Clear Sapphire

2 x AMD EPYC 9654 96-Core testing with a AMD Titanite_4G (RTI1002E BIOS) and ASPEED on Clear Linux OS 38100 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 2301263-NE-CLEARSAPP02
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
BLAS (Basic Linear Algebra Sub-Routine) Tests 2 Tests
C/C++ Compiler Tests 4 Tests
CPU Massive 15 Tests
Creator Workloads 12 Tests
Database Test Suite 3 Tests
Encoding 4 Tests
Fortran Tests 3 Tests
Game Development 4 Tests
HPC - High Performance Computing 10 Tests
Java 2 Tests
Common Kernel Benchmarks 2 Tests
Machine Learning 4 Tests
Molecular Dynamics 4 Tests
MPI Benchmarks 3 Tests
Multi-Core 16 Tests
NVIDIA GPU Compute 3 Tests
Intel oneAPI 5 Tests
OpenMPI Tests 4 Tests
Programmer / Developer System Benchmarks 3 Tests
Python 3 Tests
Renderers 2 Tests
Scientific Computing 4 Tests
Server 5 Tests
Server CPU Tests 12 Tests
Single-Threaded 2 Tests
Video Encoding 4 Tests
Common Workstation Benchmarks 2 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results 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
Clear Linux: 8490H 2P
January 23 2023
  9 Hours, 41 Minutes
Clear Linux: EPYC 9654 2P
January 25 2023
  14 Hours, 5 Minutes
Clear Linux
January 26 2023
  9 Hours, 54 Minutes
Invert Hiding All Results Option
  11 Hours, 13 Minutes

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

Clear Linux: 8490H 2P

Processor: 2 x Intel Xeon Platinum 8490H @ 3.50GHz (120 Cores / 240 Threads), Motherboard: Quanta Cloud S6Q-MB-MPS (3A10.uh BIOS), Chipset: Intel Device 1bce, Memory: 1008GB, Disk: 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007 + 960GB INTEL SSDSC2KG96, Graphics: ASPEED, Monitor: VGA HDMI, Network: 4 x Intel E810-C for QSFP + 2 x Intel X710 for 10GBASE-T

OS: Clear Linux OS 38100, Kernel: 6.1.7-1247.native (x86_64), Display Server: X Server 1.21.1.6, Compiler: GCC 12.2.1 20230112 releases/gcc-12.2.0-345-gc142b30b3a + Clang 15.0.6 + LLVM 15.0.6, File-System: ext4, Screen Resolution: 640x480

Kernel Notes: Transparent Huge Pages: always
Environment Notes: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop -fvisibility-inlines-hidden -Wl,--enable-new-dtags -std=gnu++17" FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,-sort-common -Wl,--enable-new-dtags" CFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx""
Compiler Notes: --build=x86_64-generic-linux --disable-libmpx --disable-libunwind-exceptions --disable-multiarch --disable-vtable-verify --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-clocale=gnu --enable-default-pie --enable-gnu-indirect-function --enable-gnu-indirect-function --enable-host-shared --enable-languages=c,c++,fortran,go,jit --enable-ld=default --enable-libstdcxx-pch --enable-linux-futex --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=x86-64-v3 --with-gcc-major-version-only --with-glibc-version=2.35 --with-gnu-ld --with-isl --with-pic --with-ppl=yes --with-tune=sapphirerapids --with-zstd
Processor Notes: Scaling Governor: intel_pstate performance (EPP: default) - CPU Microcode: 0x2b0000c0
Java Notes: OpenJDK Runtime Environment (build 18.0.2-internal+0-adhoc.mockbuild.corretto-18-18.0.2.9.1)
Python Notes: Python 3.11.0
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

Clear Linux: EPYC 9654 2P

Kernel Notes: Transparent Huge Pages: always
Environment Notes: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop -fvisibility-inlines-hidden -Wl,--enable-new-dtags -std=gnu++17" FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,-sort-common -Wl,--enable-new-dtags" CFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx""
Compiler Notes: --build=x86_64-generic-linux --disable-libmpx --disable-libunwind-exceptions --disable-multiarch --disable-vtable-verify --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-clocale=gnu --enable-default-pie --enable-gnu-indirect-function --enable-gnu-indirect-function --enable-host-shared --enable-languages=c,c++,fortran,go,jit --enable-ld=default --enable-libstdcxx-pch --enable-linux-futex --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=x86-64-v3 --with-gcc-major-version-only --with-glibc-version=2.35 --with-gnu-ld --with-isl --with-pic --with-ppl=yes --with-tune=sapphirerapids --with-zstd
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Java Notes: OpenJDK Runtime Environment (build 18.0.2-internal+0-adhoc.mockbuild.corretto-18-18.0.2.9.1)
Python Notes: Python 3.11.0
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Clear Linux

Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

Clear SapphireProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDisplay ServerCompilerFile-SystemScreen ResolutionClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2 x Intel Xeon Platinum 8490H @ 3.50GHz (120 Cores / 240 Threads)Quanta Cloud S6Q-MB-MPS (3A10.uh BIOS)Intel Device 1bce1008GB2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007 + 960GB INTEL SSDSC2KG96ASPEEDVGA HDMI4 x Intel E810-C for QSFP + 2 x Intel X710 for 10GBASE-TClear Linux OS 381006.1.7-1247.native (x86_64)X Server 1.21.1.6GCC 12.2.1 20230112 releases/gcc-12.2.0-345-gc142b30b3a + Clang 15.0.6 + LLVM 15.0.6ext4640x4802 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads)AMD Titanite_4G (RTI1002E BIOS)AMD Device 14a41520GB2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007Broadcom NetXtreme BCM5720 PCIeOpenBenchmarking.orgKernel Details- Transparent Huge Pages: alwaysEnvironment Details- FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop -fvisibility-inlines-hidden -Wl,--enable-new-dtags -std=gnu++17" FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,-sort-common -Wl,--enable-new-dtags" CFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx"" Compiler Details- --build=x86_64-generic-linux --disable-libmpx --disable-libunwind-exceptions --disable-multiarch --disable-vtable-verify --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-clocale=gnu --enable-default-pie --enable-gnu-indirect-function --enable-gnu-indirect-function --enable-host-shared --enable-languages=c,c++,fortran,go,jit --enable-ld=default --enable-libstdcxx-pch --enable-linux-futex --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=x86-64-v3 --with-gcc-major-version-only --with-glibc-version=2.35 --with-gnu-ld --with-isl --with-pic --with-ppl=yes --with-tune=sapphirerapids --with-zstd Processor Details- Clear Linux: 8490H 2P: Scaling Governor: intel_pstate performance (EPP: default) - CPU Microcode: 0x2b0000c0- Clear Linux: EPYC 9654 2P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110d- Clear Linux: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Details- OpenJDK Runtime Environment (build 18.0.2-internal+0-adhoc.mockbuild.corretto-18-18.0.2.9.1)Python Details- Python 3.11.0Security Details- Clear Linux: 8490H 2P: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected - Clear Linux: EPYC 9654 2P: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - Clear Linux: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear LinuxResult OverviewPhoronix Test Suite100%138%175%213%251%oneDNNONNX RuntimeNAMDminiBUDEHigh Performance Conjugate GradientStress-NGBlenderEmbreeLiquid-DSPXcompact3d Incompact3dCpuminer-OptLeelaChessZeroIntel Open Image DenoiseKvazaarZstd CompressionDaCapo BenchmarkNWChemGROMACSOSPRayNumenta Anomaly Benchmarkuvg266ASTC EncoderOpenVKLlibavif avifencPyBenchMariaDBSVT-AV1PostgreSQLRenaissancePHPBenchsimdjsonPyPerformance

Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear LinuxPer Watt Result OverviewPhoronix Test Suite100%139%178%218%257%miniBUDEZstd CompressionEmbreeStress-NGKvazaarHigh Performance Conjugate GradientSVT-AV1Liquid-DSPASTC Encoderuvg266simdjsonCpuminer-OptGROMACSOSPRayONNX RuntimeOpenVKLMariaDBPHPBenchIntel Open Image DenoiseLeelaChessZeroP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.M

Clear Sapphireonednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPUstress-ng: Context Switchingstress-ng: Memory Copyingonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonnx: yolov4 - CPU - Standardstress-ng: Mutexminibude: OpenMP - BM1minibude: OpenMP - BM1stress-ng: Malloconnx: fcn-resnet101-11 - CPU - Standardnamd: ATPase Simulation - 327,506 Atomsstress-ng: Forkinghpcg: kvazaar: Bosphorus 4K - Slowkvazaar: Bosphorus 4K - Mediumblender: Barbershop - CPU-Onlystress-ng: Vector Mathminibude: OpenMP - BM2minibude: OpenMP - BM2stress-ng: Semaphoresembree: Pathtracer ISPC - Crownliquid-dsp: 240 - 256 - 57onednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUblender: Classroom - CPU-Onlyonnx: bertsquad-12 - CPU - Standardstress-ng: Glibc C String Functionsstress-ng: SENDFILErenaissance: Akka Unbalanced Cobwebbed Treecompress-zstd: 8 - Compression Speeduvg266: Bosphorus 4K - Mediumembree: Pathtracer ISPC - Asian Dragoncpuminer-opt: Triple SHA-256, Onecoinembree: Pathtracer ISPC - Asian Dragon Objcpuminer-opt: LBC, LBRY Creditscompress-zstd: 8, Long Mode - Decompression Speedcpuminer-opt: x25xincompact3d: input.i3d 193 Cells Per Directionstress-ng: Cryptostress-ng: Matrix Mathcompress-zstd: 19 - Decompression Speedcompress-zstd: 8 - Decompression Speedastcenc: Exhaustivelczero: Eigenoidn: RT.hdr_alb_nrm.3840x2160liquid-dsp: 128 - 256 - 57compress-zstd: 19, Long Mode - Decompression Speeddacapobench: H2stress-ng: MMAPastcenc: Thoroughstress-ng: MEMFDcompress-zstd: 8, Long Mode - Compression Speedcpuminer-opt: scryptnumenta-nab: KNN CADnwchem: C240 Buckyballnumenta-nab: Earthgecko Skylineospray: particle_volume/ao/real_timeastcenc: Fastavifenc: 6renaissance: Genetic Algorithm Using Jenetics + Futurespyperformance: nbodyrenaissance: Apache Spark Bayesnumenta-nab: Bayesian Changepointpyperformance: regex_compilecompress-zstd: 19 - Compression Speednumenta-nab: Windowed Gaussianospray: gravity_spheres_volume/dim_512/ao/real_timeopenvkl: vklBenchmark ISPCdacapobench: Jythonmysqlslap: 1024pybench: Total For Average Test Timesnumenta-nab: Contextual Anomaly Detector OSEavifenc: 2pyperformance: json_loadsnumenta-nab: Relative Entropykvazaar: Bosphorus 4K - Super Fastrenaissance: Apache Spark ALSastcenc: Mediumavifenc: 0renaissance: Rand Forestkvazaar: Bosphorus 4K - Ultra Fastrenaissance: Apache Spark PageRankkvazaar: Bosphorus 4K - Very Faststress-ng: IO_uringrenaissance: ALS Movie Lenspyperformance: floatpyperformance: python_startupuvg266: Bosphorus 4K - Very Fastuvg266: Bosphorus 4K - Super Fastuvg266: Bosphorus 4K - Ultra Fastpyperformance: 2to3pyperformance: raytracesimdjson: LargeRandrenaissance: Finagle HTTP Requestspyperformance: pickle_pure_pythonpyperformance: crypto_pyaespyperformance: gopyperformance: django_templatesimdjson: PartialTweetsphpbench: PHP Benchmark Suitepyperformance: pathlibsimdjson: TopTweetstress-ng: System V Message Passingpyperformance: chaosclickhouse: 100M Rows Hits Dataset, Third Runclickhouse: 100M Rows Hits Dataset, Second Runclickhouse: 100M Rows Hits Dataset, First Run / Cold Cachebrl-cad: VGR Performance Metricstress-ng: Futexstress-ng: NUMAstress-ng: CPU Cachestress-ng: Atomicstress-ng: Socket Activitypgbench: 100 - 1000 - Read Write - Average Latencypgbench: 100 - 1000 - Read Writepgbench: 100 - 800 - Read Write - Average Latencypgbench: 100 - 800 - Read Writesvt-av1: Preset 12 - Bosphorus 4Ksvt-av1: Preset 13 - Bosphorus 4Krenaissance: Scala Dottyrenaissance: Savina Reactors.IOgromacs: MPI CPU - water_GMX50_bareonednn: IP Shapes 3D - bf16bf16bf16 - CPUonednn: IP Shapes 1D - bf16bf16bf16 - CPUonnx: ArcFace ResNet-100 - CPU - Standardonnx: GPT-2 - CPU - Standardonnx: super-resolution-10 - CPU - Standardcompress-zstd: 19, Long Mode - Compression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.56061627204245.357851.21489.12756459688403.75132.2893307.227363422967.845070.2630756447.7562.446233.2733.92147.37534773.07183.9634599.07220839432.5599.144172097666670.249854586.43836.5386819260094.312140625.3217042.92942.929.13131.34812094367114.22666674383243.65002.696.18280959203339.63551787.072635.93204.28.4382118944.2871727333332795.324944223.4685.84913631.89822.53644.0586.2751834.366.81434.6997821.58462.4532026.263.3449.021.69510497.84.27038.09611142267979266432.44237.55718.310.75970.412073.0525.401367.334646.572.142704.170.5851657.619105.653.78.3655.8956.8557.282012141.2910387.621958.610926.75.79202200011.96.7434967285.0052.2535.16530.44504.284231422167854.2715.8015.9017822.0798576.8114.2437021310.52376025224.364213.445644.99853.714.9801.519360.9766221739111201170751.62.39780112301150.2625271.941388.72302130850928.40288.9227223.043773890342.112460.128704133.0543.613564.9465.6878.011004273.56343.9518598.75938286893.79180.6866130336666670.4493991037.9620.4163633924057.663160659.4229426.54996.648.11211.28853350626184.273110740534970.97674.586.39028160310827.69823901.713840.64637.912.097983373.0498897000003821.233905646.69114.75772748.611087.84797.2166.6041466.554.26842.5697765.08482.0091677.476.0375.618.62687.6114.23.73944.53771332312570575528.62033.29416.29.64577.941862.2556.679460.577593.479.192551.077.2348522.088554.756.98.0058.4159.2258.661932231.2410803.322560.011227.35.851993673126.6634597345.4452.73097472.1063.0151.40208201.40102162.2414.6086879610.92373652258.442226.934625.09999.318.7111.970204.8191011515981645354.62.37655113443848.3425149.301540.56232127500171.07289.2997232.468779076525.722430.128634099.2085.837665.0165.5478.951003872.99342.4508561.23738317964.95181.4164130476666670.4483531052.70220.4349034041789.973745383.0929003.54933.448.07212.92153387347184.143110750004978.17584.874.17020929310982.35823465.883845.54638.312.070688153.0598868000003842.234225691.19114.78922737.601080.84793.6266.6931458.353.13942.6274672.10212.0071665.075.8387.618.27288.2116.03.64144.42721334311169674930.21233.13516.29.65478.481912.5500.662560.803584.279.762450.576.9148493.708651.957.17.8859.2760.1360.021922241.2410800.022760.311227.45.93197921312.16.6534562392.2252.23071495.2960.8651.20272067.02112224.5314.6956834911.22871275236.887207.354578.08672.417.8281.753154.169198945705541755.7OpenBenchmarking.org

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 the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.53951.0791.61852.1582.6975SE +/- 0.007878, N = 3SE +/- 0.019785, N = 9SE +/- 0.018337, N = 30.5606162.3978002.376550MIN: 0.44MIN: 1.99MIN: 1.971. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
MinAvgMaxClear Linux: 8490H 2P125583708Clear Linux: EPYC 9654 2P47492617Clear Linux46481607OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 3.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 14970 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 12350 Joules per run.3. Clear Linux: Approximate power consumption of 12337 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810Min: 0.55 / Avg: 0.56 / Max: 0.58Min: 2.31 / Avg: 2.4 / Max: 2.52Min: 2.34 / Avg: 2.38 / Max: 2.41. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -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: Context SwitchingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20M40M60M80M100MSE +/- 118201.25, N = 3SE +/- 434697.20, N = 3SE +/- 330195.92, N = 327204245.35112301150.26113443848.341. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: Context SwitchingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux40K80K120K160K200K43149.70188157.68194395.73
MinAvgMaxClear Linux: 8490H 2P135630754Clear Linux: EPYC 9654 2P44597686Clear Linux28584674OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21646 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 20691 Joules per run.3. Clear Linux: Approximate power consumption of 20425 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Context SwitchingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20M40M60M80M100MMin: 27006336.9 / Avg: 27204245.35 / Max: 27415176.32Min: 111509970.36 / Avg: 112301150.26 / Max: 113008847.13Min: 112985973.71 / Avg: 113443848.34 / Max: 114084915.671. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Memory CopyingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux5K10K15K20K25KSE +/- 8.38, N = 3SE +/- 283.80, N = 15SE +/- 102.52, N = 37851.2125271.9425149.301. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: Memory CopyingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux122436486012.5451.9052.72
MinAvgMaxClear Linux: 8490H 2P131626721Clear Linux: EPYC 9654 2P44487561Clear Linux47477546OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21703 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 16587 Joules per run.3. Clear Linux: Approximate power consumption of 16695 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Memory CopyingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4K8K12K16K20KMin: 7836.62 / Avg: 7851.21 / Max: 7865.65Min: 23947.49 / Avg: 25271.94 / Max: 28271.13Min: 24989.14 / Avg: 25149.3 / Max: 25340.251. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -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 the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux30060090012001500SE +/- 2.29, N = 3SE +/- 18.64, N = 3SE +/- 17.06, N = 3489.131388.721540.56MIN: 471.03MIN: 1325.47MIN: 1477.951. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
MinAvgMaxClear Linux: 8490H 2P131653755Clear Linux: EPYC 9654 2P47505607Clear Linux50491598OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 3.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 54868 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 42046 Joules per run.3. Clear Linux: Approximate power consumption of 41378 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux30060090012001500Min: 485.94 / Avg: 489.13 / Max: 493.58Min: 1351.5 / Avg: 1388.72 / Max: 1409.22Min: 1511.1 / Avg: 1540.56 / Max: 1570.21. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux120240360480600SE +/- 0.17, N = 3SE +/- 2.65, N = 12SE +/- 0.60, N = 35643022321. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.20930.41860.62790.83721.04650.9300.6830.533
MinAvgMaxClear Linux: 8490H 2P125606629Clear Linux: EPYC 9654 2P43442461Clear Linux53435455OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 75981 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 54818 Joules per run.3. Clear Linux: Approximate power consumption of 54233 Joules per run.
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux100200300400500Min: 563.5 / Avg: 563.83 / Max: 564Min: 283.5 / Avg: 302 / Max: 309.5Min: 231 / Avg: 232.17 / Max: 2331. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: MutexClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux30M60M90M120M150MSE +/- 124816.60, N = 3SE +/- 1710955.16, N = 3SE +/- 71254.14, N = 359688403.75130850928.40127500171.071. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: MutexClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux50K100K150K200K250K93528.29219027.60215853.94
MinAvgMaxClear Linux: 8490H 2P129638754Clear Linux: EPYC 9654 2P46597688Clear Linux47591679OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21698 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 20910 Joules per run.3. Clear Linux: Approximate power consumption of 20477 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MutexClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20M40M60M80M100MMin: 59546618.4 / Avg: 59688403.75 / Max: 59937229.66Min: 128649649 / Avg: 130850928.4 / Max: 134220461.98Min: 127362057.58 / Avg: 127500171.07 / Max: 127599640.921. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -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.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux60120180240300SE +/- 1.23, N = 7SE +/- 0.24, N = 10SE +/- 0.09, N = 10132.29288.92289.301. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgBillion Interactions/s Per Watt, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.24030.48060.72090.96121.20150.3521.0291.068
MinAvgMaxClear Linux: 8490H 2P124.4376.3582.5Clear Linux: EPYC 9654 2P48.2280.8624.0Clear Linux38.4270.8616.1OpenBenchmarking.orgWatts, Fewer Is BetterminiBUDE 20210901CPU Power Consumption Monitor1603204806408001. Clear Linux: 8490H 2P: Approximate power consumption of 3924 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 2050 Joules per run.3. Clear Linux: Approximate power consumption of 1950 Joules per run.
OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux50100150200250Min: 128.53 / Avg: 132.29 / Max: 137.28Min: 287.23 / Avg: 288.92 / Max: 289.72Min: 288.9 / Avg: 289.3 / Max: 289.741. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux16003200480064008000SE +/- 30.77, N = 7SE +/- 6.09, N = 10SE +/- 2.25, N = 103307.237223.047232.471. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux13002600390052006500Min: 3213.2 / Avg: 3307.23 / Max: 3431.94Min: 7180.66 / Avg: 7223.04 / Max: 7242.91Min: 7222.53 / Avg: 7232.47 / Max: 7243.381. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -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: MallocClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux200M400M600M800M1000MSE +/- 1102560.16, N = 3SE +/- 1462259.31, N = 3SE +/- 1128975.32, N = 3363422967.84773890342.11779076525.721. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: MallocClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux300K600K900K1200K1500K577946.031260057.851284436.46
MinAvgMaxClear Linux: 8490H 2P131629751Clear Linux: EPYC 9654 2P52614713Clear Linux47607706OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21799 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 21087 Joules per run.3. Clear Linux: Approximate power consumption of 21229 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MallocClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux140M280M420M560M700MMin: 362048900.12 / Avg: 363422967.84 / Max: 365603611.46Min: 772216774.13 / Avg: 773890342.11 / Max: 776804139.47Min: 776934311.14 / Avg: 779076525.72 / Max: 780765648.671. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux110220330440550SE +/- 2.92, N = 3SE +/- 3.00, N = 3SE +/- 1.45, N = 35072462431. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.17440.34880.52320.69760.8720.7750.4970.495
MinAvgMaxClear Linux: 8490H 2P123654682Clear Linux: EPYC 9654 2P55494524Clear Linux53491517OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 81302 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 61406 Joules per run.3. Clear Linux: Approximate power consumption of 61071 Joules per run.
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux90180270360450Min: 504 / Avg: 507.17 / Max: 513Min: 241.5 / Avg: 245.67 / Max: 251.5Min: 240.5 / Avg: 243.17 / Max: 245.51. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

NAMD

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.05920.11840.17760.23680.296SE +/- 0.00330, N = 3SE +/- 0.00038, N = 3SE +/- 0.00011, N = 30.263070.128700.12863
MinAvgMaxClear Linux: 8490H 2P123587753Clear Linux: EPYC 9654 2P55306680Clear Linux48315676OpenBenchmarking.orgWatts, Fewer Is BetterNAMD 2.14CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 19967 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 20630 Joules per run.3. Clear Linux: Approximate power consumption of 21448 Joules per run.
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux12345Min: 0.26 / Avg: 0.26 / Max: 0.27Min: 0.13 / Avg: 0.13 / Max: 0.13Min: 0.13 / Avg: 0.13 / Max: 0.13

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: ForkingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux12K24K36K48K60KSE +/- 312.44, N = 3SE +/- 4.75, N = 3SE +/- 8.76, N = 356447.754133.054099.201. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: ForkingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2040608010088.9810.2510.36
MinAvgMaxClear Linux: 8490H 2P137634747Clear Linux: EPYC 9654 2P47403476Clear Linux45396468OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21992 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 13843 Joules per run.3. Clear Linux: Approximate power consumption of 13712 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: ForkingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux10K20K30K40K50KMin: 55963.27 / Avg: 56447.75 / Max: 57031.77Min: 4125.96 / Avg: 4133.05 / Max: 4142.07Min: 4081.78 / Avg: 4099.2 / Max: 4109.461. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100SE +/- 0.14, N = 3SE +/- 0.02, N = 3SE +/- 0.42, N = 362.4543.6185.841. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHigh Performance Conjugate Gradient 3.1Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.03290.06580.09870.13160.16450.0910.0720.146
MinAvgMaxClear Linux: 8490H 2P127684720Clear Linux: EPYC 9654 2P48608663Clear Linux43588683OpenBenchmarking.orgWatts, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 127703 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 182348 Joules per run.3. Clear Linux: Approximate power consumption of 125908 Joules per run.
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1632486480Min: 62.18 / Avg: 62.45 / Max: 62.61Min: 43.57 / Avg: 43.61 / Max: 43.64Min: 85.04 / Avg: 85.84 / Max: 86.481. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: SlowClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1530456075SE +/- 0.04, N = 3SE +/- 0.31, N = 5SE +/- 0.30, N = 533.2764.9465.011. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: SlowClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.0380.0760.1140.1520.190.0670.1690.169
MinAvgMaxClear Linux: 8490H 2P136.3498.5589.1Clear Linux: EPYC 9654 2P53.9383.2589.8Clear Linux52.9384.7582.7OpenBenchmarking.orgWatts, Fewer Is BetterKvazaar 2.2CPU Power Consumption Monitor1603204806408001. Clear Linux: 8490H 2P: Approximate power consumption of 11133 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 5288 Joules per run.3. Clear Linux: Approximate power consumption of 5156 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: SlowClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265Min: 33.2 / Avg: 33.27 / Max: 33.35Min: 63.81 / Avg: 64.94 / Max: 65.45Min: 64.05 / Avg: 65.01 / Max: 65.821. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: MediumClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1530456075SE +/- 0.10, N = 3SE +/- 0.28, N = 5SE +/- 0.17, N = 533.9265.6865.541. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: MediumClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.03890.07780.11670.15560.19450.0690.1710.173
MinAvgMaxClear Linux: 8490H 2P128.1489.5589.9Clear Linux: EPYC 9654 2P52.9384.0584.9Clear Linux51.6378.5580.9OpenBenchmarking.orgWatts, Fewer Is BetterKvazaar 2.2CPU Power Consumption Monitor1603204806408001. Clear Linux: 8490H 2P: Approximate power consumption of 10769 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 5299 Joules per run.3. Clear Linux: Approximate power consumption of 5223 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: MediumClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265Min: 33.75 / Avg: 33.92 / Max: 34.09Min: 65.14 / Avg: 65.68 / Max: 66.65Min: 64.9 / Avg: 65.54 / Max: 65.821. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Barbershop - Compute: CPU-OnlyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux306090120150SE +/- 1.23, N = 3SE +/- 0.12, N = 3SE +/- 0.03, N = 3147.3778.0178.95
MinAvgMaxClear Linux: 8490H 2P128667754Clear Linux: EPYC 9654 2P51627696Clear Linux53623685OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.4CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 102731 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 52268 Joules per run.3. Clear Linux: Approximate power consumption of 52527 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Barbershop - Compute: CPU-OnlyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux306090120150Min: 146.03 / Avg: 147.37 / Max: 149.82Min: 77.82 / Avg: 78.01 / Max: 78.22Min: 78.89 / Avg: 78.95 / Max: 79.01

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 MathClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux200K400K600K800K1000KSE +/- 110.65, N = 3SE +/- 22.18, N = 3SE +/- 23.11, N = 3534773.071004273.561003872.991. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: Vector MathClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux400800120016002000848.121808.771883.36
MinAvgMaxClear Linux: 8490H 2P127631721Clear Linux: EPYC 9654 2P55555638Clear Linux39533626OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21649 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 19248 Joules per run.3. Clear Linux: Approximate power consumption of 19011 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Vector MathClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux200K400K600K800K1000KMin: 534642.7 / Avg: 534773.07 / Max: 534993.12Min: 1004229.44 / Avg: 1004273.56 / Max: 1004299.69Min: 1003833.39 / Avg: 1003872.99 / Max: 1003913.441. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -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.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux70140210280350SE +/- 1.55, N = 3SE +/- 1.68, N = 4SE +/- 1.97, N = 4183.96343.95342.451. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgBillion Interactions/s Per Watt, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.16220.32440.48660.64880.8110.3330.7090.721
MinAvgMaxClear Linux: 8490H 2P130552756Clear Linux: EPYC 9654 2P46485643Clear Linux43475638OpenBenchmarking.orgWatts, Fewer Is BetterminiBUDE 20210901CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 18226 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 9700 Joules per run.3. Clear Linux: Approximate power consumption of 9615 Joules per run.
OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux60120180240300Min: 181.85 / Avg: 183.96 / Max: 186.98Min: 341.99 / Avg: 343.95 / Max: 348.97Min: 337.24 / Avg: 342.45 / Max: 345.531. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KSE +/- 38.68, N = 3SE +/- 42.08, N = 4SE +/- 49.24, N = 44599.078598.768561.241. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux15003000450060007500Min: 4546.23 / Avg: 4599.07 / Max: 4674.43Min: 8549.72 / Avg: 8598.76 / Max: 8724.33Min: 8430.97 / Avg: 8561.24 / Max: 8638.351. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -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: SemaphoresClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux8M16M24M32M40MSE +/- 22111.46, N = 3SE +/- 86170.88, N = 3SE +/- 169599.87, N = 320839432.5538286893.7938317964.951. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: SemaphoresClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20K40K60K80K100K42351.8892423.0293551.52
MinAvgMaxClear Linux: 8490H 2P128.2492.1551.4Clear Linux: EPYC 9654 2P51.2414.3481.3Clear Linux36.4409.6478.5OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 17058 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 14361 Joules per run.3. Clear Linux: Approximate power consumption of 14199 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: SemaphoresClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux7M14M21M28M35MMin: 20810875.07 / Avg: 20839432.55 / Max: 20882953.43Min: 38115884.76 / Avg: 38286893.79 / Max: 38390923.87Min: 38102703 / Avg: 38317964.95 / Max: 38652618.091. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: CrownClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200SE +/- 0.40, N = 6SE +/- 0.50, N = 8SE +/- 0.55, N = 899.14180.69181.42MIN: 78.31 / MAX: 136.06MIN: 133.01 / MAX: 206.61MIN: 140.55 / MAX: 204.2
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: CrownClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.12490.24980.37470.49960.62450.2280.5550.550
MinAvgMaxClear Linux: 8490H 2P133435639Clear Linux: EPYC 9654 2P31325641Clear Linux26330648OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 3.13CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 5152 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 2726 Joules per run.3. Clear Linux: Approximate power consumption of 2802 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: CrownClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux306090120150Min: 98.02 / Avg: 99.14 / Max: 100.59Min: 177.64 / Avg: 180.69 / Max: 182.58Min: 178.34 / Avg: 181.42 / Max: 183.02

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 240 - Buffer Length: 256 - Filter Length: 57Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3000M6000M9000M12000M15000MSE +/- 3877427.56, N = 3SE +/- 3844187.53, N = 3SE +/- 13678856.35, N = 3720976666713033666667130476666671. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 2021.01.31Threads: 240 - Buffer Length: 256 - Filter Length: 57Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux5M10M15M20M25M11939483.4623455551.3523414150.14
MinAvgMaxClear Linux: 8490H 2P120604754Clear Linux: EPYC 9654 2P53556672Clear Linux47557666OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 2021.01.31CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 14694 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 13707 Joules per run.3. Clear Linux: Approximate power consumption of 13560 Joules per run.
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 240 - Buffer Length: 256 - Filter Length: 57Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2000M4000M6000M8000M10000MMin: 7202800000 / Avg: 7209766666.67 / Max: 7216200000Min: 13028000000 / Avg: 13033666666.67 / Max: 13041000000Min: 13033000000 / Avg: 13047666666.67 / Max: 130750000001. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -pthread -lm -lc -lliquid

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 the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.10110.20220.30330.40440.5055SE +/- 0.001245, N = 7SE +/- 0.005428, N = 15SE +/- 0.003761, N = 150.2498540.4493990.448353MIN: 0.21MIN: 0.32MIN: 0.331. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
MinAvgMaxClear Linux: 8490H 2P124458741Clear Linux: EPYC 9654 2P39363600Clear Linux30355587OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 3.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 4641 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 3706 Joules per run.3. Clear Linux: Approximate power consumption of 3649 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux12345Min: 0.25 / Avg: 0.25 / Max: 0.26Min: 0.42 / Avg: 0.45 / Max: 0.49Min: 0.43 / Avg: 0.45 / Max: 0.471. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2004006008001000SE +/- 0.10, N = 3SE +/- 10.05, N = 3SE +/- 9.48, N = 15586.441037.961052.70MIN: 563.39MIN: 990.8MIN: 955.431. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
MinAvgMaxClear Linux: 8490H 2P130653753Clear Linux: EPYC 9654 2P39506594Clear Linux16501585OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 3.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 55513 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 42522 Joules per run.3. Clear Linux: Approximate power consumption of 41563 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2004006008001000Min: 586.25 / Avg: 586.44 / Max: 586.58Min: 1023.65 / Avg: 1037.96 / Max: 1057.35Min: 985.96 / Avg: 1052.7 / Max: 1108.371. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Classroom - Compute: CPU-OnlyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux816243240SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 336.5320.4120.43
MinAvgMaxClear Linux: 8490H 2P130631754Clear Linux: EPYC 9654 2P51554677Clear Linux48539670OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.4CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 26293 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 14026 Joules per run.3. Clear Linux: Approximate power consumption of 13665 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Classroom - Compute: CPU-OnlyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux816243240Min: 36.43 / Avg: 36.53 / Max: 36.64Min: 20.29 / Avg: 20.41 / Max: 20.6Min: 20.38 / Avg: 20.43 / Max: 20.5

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2004006008001000SE +/- 11.88, N = 3SE +/- 1.53, N = 3SE +/- 3.55, N = 38686364901. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.38180.76361.14541.52721.9091.5521.6971.316
MinAvgMaxClear Linux: 8490H 2P123.3559.2577.4Clear Linux: EPYC 9654 2P50.7374.8389.2Clear Linux53.3372.3387.7OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 69899 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 46721 Joules per run.3. Clear Linux: Approximate power consumption of 46412 Joules per run.
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux150300450600750Min: 848.5 / Avg: 868 / Max: 889.5Min: 632.5 / Avg: 635.5 / Max: 637.5Min: 485.5 / Avg: 490 / Max: 4971. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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 FunctionsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux7M14M21M28M35MSE +/- 247947.31, N = 3SE +/- 288633.70, N = 8SE +/- 281854.05, N = 1519260094.3133924057.6634041789.971. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: Glibc C String FunctionsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux11K22K33K44K55K30597.4852936.1753681.46
MinAvgMaxClear Linux: 8490H 2P129629754Clear Linux: EPYC 9654 2P53641730Clear Linux41634721OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21822 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 21869 Joules per run.3. Clear Linux: Approximate power consumption of 21561 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Glibc C String FunctionsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux6M12M18M24M30MMin: 18785422.16 / Avg: 19260094.31 / Max: 19621721.87Min: 32565746.78 / Avg: 33924057.66 / Max: 34843218.77Min: 31561888.87 / Avg: 34041789.97 / Max: 34888473.291. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: SENDFILEClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux800K1600K2400K3200K4000KSE +/- 5240.70, N = 3SE +/- 31076.17, N = 3SE +/- 45875.47, N = 32140625.323160659.423745383.091. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: SENDFILEClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux130026003900520065003409.285282.436257.35
MinAvgMaxClear Linux: 8490H 2P129628752Clear Linux: EPYC 9654 2P53598695Clear Linux48599689OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21557 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 20942 Joules per run.3. Clear Linux: Approximate power consumption of 20750 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: SENDFILEClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux600K1200K1800K2400K3000KMin: 2130833.75 / Avg: 2140625.32 / Max: 2148759.72Min: 3099303.91 / Avg: 3160659.42 / Max: 3199928.57Min: 3658661.35 / Avg: 3745383.09 / Max: 3814689.671. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Akka Unbalanced Cobwebbed TreeClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux6K12K18K24K30KSE +/- 128.63, N = 3SE +/- 370.12, N = 9SE +/- 296.83, N = 317042.929426.529003.5MIN: 12570.6 / MAX: 17239.96MIN: 19199.83 / MAX: 31858.56MIN: 20271.46 / MAX: 29909.89
MinAvgMaxClear Linux: 8490H 2P128.7439.5560.1Clear Linux: EPYC 9654 2P46.9335.9434.2Clear Linux52.5329.2426.4OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 152066 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 202684 Joules per run.3. Clear Linux: Approximate power consumption of 195850 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Akka Unbalanced Cobwebbed TreeClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux5K10K15K20K25KMin: 16801.12 / Avg: 17042.87 / Max: 17239.96Min: 28522.35 / Avg: 29426.48 / Max: 31858.56Min: 28432.45 / Avg: 29003.49 / Max: 29429.61

Zstd Compression

This test measures the time needed to compress/decompress a sample input file using Zstd compression supplied by the system or otherwise externally of the test profile. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 8 - Compression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux11002200330044005500SE +/- 33.68, N = 3SE +/- 21.10, N = 3SE +/- 15.70, N = 32942.94996.64933.41. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***
OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 8 - Compression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux9001800270036004500Min: 2899.9 / Avg: 2942.9 / Max: 3009.3Min: 4954.6 / Avg: 4996.6 / Max: 5021.2Min: 4904.6 / Avg: 4933.43 / Max: 4958.61. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: MediumClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1122334455SE +/- 0.03, N = 3SE +/- 0.36, N = 4SE +/- 0.18, N = 429.1348.1148.071. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17
OpenBenchmarking.orgFrames Per Second Per Watt, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: MediumClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.02520.05040.07560.10080.1260.0530.1100.112
MinAvgMaxClear Linux: 8490H 2P129550644Clear Linux: EPYC 9654 2P47438605Clear Linux48429597OpenBenchmarking.orgWatts, Fewer Is Betteruvg266 0.4.1CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 13751 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 7555 Joules per run.3. Clear Linux: Approximate power consumption of 7183 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: MediumClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1020304050Min: 29.09 / Avg: 29.13 / Max: 29.18Min: 47.04 / Avg: 48.11 / Max: 48.62Min: 47.66 / Avg: 48.07 / Max: 48.541. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian DragonClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux50100150200250SE +/- 0.77, N = 7SE +/- 0.44, N = 9SE +/- 0.40, N = 9131.35211.29212.92MIN: 103.02 / MAX: 173.94MIN: 204.87 / MAX: 224.27MIN: 206.18 / MAX: 223.02
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian DragonClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.16670.33340.50010.66680.83350.2960.7190.741
MinAvgMaxClear Linux: 8490H 2P131444683Clear Linux: EPYC 9654 2P20294631Clear Linux17287629OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 3.13CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 4442 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 2253 Joules per run.3. Clear Linux: Approximate power consumption of 2235 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian DragonClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200Min: 128.42 / Avg: 131.35 / Max: 133.56Min: 209.43 / Avg: 211.29 / Max: 213.63Min: 210.31 / Avg: 212.92 / Max: 214.32

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, OnecoinClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux700K1400K2100K2800K3500KSE +/- 16269.33, N = 3SE +/- 36423.01, N = 5SE +/- 5126.51, N = 32094367335062633873471. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s Per Watt, More Is BetterCpuminer-Opt 3.20.3Algorithm: Triple SHA-256, OnecoinClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux110022003300440055003339.585126.055264.44
MinAvgMaxClear Linux: 8490H 2P123627754Clear Linux: EPYC 9654 2P44654757Clear Linux43643746OpenBenchmarking.orgWatts, Fewer Is BetterCpuminer-Opt 3.20.3CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21741 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 22355 Joules per run.3. Clear Linux: Approximate power consumption of 21662 Joules per run.
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Triple SHA-256, OnecoinClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux600K1200K1800K2400K3000KMin: 2063540 / Avg: 2094366.67 / Max: 2118800Min: 3284540 / Avg: 3350626 / Max: 3484940Min: 3381680 / Avg: 3387346.67 / Max: 33975801. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian Dragon ObjClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200SE +/- 1.61, N = 3SE +/- 0.52, N = 4SE +/- 0.32, N = 4114.23184.27184.14MIN: 89.56 / MAX: 148.79MIN: 179.01 / MAX: 193.84MIN: 179.27 / MAX: 193.3
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian Dragon ObjClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.15930.31860.47790.63720.79650.2910.7080.708
MinAvgMaxClear Linux: 8490H 2P132392695Clear Linux: EPYC 9654 2P23260631Clear Linux15260627OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 3.13CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 9025 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 4618 Joules per run.3. Clear Linux: Approximate power consumption of 4553 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian Dragon ObjClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux306090120150Min: 111.1 / Avg: 114.23 / Max: 116.42Min: 183.05 / Avg: 184.27 / Max: 185.29Min: 183.59 / Avg: 184.14 / Max: 184.97

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 CreditsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux200K400K600K800K1000KSE +/- 7329.77, N = 5SE +/- 256.93, N = 3SE +/- 1402.08, N = 3667438107405310750001. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s Per Watt, More Is BetterCpuminer-Opt 3.20.3Algorithm: LBC, LBRY CreditsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4008001200160020001060.181714.181741.63
MinAvgMaxClear Linux: 8490H 2P126630753Clear Linux: EPYC 9654 2P35627737Clear Linux44617727OpenBenchmarking.orgWatts, Fewer Is BetterCpuminer-Opt 3.20.3CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21531 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 21303 Joules per run.3. Clear Linux: Approximate power consumption of 21192 Joules per run.
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: LBC, LBRY CreditsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux200K400K600K800K1000KMin: 648590 / Avg: 667438 / Max: 685960Min: 1073540 / Avg: 1074053.33 / Max: 1074330Min: 1072250 / Avg: 1075000 / Max: 10768501. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Zstd Compression

This test measures the time needed to compress/decompress a sample input file using Zstd compression supplied by the system or otherwise externally of the test profile. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 8, Long Mode - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux11002200330044005500SE +/- 40.64, N = 6SE +/- 3.40, N = 3SE +/- 12.23, N = 33243.64970.94978.11. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***
OpenBenchmarking.orgMB/s Per Watt, More Is BetterZstd CompressionCompression Level: 8, Long Mode - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux5101520258.78122.79122.757
MinAvgMaxClear Linux: 8490H 2P128.0369.4416.7Clear Linux: EPYC 9654 2P32.9218.1264.4Clear Linux47.3218.8264.2OpenBenchmarking.orgWatts, Fewer Is BetterZstd CompressionCPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 14836 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 8506 Joules per run.3. Clear Linux: Approximate power consumption of 8531 Joules per run.
OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 8, Long Mode - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux9001800270036004500Min: 3135.3 / Avg: 3243.58 / Max: 3429.3Min: 4965.7 / Avg: 4970.93 / Max: 4977.3Min: 4953.8 / Avg: 4978.1 / Max: 4992.71. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***

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: x25xClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux16003200480064008000SE +/- 1.75, N = 3SE +/- 16.54, N = 3SE +/- 29.73, N = 35002.697674.587584.871. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s Per Watt, More Is BetterCpuminer-Opt 3.20.3Algorithm: x25xClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux481216207.98413.60013.661
MinAvgMaxClear Linux: 8490H 2P128627754Clear Linux: EPYC 9654 2P55564647Clear Linux53555639OpenBenchmarking.orgWatts, Fewer Is BetterCpuminer-Opt 3.20.3CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21723 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 19375 Joules per run.3. Clear Linux: Approximate power consumption of 19063 Joules per run.
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: x25xClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux13002600390052006500Min: 4999.23 / Avg: 5002.69 / Max: 5004.87Min: 7643.79 / Avg: 7674.58 / Max: 7700.43Min: 7525.41 / Avg: 7584.87 / Max: 7614.981. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Xcompact3d Incompact3d

Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810SE +/- 0.03285028, N = 6SE +/- 0.03622397, N = 6SE +/- 0.01513938, N = 86.182809596.390281604.170209291. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi
MinAvgMaxClear Linux: 8490H 2P127500749Clear Linux: EPYC 9654 2P44405660Clear Linux41344642OpenBenchmarking.orgWatts, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 5500 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 4595 Joules per run.3. Clear Linux: Approximate power consumption of 3138 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3691215Min: 6.1 / Avg: 6.18 / Max: 6.3Min: 6.26 / Avg: 6.39 / Max: 6.47Min: 4.11 / Avg: 4.17 / Max: 4.251. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi

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: CryptoClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux70K140K210K280K350KSE +/- 149.73, N = 3SE +/- 130.18, N = 3SE +/- 116.86, N = 3203339.63310827.69310982.351. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: CryptoClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux120240360480600324.94560.96567.41
MinAvgMaxClear Linux: 8490H 2P127626711Clear Linux: EPYC 9654 2P45554637Clear Linux44548630OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21694 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 19394 Joules per run.3. Clear Linux: Approximate power consumption of 18817 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CryptoClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux50K100K150K200K250KMin: 203147.96 / Avg: 203339.63 / Max: 203634.72Min: 310673.98 / Avg: 310827.69 / Max: 311086.53Min: 310762.16 / Avg: 310982.35 / Max: 311160.311. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Matrix MathClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux200K400K600K800K1000KSE +/- 234.85, N = 3SE +/- 140.19, N = 3SE +/- 14.38, N = 3551787.07823901.71823465.881. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: Matrix MathClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux30060090012001500877.211448.321461.74
MinAvgMaxClear Linux: 8490H 2P134629754Clear Linux: EPYC 9654 2P55569658Clear Linux47563651OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21806 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 19721 Joules per run.3. Clear Linux: Approximate power consumption of 19529 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Matrix MathClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux140K280K420K560K700KMin: 551399.08 / Avg: 551787.07 / Max: 552210.33Min: 823734.94 / Avg: 823901.71 / Max: 824180.29Min: 823448.14 / Avg: 823465.88 / Max: 823494.351. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

Zstd Compression

This test measures the time needed to compress/decompress a sample input file using Zstd compression supplied by the system or otherwise externally of the test profile. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 19 - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux8001600240032004000SE +/- 18.09, N = 15SE +/- 1.69, N = 7SE +/- 1.71, N = 152635.93840.63845.51. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***
OpenBenchmarking.orgMB/s Per Watt, More Is BetterZstd CompressionCompression Level: 19 - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux481216206.84616.90716.856
MinAvgMaxClear Linux: 8490H 2P128.2385.0500.1Clear Linux: EPYC 9654 2P41.2227.2298.8Clear Linux45.5228.1300.0OpenBenchmarking.orgWatts, Fewer Is BetterZstd CompressionCPU Power Consumption Monitor1302603905206501. Clear Linux: 8490H 2P: Approximate power consumption of 17505 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 9735 Joules per run.3. Clear Linux: Approximate power consumption of 9566 Joules per run.
OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 19 - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux7001400210028003500Min: 2386.9 / Avg: 2635.89 / Max: 2673.4Min: 3834.1 / Avg: 3840.59 / Max: 3847.5Min: 3833.4 / Avg: 3845.49 / Max: 3854.41. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***

OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 8 - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux10002000300040005000SE +/- 23.01, N = 3SE +/- 3.63, N = 3SE +/- 4.71, N = 33204.24637.94638.31. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***
OpenBenchmarking.orgMB/s Per Watt, More Is BetterZstd CompressionCompression Level: 8 - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux5101520257.67818.68118.602
MinAvgMaxClear Linux: 8490H 2P128.2417.3573.3Clear Linux: EPYC 9654 2P43.7248.3347.1Clear Linux34.5249.3345.4OpenBenchmarking.orgWatts, Fewer Is BetterZstd CompressionCPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 17109 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 9848 Joules per run.3. Clear Linux: Approximate power consumption of 9974 Joules per run.
OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 8 - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux8001600240032004000Min: 3159.2 / Avg: 3204.17 / Max: 3235.1Min: 4632.8 / Avg: 4637.87 / Max: 4644.9Min: 4630 / Avg: 4638.27 / Max: 4646.31. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: ExhaustiveClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3691215SE +/- 0.0035, N = 5SE +/- 0.0219, N = 6SE +/- 0.0116, N = 68.438212.097912.07061. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -flto -pthread
OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.0Preset: ExhaustiveClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.00880.01760.02640.03520.0440.0200.0390.039
MinAvgMaxClear Linux: 8490H 2P132431754Clear Linux: EPYC 9654 2P51309666Clear Linux47307667OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 6378 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 3753 Joules per run.3. Clear Linux: Approximate power consumption of 3730 Joules per run.
OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: ExhaustiveClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux48121620Min: 8.43 / Avg: 8.44 / Max: 8.45Min: 12.04 / Avg: 12.1 / Max: 12.19Min: 12.04 / Avg: 12.07 / Max: 12.11. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -flto -pthread

LeelaChessZero

LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: EigenClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3K6K9K12K15KSE +/- 139.81, N = 3SE +/- 108.40, N = 9SE +/- 49.36, N = 311894833788151. (CXX) g++ options: -flto -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -pthread
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.28Backend: EigenClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux61218243022.1022.3723.49
MinAvgMaxClear Linux: 8490H 2P128538703Clear Linux: EPYC 9654 2P46373509Clear Linux53375509OpenBenchmarking.orgWatts, Fewer Is BetterLeelaChessZero 0.28CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 198063 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 138259 Joules per run.3. Clear Linux: Approximate power consumption of 138606 Joules per run.
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: EigenClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KMin: 11667 / Avg: 11894.33 / Max: 12149Min: 7909 / Avg: 8337.11 / Max: 8807Min: 8730 / Avg: 8815.33 / Max: 89011. (CXX) g++ options: -flto -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -pthread

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.hdr_alb_nrm.3840x2160Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.9631.9262.8893.8524.815SE +/- 0.01, N = 6SE +/- 0.01, N = 5SE +/- 0.01, N = 54.283.043.05
OpenBenchmarking.orgImages / Sec Per Watt, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.hdr_alb_nrm.3840x2160Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.00250.0050.00750.010.01250.0100.0110.011
MinAvgMaxClear Linux: 8490H 2P134414726Clear Linux: EPYC 9654 2P46287571Clear Linux22278564OpenBenchmarking.orgWatts, Fewer Is BetterIntel Open Image Denoise 1.4.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 5031 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 4248 Joules per run.3. Clear Linux: Approximate power consumption of 4227 Joules per run.
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.hdr_alb_nrm.3840x2160Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810Min: 4.25 / Avg: 4.28 / Max: 4.33Min: 3.03 / Avg: 3.04 / Max: 3.05Min: 3.01 / Avg: 3.05 / Max: 3.08

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 57Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2000M4000M6000M8000M10000MSE +/- 102638026.31, N = 3SE +/- 7145161.53, N = 3SE +/- 1113552.87, N = 37172733333988970000098868000001. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 57Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4M8M12M16M20M11834462.7818932461.5919097015.65
MinAvgMaxClear Linux: 8490H 2P127606753Clear Linux: EPYC 9654 2P44522637Clear Linux42518629OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 2021.01.31CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 14748 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 12537 Joules per run.3. Clear Linux: Approximate power consumption of 12943 Joules per run.
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 57Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2000M4000M6000M8000M10000MMin: 6968800000 / Avg: 7172733333.33 / Max: 7295000000Min: 9879100000 / Avg: 9889700000 / Max: 9903300000Min: 9885400000 / Avg: 9886800000 / Max: 98890000001. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -pthread -lm -lc -lliquid

Zstd Compression

This test measures the time needed to compress/decompress a sample input file using Zstd compression supplied by the system or otherwise externally of the test profile. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 19, Long Mode - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux8001600240032004000SE +/- 4.60, N = 12SE +/- 7.60, N = 3SE +/- 2.85, N = 152795.33821.23842.21. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***
OpenBenchmarking.orgMB/s Per Watt, More Is BetterZstd CompressionCompression Level: 19, Long Mode - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux481216207.65517.68017.713
MinAvgMaxClear Linux: 8490H 2P127.0365.1426.3Clear Linux: EPYC 9654 2P37.2216.1258.7Clear Linux38.1216.9258.9OpenBenchmarking.orgWatts, Fewer Is BetterZstd CompressionCPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 15975 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 9222 Joules per run.3. Clear Linux: Approximate power consumption of 8995 Joules per run.
OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 19, Long Mode - Decompression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux7001400210028003500Min: 2745.5 / Avg: 2795.34 / Max: 2804Min: 3807.4 / Avg: 3821.2 / Max: 3833.6Min: 3811.8 / Avg: 3842.25 / Max: 3855.41. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux7001400210028003500SE +/- 30.82, N = 20SE +/- 24.47, N = 20SE +/- 21.28, N = 20249433903422
MinAvgMaxClear Linux: 8490H 2P135.0310.8402.9Clear Linux: EPYC 9654 2P45.1167.7265.1Clear Linux37.0165.0262.6OpenBenchmarking.orgWatts, Fewer Is BetterDaCapo Benchmark 9.12-MR1CPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 3000 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 1736 Joules per run.3. Clear Linux: Approximate power consumption of 1707 Joules per run.
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux6001200180024003000Min: 2297 / Avg: 2494.15 / Max: 2758Min: 3197 / Avg: 3389.9 / Max: 3637Min: 3242 / Avg: 3421.65 / Max: 3590

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: MMAPClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux12002400360048006000SE +/- 0.80, N = 3SE +/- 4.47, N = 3SE +/- 1.40, N = 34223.465646.695691.191. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: MMAPClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux481216207.15113.72713.988
MinAvgMaxClear Linux: 8490H 2P136591660Clear Linux: EPYC 9654 2P53411469Clear Linux45407464OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 20278 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 14261 Joules per run.3. Clear Linux: Approximate power consumption of 14241 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MMAPClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux10002000300040005000Min: 4222.4 / Avg: 4223.46 / Max: 4225.03Min: 5637.98 / Avg: 5646.69 / Max: 5652.8Min: 5688.66 / Avg: 5691.19 / Max: 5693.491. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: ThoroughClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux306090120150SE +/- 0.04, N = 6SE +/- 0.07, N = 7SE +/- 0.14, N = 785.85114.76114.791. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -flto -pthread
OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.0Preset: ThoroughClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.10670.21340.32010.42680.53350.2280.4740.467
MinAvgMaxClear Linux: 8490H 2P130377738Clear Linux: EPYC 9654 2P47242654Clear Linux49246653OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 4464 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 2524 Joules per run.3. Clear Linux: Approximate power consumption of 2460 Joules per run.
OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: ThoroughClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100Min: 85.7 / Avg: 85.85 / Max: 85.94Min: 114.49 / Avg: 114.76 / Max: 114.98Min: 114.25 / Avg: 114.79 / Max: 115.141. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -flto -pthread

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: MEMFDClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux8001600240032004000SE +/- 1.38, N = 3SE +/- 5.98, N = 3SE +/- 5.64, N = 33631.892748.612737.601. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: MEMFDClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2468106.2456.8176.870
MinAvgMaxClear Linux: 8490H 2P133582654Clear Linux: EPYC 9654 2P51403462Clear Linux45398455OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 20356 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 14112 Joules per run.3. Clear Linux: Approximate power consumption of 13813 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MEMFDClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux6001200180024003000Min: 3629.55 / Avg: 3631.89 / Max: 3634.33Min: 2736.89 / Avg: 2748.61 / Max: 2756.51Min: 2727.78 / Avg: 2737.6 / Max: 2747.321. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

Zstd Compression

This test measures the time needed to compress/decompress a sample input file using Zstd compression supplied by the system or otherwise externally of the test profile. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 8, Long Mode - Compression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2004006008001000SE +/- 7.71, N = 6SE +/- 10.39, N = 3SE +/- 3.44, N = 3822.51087.81080.81. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***
OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 8, Long Mode - Compression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2004006008001000Min: 799.1 / Avg: 822.45 / Max: 848.6Min: 1067.1 / Avg: 1087.83 / Max: 1099.3Min: 1073.9 / Avg: 1080.77 / Max: 1084.51. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***

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: scryptClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux10002000300040005000SE +/- 23.43, N = 3SE +/- 1.60, N = 3SE +/- 1.26, N = 33644.054797.214793.621. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s Per Watt, More Is BetterCpuminer-Opt 3.20.3Algorithm: scryptClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2468105.7948.1458.261
MinAvgMaxClear Linux: 8490H 2P131629754Clear Linux: EPYC 9654 2P42589680Clear Linux53580676OpenBenchmarking.orgWatts, Fewer Is BetterCpuminer-Opt 3.20.3CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21384 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 20025 Joules per run.3. Clear Linux: Approximate power consumption of 20310 Joules per run.
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: scryptClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux8001600240032004000Min: 3597.29 / Avg: 3644.05 / Max: 3670.09Min: 4794.02 / Avg: 4797.21 / Max: 4799.11Min: 4791.1 / Avg: 4793.62 / Max: 4795.021. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: KNN CADClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.23, N = 386.2866.6066.69
MinAvgMaxClear Linux: 8490H 2P133462704Clear Linux: EPYC 9654 2P53281501Clear Linux50280488OpenBenchmarking.orgWatts, Fewer Is BetterNumenta Anomaly Benchmark 1.1CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 41751 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 19835 Joules per run.3. Clear Linux: Approximate power consumption of 19876 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: KNN CADClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1632486480Min: 85.12 / Avg: 86.28 / Max: 86.86Min: 66.13 / Avg: 66.6 / Max: 67.24Min: 66.33 / Avg: 66.69 / Max: 67.11

NWChem

NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 BuckyballClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4008001200160020001834.31466.51458.31. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lcomex -lm -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2
OpenBenchmarking.orgWatts, Fewer Is BetterNWChem 7.0.2CPU Power Consumption MonitorClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux130260390520650Min: 131.37 / Avg: 695.57 / Max: 740.37Min: 51.41 / Avg: 664.88 / Max: 699.68Min: 52.4 / Avg: 653.87 / Max: 684.74

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Earthgecko SkylineClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1530456075SE +/- 0.64, N = 3SE +/- 0.80, N = 15SE +/- 0.74, N = 366.8154.2753.14
MinAvgMaxClear Linux: 8490H 2P124.8412.0552.8Clear Linux: EPYC 9654 2P44.9251.5420.7Clear Linux44.0249.3416.1OpenBenchmarking.orgWatts, Fewer Is BetterNumenta Anomaly Benchmark 1.1CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 29254 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 14637 Joules per run.3. Clear Linux: Approximate power consumption of 14459 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Earthgecko SkylineClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265Min: 65.84 / Avg: 66.81 / Max: 68.03Min: 52.22 / Avg: 54.27 / Max: 64.6Min: 52.3 / Avg: 53.14 / Max: 54.61

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific 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.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: particle_volume/ao/real_timeClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1020304050SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 334.7042.5742.63
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.10Benchmark: particle_volume/ao/real_timeClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.01690.03380.05070.06760.08450.0510.0730.075
MinAvgMaxClear Linux: 8490H 2P128687728Clear Linux: EPYC 9654 2P47584602Clear Linux50571588OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.10CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 158626 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 127930 Joules per run.3. Clear Linux: Approximate power consumption of 124782 Joules per run.
OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: particle_volume/ao/real_timeClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux918273645Min: 34.64 / Avg: 34.7 / Max: 34.73Min: 42.54 / Avg: 42.57 / Max: 42.61Min: 42.54 / Avg: 42.63 / Max: 42.72

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2004006008001000SE +/- 6.31, N = 6SE +/- 4.28, N = 6SE +/- 2.57, N = 5821.58765.08672.101. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -flto -pthread
OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.0Preset: FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.62031.24061.86092.48123.10152.0862.7572.347
MinAvgMaxClear Linux: 8490H 2P130394668Clear Linux: EPYC 9654 2P47278543Clear Linux44286534OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 4923 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 3562 Joules per run.3. Clear Linux: Approximate power consumption of 3781 Joules per run.
OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux140280420560700Min: 805.48 / Avg: 821.58 / Max: 850.07Min: 749.32 / Avg: 765.08 / Max: 775.12Min: 665.6 / Avg: 672.1 / Max: 678.891. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -flto -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: 6Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.55191.10381.65572.20762.7595SE +/- 0.005, N = 10SE +/- 0.002, N = 10SE +/- 0.002, N = 72.4532.0092.0071. (CXX) g++ options: -O3 -fPIC -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lm
MinAvgMaxClear Linux: 8490H 2P125287694Clear Linux: EPYC 9654 2P44147408Clear Linux44145352OpenBenchmarking.orgWatts, Fewer Is Betterlibavif avifenc 0.11CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 1892 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 999 Joules per run.3. Clear Linux: Approximate power consumption of 1429 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810Min: 2.42 / Avg: 2.45 / Max: 2.48Min: 2 / Avg: 2.01 / Max: 2.02Min: 2 / Avg: 2.01 / Max: 2.021. (CXX) g++ options: -O3 -fPIC -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lm

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Genetic Algorithm Using Jenetics + FuturesClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux400800120016002000SE +/- 23.45, N = 3SE +/- 7.45, N = 3SE +/- 5.95, N = 32026.21677.41665.0MIN: 1895.15 / MAX: 2139.35MIN: 1620.99 / MAX: 1745.53MIN: 1554.78 / MAX: 1738.21
MinAvgMaxClear Linux: 8490H 2P135.7362.8407.1Clear Linux: EPYC 9654 2P45.5225.7275.2Clear Linux45.7222.3251.4OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 38816 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 20091 Joules per run.3. Clear Linux: Approximate power consumption of 19713 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Genetic Algorithm Using Jenetics + FuturesClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux400800120016002000Min: 1990.9 / Avg: 2026.23 / Max: 2070.61Min: 1662.79 / Avg: 1677.39 / Max: 1687.29Min: 1655.62 / Avg: 1664.99 / Max: 1676.02

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: nbodyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100SE +/- 0.17, N = 3SE +/- 0.18, N = 3SE +/- 0.07, N = 363.376.075.8
MinAvgMaxClear Linux: 8490H 2P132.7234.4343.7Clear Linux: EPYC 9654 2P51.6195.9218.5Clear Linux50.6195.1217.7OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor801602403204001. Clear Linux: 8490H 2P: Approximate power consumption of 6642 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 6334 Joules per run.3. Clear Linux: Approximate power consumption of 6307 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: nbodyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1530456075Min: 63 / Avg: 63.33 / Max: 63.5Min: 75.7 / Avg: 75.97 / Max: 76.3Min: 75.7 / Avg: 75.77 / Max: 75.9

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark BayesClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux100200300400500SE +/- 1.22, N = 3SE +/- 1.01, N = 3SE +/- 4.04, N = 3449.0375.6387.6MIN: 295.86 / MAX: 720.71MIN: 244.91 / MAX: 682.26MIN: 236.1 / MAX: 595.28
MinAvgMaxClear Linux: 8490H 2P133463671Clear Linux: EPYC 9654 2P51290456Clear Linux51291447OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 12955 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 7054 Joules per run.3. Clear Linux: Approximate power consumption of 7180 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark BayesClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux80160240320400Min: 446.55 / Avg: 448.99 / Max: 450.28Min: 373.74 / Avg: 375.64 / Max: 377.18Min: 381.76 / Avg: 387.58 / Max: 395.35

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Bayesian ChangepointClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux510152025SE +/- 0.25, N = 15SE +/- 0.18, N = 3SE +/- 0.17, N = 1521.7018.6318.27
MinAvgMaxClear Linux: 8490H 2P129.9360.5549.2Clear Linux: EPYC 9654 2P51.7216.5372.0Clear Linux52.3217.2388.0OpenBenchmarking.orgWatts, Fewer Is BetterNumenta Anomaly Benchmark 1.1CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 9277 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 5124 Joules per run.3. Clear Linux: Approximate power consumption of 4866 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Bayesian ChangepointClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux510152025Min: 20.52 / Avg: 21.69 / Max: 23.64Min: 18.31 / Avg: 18.63 / Max: 18.94Min: 17.01 / Avg: 18.27 / Max: 19.1

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: regex_compileClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.12, N = 3104.087.688.2
MinAvgMaxClear Linux: 8490H 2P136.3234.3320.1Clear Linux: EPYC 9654 2P46.2200.0218.6Clear Linux47.9199.0217.8OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor801602403204001. Clear Linux: 8490H 2P: Approximate power consumption of 6249 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 7733 Joules per run.3. Clear Linux: Approximate power consumption of 7762 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: regex_compileClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100Min: 104 / Avg: 104 / Max: 104Min: 87.6 / Avg: 87.63 / Max: 87.7Min: 88 / Avg: 88.17 / Max: 88.4

Zstd Compression

This test measures the time needed to compress/decompress a sample input file using Zstd compression supplied by the system or otherwise externally of the test profile. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 19 - Compression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux306090120150SE +/- 0.74, N = 15SE +/- 1.00, N = 7SE +/- 0.96, N = 1597.8114.2116.01. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***
OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 19 - Compression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100Min: 92.8 / Avg: 97.83 / Max: 101.9Min: 109.2 / Avg: 114.16 / Max: 117.9Min: 111 / Avg: 116.03 / Max: 122.61. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Windowed GaussianClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.96081.92162.88243.84324.804SE +/- 0.018, N = 8SE +/- 0.023, N = 9SE +/- 0.029, N = 94.2703.7393.641
MinAvgMaxClear Linux: 8490H 2P129.0290.8520.5Clear Linux: EPYC 9654 2P32.7164.1355.7Clear Linux44.8163.3349.9OpenBenchmarking.orgWatts, Fewer Is BetterNumenta Anomaly Benchmark 1.1CPU Power Consumption Monitor1302603905206501. Clear Linux: 8490H 2P: Approximate power consumption of 2472 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 1294 Joules per run.3. Clear Linux: Approximate power consumption of 1288 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Windowed GaussianClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810Min: 4.17 / Avg: 4.27 / Max: 4.35Min: 3.64 / Avg: 3.74 / Max: 3.88Min: 3.48 / Avg: 3.64 / Max: 3.74

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific 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.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: gravity_spheres_volume/dim_512/ao/real_timeClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1020304050SE +/- 0.15, N = 3SE +/- 0.14, N = 3SE +/- 0.08, N = 338.1044.5444.43
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.10Benchmark: gravity_spheres_volume/dim_512/ao/real_timeClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.01760.03520.05280.07040.0880.0580.0760.078
MinAvgMaxClear Linux: 8490H 2P120662740Clear Linux: EPYC 9654 2P19583627Clear Linux19573614OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.10CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 81924 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 69015 Joules per run.3. Clear Linux: Approximate power consumption of 67772 Joules per run.
OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.10Benchmark: gravity_spheres_volume/dim_512/ao/real_timeClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux918273645Min: 37.79 / Avg: 38.1 / Max: 38.29Min: 44.31 / Avg: 44.54 / Max: 44.78Min: 44.34 / Avg: 44.43 / Max: 44.59

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 1.3.1Benchmark: vklBenchmark ISPCClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux30060090012001500SE +/- 4.33, N = 3SE +/- 2.85, N = 3SE +/- 16.23, N = 3114213321334MIN: 194 / MAX: 5100MIN: 333 / MAX: 4562MIN: 331 / MAX: 5092
OpenBenchmarking.orgItems / Sec Per Watt, More Is BetterOpenVKL 1.3.1Benchmark: vklBenchmark ISPCClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.60951.2191.82852.4383.04752.0422.7092.703
OpenBenchmarking.orgWatts, Fewer Is BetterOpenVKL 1.3.1CPU Power Consumption MonitorClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux130260390520650Min: 62.29 / Avg: 559.18 / Max: 732.51Min: 40.7 / Avg: 491.72 / Max: 665.25Min: 52.67 / Avg: 493.46 / Max: 664.31. Clear Linux: 8490H 2P: Approximate power consumption of 229636 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 202915 Joules per run.3. Clear Linux: Approximate power consumption of 202153 Joules per run.
OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 1.3.1Benchmark: vklBenchmark ISPCClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2004006008001000Min: 1135 / Avg: 1142.33 / Max: 1150Min: 1329 / Avg: 1332.33 / Max: 1338Min: 1302 / Avg: 1334.33 / Max: 1353

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux7001400210028003500SE +/- 7.23, N = 8SE +/- 9.49, N = 8SE +/- 7.25, N = 8267931253111
MinAvgMaxClear Linux: 8490H 2P129.9278.1397.8Clear Linux: EPYC 9654 2P50.0146.1248.5Clear Linux40.2144.6248.4OpenBenchmarking.orgWatts, Fewer Is BetterDaCapo Benchmark 9.12-MR1CPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 2260 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 1242 Joules per run.3. Clear Linux: Approximate power consumption of 1247 Joules per run.
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux5001000150020002500Min: 2644 / Avg: 2678.5 / Max: 2707Min: 3101 / Avg: 3124.5 / Max: 3174Min: 3084 / Avg: 3111 / Max: 3145

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.8.2Clients: 1024Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2004006008001000SE +/- 5.90, N = 3SE +/- 10.04, N = 3SE +/- 7.35, N = 37927056961. (CXX) g++ options: -fPIC -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -pie -fstack-protector -shared -lrt -lpthread -lz -ldl -lm -lstdc++
OpenBenchmarking.orgQueries Per Second Per Watt, More Is BetterMariaDB 10.8.2Clients: 1024Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.53781.07561.61342.15122.6891.9522.3902.390
MinAvgMaxClear Linux: 8490H 2P135.5405.8421.5Clear Linux: EPYC 9654 2P50.8295.0315.3Clear Linux52.9291.2340.5OpenBenchmarking.orgWatts, Fewer Is BetterMariaDB 10.8.2CPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 111859 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 89788 Joules per run.3. Clear Linux: Approximate power consumption of 89502 Joules per run.
OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.8.2Clients: 1024Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux140280420560700Min: 786.16 / Avg: 792.06 / Max: 803.86Min: 690.61 / Avg: 704.67 / Max: 724.11Min: 681.2 / Avg: 695.89 / Max: 703.731. (CXX) g++ options: -fPIC -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -pie -fstack-protector -shared -lrt -lpthread -lz -ldl -lm -lstdc++

PyBench

This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux160320480640800SE +/- 0.82, N = 4SE +/- 1.20, N = 3SE +/- 2.60, N = 3664755749
MinAvgMaxClear Linux: 8490H 2P130.9214.4298.5Clear Linux: EPYC 9654 2P48.3186.5218.5Clear Linux39.4186.8217.8OpenBenchmarking.orgWatts, Fewer Is BetterPyBench 2018-02-16CPU Power Consumption Monitor801602403204001. Clear Linux: 8490H 2P: Approximate power consumption of 4234 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 4165 Joules per run.3. Clear Linux: Approximate power consumption of 4109 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux130260390520650Min: 662 / Avg: 664 / Max: 666Min: 753 / Avg: 755.33 / Max: 757Min: 745 / Avg: 749.33 / Max: 754

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Contextual Anomaly Detector OSEClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux816243240SE +/- 0.39, N = 4SE +/- 0.04, N = 3SE +/- 0.10, N = 332.4428.6230.21
MinAvgMaxClear Linux: 8490H 2P133.1348.3540.7Clear Linux: EPYC 9654 2P49.1212.0374.1Clear Linux46.3209.7375.9OpenBenchmarking.orgWatts, Fewer Is BetterNumenta Anomaly Benchmark 1.1CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 12800 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 7066 Joules per run.3. Clear Linux: Approximate power consumption of 7270 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Contextual Anomaly Detector OSEClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux714212835Min: 31.56 / Avg: 32.44 / Max: 33.29Min: 28.56 / Avg: 28.62 / Max: 28.7Min: 30.07 / Avg: 30.21 / Max: 30.41

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: 2Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux918273645SE +/- 0.15, N = 3SE +/- 0.32, N = 3SE +/- 0.15, N = 337.5633.2933.141. (CXX) g++ options: -O3 -fPIC -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lm
MinAvgMaxClear Linux: 8490H 2P133397737Clear Linux: EPYC 9654 2P53245461Clear Linux45246472OpenBenchmarking.orgWatts, Fewer Is Betterlibavif avifenc 0.11CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 16675 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 9230 Joules per run.3. Clear Linux: Approximate power consumption of 9268 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 2Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux816243240Min: 37.4 / Avg: 37.56 / Max: 37.86Min: 32.76 / Avg: 33.29 / Max: 33.87Min: 32.97 / Avg: 33.13 / Max: 33.431. (CXX) g++ options: -O3 -fPIC -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lm

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: json_loadsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux510152025SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 318.316.216.2
MinAvgMaxClear Linux: 8490H 2P133.2234.5348.4Clear Linux: EPYC 9654 2P48.4197.5218.3Clear Linux29.2196.6217.4OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 7818 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 6846 Joules per run.3. Clear Linux: Approximate power consumption of 6751 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: json_loadsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux510152025Min: 18.3 / Avg: 18.3 / Max: 18.3Min: 16.2 / Avg: 16.23 / Max: 16.3Min: 16.2 / Avg: 16.23 / Max: 16.3

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Relative EntropyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3691215SE +/- 0.038, N = 5SE +/- 0.056, N = 5SE +/- 0.098, N = 610.7599.6459.654
MinAvgMaxClear Linux: 8490H 2P128.3318.6527.6Clear Linux: EPYC 9654 2P51.0192.9361.8Clear Linux43.8194.1355.8OpenBenchmarking.orgWatts, Fewer Is BetterNumenta Anomaly Benchmark 1.1CPU Power Consumption Monitor1302603905206501. Clear Linux: 8490H 2P: Approximate power consumption of 4779 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 2701 Joules per run.3. Clear Linux: Approximate power consumption of 2686 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Relative EntropyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3691215Min: 10.63 / Avg: 10.76 / Max: 10.84Min: 9.51 / Avg: 9.65 / Max: 9.83Min: 9.5 / Avg: 9.65 / Max: 10.13

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Super FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100SE +/- 0.55, N = 10SE +/- 0.64, N = 6SE +/- 0.24, N = 670.4177.9478.481. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Super FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.06860.13720.20580.27440.3430.1710.2990.305
MinAvgMaxClear Linux: 8490H 2P132.3411.3535.0Clear Linux: EPYC 9654 2P41.7260.9387.9Clear Linux49.2257.3387.1OpenBenchmarking.orgWatts, Fewer Is BetterKvazaar 2.2CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 5182 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 3175 Joules per run.3. Clear Linux: Approximate power consumption of 3130 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Super FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1530456075Min: 68.57 / Avg: 70.41 / Max: 74.14Min: 75.59 / Avg: 77.94 / Max: 79.9Min: 77.41 / Avg: 78.48 / Max: 78.91. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark ALSClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux400800120016002000SE +/- 5.16, N = 3SE +/- 21.37, N = 3SE +/- 20.05, N = 32073.01862.21912.5MIN: 1941.42 / MAX: 3230.55MIN: 1666.18 / MAX: 2218.13MIN: 1772.33 / MAX: 2265.01
MinAvgMaxClear Linux: 8490H 2P131.3398.0426.5Clear Linux: EPYC 9654 2P27.1236.6259.0Clear Linux48.6235.0255.2OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 30510 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 16880 Joules per run.3. Clear Linux: Approximate power consumption of 17078 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark ALSClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux400800120016002000Min: 2067.01 / Avg: 2073 / Max: 2083.27Min: 1819.72 / Avg: 1862.16 / Max: 1887.64Min: 1887.8 / Avg: 1912.49 / Max: 1952.2

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: MediumClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux120240360480600SE +/- 1.59, N = 7SE +/- 0.92, N = 8SE +/- 1.40, N = 8525.40556.68500.661. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -flto -pthread
OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.0Preset: MediumClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.68471.36942.05412.73883.42351.7323.0432.763
MinAvgMaxClear Linux: 8490H 2P131303664Clear Linux: EPYC 9654 2P41183562Clear Linux48181565OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 2817 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 1555 Joules per run.3. Clear Linux: Approximate power consumption of 1586 Joules per run.
OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: MediumClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux100200300400500Min: 520.43 / Avg: 525.4 / Max: 533.61Min: 551.89 / Avg: 556.68 / Max: 559.31Min: 494.41 / Avg: 500.66 / Max: 504.671. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -flto -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: 0Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1530456075SE +/- 0.11, N = 3SE +/- 0.17, N = 3SE +/- 0.08, N = 367.3360.5860.801. (CXX) g++ options: -O3 -fPIC -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lm
MinAvgMaxClear Linux: 8490H 2P134433724Clear Linux: EPYC 9654 2P33274481Clear Linux45273476OpenBenchmarking.orgWatts, Fewer Is Betterlibavif avifenc 0.11CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 31152 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 17815 Joules per run.3. Clear Linux: Approximate power consumption of 17860 Joules per run.
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 0Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265Min: 67.11 / Avg: 67.33 / Max: 67.49Min: 60.31 / Avg: 60.58 / Max: 60.9Min: 60.64 / Avg: 60.8 / Max: 60.91. (CXX) g++ options: -O3 -fPIC -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -lm

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Random ForestClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux140280420560700SE +/- 6.37, N = 5SE +/- 2.61, N = 3SE +/- 5.64, N = 6646.5593.4584.2MIN: 581.7 / MAX: 773.51MIN: 536.42 / MAX: 709.21MIN: 512.73 / MAX: 725.6
MinAvgMaxClear Linux: 8490H 2P137.8365.9407.6Clear Linux: EPYC 9654 2P46.9212.8248.7Clear Linux50.9210.9246.6OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 14343 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 8086 Joules per run.3. Clear Linux: Approximate power consumption of 7767 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Random ForestClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux110220330440550Min: 632.22 / Avg: 646.47 / Max: 670.02Min: 588.18 / Avg: 593.38 / Max: 596.42Min: 563.32 / Avg: 584.19 / Max: 598.89

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Ultra FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100SE +/- 0.65, N = 6SE +/- 0.65, N = 6SE +/- 0.81, N = 672.1479.1979.761. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Ultra FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.07610.15220.22830.30440.38050.1890.3330.338
MinAvgMaxClear Linux: 8490H 2P128.6381.9492.9Clear Linux: EPYC 9654 2P38.9237.7354.4Clear Linux29.2235.9355.0OpenBenchmarking.orgWatts, Fewer Is BetterKvazaar 2.2CPU Power Consumption Monitor1302603905206501. Clear Linux: 8490H 2P: Approximate power consumption of 4774 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 2853 Joules per run.3. Clear Linux: Approximate power consumption of 2792 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Ultra FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1530456075Min: 70.24 / Avg: 72.14 / Max: 74.22Min: 76.96 / Avg: 79.19 / Max: 81.8Min: 77.11 / Avg: 79.76 / Max: 81.771. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark PageRankClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux6001200180024003000SE +/- 29.19, N = 3SE +/- 23.16, N = 15SE +/- 28.19, N = 152704.12551.02450.5MIN: 2424.36 / MAX: 2742.43MIN: 2107.47 / MAX: 2809.93MIN: 1985.5 / MAX: 2709.64
MinAvgMaxClear Linux: 8490H 2P131.4375.6398.6Clear Linux: EPYC 9654 2P46.5221.7246.5Clear Linux41.9220.0244.6OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 35181 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 19138 Joules per run.3. Clear Linux: Approximate power consumption of 18464 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark PageRankClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux5001000150020002500Min: 2646.81 / Avg: 2704.1 / Max: 2742.43Min: 2365.39 / Avg: 2551.03 / Max: 2712.5Min: 2273.38 / Avg: 2450.45 / Max: 2602.11

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Very FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100SE +/- 0.66, N = 6SE +/- 0.64, N = 6SE +/- 0.55, N = 670.5877.2376.911. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Very FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.06230.12460.18690.24920.31150.1630.2770.276
MinAvgMaxClear Linux: 8490H 2P128.8433.2568.4Clear Linux: EPYC 9654 2P48.0279.1424.2Clear Linux44.6278.4421.2OpenBenchmarking.orgWatts, Fewer Is BetterKvazaar 2.2CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 5559 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 3349 Joules per run.3. Clear Linux: Approximate power consumption of 3387 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Very FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1530456075Min: 68.96 / Avg: 70.58 / Max: 72.71Min: 75.25 / Avg: 77.23 / Max: 79.24Min: 74.61 / Avg: 76.91 / Max: 78.471. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpthread -lm -lrt

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_uringClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux11K22K33K44K55KSE +/- 35.22, N = 3SE +/- 107.85, N = 3SE +/- 203.39, N = 351657.6148522.0848493.701. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: IO_uringClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200132.17187.90190.17
MinAvgMaxClear Linux: 8490H 2P135.7390.8440.7Clear Linux: EPYC 9654 2P53.4258.2323.3Clear Linux45.1255.0332.6OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor1202403604806001. Clear Linux: 8490H 2P: Approximate power consumption of 13549 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 9038 Joules per run.3. Clear Linux: Approximate power consumption of 8840 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: IO_uringClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux9K18K27K36K45KMin: 51587.17 / Avg: 51657.61 / Max: 51693.87Min: 48321.86 / Avg: 48522.08 / Max: 48691.7Min: 48216.27 / Avg: 48493.7 / Max: 48890.051. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: ALS Movie LensClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KSE +/- 95.99, N = 4SE +/- 62.83, N = 12SE +/- 118.91, N = 99105.68554.78651.9MIN: 8667.04 / MAX: 9936.24MIN: 7473.52 / MAX: 10039.49MIN: 7297.19 / MAX: 9790.29
MinAvgMaxClear Linux: 8490H 2P132.3396.2412.6Clear Linux: EPYC 9654 2P43.0231.5251.3Clear Linux46.6229.5250.0OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 79734 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 43968 Joules per run.3. Clear Linux: Approximate power consumption of 44089 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: ALS Movie LensClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux16003200480064008000Min: 8850.79 / Avg: 9105.56 / Max: 9316.84Min: 8080.41 / Avg: 8554.71 / Max: 8819.13Min: 8035.98 / Avg: 8651.9 / Max: 9175.04

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: floatClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.15, N = 353.756.957.1
MinAvgMaxClear Linux: 8490H 2P133.2316.3359.5Clear Linux: EPYC 9654 2P48.1191.6218.7Clear Linux43.4190.5217.7OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 7906 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 5046 Joules per run.3. Clear Linux: Approximate power consumption of 5017 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: floatClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1122334455Min: 53.6 / Avg: 53.67 / Max: 53.8Min: 56.8 / Avg: 56.9 / Max: 57Min: 56.8 / Avg: 57.07 / Max: 57.3

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: python_startupClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810SE +/- 0.01, N = 3SE +/- 0.10, N = 3SE +/- 0.00, N = 38.368.007.88
MinAvgMaxClear Linux: 8490H 2P134.3351.1366.5Clear Linux: EPYC 9654 2P52.1208.9218.7Clear Linux49.5207.7217.5OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 26802 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 15181 Joules per run.3. Clear Linux: Approximate power consumption of 14955 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: python_startupClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3691215Min: 8.35 / Avg: 8.36 / Max: 8.38Min: 7.85 / Avg: 8 / Max: 8.19Min: 7.88 / Avg: 7.88 / Max: 7.88

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Very FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265SE +/- 0.39, N = 5SE +/- 0.40, N = 5SE +/- 0.36, N = 555.8958.4159.271. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17
OpenBenchmarking.orgFrames Per Second Per Watt, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Very FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.04460.08920.13380.17840.2230.1220.1920.198
MinAvgMaxClear Linux: 8490H 2P136.3458.0573.2Clear Linux: EPYC 9654 2P45.5303.8423.5Clear Linux40.8298.6418.8OpenBenchmarking.orgWatts, Fewer Is Betteruvg266 0.4.1CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 6778 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 4497 Joules per run.3. Clear Linux: Approximate power consumption of 4360 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Very FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1224364860Min: 54.82 / Avg: 55.89 / Max: 56.8Min: 57.43 / Avg: 58.41 / Max: 59.84Min: 58.09 / Avg: 59.27 / Max: 60.161. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Super FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265SE +/- 0.44, N = 5SE +/- 0.53, N = 5SE +/- 0.39, N = 1556.8559.2260.131. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17
OpenBenchmarking.orgFrames Per Second Per Watt, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Super FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.04610.09220.13830.18440.23050.1290.2040.205
MinAvgMaxClear Linux: 8490H 2P130.9441.2561.4Clear Linux: EPYC 9654 2P44.9290.5408.2Clear Linux44.6292.7408.1OpenBenchmarking.orgWatts, Fewer Is Betteruvg266 0.4.1CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 6618 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 4125 Joules per run.3. Clear Linux: Approximate power consumption of 4176 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Super FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1224364860Min: 55.63 / Avg: 56.85 / Max: 58Min: 57.95 / Avg: 59.22 / Max: 60.63Min: 58.13 / Avg: 60.13 / Max: 63.651. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Ultra FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265SE +/- 0.52, N = 5SE +/- 0.53, N = 15SE +/- 0.62, N = 1557.2858.6660.021. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17
OpenBenchmarking.orgFrames Per Second Per Watt, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Ultra FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.04970.09940.14910.19880.24850.1330.2140.221
MinAvgMaxClear Linux: 8490H 2P132.7430.9534.1Clear Linux: EPYC 9654 2P47.3273.5384.2Clear Linux41.9271.5377.2OpenBenchmarking.orgWatts, Fewer Is Betteruvg266 0.4.1CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 6378 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 3975 Joules per run.3. Clear Linux: Approximate power consumption of 3838 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Ultra FastClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1224364860Min: 56.6 / Avg: 57.28 / Max: 59.36Min: 56.83 / Avg: 58.66 / Max: 65.12Min: 57.47 / Avg: 60.02 / Max: 66.211. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: 2to3Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3201193192
MinAvgMaxClear Linux: 8490H 2P134.9333.4358.4Clear Linux: EPYC 9654 2P53.1201.9218.6Clear Linux26.1200.6217.5OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 16894 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 8412 Joules per run.3. Clear Linux: Approximate power consumption of 8360 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: 2to3Clear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200Min: 201 / Avg: 201 / Max: 201Min: 192 / Avg: 192.67 / Max: 193Min: 192 / Avg: 192 / Max: 192

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: raytraceClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux50100150200250SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3214223224
MinAvgMaxClear Linux: 8490H 2P131.4234.0350.5Clear Linux: EPYC 9654 2P46.2202.2219.0Clear Linux39.8201.6218.0OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 10063 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 8895 Joules per run.3. Clear Linux: Approximate power consumption of 8802 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: raytraceClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200Min: 214 / Avg: 214.33 / Max: 215Min: 223 / Avg: 223 / Max: 223Min: 224 / Avg: 224 / Max: 224

simdjson

This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: LargeRandomClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.29030.58060.87091.16121.4515SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.291.241.241. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17
OpenBenchmarking.orgGB/s Per Watt, More Is Bettersimdjson 2.0Throughput Test: LargeRandomClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.00140.00280.00420.00560.0070.0040.0060.006
MinAvgMaxClear Linux: 8490H 2P130.6333.3358.9Clear Linux: EPYC 9654 2P35.8206.7220.5Clear Linux45.5205.7219.1OpenBenchmarking.orgWatts, Fewer Is Bettersimdjson 2.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 18778 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 11298 Joules per run.3. Clear Linux: Approximate power consumption of 11315 Joules per run.
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: LargeRandomClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810Min: 1.29 / Avg: 1.29 / Max: 1.29Min: 1.24 / Avg: 1.24 / Max: 1.24Min: 1.24 / Avg: 1.24 / Max: 1.241. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Finagle HTTP RequestsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KSE +/- 131.33, N = 12SE +/- 100.96, N = 6SE +/- 113.71, N = 410387.610803.310800.0MIN: 8501.57 / MAX: 11367.76MIN: 9487.14 / MAX: 11273.39MIN: 9740.37 / MAX: 11088.94
OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption MonitorClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux120240360480600Min: 135.75 / Avg: 640.36 / Max: 702.13Min: 47.66 / Avg: 503.08 / Max: 575.15Min: 30.3 / Avg: 494.69 / Max: 554.761. Clear Linux: 8490H 2P: Approximate power consumption of 85541 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 69845 Joules per run.3. Clear Linux: Approximate power consumption of 68639 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Finagle HTTP RequestsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KMin: 9658.22 / Avg: 10387.55 / Max: 11238.54Min: 10405.08 / Avg: 10803.3 / Max: 11093.38Min: 10482.55 / Avg: 10800.02 / Max: 11022.18

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pickle_pure_pythonClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux50100150200250SE +/- 0.00, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3219225227
MinAvgMaxClear Linux: 8490H 2P135.4232.3296.8Clear Linux: EPYC 9654 2P51.5196.0218.9Clear Linux48.3196.1217.9OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor701402102803501. Clear Linux: 8490H 2P: Approximate power consumption of 7123 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 6141 Joules per run.3. Clear Linux: Approximate power consumption of 6143 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pickle_pure_pythonClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200Min: 219 / Avg: 219 / Max: 219Min: 224 / Avg: 225 / Max: 226Min: 226 / Avg: 227 / Max: 228

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: crypto_pyaesClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265SE +/- 0.03, N = 3SE +/- 0.12, N = 3SE +/- 0.03, N = 358.660.060.3
MinAvgMaxClear Linux: 8490H 2P134.5231.9299.4Clear Linux: EPYC 9654 2P46.7192.8218.4Clear Linux46.5191.2217.6OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor801602403204001. Clear Linux: 8490H 2P: Approximate power consumption of 6184 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 5206 Joules per run.3. Clear Linux: Approximate power consumption of 5227 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: crypto_pyaesClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1224364860Min: 58.6 / Avg: 58.63 / Max: 58.7Min: 59.8 / Avg: 60 / Max: 60.2Min: 60.2 / Avg: 60.27 / Max: 60.3

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: goClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux306090120150SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3109112112
MinAvgMaxClear Linux: 8490H 2P130.6234.3354.5Clear Linux: EPYC 9654 2P52.8191.0218.3Clear Linux42.2189.4217.4OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 6013 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 4839 Joules per run.3. Clear Linux: Approximate power consumption of 4924 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: goClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20406080100Min: 109 / Avg: 109 / Max: 109Min: 112 / Avg: 112.33 / Max: 113Min: 112 / Avg: 112 / Max: 112

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: django_templateClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux612182430SE +/- 0.15, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 326.727.327.4
MinAvgMaxClear Linux: 8490H 2P132.5289.8356.1Clear Linux: EPYC 9654 2P44.9195.9219.0Clear Linux48.1195.4217.8OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 9757 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 6270 Joules per run.3. Clear Linux: Approximate power consumption of 6253 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: django_templateClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux612182430Min: 26.5 / Avg: 26.73 / Max: 27Min: 27.2 / Avg: 27.27 / Max: 27.3Min: 27.3 / Avg: 27.37 / Max: 27.4

simdjson

This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: PartialTweetsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1.33432.66864.00295.33726.6715SE +/- 0.01, N = 3SE +/- 0.06, N = 5SE +/- 0.00, N = 35.795.855.931. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17
OpenBenchmarking.orgGB/s Per Watt, More Is Bettersimdjson 2.0Throughput Test: PartialTweetsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.00650.0130.01950.0260.03250.0170.0280.029
MinAvgMaxClear Linux: 8490H 2P127.7332.4349.6Clear Linux: EPYC 9654 2P48.0210.1222.5Clear Linux29.7208.0218.8OpenBenchmarking.orgWatts, Fewer Is Bettersimdjson 2.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 22823 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 14877 Joules per run.3. Clear Linux: Approximate power consumption of 14839 Joules per run.
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: PartialTweetsClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810Min: 5.78 / Avg: 5.79 / Max: 5.8Min: 5.6 / Avg: 5.85 / Max: 5.91Min: 5.93 / Avg: 5.93 / Max: 5.941. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17

PHPBench

PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux400K800K1200K1600K2000KSE +/- 7368.91, N = 5SE +/- 4431.47, N = 5SE +/- 19191.23, N = 5202200019936731979213
OpenBenchmarking.orgScore Per Watt, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10K9741.0611611.6811452.08
MinAvgMaxClear Linux: 8490H 2P131.8207.6271.9Clear Linux: EPYC 9654 2P51.9171.7218.7Clear Linux35.2172.8217.7OpenBenchmarking.orgWatts, Fewer Is BetterPHPBench 0.8.1CPU Power Consumption Monitor701402102803501. Clear Linux: 8490H 2P: Approximate power consumption of 2906 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 2472 Joules per run.3. Clear Linux: Approximate power consumption of 2489 Joules per run.
OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux400K800K1200K1600K2000KMin: 2008348 / Avg: 2022000.2 / Max: 2041160Min: 1981433 / Avg: 1993673.4 / Max: 2009081Min: 1905195 / Avg: 1979213 / Max: 2009823

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pathlibClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 311.912.012.1
MinAvgMaxClear Linux: 8490H 2P135.8335.2363.0Clear Linux: EPYC 9654 2P50.8200.6218.5Clear Linux44.4199.5217.6OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 13296 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 8158 Joules per run.3. Clear Linux: Approximate power consumption of 8180 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pathlibClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux48121620Min: 11.9 / Avg: 11.9 / Max: 11.9Min: 12 / Avg: 12 / Max: 12Min: 12 / Avg: 12.07 / Max: 12.1

simdjson

This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: TopTweetClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 36.746.666.651. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17
OpenBenchmarking.orgGB/s Per Watt, More Is Bettersimdjson 2.0Throughput Test: TopTweetClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.00720.01440.02160.02880.0360.0200.0320.032
MinAvgMaxClear Linux: 8490H 2P129.7333.6351.1Clear Linux: EPYC 9654 2P42.9209.4219.5Clear Linux52.4208.1218.0OpenBenchmarking.orgWatts, Fewer Is Bettersimdjson 2.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 23015 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 14865 Joules per run.3. Clear Linux: Approximate power consumption of 14635 Joules per run.
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 2.0Throughput Test: TopTweetClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3691215Min: 6.73 / Avg: 6.74 / Max: 6.75Min: 6.64 / Avg: 6.66 / Max: 6.67Min: 6.6 / Avg: 6.65 / Max: 6.671. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17

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 PassingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux7M14M21M28M35MSE +/- 273980.47, N = 10SE +/- 37069.81, N = 3SE +/- 35838.24, N = 334967285.0034597345.4434562392.221. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: System V Message PassingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20K40K60K80K100K62111.2589800.1790475.76
MinAvgMaxClear Linux: 8490H 2P128563632Clear Linux: EPYC 9654 2P53385441Clear Linux47382432OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 19198 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 13484 Joules per run.3. Clear Linux: Approximate power consumption of 13243 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: System V Message PassingClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux6M12M18M24M30MMin: 34638976.03 / Avg: 34967285 / Max: 37431581.93Min: 34527176.82 / Avg: 34597345.44 / Max: 34653161.04Min: 34497342.26 / Avg: 34562392.22 / Max: 34620984.711. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: chaosClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1224364860SE +/- 0.03, N = 3SE +/- 0.15, N = 3SE +/- 0.09, N = 352.252.752.2
MinAvgMaxClear Linux: 8490H 2P123.5232.7353.5Clear Linux: EPYC 9654 2P46.0190.7218.7Clear Linux34.0189.2217.8OpenBenchmarking.orgWatts, Fewer Is BetterPyPerformance 1.0.0CPU Power Consumption Monitor1002003004005001. Clear Linux: 8490H 2P: Approximate power consumption of 5895 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 4832 Joules per run.3. Clear Linux: Approximate power consumption of 4731 Joules per run.
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: chaosClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1122334455Min: 52.1 / Avg: 52.17 / Max: 52.2Min: 52.5 / Avg: 52.73 / Max: 53Min: 52 / Avg: 52.17 / Max: 52.3

Meta Performance Per Watts

OpenBenchmarking.orgPerformance Per Watts, More Is BetterMeta Performance Per WattsPerformance Per WattsClear Linux: EPYC 9654 2PClear Linux110022003300440055004937.644923.04

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunClear Linux: 8490H 2P120240360480600SE +/- 1.04, N = 3535.16MIN: 60.06 / MAX: 6000
OpenBenchmarking.orgQueries Per Minute, Geo Mean Per Watt, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunClear Linux: 8490H 2P0.29180.58360.87541.16721.4591.297
MinAvgMaxClear Linux: 8490H 2P128413678OpenBenchmarking.orgWatts, Fewer Is BetterClickHouse 22.12.3.5CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 89529 Joules per run.
OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunClear Linux: 8490H 2P90180270360450Min: 533.26 / Avg: 535.16 / Max: 536.83

100M Rows Hits Dataset, Third Run

Clear Linux: EPYC 9654 2P: The test run did not produce a result.

Clear Linux: The test run did not produce a result.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunClear Linux: 8490H 2P110220330440550SE +/- 2.62, N = 3530.44MIN: 63.9 / MAX: 5454.55
OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunClear Linux: 8490H 2P90180270360450Min: 525.25 / Avg: 530.44 / Max: 533.65

100M Rows Hits Dataset, Second Run

Clear Linux: EPYC 9654 2P: The test run did not produce a result.

Clear Linux: The test run did not produce a result.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheClear Linux: 8490H 2P110220330440550SE +/- 1.60, N = 3504.28MIN: 60.73 / MAX: 5454.55
OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheClear Linux: 8490H 2P90180270360450Min: 501.19 / Avg: 504.28 / Max: 506.55

100M Rows Hits Dataset, First Run / Cold Cache

Clear Linux: EPYC 9654 2P: The test run did not produce a result.

Clear Linux: The test run did not produce a result.

BRL-CAD

BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.34VGR Performance Metric8490H 2P900K1800K2700K3600K4500K42314221. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -std=c++14 -fvisibility=hidden -fno-strict-aliasing -fno-common -ftemplate-depth-128 -ggdb3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lregex_brl -lz_brl -lnetpbm -ldl -lm -ltk8.6 -lSM -lICE -lXext -lXi
OpenBenchmarking.orgVGR Performance Metric Per Watt, More Is BetterBRL-CAD 7.34VGR Performance MetricClear Linux: 8490H 2P3K6K9K12K15K12413.45
OpenBenchmarking.orgWatts, Fewer Is BetterBRL-CAD 7.34CPU Power Consumption MonitorClear Linux: 8490H 2P90180270360450Min: 130.75 / Avg: 340.87 / Max: 528.56

VGR Performance Metric

Clear Linux: EPYC 9654 2P: The test run did not produce a result.

CPU Power Consumption Monitor

OpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System MonitoringClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux130260390520650Min: 62.29 / Avg: 477.92 / Max: 756.4Min: 19 / Avg: 320.04 / Max: 757.24Min: 15.47 / Avg: 350.9 / Max: 746.4

Stress-NG

MinAvgMaxClear Linux: 8490H 2P131.4527.8588.3Clear Linux: EPYC 9654 2P51.8340.4390.5Clear Linux43.3337.8381.2OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor160320480640800

OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: FutexClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10K318.059100.349091.38

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: FutexClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux700K1400K2100K2800K3500KSE +/- 3457.07, N = 15SE +/- 26901.92, N = 3SE +/- 29815.22, N = 3167854.273097472.103071495.291. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: FutexClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10K318.059100.349091.38
MinAvgMaxClear Linux: 8490H 2P131.4527.8588.3Clear Linux: EPYC 9654 2P51.8340.4390.5Clear Linux43.3337.8381.2OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor1603204806408001. Clear Linux: 8490H 2P: Approximate power consumption of 19774 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 11913 Joules per run.3. Clear Linux: Approximate power consumption of 11825 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: FutexClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux500K1000K1500K2000K2500KMin: 153007.67 / Avg: 167854.27 / Max: 197164.13Min: 3046903.1 / Avg: 3097472.1 / Max: 3138669.55Min: 3031565.54 / Avg: 3071495.29 / Max: 3129814.471. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

MinAvgMaxClear Linux: 8490H 2P132634755Clear Linux: EPYC 9654 2P50477559Clear Linux37481557OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor2004006008001000

OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: NUMAClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.02970.05940.08910.11880.14850.0250.1320.127

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: NUMAClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1428425670SE +/- 0.02, N = 3SE +/- 0.12, N = 3SE +/- 1.05, N = 1515.8063.0160.861. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: NUMAClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.02970.05940.08910.11880.14850.0250.1320.127
MinAvgMaxClear Linux: 8490H 2P132634755Clear Linux: EPYC 9654 2P50477559Clear Linux37481557OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 22173 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 16689 Joules per run.3. Clear Linux: Approximate power consumption of 16517 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: NUMAClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1224364860Min: 15.77 / Avg: 15.8 / Max: 15.84Min: 62.78 / Avg: 63.01 / Max: 63.14Min: 52.75 / Avg: 60.86 / Max: 63.421. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

MinAvgMaxClear Linux: 8490H 2P130349755Clear Linux: EPYC 9654 2P52215449Clear Linux46214471OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor2004006008001000

OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: CPU CacheClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.0540.1080.1620.2160.270.0460.2390.240

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU CacheClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1224364860SE +/- 0.01, N = 3SE +/- 0.21, N = 3SE +/- 0.83, N = 1515.9051.4051.201. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: CPU CacheClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.0540.1080.1620.2160.270.0460.2390.240
MinAvgMaxClear Linux: 8490H 2P130349755Clear Linux: EPYC 9654 2P52215449Clear Linux46214471OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 12221 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 7592 Joules per run.3. Clear Linux: Approximate power consumption of 7352 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU CacheClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1020304050Min: 15.89 / Avg: 15.9 / Max: 15.91Min: 50.99 / Avg: 51.4 / Max: 51.72Min: 46.07 / Avg: 51.2 / Max: 56.371. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

MinAvgMaxClear Linux: 8490H 2P126.9508.2559.9Clear Linux: EPYC 9654 2P46.4325.6374.0Clear Linux42.8320.2363.3OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor140280420560700

OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: AtomicClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux200400600800100035.07639.41849.69

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: AtomicClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux60K120K180K240K300KSE +/- 235.40, N = 15SE +/- 15718.21, N = 15SE +/- 31981.50, N = 1517822.07208201.40272067.021. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: AtomicClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux200400600800100035.07639.41849.69
MinAvgMaxClear Linux: 8490H 2P126.9508.2559.9Clear Linux: EPYC 9654 2P46.4325.6374.0Clear Linux42.8320.2363.3OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor1402804205607001. Clear Linux: 8490H 2P: Approximate power consumption of 18294 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 11179 Joules per run.3. Clear Linux: Approximate power consumption of 10972 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: AtomicClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux50K100K150K200K250KMin: 16465.49 / Avg: 17822.07 / Max: 20208.75Min: 157464.02 / Avg: 208201.4 / Max: 364689.25Min: 157576.8 / Avg: 272067.02 / Max: 454245.361. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

MinAvgMaxClear Linux: 8490H 2P128634755Clear Linux: EPYC 9654 2P52616704Clear Linux29610692OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor2004006008001000

OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: Socket ActivityClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200155.58165.83184.04

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Socket ActivityClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20K40K60K80K100KSE +/- 8680.33, N = 12SE +/- 877.37, N = 8SE +/- 9969.26, N = 1598576.81102162.24112224.531. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.14.06Test: Socket ActivityClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200155.58165.83184.04
MinAvgMaxClear Linux: 8490H 2P128634755Clear Linux: EPYC 9654 2P52616704Clear Linux29610692OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.14.06CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 21701 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 21177 Joules per run.3. Clear Linux: Approximate power consumption of 20774 Joules per run.
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Socket ActivityClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux20K40K60K80K100KMin: 78703.14 / Avg: 98576.81 / Max: 192303.11Min: 98580.64 / Avg: 102162.24 / Max: 105521.48Min: 95327.5 / Avg: 112224.53 / Max: 251376.721. (CC) gcc options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -O2 -std=gnu99 -lm -fuse-ld=gold -laio -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lkmod -lrt -lz -pthread

PostgreSQL

MinAvgMaxClear Linux: 8490H 2P135.9510.3591.0Clear Linux: EPYC 9654 2P45.6404.3483.9Clear Linux48.5395.5482.8OpenBenchmarking.orgWatts, Fewer Is BetterPostgreSQL 15CPU Power Consumption Monitor160320480640800

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average LatencyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux48121620SE +/- 0.01, N = 3SE +/- 0.32, N = 12SE +/- 0.30, N = 1214.2414.6114.701. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpgcommon -lpgport -lpq -lm
MinAvgMaxClear Linux: 8490H 2P135.9510.3591.0Clear Linux: EPYC 9654 2P45.6404.3483.9Clear Linux48.5395.5482.8OpenBenchmarking.orgWatts, Fewer Is BetterPostgreSQL 15CPU Power Consumption Monitor1603204806408001. Clear Linux: 8490H 2P: Approximate power consumption of 72796 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 56704 Joules per run.3. Clear Linux: Approximate power consumption of 55536 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average LatencyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux48121620Min: 14.23 / Avg: 14.24 / Max: 14.26Min: 13.31 / Avg: 14.61 / Max: 16.84Min: 13.25 / Avg: 14.7 / Max: 16.51. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read WriteClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux15K30K45K60K75KSE +/- 51.90, N = 3SE +/- 1422.48, N = 12SE +/- 1359.28, N = 127021368796683491. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read WriteClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux12K24K36K48K60KMin: 70118.59 / Avg: 70212.71 / Max: 70297.67Min: 59394 / Avg: 68796.1 / Max: 75151.31Min: 60620.71 / Avg: 68349.24 / Max: 75466.881. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpgcommon -lpgport -lpq -lm

MinAvgMaxClear Linux: 8490H 2P136.4499.1588.0Clear Linux: EPYC 9654 2P51.7397.2482.8Clear Linux46.9384.3461.7OpenBenchmarking.orgWatts, Fewer Is BetterPostgreSQL 15CPU Power Consumption Monitor160320480640800

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average LatencyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3691215SE +/- 0.03, N = 3SE +/- 0.30, N = 9SE +/- 0.15, N = 310.5210.9211.231. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpgcommon -lpgport -lpq -lm
MinAvgMaxClear Linux: 8490H 2P136.4499.1588.0Clear Linux: EPYC 9654 2P51.7397.2482.8Clear Linux46.9384.3461.7OpenBenchmarking.orgWatts, Fewer Is BetterPostgreSQL 15CPU Power Consumption Monitor1603204806408001. Clear Linux: 8490H 2P: Approximate power consumption of 71042 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 55869 Joules per run.3. Clear Linux: Approximate power consumption of 54446 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average LatencyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3691215Min: 10.47 / Avg: 10.52 / Max: 10.57Min: 9.98 / Avg: 10.92 / Max: 12.67Min: 10.94 / Avg: 11.23 / Max: 11.441. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read WriteClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux16K32K48K64K80KSE +/- 214.33, N = 3SE +/- 1898.65, N = 9SE +/- 963.04, N = 37602573652712751. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read WriteClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux13K26K39K52K65KMin: 75693.49 / Avg: 76024.87 / Max: 76426.05Min: 63149.06 / Avg: 73652.5 / Max: 80179.24Min: 69938.95 / Avg: 71274.54 / Max: 73144.211. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -lpgcommon -lpgport -lpq -lm

SVT-AV1

MinAvgMaxClear Linux: 8490H 2P130.2298.2506.4Clear Linux: EPYC 9654 2P49.1164.4331.4Clear Linux42.0166.1322.5OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 1.4CPU Power Consumption Monitor130260390520650

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 4KClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.35370.70741.06111.41481.76850.7521.5721.427

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 4KClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux60120180240300SE +/- 1.05, N = 6SE +/- 8.59, N = 15SE +/- 5.47, N = 15224.36258.44236.891. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 4KClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.35370.70741.06111.41481.76850.7521.5721.427
MinAvgMaxClear Linux: 8490H 2P130.2298.2506.4Clear Linux: EPYC 9654 2P49.1164.4331.4Clear Linux42.0166.1322.5OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 1.4CPU Power Consumption Monitor1302603905206501. Clear Linux: 8490H 2P: Approximate power consumption of 3181 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 1370 Joules per run.3. Clear Linux: Approximate power consumption of 1417 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 4KClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux50100150200250Min: 221.8 / Avg: 224.36 / Max: 228.96Min: 206.79 / Avg: 258.44 / Max: 321.3Min: 203 / Avg: 236.89 / Max: 269.411. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

MinAvgMaxClear Linux: 8490H 2P123.8286.0480.6Clear Linux: EPYC 9654 2P47.2157.5308.6Clear Linux41.6159.4306.2OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 1.4CPU Power Consumption Monitor120240360480600

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 4KClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.32420.64840.97261.29681.6210.7461.4411.301

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 4KClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux50100150200250SE +/- 2.96, N = 15SE +/- 7.24, N = 15SE +/- 5.97, N = 15213.45226.93207.351. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 4KClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.32420.64840.97261.29681.6210.7461.4411.301
MinAvgMaxClear Linux: 8490H 2P123.8286.0480.6Clear Linux: EPYC 9654 2P47.2157.5308.6Clear Linux41.6159.4306.2OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 1.4CPU Power Consumption Monitor1202403604806001. Clear Linux: 8490H 2P: Approximate power consumption of 2440 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 1228 Joules per run.3. Clear Linux: Approximate power consumption of 1285 Joules per run.
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 4KClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux4080120160200Min: 172.56 / Avg: 213.45 / Max: 218.42Min: 148.46 / Avg: 226.93 / Max: 257.19Min: 148.11 / Avg: 207.35 / Max: 241.071. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Renaissance

MinAvgMaxClear Linux: 8490H 2P134.5361.1422.7Clear Linux: EPYC 9654 2P42.1213.9263.3Clear Linux52.5211.8260.1OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption Monitor110220330440550

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Scala DottyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux140280420560700SE +/- 10.05, N = 15SE +/- 10.77, N = 15SE +/- 2.60, N = 3644.9625.0578.0MIN: 515.92 / MAX: 950.96MIN: 492.3 / MAX: 932.64MIN: 492.04 / MAX: 909.55
MinAvgMaxClear Linux: 8490H 2P134.5361.1422.7Clear Linux: EPYC 9654 2P42.1213.9263.3Clear Linux52.5211.8260.1OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption Monitor1102203304405501. Clear Linux: 8490H 2P: Approximate power consumption of 18679 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 10410 Joules per run.3. Clear Linux: Approximate power consumption of 10306 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Scala DottyClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux110220330440550Min: 609.34 / Avg: 644.92 / Max: 699.98Min: 576.49 / Avg: 625.04 / Max: 692.46Min: 572.89 / Avg: 578.04 / Max: 581.27

OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption MonitorClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux120240360480600Min: 126.9 / Avg: 396.68 / Max: 689.1Min: 53.78 / Avg: 248.38 / Max: 389.37Min: 41.8 / Avg: 237.74 / Max: 417.37

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Savina Reactors.IOClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KSE +/- 138.98, N = 12SE +/- 101.90, N = 3SE +/- 274.01, N = 129853.79999.38672.4MIN: 9315.85 / MAX: 17034.6MIN: 9868.41 / MAX: 17072.77MIN: 7489.86 / MAX: 15895.97
OpenBenchmarking.orgWatts, Fewer Is BetterRenaissance 0.14CPU Power Consumption MonitorClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux120240360480600Min: 126.9 / Avg: 396.68 / Max: 689.1Min: 53.78 / Avg: 248.38 / Max: 389.37Min: 41.8 / Avg: 237.74 / Max: 417.371. Clear Linux: 8490H 2P: Approximate power consumption of 56427 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 35436 Joules per run.3. Clear Linux: Approximate power consumption of 29599 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Savina Reactors.IOClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KMin: 9315.85 / Avg: 9853.66 / Max: 10835.03Min: 9868.41 / Avg: 9999.33 / Max: 10200.05Min: 7489.86 / Avg: 8672.35 / Max: 10203.39

GROMACS

MinAvgMaxClear Linux: 8490H 2P124613754Clear Linux: EPYC 9654 2P49522697Clear Linux45530711OpenBenchmarking.orgWatts, Fewer Is BetterGROMACS 2022.1CPU Power Consumption Monitor2004006008001000

OpenBenchmarking.orgNs Per Day Per Watt, More Is BetterGROMACS 2022.1Implementation: MPI CPU - Input: water_GMX50_bareClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.00810.01620.02430.03240.04050.0240.0360.034

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2022.1Implementation: MPI CPU - Input: water_GMX50_bareClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux510152025SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.90, N = 1214.9818.7117.831. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17
OpenBenchmarking.orgNs Per Day Per Watt, More Is BetterGROMACS 2022.1Implementation: MPI CPU - Input: water_GMX50_bareClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.00810.01620.02430.03240.04050.0240.0360.034
MinAvgMaxClear Linux: 8490H 2P124613754Clear Linux: EPYC 9654 2P49522697Clear Linux45530711OpenBenchmarking.orgWatts, Fewer Is BetterGROMACS 2022.1CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 17164 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 14086 Joules per run.3. Clear Linux: Approximate power consumption of 14224 Joules per run.
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2022.1Implementation: MPI CPU - Input: water_GMX50_bareClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux510152025Min: 14.94 / Avg: 14.98 / Max: 15.02Min: 18.69 / Avg: 18.71 / Max: 18.72Min: 7.92 / Avg: 17.83 / Max: 18.791. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17

oneDNN

MinAvgMaxClear Linux: 8490H 2P123530750Clear Linux: EPYC 9654 2P33393600Clear Linux42389611OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 3.0CPU Power Consumption Monitor2004006008001000

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.44330.88661.32991.77322.2165SE +/- 0.00444, N = 5SE +/- 0.02694, N = 15SE +/- 0.08522, N = 151.519361.970201.75315MIN: 1.15MIN: 1.35MIN: 1.221. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
MinAvgMaxClear Linux: 8490H 2P123530750Clear Linux: EPYC 9654 2P33393600Clear Linux42389611OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 3.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 7095 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 5211 Joules per run.3. Clear Linux: Approximate power consumption of 5131 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810Min: 1.51 / Avg: 1.52 / Max: 1.54Min: 1.73 / Avg: 1.97 / Max: 2.12Min: 1.45 / Avg: 1.75 / Max: 2.321. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

MinAvgMaxClear Linux: 8490H 2P125580752Clear Linux: EPYC 9654 2P43433629Clear Linux43428606OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 3.0CPU Power Consumption Monitor2004006008001000

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1.08432.16863.25294.33725.4215SE +/- 0.004589, N = 4SE +/- 0.177585, N = 15SE +/- 0.272228, N = 120.9766224.8191004.169190MIN: 0.82MIN: 2.71MIN: 2.461. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
MinAvgMaxClear Linux: 8490H 2P125580752Clear Linux: EPYC 9654 2P43433629Clear Linux43428606OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 3.0CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 11303 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 8282 Joules per run.3. Clear Linux: Approximate power consumption of 8206 Joules per run.
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPUClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux246810Min: 0.97 / Avg: 0.98 / Max: 0.99Min: 3.87 / Avg: 4.82 / Max: 6.5Min: 3.41 / Avg: 4.17 / Max: 6.181. (CXX) g++ options: -O3 -march=native -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

ONNX Runtime

MinAvgMaxClear Linux: 8490H 2P72614636Clear Linux: EPYC 9654 2P48447467Clear Linux44440460OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption Monitor2004006008001000

OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.63771.27541.91312.55083.18852.8342.5732.030

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux400800120016002000SE +/- 17.61, N = 6SE +/- 26.51, N = 11SE +/- 5.33, N = 3173911518941. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux0.63771.27541.91312.55083.18852.8342.5732.030
MinAvgMaxClear Linux: 8490H 2P72614636Clear Linux: EPYC 9654 2P48447467Clear Linux44440460OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption Monitor20040060080010001. Clear Linux: 8490H 2P: Approximate power consumption of 76795 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 55396 Joules per run.3. Clear Linux: Approximate power consumption of 54892 Joules per run.
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux30060090012001500Min: 1655.5 / Avg: 1738.58 / Max: 1776Min: 886 / Avg: 1150.55 / Max: 1187.5Min: 886.5 / Avg: 894.33 / Max: 904.51. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption MonitorClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux80160240320400Min: 126.24 / Avg: 432.84 / Max: 457.84Min: 40.77 / Avg: 244.28 / Max: 253.15Min: 34.17 / Avg: 243.48 / Max: 251.53

OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux61218243025.6924.4823.43

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KSE +/- 299.17, N = 12SE +/- 178.04, N = 12SE +/- 7.78, N = 311120598157051. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux61218243025.6924.4823.43
OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption MonitorClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux80160240320400Min: 126.24 / Avg: 432.84 / Max: 457.84Min: 40.77 / Avg: 244.28 / Max: 253.15Min: 34.17 / Avg: 243.48 / Max: 251.531. Clear Linux: 8490H 2P: Approximate power consumption of 54033 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 30352 Joules per run.3. Clear Linux: Approximate power consumption of 30435 Joules per run.
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KMin: 10594.5 / Avg: 11120.13 / Max: 14369Min: 5713 / Avg: 5981.42 / Max: 7936.5Min: 5689.5 / Avg: 5705 / Max: 57141. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

MinAvgMaxClear Linux: 8490H 2P128.5432.6443.5Clear Linux: EPYC 9654 2P43.1256.6267.4Clear Linux46.1252.3261.9OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption Monitor120240360480600

OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux61218243027.0625.1421.48

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux3K6K9K12K15KSE +/- 76.22, N = 3SE +/- 323.96, N = 9SE +/- 28.14, N = 311707645354171. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux61218243027.0625.1421.48
MinAvgMaxClear Linux: 8490H 2P128.5432.6443.5Clear Linux: EPYC 9654 2P43.1256.6267.4Clear Linux46.1252.3261.9OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.11CPU Power Consumption Monitor1202403604806001. Clear Linux: 8490H 2P: Approximate power consumption of 53928 Joules per run.2. Clear Linux: EPYC 9654 2P: Approximate power consumption of 31767 Joules per run.3. Clear Linux: Approximate power consumption of 31363 Joules per run.
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: StandardClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux2K4K6K8K10KMin: 11576.5 / Avg: 11707 / Max: 11840.5Min: 5296.5 / Avg: 6452.78 / Max: 7381Min: 5371 / Avg: 5416.67 / Max: 54681. (CXX) g++ options: -O3 -pipe -fexceptions -m64 -ffat-lto-objects -fno-trapping-math -mrelax-cmpxchg-loop -std=gnu++17 -ffunction-sections -fdata-sections -march=native -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

Zstd Compression

This test measures the time needed to compress/decompress a sample input file using Zstd compression supplied by the system or otherwise externally of the test profile. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 19, Long Mode - Compression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1326395265SE +/- 0.86, N = 12SE +/- 0.12, N = 3SE +/- 0.96, N = 1551.654.655.71. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***
OpenBenchmarking.orgMB/s, More Is BetterZstd CompressionCompression Level: 19, Long Mode - Compression SpeedClear Linux: 8490H 2PClear Linux: EPYC 9654 2PClear Linux1122334455Min: 47.4 / Avg: 51.57 / Max: 57.2Min: 54.4 / Avg: 54.63 / Max: 54.8Min: 48.4 / Avg: 55.67 / Max: 59.51. *** zstd command line interface 64-bits v1.5.2, by Yann Collet ***

158 Results Shown

oneDNN
Stress-NG:
  Context Switching
  Memory Copying
oneDNN
ONNX Runtime
Stress-NG
miniBUDE:
  OpenMP - BM1:
    Billion Interactions/s
    GFInst/s
Stress-NG
ONNX Runtime
NAMD
Stress-NG
High Performance Conjugate Gradient
Kvazaar:
  Bosphorus 4K - Slow
  Bosphorus 4K - Medium
Blender
Stress-NG
miniBUDE:
  OpenMP - BM2:
    Billion Interactions/s
    GFInst/s
Stress-NG
Embree
Liquid-DSP
oneDNN:
  Convolution Batch Shapes Auto - bf16bf16bf16 - CPU
  Recurrent Neural Network Training - bf16bf16bf16 - CPU
Blender
ONNX Runtime
Stress-NG:
  Glibc C String Functions
  SENDFILE
Renaissance
Zstd Compression
uvg266
Embree
Cpuminer-Opt
Embree
Cpuminer-Opt
Zstd Compression
Cpuminer-Opt
Xcompact3d Incompact3d
Stress-NG:
  Crypto
  Matrix Math
Zstd Compression:
  19 - Decompression Speed
  8 - Decompression Speed
ASTC Encoder
LeelaChessZero
Intel Open Image Denoise
Liquid-DSP
Zstd Compression
DaCapo Benchmark
Stress-NG
ASTC Encoder
Stress-NG
Zstd Compression
Cpuminer-Opt
Numenta Anomaly Benchmark
NWChem
Numenta Anomaly Benchmark
OSPRay
ASTC Encoder
libavif avifenc
Renaissance
PyPerformance
Renaissance
Numenta Anomaly Benchmark
PyPerformance
Zstd Compression
Numenta Anomaly Benchmark
OSPRay
OpenVKL
DaCapo Benchmark
MariaDB
PyBench
Numenta Anomaly Benchmark
libavif avifenc
PyPerformance
Numenta Anomaly Benchmark
Kvazaar
Renaissance
ASTC Encoder
libavif avifenc
Renaissance
Kvazaar
Renaissance
Kvazaar
Stress-NG
Renaissance
PyPerformance:
  float
  python_startup
uvg266:
  Bosphorus 4K - Very Fast
  Bosphorus 4K - Super Fast
  Bosphorus 4K - Ultra Fast
PyPerformance:
  2to3
  raytrace
simdjson
Renaissance
PyPerformance:
  pickle_pure_python
  crypto_pyaes
  go
  django_template
simdjson
PHPBench
PyPerformance
simdjson
Stress-NG
PyPerformance
Meta Performance Per Watts
ClickHouse:
  100M Rows Hits Dataset, Third Run
  100M Rows Hits Dataset, Second Run
  100M Rows Hits Dataset, First Run / Cold Cache
BRL-CAD
CPU Power Consumption Monitor:
  Phoronix Test Suite System Monitoring
  CPU Power Consumption Monitor
  Futex
Stress-NG
Stress-NG:
  CPU Power Consumption Monitor
  NUMA
Stress-NG
Stress-NG:
  CPU Power Consumption Monitor
  CPU Cache
Stress-NG
Stress-NG:
  CPU Power Consumption Monitor
  Atomic
Stress-NG
Stress-NG:
  CPU Power Consumption Monitor
  Socket Activity
Stress-NG
PostgreSQL
PostgreSQL:
  100 - 1000 - Read Write - Average Latency
  100 - 1000 - Read Write
PostgreSQL
PostgreSQL:
  100 - 800 - Read Write - Average Latency
  100 - 800 - Read Write
SVT-AV1:
  CPU Power Consumption Monitor
  Preset 12 - Bosphorus 4K
SVT-AV1
SVT-AV1:
  CPU Power Consumption Monitor
  Preset 13 - Bosphorus 4K
SVT-AV1
Renaissance
Renaissance
Renaissance
Renaissance
GROMACS:
  CPU Power Consumption Monitor
  MPI CPU - water_GMX50_bare
GROMACS
oneDNN
oneDNN
oneDNN
oneDNN
ONNX Runtime:
  CPU Power Consumption Monitor
  ArcFace ResNet-100 - CPU - Standard
ONNX Runtime
ONNX Runtime:
  CPU Power Consumption Monitor
  GPT-2 - CPU - Standard
ONNX Runtime
ONNX Runtime:
  CPU Power Consumption Monitor
  super-resolution-10 - CPU - Standard
ONNX Runtime
Zstd Compression

Clear Linux: 8490H 2P

Processor: 2 x Intel Xeon Platinum 8490H @ 3.50GHz (120 Cores / 240 Threads), Motherboard: Quanta Cloud S6Q-MB-MPS (3A10.uh BIOS), Chipset: Intel Device 1bce, Memory: 1008GB, Disk: 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007 + 960GB INTEL SSDSC2KG96, Graphics: ASPEED, Monitor: VGA HDMI, Network: 4 x Intel E810-C for QSFP + 2 x Intel X710 for 10GBASE-T

OS: Clear Linux OS 38100, Kernel: 6.1.7-1247.native (x86_64), Display Server: X Server 1.21.1.6, Compiler: GCC 12.2.1 20230112 releases/gcc-12.2.0-345-gc142b30b3a + Clang 15.0.6 + LLVM 15.0.6, File-System: ext4, Screen Resolution: 640x480

Kernel Notes: Transparent Huge Pages: always
Environment Notes: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop -fvisibility-inlines-hidden -Wl,--enable-new-dtags -std=gnu++17" FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,-sort-common -Wl,--enable-new-dtags" CFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx""
Compiler Notes: --build=x86_64-generic-linux --disable-libmpx --disable-libunwind-exceptions --disable-multiarch --disable-vtable-verify --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-clocale=gnu --enable-default-pie --enable-gnu-indirect-function --enable-gnu-indirect-function --enable-host-shared --enable-languages=c,c++,fortran,go,jit --enable-ld=default --enable-libstdcxx-pch --enable-linux-futex --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=x86-64-v3 --with-gcc-major-version-only --with-glibc-version=2.35 --with-gnu-ld --with-isl --with-pic --with-ppl=yes --with-tune=sapphirerapids --with-zstd
Processor Notes: Scaling Governor: intel_pstate performance (EPP: default) - CPU Microcode: 0x2b0000c0
Java Notes: OpenJDK Runtime Environment (build 18.0.2-internal+0-adhoc.mockbuild.corretto-18-18.0.2.9.1)
Python Notes: Python 3.11.0
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 23 January 2023 10:51 by user phoronix.

Clear Linux: EPYC 9654 2P

Kernel Notes: Transparent Huge Pages: always
Environment Notes: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop -fvisibility-inlines-hidden -Wl,--enable-new-dtags -std=gnu++17" FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,-sort-common -Wl,--enable-new-dtags" CFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx""
Compiler Notes: --build=x86_64-generic-linux --disable-libmpx --disable-libunwind-exceptions --disable-multiarch --disable-vtable-verify --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-clocale=gnu --enable-default-pie --enable-gnu-indirect-function --enable-gnu-indirect-function --enable-host-shared --enable-languages=c,c++,fortran,go,jit --enable-ld=default --enable-libstdcxx-pch --enable-linux-futex --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=x86-64-v3 --with-gcc-major-version-only --with-glibc-version=2.35 --with-gnu-ld --with-isl --with-pic --with-ppl=yes --with-tune=sapphirerapids --with-zstd
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Java Notes: OpenJDK Runtime Environment (build 18.0.2-internal+0-adhoc.mockbuild.corretto-18-18.0.2.9.1)
Python Notes: Python 3.11.0
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 25 January 2023 10:46 by user phoronix.

Clear Linux

Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Clear Linux OS 38100, Kernel: 6.1.7-1247.native (x86_64), Display Server: X Server 1.21.1.6, Compiler: GCC 12.2.1 20230112 releases/gcc-12.2.0-345-gc142b30b3a + Clang 15.0.6 + LLVM 15.0.6, File-System: ext4, Screen Resolution: 640x480

Kernel Notes: Transparent Huge Pages: always
Environment Notes: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop -fvisibility-inlines-hidden -Wl,--enable-new-dtags -std=gnu++17" FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,-sort-common -Wl,--enable-new-dtags" CFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z,now -Wl,-z,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mrelax-cmpxchg-loop" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx""
Compiler Notes: --build=x86_64-generic-linux --disable-libmpx --disable-libunwind-exceptions --disable-multiarch --disable-vtable-verify --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-clocale=gnu --enable-default-pie --enable-gnu-indirect-function --enable-gnu-indirect-function --enable-host-shared --enable-languages=c,c++,fortran,go,jit --enable-ld=default --enable-libstdcxx-pch --enable-linux-futex --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=x86-64-v3 --with-gcc-major-version-only --with-glibc-version=2.35 --with-gnu-ld --with-isl --with-pic --with-ppl=yes --with-tune=sapphirerapids --with-zstd
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Java Notes: OpenJDK Runtime Environment (build 18.0.2-internal+0-adhoc.mockbuild.corretto-18-18.0.2.9.1)
Python Notes: Python 3.11.0
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 26 January 2023 01:38 by user phoronix.