3950X April

AMD Ryzen 9 3950X 16-Core testing with a ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS) and Sapphire AMD Radeon RX 470/480/570/570X/580/580X/590 4GB on Ubuntu 20.04 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 2204120-NE-3950XAPRI19
Jump To Table - Results

View

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

Limit displaying results to tests within:

AV1 3 Tests
C/C++ Compiler Tests 3 Tests
CPU Massive 2 Tests
Creator Workloads 6 Tests
Encoding 3 Tests
Common Kernel Benchmarks 2 Tests
Multi-Core 8 Tests
Intel oneAPI 3 Tests
Raytracing 2 Tests
Renderers 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
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
A
April 11 2022
  1 Hour, 26 Minutes
B
April 11 2022
  4 Hours, 15 Minutes
3
April 12 2022
  4 Hours, 4 Minutes
Invert Hiding All Results Option
  3 Hours, 15 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):


3950X AprilProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerVulkanCompilerFile-SystemScreen ResolutionAB3AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS)AMD Starship/Matisse16GBSamsung SSD 970 EVO 250GBSapphire AMD Radeon RX 470/480/570/570X/580/580X/590 4GB (1260/1750MHz)AMD Ellesmere HDMI AudioDELL S2409WIntel I211 + Realtek RTL8822BE 802.11a/b/g/n/acUbuntu 20.045.11.0-43-generic (x86_64)GNOME Shell 3.36.4X Server 1.20.131.2.128GCC 9.3.0ext41920x1080OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8701021Java Details- OpenJDK Runtime Environment (build 11.0.13+8-Ubuntu-0ubuntu1.20.04)Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

AB3Result OverviewPhoronix Test Suite100%101%102%103%OSPrayParallel BZIP2 Compressionperf-benchoneDNNTimed MPlayer CompilationFacebook RocksDBAOM AV1libavif avifenclibgav1Java JMHOSPray Studio

3950X Aprilospray: particle_volume/pathtracer/real_timeospray: particle_volume/scivis/real_timeospray: particle_volume/ao/real_timejava-jmh: Throughputospray-studio: 1 - 1080p - 16 - Path Tracerlibgav1: Chimera 1080p 10-bitospray: gravity_spheres_volume/dim_512/scivis/real_timeospray: gravity_spheres_volume/dim_512/ao/real_timeaom-av1: Speed 0 Two-Pass - Bosphorus 4Kospray-studio: 3 - 1080p - 1 - Path Traceraom-av1: Speed 4 Two-Pass - Bosphorus 4Kospray-studio: 3 - 1080p - 16 - Path Tracerospray: gravity_spheres_volume/dim_512/pathtracer/real_timeospray-studio: 1 - 1080p - 1 - Path Tracerospray-studio: 2 - 1080p - 1 - Path Traceraom-av1: Speed 6 Realtime - Bosphorus 4Kavifenc: 0ospray-studio: 2 - 1080p - 16 - Path Traceronednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUospray-studio: 3 - 1080p - 32 - Path Traceronednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUospray-studio: 2 - 1080p - 32 - Path Tracerospray-studio: 1 - 1080p - 32 - Path Traceraom-av1: Speed 6 Realtime - Bosphorus 1080paom-av1: Speed 6 Two-Pass - Bosphorus 4Kaom-av1: Speed 4 Two-Pass - Bosphorus 1080plibgav1: Chimera 1080procksdb: Read While Writingrocksdb: Update Randrocksdb: Read Rand Write Randrocksdb: Rand Readlibgav1: Summer Nature 4Kavifenc: 2aom-av1: Speed 10 Realtime - Bosphorus 4Kaom-av1: Speed 0 Two-Pass - Bosphorus 1080pperf-bench: Futex Lock-Piaom-av1: Speed 9 Realtime - Bosphorus 4Kperf-bench: Epoll Waitperf-bench: Futex Hashaom-av1: Speed 6 Two-Pass - Bosphorus 1080pbuild-mplayer: Time To Compileonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUperf-bench: Memcpy 1MBlibgav1: Summer Nature 1080paom-av1: Speed 8 Realtime - Bosphorus 4Konednn: IP Shapes 1D - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUaom-av1: Speed 10 Realtime - Bosphorus 1080pperf-bench: Sched Pipeonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUperf-bench: Memset 1MBavifenc: 6, Losslessonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUavifenc: 6aom-av1: Speed 8 Realtime - Bosphorus 1080ponednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUavifenc: 10, Losslessaom-av1: Speed 9 Realtime - Bosphorus 1080pcompress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compressionperf-bench: Syscall Basiconednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUAB3241.18722.413423.465632629111123.5532956651.812.466192.559350.1922166.03353124.063731854190414.47112.112303845135.115129.545117.44769502677.162653.012600.0666843653888.5210.9210.28146.78322205365477523688228193347163.7756.07449.580.4548249.3929976479934028.2621.9514.170831.8473914.470071198.4141.474.660611.29867124.994038830.7894430.67941573.5202239.99811.04470.7905376.45995.7720.616423.24135.622115.94.342204878734.375952.5431241.55122.520923.590832638201689.9182949851.692.508562.613990.1822196.00352994.060951852190614.35112.407302835149.075136.735119.53768912638.952596.922610.4867035652918.3710.9410.29147.37325400065624823831898132905263.6756.38151.470.4546151.5132098481977528.2622.0654.296551.8474314.439690198.0340.214.663991.30043123.894037600.7921540.68081372.6136839.81410.64790.7880606.55594.4920.575923.03975.666118.044.400215921794.400322.55214241.93122.583023.643232625899282.1112954151.752.260832.351830.1822236.00354044.050741850190714.64112.243302965093.465075.495091.02768922581.962556.632540.7166701652968.4510.9410.27146.99325616065480623806138150045963.7156.01952.260.4446150.8533429482352328.1121.9654.218571.8470614.871727198.4840.534.650931.30085122.974032270.7955310.70167669.6814759.89910.28180.7893306.56596.4920.525022.98905.596117.744.348204074334.372832.56316OpenBenchmarking.org

OSPray

OpenBenchmarking.orgItems Per Second, More Is BetterOSPray 2.9Benchmark: particle_volume/pathtracer/real_timeAB350100150200250SE +/- 0.20, N = 3SE +/- 0.34, N = 3241.19241.55241.93

OpenBenchmarking.orgItems Per Second, More Is BetterOSPray 2.9Benchmark: particle_volume/scivis/real_timeAB3510152025SE +/- 0.04, N = 3SE +/- 0.05, N = 322.4122.5222.58

OpenBenchmarking.orgItems Per Second, More Is BetterOSPray 2.9Benchmark: particle_volume/ao/real_timeAB3612182430SE +/- 0.02, N = 3SE +/- 0.01, N = 323.4723.5923.64

Java JMH

This very basic test profile runs the stock benchmark of the Java JMH benchmark via Maven. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/s, More Is BetterJava JMHThroughputAB37000M14000M21000M28000M35000M32629111123.5532638201689.9232625899282.11

OSPray Studio

Intel OSPray Studio is an open-source, interactive visualization and ray-tracing software package. OSPray Studio makes use of Intel OSPray, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPray Studio 0.10Camera: 1 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path TracerAB36K12K18K24K30KSE +/- 13.72, N = 32956629498295411. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.17Video Input: Chimera 1080p 10-bitAB31224364860SE +/- 0.10, N = 3SE +/- 0.01, N = 351.8151.6951.751. (CXX) g++ options: -O3 -lpthread -lrt

OSPray

OpenBenchmarking.orgItems Per Second, More Is BetterOSPray 2.9Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeAB30.56441.12881.69322.25762.822SE +/- 0.00142, N = 3SE +/- 0.00092, N = 32.466192.508562.26083

OpenBenchmarking.orgItems Per Second, More Is BetterOSPray 2.9Benchmark: gravity_spheres_volume/dim_512/ao/real_timeAB30.58811.17621.76432.35242.9405SE +/- 0.00062, N = 3SE +/- 0.00623, N = 32.559352.613992.35183

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4KAB30.04280.08560.12840.17120.214SE +/- 0.00, N = 3SE +/- 0.00, N = 40.190.180.181. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OSPray Studio

Intel OSPray Studio is an open-source, interactive visualization and ray-tracing software package. OSPray Studio makes use of Intel OSPray, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPray Studio 0.10Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path TracerAB35001000150020002500SE +/- 1.76, N = 32216221922231. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4KAB3246810SE +/- 0.01, N = 3SE +/- 0.02, N = 36.036.006.001. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OSPray Studio

Intel OSPray Studio is an open-source, interactive visualization and ray-tracing software package. OSPray Studio makes use of Intel OSPray, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPray Studio 0.10Camera: 3 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path TracerAB38K16K24K32K40KSE +/- 21.28, N = 3SE +/- 45.40, N = 33531235299354041. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread

OSPray

OpenBenchmarking.orgItems Per Second, More Is BetterOSPray 2.9Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeAB30.91431.82862.74293.65724.5715SE +/- 0.00423, N = 3SE +/- 0.00418, N = 34.063734.060954.05074

OSPray Studio

Intel OSPray Studio is an open-source, interactive visualization and ray-tracing software package. OSPray Studio makes use of Intel OSPray, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPray Studio 0.10Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path TracerAB3400800120016002000SE +/- 0.67, N = 3SE +/- 0.67, N = 31854185218501. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterOSPray Studio 0.10Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path TracerAB3400800120016002000SE +/- 1.86, N = 3SE +/- 2.65, N = 31904190619071. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KAB348121620SE +/- 0.11, N = 15SE +/- 0.18, N = 314.4714.3514.641. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -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 0.10Encoder Speed: 0AB3306090120150SE +/- 0.41, N = 3SE +/- 0.34, N = 3112.11112.41112.241. (CXX) g++ options: -O3 -fPIC -lm

OSPray Studio

Intel OSPray Studio is an open-source, interactive visualization and ray-tracing software package. OSPray Studio makes use of Intel OSPray, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPray Studio 0.10Camera: 2 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path TracerAB37K14K21K28K35KSE +/- 17.17, N = 3SE +/- 11.15, N = 33038430283302961. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUAB311002200330044005500SE +/- 12.81, N = 3SE +/- 56.24, N = 35135.115149.075093.46MIN: 5111.68MIN: 5102.61MIN: 4939.211. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUAB311002200330044005500SE +/- 21.94, N = 3SE +/- 15.57, N = 35129.545136.735075.49MIN: 5095.07MIN: 5068.06MIN: 5015.991. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUAB311002200330044005500SE +/- 34.54, N = 3SE +/- 19.61, N = 35117.445119.535091.02MIN: 5086.78MIN: 5034.39MIN: 5027.141. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OSPray Studio

Intel OSPray Studio is an open-source, interactive visualization and ray-tracing software package. OSPray Studio makes use of Intel OSPray, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPray Studio 0.10Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path TracerAB316K32K48K64K80KSE +/- 55.43, N = 3SE +/- 50.93, N = 37695076891768921. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUAB36001200180024003000SE +/- 26.03, N = 3SE +/- 28.67, N = 32677.162638.952581.96MIN: 2661.56MIN: 2574.31MIN: 2502.811. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUAB36001200180024003000SE +/- 20.31, N = 3SE +/- 15.15, N = 32653.012596.922556.63MIN: 2610.08MIN: 2539.82MIN: 2510.011. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUAB36001200180024003000SE +/- 41.03, N = 3SE +/- 7.94, N = 32600.062610.482540.71MIN: 2582.7MIN: 2529.82MIN: 2495.841. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OSPray Studio

Intel OSPray Studio is an open-source, interactive visualization and ray-tracing software package. OSPray Studio makes use of Intel OSPray, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPray Studio 0.10Camera: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path TracerAB314K28K42K56K70KSE +/- 63.26, N = 3SE +/- 128.74, N = 36684367035667011. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterOSPray Studio 0.10Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path TracerAB314K28K42K56K70KSE +/- 76.21, N = 3SE +/- 54.01, N = 36538865291652961. (CXX) g++ options: -O3 -lm -ldl -lpthread -pthread

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080pAB3246810SE +/- 0.03, N = 3SE +/- 0.02, N = 38.528.378.451. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4KAB33691215SE +/- 0.04, N = 3SE +/- 0.02, N = 310.9210.9410.941. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080pAB33691215SE +/- 0.02, N = 3SE +/- 0.01, N = 310.2810.2910.271. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.17Video Input: Chimera 1080pAB3306090120150SE +/- 0.37, N = 3SE +/- 0.08, N = 3146.78147.37146.991. (CXX) g++ options: -O3 -lpthread -lrt

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.0.1Test: Read While WritingAB3700K1400K2100K2800K3500KSE +/- 9707.64, N = 3SE +/- 10292.52, N = 33222053325400032561601. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.0.1Test: Update RandomAB3140K280K420K560K700KSE +/- 1055.36, N = 3SE +/- 1748.11, N = 36547756562486548061. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.0.1Test: Read Random Write RandomAB3500K1000K1500K2000K2500KSE +/- 18963.04, N = 3SE +/- 5611.13, N = 32368822238318923806131. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.0.1Test: Random ReadAB320M40M60M80M100MSE +/- 886019.83, N = 3SE +/- 310806.70, N = 38193347181329052815004591. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.17Video Input: Summer Nature 4KAB31428425670SE +/- 0.06, N = 3SE +/- 0.07, N = 363.7763.6763.711. (CXX) g++ options: -O3 -lpthread -lrt

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 0.10Encoder Speed: 2AB31326395265SE +/- 0.26, N = 3SE +/- 0.08, N = 356.0756.3856.021. (CXX) g++ options: -O3 -fPIC -lm

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4KAB31224364860SE +/- 0.66, N = 15SE +/- 0.82, N = 1549.5851.4752.261. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080pAB30.10130.20260.30390.40520.5065SE +/- 0.00, N = 3SE +/- 0.00, N = 30.450.450.441. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex Lock-PiAB3100200300400500SE +/- 5.21, N = 6SE +/- 5.17, N = 34824614611. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KAB31224364860SE +/- 0.52, N = 15SE +/- 0.75, N = 349.3951.5150.851. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Epoll WaitAB37K14K21K28K35KSE +/- 344.15, N = 3SE +/- 354.56, N = 32997632098334291. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex HashAB31000K2000K3000K4000K5000KSE +/- 3374.73, N = 3SE +/- 6656.67, N = 34799340481977548235231. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080pAB3714212835SE +/- 0.04, N = 3SE +/- 0.07, N = 328.2628.2628.111. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Timed MPlayer Compilation

This test times how long it takes to build the MPlayer open-source media player program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MPlayer Compilation 1.5Time To CompileAB3510152025SE +/- 0.06, N = 3SE +/- 0.02, N = 321.9522.0721.97

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUAB30.96671.93342.90013.86684.8335SE +/- 0.02203, N = 3SE +/- 0.02789, N = 34.170834.296554.21857MIN: 3.93MIN: 3.93MIN: 3.91. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUAB30.41570.83141.24711.66282.0785SE +/- 0.00549, N = 3SE +/- 0.00506, N = 31.847391.847431.84706MIN: 1.78MIN: 1.78MIN: 1.791. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memcpy 1MBAB348121620SE +/- 0.20, N = 3SE +/- 0.09, N = 314.4714.4414.871. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.17Video Input: Summer Nature 1080pAB34080120160200SE +/- 0.14, N = 3SE +/- 0.16, N = 3198.41198.03198.481. (CXX) g++ options: -O3 -lpthread -lrt

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4KAB3918273645SE +/- 0.41, N = 3SE +/- 0.66, N = 341.4740.2140.531. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUAB31.04942.09883.14824.19765.247SE +/- 0.00368, N = 3SE +/- 0.00958, N = 34.660614.663994.65093MIN: 4.46MIN: 4.46MIN: 4.431. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUAB30.29270.58540.87811.17081.4635SE +/- 0.00156, N = 3SE +/- 0.00084, N = 31.298671.300431.30085MIN: 1.28MIN: 1.28MIN: 1.281. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080pAB3306090120150SE +/- 0.98, N = 14SE +/- 1.64, N = 4124.99123.89122.971. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched PipeAB390K180K270K360K450KSE +/- 3062.78, N = 3SE +/- 2490.82, N = 34038834037604032271. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUAB30.1790.3580.5370.7160.895SE +/- 0.004119, N = 3SE +/- 0.000562, N = 30.7894430.7921540.795531MIN: 0.74MIN: 0.74MIN: 0.751. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUAB30.15790.31580.47370.63160.7895SE +/- 0.001462, N = 3SE +/- 0.005860, N = 30.6794150.6808130.701676MIN: 0.62MIN: 0.61MIN: 0.631. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memset 1MBAB31632486480SE +/- 0.10, N = 3SE +/- 0.77, N = 773.5272.6169.681. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma

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 0.10Encoder Speed: 6, LosslessAB33691215SE +/- 0.019, N = 3SE +/- 0.107, N = 39.9989.8149.8991. (CXX) g++ options: -O3 -fPIC -lm

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUAB33691215SE +/- 0.02, N = 3SE +/- 0.03, N = 311.0410.6510.28MIN: 10.91MIN: 10.46MIN: 9.941. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUAB30.17790.35580.53370.71160.8895SE +/- 0.004439, N = 3SE +/- 0.006963, N = 30.7905370.7880600.789330MIN: 0.75MIN: 0.73MIN: 0.731. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

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 0.10Encoder Speed: 6AB3246810SE +/- 0.010, N = 3SE +/- 0.013, N = 36.4596.5556.5651. (CXX) g++ options: -O3 -fPIC -lm

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080pAB320406080100SE +/- 0.34, N = 3SE +/- 0.30, N = 395.7794.4996.491. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUAB3510152025SE +/- 0.00, N = 3SE +/- 0.00, N = 320.6220.5820.53MIN: 20.48MIN: 20.41MIN: 20.371. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUAB3612182430SE +/- 0.02, N = 3SE +/- 0.01, N = 323.2423.0422.99MIN: 23.08MIN: 22.84MIN: 22.711. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

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 0.10Encoder Speed: 10, LosslessAB31.27492.54983.82475.09966.3745SE +/- 0.046, N = 3SE +/- 0.019, N = 35.6225.6665.5961. (CXX) g++ options: -O3 -fPIC -lm

AOM AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.3Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080pAB3306090120150SE +/- 0.58, N = 3SE +/- 0.43, N = 3115.90118.04117.741. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Parallel BZIP2 Compression

This test measures the time needed to compress a file (FreeBSD-13.0-RELEASE-amd64-memstick.img) using Parallel BZIP2 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.13FreeBSD-13.0-RELEASE-amd64-memstick.img CompressionAB30.991.982.973.964.95SE +/- 0.013, N = 3SE +/- 0.060, N = 44.3424.4004.3481. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Syscall BasicAB35M10M15M20M25MSE +/- 156358.13, N = 3SE +/- 147824.81, N = 32048787321592179204074331. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -O6 -ggdb3 -funwind-tables -std=gnu99 -fPIC -lnuma

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUAB30.99011.98022.97033.96044.9505SE +/- 0.02200, N = 3SE +/- 0.01112, N = 34.375954.400324.37283MIN: 4.28MIN: 4.28MIN: 4.261. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUAB30.57671.15341.73012.30682.8835SE +/- 0.00703, N = 3SE +/- 0.01307, N = 32.543102.552142.56316MIN: 2.48MIN: 2.44MIN: 2.461. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -lpthread -ldl

Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU

A: The test run did not produce a result.

B: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

3: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU

A: The test run did not produce a result.

B: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

3: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU

A: The test run did not produce a result.

B: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

3: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU

A: The test run did not produce a result.

B: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

3: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU

A: The test run did not produce a result.

B: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

3: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU

A: The test run did not produce a result.

B: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

3: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

70 Results Shown

OSPray:
  particle_volume/pathtracer/real_time
  particle_volume/scivis/real_time
  particle_volume/ao/real_time
Java JMH
OSPray Studio
libgav1
OSPray:
  gravity_spheres_volume/dim_512/scivis/real_time
  gravity_spheres_volume/dim_512/ao/real_time
AOM AV1
OSPray Studio
AOM AV1
OSPray Studio
OSPray
OSPray Studio:
  1 - 1080p - 1 - Path Tracer
  2 - 1080p - 1 - Path Tracer
AOM AV1
libavif avifenc
OSPray Studio
oneDNN:
  Recurrent Neural Network Training - bf16bf16bf16 - CPU
  Recurrent Neural Network Training - f32 - CPU
  Recurrent Neural Network Training - u8s8f32 - CPU
OSPray Studio
oneDNN:
  Recurrent Neural Network Inference - f32 - CPU
  Recurrent Neural Network Inference - bf16bf16bf16 - CPU
  Recurrent Neural Network Inference - u8s8f32 - CPU
OSPray Studio:
  2 - 1080p - 32 - Path Tracer
  1 - 1080p - 32 - Path Tracer
AOM AV1:
  Speed 6 Realtime - Bosphorus 1080p
  Speed 6 Two-Pass - Bosphorus 4K
  Speed 4 Two-Pass - Bosphorus 1080p
libgav1
Facebook RocksDB:
  Read While Writing
  Update Rand
  Read Rand Write Rand
  Rand Read
libgav1
libavif avifenc
AOM AV1:
  Speed 10 Realtime - Bosphorus 4K
  Speed 0 Two-Pass - Bosphorus 1080p
perf-bench
AOM AV1
perf-bench:
  Epoll Wait
  Futex Hash
AOM AV1
Timed MPlayer Compilation
oneDNN:
  Deconvolution Batch shapes_1d - f32 - CPU
  Deconvolution Batch shapes_1d - u8s8f32 - CPU
perf-bench
libgav1
AOM AV1
oneDNN:
  IP Shapes 1D - f32 - CPU
  IP Shapes 1D - u8s8f32 - CPU
AOM AV1
perf-bench
oneDNN:
  Matrix Multiply Batch Shapes Transformer - f32 - CPU
  Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU
perf-bench
libavif avifenc
oneDNN:
  IP Shapes 3D - f32 - CPU
  IP Shapes 3D - u8s8f32 - CPU
libavif avifenc
AOM AV1
oneDNN:
  Convolution Batch Shapes Auto - f32 - CPU
  Convolution Batch Shapes Auto - u8s8f32 - CPU
libavif avifenc
AOM AV1
Parallel BZIP2 Compression
perf-bench
oneDNN:
  Deconvolution Batch shapes_3d - f32 - CPU
  Deconvolution Batch shapes_3d - u8s8f32 - CPU