.8431.31241.9288721792.6415.08131127.425.27435575564.9617.255294.05405.81616.23341.82356810.0910.149285.699.0266.999.0831.457.4175.38240.839641179631.151171010.39260.685.48603.5823.12610.2110.3647.4273.708581.37422523.7930.4610.29104.5140.9537146.479316.815959.4485164.79743165715.3212.67226.4255.6312.060212.64124.79198.0129199.73024137776.2231.8933.385329.948148.8330.5523.5574155.13773228524.6642.4309102170204.75125.34557.9776112.2815109.2312.749978.388922.54492463565.68409.85516.39428.6419.54793.820497053.41020.730.250.660.680.691.88178.227457.7434.81140.06932.25464.10336.90249.0562598.0525.3098765.888332035501.441960.714103283.7731954.78192.2315706.92152.469.632717409.388500.1728582.22215163.9969896.4116614.2886726.13592.7861203.2146.9890998.926917.3237904.327932169.448.98773.0746.0823257520000047.21948.28650.257945.810862048.56447.65938.52.43148.143182024070.668.7649.22820.31041221764.01976617.7111.4965507.924145.8384210.425280601064028.712106.7340572.792806.1642.1747.952.9108.91148.6134.91617.0345212604.7485915026.613924963.4385.99.7239.81131.57479.65617.74059.708541.8416.0762709.03181.455633.059511.22361.479572484.9113.64286371966.3752.781254.426154.5749.0131.49240.5201759552.6215.01130.82126.515.25072575708.5317.326292.1406.70616.26841.85230710.0610.094286.849.0366.899.1331.627.4575.50239.59631.741172010.44261.265.505603.3323.22598.97110.4947.6274.011781.04422525.1430.5810.29104.1141.0572146.110816.767859.619164.7215.3312.68226.5955.4712.028812.64124.75828.0151199.8176.2531.8333.322830.004148.8530.5123.5672155.05813851724.6342.4149102180205.03125.16537.989112.2567109.3812.746378.408822.55042465656.19410.30716.41228.6719.52793.8820495053.40520.730.250.660.680.691.278955.4712556.97632.87541.13434.54860.33635.25748.982620.22125.294168.545501531955.882112.413796353.9930793.53184.7716085.48149.7910.142563831.678961.227788.51205666.1268310.6616361.6483591.653456.5358814.3845.6487587.666686.2538610.177654922.058.84770.8946.6243254433333348.24961.48631.338175.810566748.21548.4539.262.49151.763114323571.268.8049.490820.20261197864.37376562.05110.2666302.864147.6582789.4251943310577.328.332093.7040197.62840.1642.0748.232.86107.41148.5934.60609.8345213974.7308614856.1013767394.3384.939.82389.70130.149474.53618.13639.676141.4266.0192730.30182.256932.91411.17359.999032481.61114.40376344621.5652.416254.862155.8148.6231.55240.092.6315.12130.06127.375.28794571674.6817.209293.29404.24716.17042.07700710.1210.152287.339.0767.269.1231.567.4175.10240.03628.471166010.44260.025.497600.8723.132603.19110.0247.6073.766681.309222434.6930.4810.33104.2740.9084146.641416.825859.4138164.2615.3712.71225.8955.4912.031312.67125.04697.9966199.3576.3731.8533.340429.988648.7730.5023.596155.3124.6742.3627102017204.73125.26917.9824112.1223109.3012.763478.306222.52102466594.73410.14116.40528.6519.53193.7920510053.43980.730.250.660.680.691.741216.2239655.87533.61242.57533.67961.81236.68849.5002564.20625.914OpenBenchmarking.org

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: IO_uringbaaa2K4K6K8K10K4168.548765.885126.991. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

spaCy

The spaCy library is an open-source solution for advanced neural language processing (NLP). The spaCy library leverages Python and is a leading neural language processing solution. This test profile times the spaCy CPU performance with various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgtokens/sec, More Is BetterspaCy 3.4.1Model: en_core_web_trfbaaa2004006008001000550833832

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Glibc C String Functionsbaaa400K800K1200K1600K2000K1531955.882035501.442048965.251. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Garlicoinbaaa5001000150020002500SE +/- 24.17, N = 152112.411960.711939.651. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Garlicoinbaaa400800120016002000Min: 1974.54 / Avg: 2112.41 / Max: 2248.241. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Context Switchingbaaa900K1800K2700K3600K4500K3796353.994103283.774076744.521. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU Stressbaaa7K14K21K28K35K30793.5331954.7832900.391. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Glibc Qsort Data Sortingbaaa4080120160200184.77192.23180.771. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha2_atoncebaaa4K8K12K16K20KSE +/- 116.42, N = 1516085.4815706.9216686.571. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: t1ha2_atoncebaaa816243240SE +/- 0.28, N = 1533.6834.5532.251. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha2_atoncebaaa3K6K9K12K15KMin: 15518.5 / Avg: 16085.48 / Max: 16734.811. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU Cachebaaa4080120160200149.79152.46158.891. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 1 - Resolution: 4Kbaaa3691215SE +/- 0.04, N = 310.149.6310.211. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 1 - Resolution: 4Kbaaa3691215Min: 10.06 / Avg: 10.14 / Max: 10.211. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Futexbaaa600K1200K1800K2400K3000K2563831.672717409.382600010.051. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Socket Activitybaaa2K4K6K8K10K8961.208500.178595.201. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash32 x86_64 AVXbaaa6K12K18K24K30KSE +/- 369.00, N = 327788.5128582.2229093.031. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: FarmHash32 x86_64 AVXbaaa1020304050SE +/- 0.70, N = 342.5841.1340.071. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash32 x86_64 AVXbaaa5K10K15K20K25KMin: 27350.75 / Avg: 27788.51 / Max: 28521.941. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: SENDFILEbaaa50K100K150K200K250K205666.12215163.99215308.141. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha0_aes_avx2 x86_64baaa15K30K45K60K75KSE +/- 549.14, N = 1568310.6669896.4166796.461. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: t1ha0_aes_avx2 x86_64baaa816243240SE +/- 0.30, N = 1533.6132.8834.811. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha0_aes_avx2 x86_64baaa12K24K36K48K60KMin: 64675.04 / Avg: 68310.66 / Max: 71530.281. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash128baaa4K8K12K16K20KSE +/- 128.88, N = 1516361.6416614.2815884.241. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: FarmHash128baaa1428425670SE +/- 0.72, N = 1561.8160.3464.101. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash128baaa3K6K9K12K15KMin: 15711.76 / Avg: 16361.64 / Max: 17016.291. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 100baaa20K40K60K80K100K83591.6586726.1087222.681. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Memory Copyingbaaa80016002400320040003456.533592.783599.421. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Matrix Mathbaaa13K26K39K52K65K58814.3861203.2161078.061. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080pbaaa1122334455SE +/- 0.20, N = 345.6446.9845.201. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080pbaaa1020304050Min: 45.33 / Avg: 45.64 / Max: 46.021. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Vector Mathbaaa20K40K60K80K100K87587.6690998.9290999.651. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: fasthash32baaa15003000450060007500SE +/- 77.96, N = 46686.256917.326658.921. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: fasthash32baaa816243240SE +/- 0.48, N = 436.6935.2636.901. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: fasthash32baaa12002400360048006000Min: 6583.69 / Avg: 6686.25 / Max: 6918.61. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: MeowHash x86_64 AES-NIbaaa8K16K24K32K40KSE +/- 285.07, N = 1538610.1737904.3237183.281. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: MeowHash x86_64 AES-NIbaaa1326395265SE +/- 0.39, N = 1555.8856.9857.741. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: MeowHash x86_64 AES-NIbaaa7K14K21K28K35KMin: 37361.23 / Avg: 38610.17 / Max: 40418.621. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: System V Message Passingbaaa2M4M6M8M10M7654922.057932169.447925418.151. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 80baaa3691215SE +/- 0.03, N = 38.848.988.681. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 80baaa3691215Min: 8.78 / Avg: 8.84 / Max: 8.91. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MEMFDbaaa170340510680850770.89773.07747.311. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

EnCodec

EnCodec is a Facebook/Meta developed AI means of compressing audio files using High Fidelity Neural Audio Compression. EnCodec is designed to provide codec compression at 6 kbps using their novel AI-powered compression technique. The test profile uses a lengthy JFK speech as the audio input for benchmarking and the performance measurement is measuring the time to encode the EnCodec file from WAV. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 1.5 kbpsbaaa112233445546.6246.0845.14

nekRS

nekRS is an open-source Navier Stokes solver based on the spectral element method. NekRS supports both CPU and GPU/accelerator support though this test profile is currently configured for CPU execution. NekRS is part of Nek5000 of the Mathematics and Computer Science MCS at Argonne National Laboratory. This nekRS benchmark is primarily relevant to large core count HPC servers and otherwise may be very time consuming. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFLOP/s, More Is BetternekRS 22.0Input: TurboPipe Periodicbaaa7000M14000M21000M28000M35000MSE +/- 20784155.29, N = 33254433333332575200000315599000001. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -lmpi_cxx -lmpi
OpenBenchmarking.orgFLOP/s, More Is BetternekRS 22.0Input: TurboPipe Periodicbaaa6000M12000M18000M24000M30000MMin: 32512500000 / Avg: 32544333333.33 / Max: 325834000001. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -lmpi_cxx -lmpi

JPEG XL Decoding libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: 1baaa1122334455SE +/- 0.56, N = 348.2447.2148.69
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: 1baaa1020304050Min: 47.14 / Avg: 48.24 / Max: 48.91

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: polar_nocomputebaaa2004006008001000SE +/- 3.01, N = 3961.48948.28977.721. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: polar_nocomputebaaa2004006008001000Min: 957.36 / Avg: 961.48 / Max: 967.351. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: INIVOL and Fluid Structure Interaction Drop Containerbaaa140280420560700SE +/- 0.82, N = 3631.33650.25633.97
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: INIVOL and Fluid Structure Interaction Drop Containerbaaa110220330440550Min: 629.8 / Avg: 631.33 / Max: 632.59

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1Mbaaa2K4K6K8K10KSE +/- 88.16, N = 38175.87945.87982.61. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1Mbaaa14002800420056007000Min: 8048 / Avg: 8175.8 / Max: 8344.91. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Skeincoinbaaa20K40K60K80K100KSE +/- 1171.67, N = 31056671086201068801. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Skeincoinbaaa20K40K60K80K100KMin: 104490 / Avg: 105666.67 / Max: 1080101. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

EnCodec

EnCodec is a Facebook/Meta developed AI means of compressing audio files using High Fidelity Neural Audio Compression. EnCodec is designed to provide codec compression at 6 kbps using their novel AI-powered compression technique. The test profile uses a lengthy JFK speech as the audio input for benchmarking and the performance measurement is measuring the time to encode the EnCodec file from WAV. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 6 kbpsbaaa112233445548.2248.5647.33

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 3 kbpsbaaa112233445548.4547.6647.23

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 1 - Resolution: 1080pbaaa918273645SE +/- 0.13, N = 339.2638.5039.481. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 1 - Resolution: 1080pbaaa816243240Min: 39.06 / Avg: 39.26 / Max: 39.511. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 3 - Resolution: 4Kbaaa0.56031.12061.68092.24122.8015SE +/- 0.01, N = 32.492.432.461. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 3 - Resolution: 4Kbaaa246810Min: 2.47 / Avg: 2.49 / Max: 2.511. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: SHA3-256baaa306090120150SE +/- 1.41, N = 7151.76148.14149.081. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: SHA3-256baaa6001200180024003000SE +/- 19.10, N = 72564.212620.222598.051. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: SHA3-256baaa306090120150Min: 147.24 / Avg: 151.76 / Max: 157.611. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: LBC, LBRY Creditsbaaa7K14K21K28K35KSE +/- 18.56, N = 33114331820312601. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: LBC, LBRY Creditsbaaa6K12K18K24K30KMin: 31120 / Avg: 31143.33 / Max: 311801. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: wyhashbaaa5K10K15K20K25KSE +/- 338.95, N = 323571.2624070.6623915.751. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: wyhashbaaa612182430SE +/- 0.37, N = 325.9125.2925.311. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: wyhashbaaa4K8K12K16K20KMin: 23093.48 / Avg: 23571.26 / Max: 24226.631. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 90baaa246810SE +/- 0.08, N = 38.808.768.621. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 90baaa3691215Min: 8.71 / Avg: 8.8 / Max: 8.951. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Streambaaa112233445549.4949.2348.49

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Streambaaa51015202520.2020.3120.62

spaCy

The spaCy library is an open-source solution for advanced neural language processing (NLP). The spaCy library leverages Python and is a leading neural language processing solution. This test profile times the spaCy CPU performance with various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgtokens/sec, More Is BetterspaCy 3.4.1Model: en_core_web_lgbaaa3K6K9K12K15K119781221712196

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 2baaa1428425670SE +/- 0.20, N = 364.3764.0263.121. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 2baaa1326395265Min: 63.98 / Avg: 64.37 / Max: 64.671. (CXX) g++ options: -O3 -fPIC -lm

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500baaa16K32K48K64K80K76562.0576617.7075259.811. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080pbaaa20406080100SE +/- 0.74, N = 3110.26111.49109.521. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080pbaaa20406080100Min: 109.19 / Avg: 110.26 / Max: 111.671. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 1000baaa14K28K42K56K70K66302.8665507.9265135.631. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUbaaa9001800270036004500SE +/- 9.23, N = 34147.654145.834075.57MIN: 4122.17MIN: 4135.73MIN: 4066.651. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUbaaa7001400210028003500Min: 4131.73 / Avg: 4147.65 / Max: 4163.721. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 200baaa20K40K60K80K100K82789.4284210.4283513.141. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Blake-2 Sbaaa110K220K330K440K550KSE +/- 4099.93, N = 105194335280605246801. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Blake-2 Sbaaa90K180K270K360K450KMin: 485330 / Avg: 519433 / Max: 5305801. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Wownero - Hash Count: 1Mbaaa2K4K6K8K10KSE +/- 4.48, N = 310577.310640.010466.61. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Wownero - Hash Count: 1Mbaaa2K4K6K8K10KMin: 10571.2 / Avg: 10577.27 / Max: 105861. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: GoogLeNetbaaa71421283528.3328.7128.79

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: av1_grain_lapbaaa5001000150020002500SE +/- 8.76, N = 32093.702106.732073.691. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: av1_grain_lapbaaa400800120016002000Min: 2076.49 / Avg: 2093.7 / Max: 2105.11. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Forkingbaaa9K18K27K36K45K40197.6040572.7940818.921. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: hdr_lutbaaa6001200180024003000SE +/- 46.08, N = 32840.162806.162797.231. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: hdr_lutbaaa5001000150020002500Min: 2790.89 / Avg: 2840.16 / Max: 2932.251. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080pbaaa1020304050SE +/- 0.26, N = 342.0742.1741.571. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080pbaaa918273645Min: 41.57 / Avg: 42.07 / Max: 42.451. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4Kbaaa1122334455SE +/- 0.11, N = 348.2347.9547.551. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4Kbaaa1020304050Min: 48.01 / Avg: 48.23 / Max: 48.41. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 2 - Resolution: 4Kbaaa0.65251.3051.95752.613.2625SE +/- 0.01, N = 32.862.902.891. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 2 - Resolution: 4Kbaaa246810Min: 2.85 / Avg: 2.86 / Max: 2.871. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080pbaaa20406080100SE +/- 0.48, N = 3107.41108.91108.471. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080pbaaa20406080100Min: 106.7 / Avg: 107.41 / Max: 108.331. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

JPEG XL Decoding libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: Allbaaa306090120150SE +/- 0.27, N = 3148.59148.61146.63
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: Allbaaa306090120150Min: 148.18 / Avg: 148.59 / Max: 149.1

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4Kbaaa816243240SE +/- 0.08, N = 334.6034.9134.461. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4Kbaaa714212835Min: 34.48 / Avg: 34.6 / Max: 34.741. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streambaaa130260390520650609.83617.03617.65

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Myriad-Groestlbaaa5K10K15K20K25KSE +/- 153.44, N = 32139721260211301. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Myriad-Groestlbaaa4K8K12K16K20KMin: 21090 / Avg: 21396.67 / Max: 215601. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUbaaa1.06842.13683.20524.27365.342SE +/- 0.01435, N = 34.730864.748594.69026MIN: 4.49MIN: 4.55MIN: 4.491. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUbaaa246810Min: 4.72 / Avg: 4.73 / Max: 4.761. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: Spooky32baaa3K6K9K12K15KSE +/- 98.93, N = 1514856.1015026.6014909.021. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: Spooky32baaa1122334455SE +/- 0.31, N = 1549.5048.9849.061. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: Spooky32baaa3K6K9K12K15KMin: 14365.69 / Avg: 14856.1 / Max: 15403.121. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Mallocbaaa3M6M9M12M15M13767394.3313924963.4313779284.621. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080pbaaa20406080100SE +/- 0.06, N = 384.9385.9085.221. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080pbaaa1632486480Min: 84.83 / Avg: 84.93 / Max: 85.051. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streambaaa36912159.82389.72309.7133

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 3 - Resolution: 1080pbaaa3691215SE +/- 0.10, N = 39.709.819.711. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 3 - Resolution: 1080pbaaa3691215Min: 9.51 / Avg: 9.7 / Max: 9.81. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 0baaa306090120150SE +/- 0.52, N = 3130.15131.57130.361. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 0baaa20406080100Min: 129.47 / Avg: 130.15 / Max: 131.161. (CXX) g++ options: -O3 -fPIC -lm

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: deband_heavybaaa100200300400500SE +/- 3.08, N = 3474.53479.65476.591. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: deband_heavybaaa90180270360450Min: 469.5 / Avg: 474.53 / Max: 480.141. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streambaaa130260390520650618.14617.74611.84

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streambaaa36912159.67619.70859.7748

Y-Cruncher

Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.7.10.9513Pi Digits To Calculate: 1Bbaaa1020304050SE +/- 0.11, N = 341.4341.8441.52
OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.7.10.9513Pi Digits To Calculate: 1Bbaaa918273645Min: 41.3 / Avg: 41.43 / Max: 41.64

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6baaa246810SE +/- 0.028, N = 36.0196.0766.0491. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6baaa246810Min: 5.96 / Avg: 6.02 / Max: 6.051. (CXX) g++ options: -O3 -fPIC -lm

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Ringcoinbaaa6001200180024003000SE +/- 7.56, N = 32730.302709.032705.271. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Ringcoinbaaa5001000150020002500Min: 2719.2 / Avg: 2730.3 / Max: 2744.751. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streambaaa4080120160200182.26181.46180.61

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streambaaa81624324032.9133.0633.21

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 2 - Resolution: 1080pbaaa3691215SE +/- 0.06, N = 311.1711.2211.121. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 2 - Resolution: 1080pbaaa3691215Min: 11.05 / Avg: 11.17 / Max: 11.231. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Execution Timebaaa80160240320400360.00361.48363.221. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUbaaa5001000150020002500SE +/- 1.05, N = 32481.612484.902503.12MIN: 2470.87MIN: 2478.59MIN: 2492.431. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUbaaa400800120016002000Min: 2480.52 / Avg: 2481.61 / Max: 2483.711. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Streambaaa306090120150114.40113.64113.46

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Mutexbaaa1.4M2.8M4.2M5.6M7M6344621.566371966.376319425.861. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Streambaaa122436486052.4252.7852.85

EnCodec

EnCodec is a Facebook/Meta developed AI means of compressing audio files using High Fidelity Neural Audio Compression. EnCodec is designed to provide codec compression at 6 kbps using their novel AI-powered compression technique. The test profile uses a lengthy JFK speech as the audio input for benchmarking and the performance measurement is measuring the time to encode the EnCodec file from WAV. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 24 kbpsbaaa122436486054.8654.4354.42

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Platformbaaa306090120150SE +/- 0.25, N = 3155.81154.57155.111. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Platformbaaa306090120150Min: 155.43 / Avg: 155.81 / Max: 156.281. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Platformbaaa1122334455SE +/- 0.08, N = 348.6249.0148.841. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Platformbaaa1020304050Min: 48.47 / Avg: 48.62 / Max: 48.741. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Platformbaaa714212835SE +/- 0.03, N = 331.5531.4931.311. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Platformbaaa714212835Min: 31.51 / Avg: 31.55 / Max: 31.61. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Platformbaaa50100150200250SE +/- 0.20, N = 3240.09240.52241.931. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Platformbaaa4080120160200Min: 239.74 / Avg: 240.09 / Max: 240.431. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 5 - Resolution: 1080pbaaa0.5941.1881.7822.3762.97SE +/- 0.00, N = 32.632.622.641. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 5 - Resolution: 1080pbaaa246810Min: 2.63 / Avg: 2.63 / Max: 2.641. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080pbaaa48121620SE +/- 0.08, N = 315.1215.0115.081. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080pbaaa48121620Min: 14.99 / Avg: 15.12 / Max: 15.251. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Rubber O-Ring Seal Installationbaaa306090120150SE +/- 0.35, N = 3130.06130.82131.00
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Rubber O-Ring Seal Installationbaaa20406080100Min: 129.66 / Avg: 130.06 / Max: 130.76

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bumper Beambaaa306090120150SE +/- 0.16, N = 3127.37126.51127.42
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bumper Beambaaa20406080100Min: 127.18 / Avg: 127.37 / Max: 127.68

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUbaaa1.18982.37963.56944.75925.949SE +/- 0.00411, N = 35.287945.250725.27435MIN: 5.2MIN: 5.17MIN: 5.181. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUbaaa246810Min: 5.28 / Avg: 5.29 / Max: 5.31. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Atomicbaaa120K240K360K480K600K571674.68575708.53575564.961. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

FLAC Audio Encoding

This test times how long it takes to encode a sample WAV file to FLAC audio format ten times using the --best preset settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.4WAV To FLACbaaa48121620SE +/- 0.04, N = 517.2117.3317.261. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.4WAV To FLACbaaa48121620Min: 17.06 / Avg: 17.21 / Max: 17.321. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MMAPbaaa60120180240300293.29292.10294.051. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

miniBUDE

MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1baaa90180270360450SE +/- 0.95, N = 3404.25406.71405.821. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1baaa70140210280350Min: 402.9 / Avg: 404.25 / Max: 406.091. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1baaa48121620SE +/- 0.04, N = 316.1716.2716.231. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1baaa48121620Min: 16.12 / Avg: 16.17 / Max: 16.241. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh Timebaaa102030405042.0841.8541.821. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: ResNet-50baaa369121510.1210.0610.09

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6, Losslessbaaa3691215SE +/- 0.04, N = 310.1510.0910.151. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6, Losslessbaaa3691215Min: 10.09 / Avg: 10.15 / Max: 10.211. (CXX) g++ options: -O3 -fPIC -lm

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bird Strike on Windshieldbaaa60120180240300SE +/- 0.10, N = 3287.33286.84285.69
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bird Strike on Windshieldbaaa50100150200250Min: 287.2 / Avg: 287.33 / Max: 287.52

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 90baaa3691215SE +/- 0.02, N = 39.079.039.021. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 90baaa3691215Min: 9.03 / Avg: 9.07 / Max: 9.111. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Livebaaa1530456075SE +/- 0.65, N = 367.2666.8966.991. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Livebaaa1326395265Min: 65.97 / Avg: 67.26 / Max: 67.991. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 80baaa3691215SE +/- 0.02, N = 39.129.139.081. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 80baaa3691215Min: 9.09 / Avg: 9.12 / Max: 9.171. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Video On Demandbaaa714212835SE +/- 0.07, N = 331.5631.6231.451. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Video On Demandbaaa714212835Min: 31.43 / Avg: 31.56 / Max: 31.671. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4Kbaaa246810SE +/- 0.01, N = 37.417.457.411. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4Kbaaa3691215Min: 7.4 / Avg: 7.41 / Max: 7.421. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Livebaaa20406080100SE +/- 0.73, N = 375.1075.5075.381. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Livebaaa1530456075Min: 74.28 / Avg: 75.1 / Max: 76.551. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Video On Demandbaaa50100150200250SE +/- 0.53, N = 3240.03239.59240.841. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Video On Demandbaaa4080120160200Min: 239.17 / Avg: 240.03 / Max: 241.011. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: x25xbaaa140280420560700SE +/- 0.96, N = 3628.47631.74631.151. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: x25xbaaa110220330440550Min: 627.08 / Avg: 628.47 / Max: 630.321. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Deepcoinbaaa3K6K9K12K15KSE +/- 5.77, N = 31166011720117101. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Deepcoinbaaa2K4K6K8K10KMin: 11650 / Avg: 11660 / Max: 116701. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: ResNet-50baaa3691215SE +/- 0.02, N = 310.4410.4410.39
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: ResNet-50baaa3691215Min: 10.42 / Avg: 10.44 / Max: 10.47

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: NUMAbaaa60120180240300260.02261.26260.681. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 10, Losslessbaaa1.23862.47723.71584.95446.193SE +/- 0.026, N = 35.4975.5055.4801. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 10, Losslessbaaa246810Min: 5.45 / Avg: 5.5 / Max: 5.541. (CXX) g++ options: -O3 -fPIC -lm

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Magibaaa130260390520650SE +/- 0.50, N = 3600.87603.33603.581. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Magibaaa110220330440550Min: 599.98 / Avg: 600.87 / Max: 601.721. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4Kbaaa612182430SE +/- 0.06, N = 323.1323.2023.101. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4Kbaaa510152025Min: 23.02 / Avg: 23.13 / Max: 23.211. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: hdr_peakdetectbaaa6001200180024003000SE +/- 4.71, N = 32603.192598.972610.201. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
OpenBenchmarking.orgFPS, More Is BetterLibplacebo 5.229.1Test: hdr_peakdetectbaaa5001000150020002500Min: 2594.27 / Avg: 2603.19 / Max: 2610.281. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: AlexNetbaaa20406080100SE +/- 0.06, N = 3110.02110.49110.36
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: AlexNetbaaa20406080100Min: 109.95 / Avg: 110.02 / Max: 110.13

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4Kbaaa1122334455SE +/- 0.12, N = 347.6047.6247.421. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4Kbaaa1020304050Min: 47.36 / Avg: 47.6 / Max: 47.731. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streambaaa163248648073.7774.0173.71

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streambaaa2040608010081.3181.0481.37

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Cryptobaaa5K10K15K20K25K22434.6922525.1422523.791. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: GoogLeNetbaaa714212835SE +/- 0.01, N = 330.4830.5830.46
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: GoogLeNetbaaa714212835Min: 30.46 / Avg: 30.48 / Max: 30.49

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: ResNet-50baaa3691215SE +/- 0.01, N = 310.3310.2910.29
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: ResNet-50baaa3691215Min: 10.32 / Avg: 10.33 / Max: 10.35

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Cell Phone Drop Testbaaa20406080100SE +/- 0.05, N = 3104.27104.11104.51
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Cell Phone Drop Testbaaa20406080100Min: 104.21 / Avg: 104.27 / Max: 104.36

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streambaaa91827364540.9141.0640.95

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streambaaa306090120150146.64146.11146.48

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streambaaa4812162016.8316.7716.82

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streambaaa132639526559.4159.6259.45

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Uploadbaaa4080120160200SE +/- 0.31, N = 3164.26164.72164.801. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Uploadbaaa306090120150Min: 163.66 / Avg: 164.26 / Max: 164.681. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Uploadbaaa48121620SE +/- 0.03, N = 315.3715.3315.321. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Uploadbaaa48121620Min: 15.33 / Avg: 15.37 / Max: 15.431. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4Kbaaa3691215SE +/- 0.06, N = 312.7112.6812.671. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4Kbaaa48121620Min: 12.6 / Avg: 12.71 / Max: 12.811. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: scryptbaaa50100150200250SE +/- 0.18, N = 3225.89226.59226.421. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: scryptbaaa4080120160200Min: 225.69 / Avg: 225.89 / Max: 226.251. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: AlexNetbaaa1224364860SE +/- 0.04, N = 355.4955.4755.63
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: AlexNetbaaa1122334455Min: 55.44 / Avg: 55.49 / Max: 55.57

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUbaaa3691215SE +/- 0.01, N = 312.0312.0312.06MIN: 11.91MIN: 11.93MIN: 11.961. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUbaaa48121620Min: 12.02 / Avg: 12.03 / Max: 12.051. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Uploadbaaa3691215SE +/- 0.06, N = 312.6712.6412.641. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Uploadbaaa48121620Min: 12.56 / Avg: 12.67 / Max: 12.771. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streambaaa306090120150125.05124.76124.79

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streambaaa2468107.99668.01518.0129

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Uploadbaaa4080120160200SE +/- 0.94, N = 3199.35199.81199.731. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Uploadbaaa4080120160200Min: 197.77 / Avg: 199.35 / Max: 201.021. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: AlexNetbaaa20406080100SE +/- 0.07, N = 376.3776.2576.22
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: AlexNetbaaa1530456075Min: 76.26 / Avg: 76.37 / Max: 76.49

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: GoogLeNetbaaa714212835SE +/- 0.03, N = 331.8531.8331.89
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: GoogLeNetbaaa714212835Min: 31.79 / Avg: 31.85 / Max: 31.91

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Streambaaa81624324033.3433.3233.39

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Streambaaa71421283529.9930.0029.95

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Video On Demandbaaa1122334455SE +/- 0.09, N = 348.7748.8548.831. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Video On Demandbaaa1020304050Min: 48.61 / Avg: 48.77 / Max: 48.941. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: GoogLeNetbaaa714212835SE +/- 0.01, N = 330.5030.5130.55
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: GoogLeNetbaaa714212835Min: 30.49 / Avg: 30.5 / Max: 30.51

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Streambaaa61218243023.6023.5723.56

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Video On Demandbaaa306090120150SE +/- 0.30, N = 3155.31155.06155.141. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Video On Demandbaaa306090120150Min: 154.78 / Avg: 155.31 / Max: 155.821. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Livebaaa612182430SE +/- 0.05, N = 324.6724.6324.661. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Livebaaa612182430Min: 24.58 / Avg: 24.67 / Max: 24.751. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Streambaaa102030405042.3642.4142.43

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Quad SHA-256, Pyritebaaa20K40K60K80K100KSE +/- 58.12, N = 31020171021801021701. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Quad SHA-256, Pyritebaaa20K40K60K80K100KMin: 101910 / Avg: 102016.67 / Max: 1021101. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Livebaaa4080120160200SE +/- 0.39, N = 3204.73205.03204.751. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Livebaaa4080120160200Min: 204.08 / Avg: 204.73 / Max: 205.411. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streambaaa306090120150125.27125.17125.35

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streambaaa2468107.98247.98907.9776

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streambaaa306090120150112.12112.26112.28

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: AlexNetbaaa20406080100SE +/- 0.06, N = 3109.30109.38109.23
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: AlexNetbaaa20406080100Min: 109.18 / Avg: 109.3 / Max: 109.36

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streambaaa369121512.7612.7512.75

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streambaaa2040608010078.3178.4178.39

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUbaaa510152025SE +/- 0.02, N = 322.5222.5522.54MIN: 21.81MIN: 22.11MIN: 22.241. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUbaaa510152025Min: 22.48 / Avg: 22.52 / Max: 22.561. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Semaphoresbaaa500K1000K1500K2000K2500K2466594.732465656.192463565.681. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

miniBUDE

MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2baaa90180270360450SE +/- 0.03, N = 3410.14410.31409.861. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2baaa70140210280350Min: 410.08 / Avg: 410.14 / Max: 410.181. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2baaa48121620SE +/- 0.00, N = 316.4116.4116.391. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2baaa48121620Min: 16.4 / Avg: 16.41 / Max: 16.411. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: GoogLeNetbaaa71421283528.6528.6728.64

Y-Cruncher

Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.7.10.9513Pi Digits To Calculate: 500Mbaaa510152025SE +/- 0.02, N = 319.5319.5319.55
OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.7.10.9513Pi Digits To Calculate: 500Mbaaa510152025Min: 19.5 / Avg: 19.53 / Max: 19.56

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: AlexNetbaaa20406080100SE +/- 0.06, N = 393.7993.8893.80
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: AlexNetbaaa20406080100Min: 93.68 / Avg: 93.79 / Max: 93.86

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Triple SHA-256, Onecoinbaaa40K80K120K160K200KSE +/- 26.46, N = 32051002049502049701. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Triple SHA-256, Onecoinbaaa40K80K120K160K200KMin: 205060 / Avg: 205100 / Max: 2051501. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streambaaa122436486053.4453.4153.41

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080pbaaa0.16430.32860.49290.65720.8215SE +/- 0.00, N = 30.730.730.731. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080pbaaa246810Min: 0.73 / Avg: 0.73 / Max: 0.731. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4Kbaaa0.05630.11260.16890.22520.2815SE +/- 0.00, N = 30.250.250.251. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4Kbaaa12345Min: 0.25 / Avg: 0.25 / Max: 0.251. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 5 - Resolution: 4Kbaaa0.14850.2970.44550.5940.7425SE +/- 0.00, N = 30.660.660.661. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 5 - Resolution: 4Kbaaa246810Min: 0.66 / Avg: 0.66 / Max: 0.661. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 100baaa0.1530.3060.4590.6120.765SE +/- 0.00, N = 30.680.680.681. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 100baaa246810Min: 0.68 / Avg: 0.68 / Max: 0.691. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 100baaa0.15530.31060.46590.62120.7765SE +/- 0.01, N = 30.690.690.691. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 100baaa246810Min: 0.68 / Avg: 0.69 / Max: 0.71. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUbaaa0.42340.84681.27021.69362.117SE +/- 0.12895, N = 151.741211.278951.88170MIN: 1.02MIN: 1.15MIN: 1.761. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUbaaa246810Min: 1.08 / Avg: 1.74 / Max: 2.731. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUbaaa246810SE +/- 0.14562, N = 156.223965.471258.22740MIN: 4.61MIN: 4.57MIN: 71. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUbaaa3691215Min: 5.48 / Avg: 6.22 / Max: 7.351. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

175 Results Shown

Stress-NG
spaCy
Stress-NG
Cpuminer-Opt
Stress-NG:
  Context Switching
  CPU Stress
  Glibc Qsort Data Sorting
SMHasher
Stress-NG
QuadRay
Stress-NG:
  Futex
  Socket Activity
SMHasher
Stress-NG
SMHasher:
  t1ha0_aes_avx2 x86_64
  FarmHash128
nginx
Stress-NG:
  Memory Copying
  Matrix Math
AOM AV1
Stress-NG
SMHasher:
  fasthash32
  MeowHash x86_64 AES-NI
Stress-NG
JPEG XL libjxl
Stress-NG
EnCodec
nekRS
JPEG XL Decoding libjxl
Libplacebo
OpenRadioss
Xmrig
Cpuminer-Opt
EnCodec:
  6 kbps
  3 kbps
QuadRay:
  1 - 1080p
  3 - 4K
SMHasher
Cpuminer-Opt
SMHasher
JPEG XL libjxl
Neural Magic DeepSparse:
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream:
    ms/batch
    items/sec
spaCy
libavif avifenc
nginx
AOM AV1
nginx
oneDNN
nginx
Cpuminer-Opt
Xmrig
TensorFlow
Libplacebo
Stress-NG
Libplacebo
AOM AV1:
  Speed 6 Two-Pass - Bosphorus 1080p
  Speed 10 Realtime - Bosphorus 4K
QuadRay
AOM AV1
JPEG XL Decoding libjxl
AOM AV1
Neural Magic DeepSparse
Cpuminer-Opt
oneDNN
SMHasher
Stress-NG
AOM AV1
Neural Magic DeepSparse
QuadRay
libavif avifenc
Libplacebo
Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream:
    ms/batch
    items/sec
Y-Cruncher
libavif avifenc
Cpuminer-Opt
Neural Magic DeepSparse:
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
QuadRay
OpenFOAM
oneDNN
Neural Magic DeepSparse
Stress-NG
Neural Magic DeepSparse
EnCodec
FFmpeg:
  libx264 - Platform:
    Seconds
    FPS
  libx265 - Platform:
    FPS
    Seconds
QuadRay
AOM AV1
OpenRadioss:
  Rubber O-Ring Seal Installation
  Bumper Beam
oneDNN
Stress-NG
FLAC Audio Encoding
Stress-NG
miniBUDE:
  OpenMP - BM1:
    GFInst/s
    Billion Interactions/s
OpenFOAM
TensorFlow
libavif avifenc
OpenRadioss
JPEG XL libjxl
FFmpeg
JPEG XL libjxl
FFmpeg
AOM AV1
FFmpeg:
  libx265 - Live
  libx265 - Video On Demand
Cpuminer-Opt:
  x25x
  Deepcoin
TensorFlow
Stress-NG
libavif avifenc
Cpuminer-Opt
AOM AV1
Libplacebo
TensorFlow
AOM AV1
Neural Magic DeepSparse:
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream:
    ms/batch
    items/sec
Stress-NG
TensorFlow:
  CPU - 32 - GoogLeNet
  CPU - 32 - ResNet-50
OpenRadioss
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream:
    ms/batch
    items/sec
FFmpeg:
  libx265 - Upload:
    Seconds
    FPS
AOM AV1
Cpuminer-Opt
TensorFlow
oneDNN
FFmpeg
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream:
    ms/batch
    items/sec
FFmpeg
TensorFlow:
  CPU - 32 - AlexNet
  CPU - 16 - GoogLeNet
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream:
    ms/batch
    items/sec
FFmpeg
TensorFlow
Neural Magic DeepSparse
FFmpeg:
  libx264 - Video On Demand
  libx264 - Live
Neural Magic DeepSparse
Cpuminer-Opt
FFmpeg
Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream:
    ms/batch
    items/sec
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream:
    items/sec
TensorFlow
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream:
    ms/batch
    items/sec
oneDNN
Stress-NG
miniBUDE:
  OpenMP - BM2:
    GFInst/s
    Billion Interactions/s
TensorFlow
Y-Cruncher
TensorFlow
Cpuminer-Opt
Neural Magic DeepSparse
AOM AV1:
  Speed 0 Two-Pass - Bosphorus 1080p
  Speed 0 Two-Pass - Bosphorus 4K
QuadRay
JPEG XL libjxl:
  JPEG - 100
  PNG - 100
oneDNN:
  Matrix Multiply Batch Shapes Transformer - f32 - CPU
  Deconvolution Batch shapes_1d - f32 - CPU

a

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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 Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701021
Graphics Notes: BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D0500100-102
Python Notes: Python 3.10.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + 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 Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 17 November 2022 07:16 by user system.

aa

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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 Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701021
Graphics Notes: BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D0500100-102
Python Notes: Python 3.10.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + 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 Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 17 November 2022 11:45 by user system.

b

Processor: AMD Ryzen 9 3900XT 12-Core @ 3.80GHz (12 Cores / 24 Threads), Motherboard: MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.B3 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 500GB Seagate FireCuda 520 SSD ZP500GM30002, Graphics: AMD Radeon RX 56/64 8GB (1630/945MHz), Audio: AMD Vega 10 HDMI Audio, Monitor: ASUS MG28U, Network: Realtek Device 2600 + Realtek Killer E3000 2.5GbE + Intel Wi-Fi 6 AX200

OS: Ubuntu 22.04, Kernel: 5.15.0-47-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.42), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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 Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701021
Graphics Notes: BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D0500100-102
Python Notes: Python 3.10.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + 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 Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 17 November 2022 19:11 by user system.