System76 Thelio Kernels Linux 6.8

Benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2401218-PTS-SYSTEM7660
Jump To Table - Results

View

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

Limit displaying results to tests within:

AV1 2 Tests
Timed Code Compilation 4 Tests
C/C++ Compiler Tests 6 Tests
CPU Massive 11 Tests
Creator Workloads 5 Tests
Encoding 3 Tests
HPC - High Performance Computing 4 Tests
Common Kernel Benchmarks 2 Tests
Machine Learning 4 Tests
Multi-Core 10 Tests
NVIDIA GPU Compute 2 Tests
Programmer / Developer System Benchmarks 4 Tests
Python Tests 5 Tests
Server 2 Tests
Server CPU Tests 7 Tests
Video Encoding 2 Tests

Statistics

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

Graph Settings

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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
v6.6.6 Stock
January 18
  8 Hours, 38 Minutes
v6.7
January 19
  6 Hours, 14 Minutes
v6.8 20 Jan
January 20
  6 Hours
Invert Hiding All Results Option
  6 Hours, 57 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):

v6.6.6 Stock

Processor: AMD Ryzen Threadripper 7980X 64-Cores @ 7.79GHz (64 Cores / 128 Threads), Motherboard: System76 Thelio Major (FA Z5 BIOS), Chipset: AMD Device 14a4, Memory: 4 x 32 GB DRAM-4800MT/s Micron MTC20F1045S1RC48BA2, Disk: 1000GB CT1000T700SSD5, Graphics: NAVI31 45GB (1760/1124MHz), Audio: AMD Device 14cc, Monitor: ASUS VP28U, Network: Aquantia Device 14c0 + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Pop 22.04, Kernel: 6.6.6-76060606-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, OpenGL: 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.54), Vulkan: 1.3.267, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105
Python Notes: Python 3.10.12
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + 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 / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

v6.7

OS: Pop 22.04, Kernel: 6.7.0-060700-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, OpenGL: 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.56), Vulkan: 1.3.267, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160

v6.8 20 Jan

OS: Pop 22.04, Kernel: 6.7.0-060700daily20240120-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, OpenGL: 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.57), Vulkan: 1.3.267, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160

System76 Thelio Kernels Linux 6.8ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen Resolutionv6.6.6 Stockv6.7v6.8 20 JanAMD Ryzen Threadripper 7980X 64-Cores @ 7.79GHz (64 Cores / 128 Threads)System76 Thelio Major (FA Z5 BIOS)AMD Device 14a44 x 32 GB DRAM-4800MT/s Micron MTC20F1045S1RC48BA21000GB CT1000T700SSD5NAVI31 45GB (1760/1124MHz)AMD Device 14ccASUS VP28UAquantia Device 14c0 + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX210/AX211/AX411Pop 22.046.6.6-76060606-generic (x86_64)GNOME Shell 42.5X Server 1.21.1.44.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.54)1.3.267GCC 11.4.0ext43840x21606.7.0-060700-generic (x86_64)4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.56)6.7.0-060700daily20240120-generic (x86_64)4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.57)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105Python Details- Python 3.10.12Security Details- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + 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 / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

v6.6.6 Stockv6.7v6.8 20 JanResult OverviewPhoronix Test Suite100%109%119%128%TensorFlowOpenVINOTimed LLVM CompilationPostgreSQLTimed Linux Kernel CompilationPyTorchStress-NGSpeedbBlenderTimed Node.js Compilation7-Zip CompressionSVT-AV1LeelaChessZeroQuicksilverQuantLibTimed PHP CompilationFLAC Audio Encodingrav1eY-Cruncher

System76 Thelio Kernels Linux 6.8tensorflow: CPU - 512 - ResNet-50openvino: Machine Translation EN To DE FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUstress-ng: Forkingbuild-llvm: Ninjaopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUstress-ng: SENDFILEstress-ng: MEMFDbuild-linux-kernel: allmodconfigstress-ng: Mutexopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUstress-ng: Cloningstress-ng: System V Message Passingpytorch: CPU - 1 - ResNet-50speedb: Read Rand Write Randpytorch: CPU - 16 - ResNet-50pytorch: CPU - 512 - ResNet-50speedb: Update Randstress-ng: MMAPstress-ng: Pipestress-ng: Glibc Qsort Data Sortingstress-ng: Pollstress-ng: CPU Stressstress-ng: Context Switchingopenvino: Handwritten English Recognition FP16-INT8 - CPUopenvino: Handwritten English Recognition FP16-INT8 - CPUpgbench: 100 - 1000 - Read Onlypgbench: 100 - 1000 - Read Only - Average Latencystress-ng: NUMAstress-ng: Mixed Schedulerstress-ng: Function Callblender: Barbershop - CPU-Onlysvt-av1: Preset 12 - Bosphorus 4Kbuild-linux-kernel: defconfigspeedb: Rand Readopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUspeedb: Read While Writingstress-ng: Floating Pointstress-ng: Cryptostress-ng: Hashstress-ng: Vector Mathstress-ng: Vector Shufflestress-ng: Wide Vector Mathstress-ng: Memory Copyingcompress-7zip: Decompression Ratingstress-ng: Vector Floating Pointstress-ng: Matrix Mathquicksilver: CORAL2 P1stress-ng: Mallocbuild-nodejs: Time To Compileopenvino: Face Detection Retail FP16-INT8 - CPUstress-ng: Fused Multiply-Addopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUquicksilver: CTS2stress-ng: Zlibblender: BMW27 - CPU-Onlysvt-av1: Preset 13 - Bosphorus 4Ktensorflow: CPU - 1 - ResNet-50stress-ng: Glibc C String Functionscompress-7zip: Compression Ratinglczero: Eigensvt-av1: Preset 8 - Bosphorus 4Kquantlib: Multi-Threadedstress-ng: AVX-512 VNNIbuild-php: Time To Compilequantlib: Single-Threadedsvt-av1: Preset 4 - Bosphorus 4Krav1e: 6encode-flac: WAV To FLACstress-ng: Pthreadrav1e: 1build-llvm: Unix Makefilesrav1e: 5rav1e: 10y-cruncher: 500My-cruncher: 1Bquicksilver: CORAL2 P2stress-ng: Atomicstress-ng: x86_64 RdRandopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP16 - CPUstress-ng: Matrix 3D Mathstress-ng: Semaphoresstress-ng: CPU Cachestress-ng: IO_uringstress-ng: AVL Treestress-ng: Futexpgbench: 100 - 1000 - Read Write - Average Latencypgbench: 100 - 1000 - Read Writev6.6.6 Stockv6.7v6.8 20 Jan48.27206.02154.9157664.20162.9611271.0425.08816781.54730.62250.16523109393.808.413740.505992.4526211336.6957.32310107746.8646.384446501037.8327118341.211894.287394187.64185200.3820153249.5636.901728.8838869750.2571036.7753828.6659193.78163.38183.37223.905453059822113672.151346725426744.36183183.9917033307.63485001.3455853.543230461.5024004.53556761228832.13385258.1827280000150760521.22121.21413960.2270244822.750.394.366.377.697959.974926.8383.26383.15233266677881.1017.76188.7167.1973865789.5635639536588.830266288.48158371.9427.1214354.99.0817.37012.524125130.591.369181.6465.62017.6015.46311.02322070000319.9411327.06249.98128.732308.9893013794.861268835.491486513.95432.833008388.9834.6892935690.57354.5490.0274065.57128.9201561.1720.421002168.85733.70213.96824181954.447.364254.676798.1026702772.5952.69334863243.2943.194629171042.8428997709.971953.717499767.91194017.2821313434.9034.871828.7740047400.2501068.0955946.6961613.76157.37186.24923.605468123627117409.531304024427569.66188548.4517551780.14499009.4957494.653320137.0224695.64570883234027.64396041.2628040000154494959.66118.04814310.2572100478.140.384.256.217.508161.265045.5185.22374.51238766678041.1217.38192.3057.1275382089.0436339735890.360269955.58282501.9226.6944410.49.2087.39812.369125537.871.384180.1875.67217.5635.48011.05622036667319.6411326.12169.15188.834379.78100143123.741271089.942249186.66635.153014245.4931.9693189990.59355.3589.8270085.01129.3951556.0620.48911584.27626.37222.59026762751.757.334274.686732.4428955467.6352.39338298543.1643.03478743969.1028177842.382020.697868494.17196549.8121334268.5935.021821.2540672490.2461081.6956064.4761542.13157.47179.57024.476467824743116852.121318231127614.73189063.3317563822.26500004.3857552.233328142.0924217.05572516235304.73395555.2627913333150399862.32118.90414329.9671381983.240.384.256.227.508146.275040.8185.25374.31236633338060.1417.40188.2307.2774823547.6135903936388.956270826.18295943.7726.9154422.29.1257.46512.375124066.291.380179.8535.66617.7165.44411.01922086667319.5611328.36162.79196.254331.88106517080.671289645.531779056.72638.583115686.2128.96334526OpenBenchmarking.org

TensorFlow

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

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 512 - Model: ResNet-50v6.6.6 Stockv6.7v6.8 20 Jan20406080100SE +/- 0.22, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 348.2790.5790.59

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Machine Translation EN To DE FP16 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan80160240320400SE +/- 1.84, N = 3SE +/- 1.14, N = 3SE +/- 0.50, N = 3206.02354.54355.351. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Machine Translation EN To DE FP16 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan306090120150SE +/- 1.37, N = 3SE +/- 0.29, N = 3SE +/- 0.12, N = 3154.9190.0289.82MIN: 45.52 / MAX: 256.28MIN: 35.71 / MAX: 120.27MIN: 37.27 / MAX: 122.761. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Forkingv6.6.6 Stockv6.7v6.8 20 Jan16K32K48K64K80KSE +/- 621.24, N = 15SE +/- 80.89, N = 3SE +/- 551.50, N = 1557664.2074065.5770085.011. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

Timed LLVM Compilation

This test times how long it takes to compile/build the LLVM compiler stack. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Ninjav6.6.6 Stockv6.7v6.8 20 Jan4080120160200SE +/- 1.94, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3162.96128.92129.40

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan30060090012001500SE +/- 12.47, N = 15SE +/- 1.78, N = 3SE +/- 6.55, N = 31271.041561.171556.061. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan612182430SE +/- 0.22, N = 15SE +/- 0.02, N = 3SE +/- 0.08, N = 325.0820.4220.48MIN: 11.1 / MAX: 59.25MIN: 11.89 / MAX: 42.47MIN: 9.39 / MAX: 41.181. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: SENDFILEv6.6.6 Stockv6.7v6.8 20 Jan200K400K600K800K1000KSE +/- 5266.72, N = 3SE +/- 4805.80, N = 3SE +/- 5909.38, N = 3816781.541002168.85911584.271. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: MEMFDv6.6.6 Stockv6.7v6.8 20 Jan160320480640800SE +/- 0.47, N = 3SE +/- 0.59, N = 3SE +/- 0.42, N = 3730.62733.70626.371. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: allmodconfigv6.6.6 Stockv6.7v6.8 20 Jan50100150200250SE +/- 4.81, N = 9SE +/- 0.47, N = 3SE +/- 0.17, N = 3250.17213.97222.59

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Mutexv6.6.6 Stockv6.7v6.8 20 Jan6M12M18M24M30MSE +/- 17585.48, N = 3SE +/- 9399.14, N = 3SE +/- 108268.55, N = 323109393.8024181954.4426762751.751. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan246810SE +/- 0.07, N = 15SE +/- 0.00, N = 3SE +/- 0.01, N = 38.417.367.33MIN: 3.83 / MAX: 32.93MIN: 3.93 / MAX: 27.08MIN: 4.02 / MAX: 31.111. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan9001800270036004500SE +/- 35.82, N = 15SE +/- 1.62, N = 3SE +/- 7.29, N = 33740.504254.674274.681. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Cloningv6.6.6 Stockv6.7v6.8 20 Jan15003000450060007500SE +/- 50.66, N = 15SE +/- 1.62, N = 3SE +/- 4.20, N = 35992.456798.106732.441. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: System V Message Passingv6.6.6 Stockv6.7v6.8 20 Jan6M12M18M24M30MSE +/- 24713.27, N = 3SE +/- 35321.93, N = 3SE +/- 346710.82, N = 326211336.6926702772.5928955467.631. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

PyTorch

This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Currently this test profile is catered to CPU-based testing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 1 - Model: ResNet-50v6.6.6 Stockv6.7v6.8 20 Jan1326395265SE +/- 0.26, N = 3SE +/- 0.25, N = 3SE +/- 0.14, N = 357.3252.6952.39MIN: 52.6 / MAX: 59.56MIN: 48.03 / MAX: 54.61MIN: 47.83 / MAX: 54.23

Speedb

Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Read Random Write Randomv6.6.6 Stockv6.7v6.8 20 Jan700K1400K2100K2800K3500KSE +/- 10648.79, N = 3SE +/- 14008.04, N = 3SE +/- 35242.08, N = 33101077334863233829851. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

PyTorch

This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Currently this test profile is catered to CPU-based testing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 16 - Model: ResNet-50v6.6.6 Stockv6.7v6.8 20 Jan1122334455SE +/- 0.23, N = 3SE +/- 0.20, N = 3SE +/- 0.24, N = 346.8643.2943.16MIN: 43.35 / MAX: 48.24MIN: 40.69 / MAX: 44.45MIN: 39.88 / MAX: 44.49

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 512 - Model: ResNet-50v6.6.6 Stockv6.7v6.8 20 Jan1122334455SE +/- 0.21, N = 3SE +/- 0.06, N = 3SE +/- 0.27, N = 346.3843.1943.03MIN: 43.06 / MAX: 47.63MIN: 40.2 / MAX: 44.13MIN: 39.58 / MAX: 44.49

Speedb

Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Update Randomv6.6.6 Stockv6.7v6.8 20 Jan100K200K300K400K500KSE +/- 1868.95, N = 3SE +/- 227.10, N = 3SE +/- 2903.23, N = 34446504629174787431. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: MMAPv6.6.6 Stockv6.7v6.8 20 Jan2004006008001000SE +/- 0.43, N = 3SE +/- 0.51, N = 3SE +/- 0.31, N = 31037.831042.84969.101. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Pipev6.6.6 Stockv6.7v6.8 20 Jan6M12M18M24M30MSE +/- 38316.59, N = 3SE +/- 293042.41, N = 5SE +/- 66944.79, N = 327118341.2128997709.9728177842.381. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Glibc Qsort Data Sortingv6.6.6 Stockv6.7v6.8 20 Jan400800120016002000SE +/- 3.32, N = 3SE +/- 2.29, N = 3SE +/- 1.64, N = 31894.281953.712020.691. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Pollv6.6.6 Stockv6.7v6.8 20 Jan2M4M6M8M10MSE +/- 8703.67, N = 3SE +/- 7207.98, N = 3SE +/- 1265.21, N = 37394187.647499767.917868494.171. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: CPU Stressv6.6.6 Stockv6.7v6.8 20 Jan40K80K120K160K200KSE +/- 1715.48, N = 3SE +/- 511.22, N = 3SE +/- 672.45, N = 3185200.38194017.28196549.811. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Context Switchingv6.6.6 Stockv6.7v6.8 20 Jan5M10M15M20M25MSE +/- 190129.10, N = 15SE +/- 193579.79, N = 15SE +/- 121700.79, N = 320153249.5621313434.9021334268.591. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan816243240SE +/- 0.02, N = 3SE +/- 0.24, N = 3SE +/- 0.08, N = 336.9034.8735.02MIN: 20.64 / MAX: 64.4MIN: 20.35 / MAX: 61.86MIN: 20.04 / MAX: 63.631. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan400800120016002000SE +/- 0.64, N = 3SE +/- 12.63, N = 3SE +/- 4.34, N = 31728.881828.771821.251. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Onlyv6.6.6 Stockv6.7v6.8 20 Jan900K1800K2700K3600K4500KSE +/- 19489.13, N = 3SE +/- 12389.83, N = 3SE +/- 32554.05, N = 33886975400474040672491. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latencyv6.6.6 Stockv6.7v6.8 20 Jan0.05780.11560.17340.23120.289SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 30.2570.2500.2461. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: NUMAv6.6.6 Stockv6.7v6.8 20 Jan2004006008001000SE +/- 1.80, N = 3SE +/- 0.24, N = 3SE +/- 3.43, N = 31036.771068.091081.691. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Mixed Schedulerv6.6.6 Stockv6.7v6.8 20 Jan12K24K36K48K60KSE +/- 166.89, N = 3SE +/- 246.60, N = 3SE +/- 72.36, N = 353828.6655946.6956064.471. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Function Callv6.6.6 Stockv6.7v6.8 20 Jan13K26K39K52K65KSE +/- 468.48, N = 3SE +/- 22.20, N = 3SE +/- 32.44, N = 359193.7861613.7661542.131. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

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 4.0Blend File: Barbershop - Compute: CPU-Onlyv6.6.6 Stockv6.7v6.8 20 Jan4080120160200SE +/- 0.10, N = 3SE +/- 0.19, N = 3SE +/- 0.11, N = 3163.38157.37157.47

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.8Encoder Mode: Preset 12 - Input: Bosphorus 4Kv6.6.6 Stockv6.7v6.8 20 Jan4080120160200SE +/- 1.21, N = 3SE +/- 0.60, N = 3SE +/- 1.68, N = 14183.37186.25179.571. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: defconfigv6.6.6 Stockv6.7v6.8 20 Jan612182430SE +/- 0.22, N = 6SE +/- 0.27, N = 4SE +/- 0.28, N = 423.9123.6124.48

Speedb

Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Random Readv6.6.6 Stockv6.7v6.8 20 Jan100M200M300M400M500MSE +/- 2124225.27, N = 3SE +/- 358460.93, N = 3SE +/- 660803.31, N = 34530598224681236274678247431. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan30K60K90K120K150KSE +/- 181.97, N = 3SE +/- 316.75, N = 3SE +/- 450.26, N = 3113672.15117409.53116852.121. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

Speedb

Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Read While Writingv6.6.6 Stockv6.7v6.8 20 Jan3M6M9M12M15MSE +/- 134947.32, N = 6SE +/- 144991.37, N = 4SE +/- 184754.10, N = 31346725413040244131823111. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Floating Pointv6.6.6 Stockv6.7v6.8 20 Jan6K12K18K24K30KSE +/- 16.76, N = 3SE +/- 32.79, N = 3SE +/- 31.86, N = 326744.3627569.6627614.731. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Cryptov6.6.6 Stockv6.7v6.8 20 Jan40K80K120K160K200KSE +/- 260.91, N = 3SE +/- 161.24, N = 3SE +/- 303.68, N = 3183183.99188548.45189063.331. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Hashv6.6.6 Stockv6.7v6.8 20 Jan4M8M12M16M20MSE +/- 3192.96, N = 3SE +/- 3053.25, N = 3SE +/- 6056.44, N = 317033307.6317551780.1417563822.261. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector Mathv6.6.6 Stockv6.7v6.8 20 Jan110K220K330K440K550KSE +/- 152.98, N = 3SE +/- 258.25, N = 3SE +/- 393.56, N = 3485001.34499009.49500004.381. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector Shufflev6.6.6 Stockv6.7v6.8 20 Jan12K24K36K48K60KSE +/- 29.85, N = 3SE +/- 7.11, N = 3SE +/- 33.42, N = 355853.5457494.6557552.231. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Wide Vector Mathv6.6.6 Stockv6.7v6.8 20 Jan700K1400K2100K2800K3500KSE +/- 830.00, N = 3SE +/- 1754.36, N = 3SE +/- 2085.07, N = 33230461.503320137.023328142.091. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Memory Copyingv6.6.6 Stockv6.7v6.8 20 Jan5K10K15K20K25KSE +/- 17.95, N = 3SE +/- 57.39, N = 3SE +/- 66.11, N = 324004.5324695.6424217.051. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Decompression Ratingv6.6.6 Stockv6.7v6.8 20 Jan120K240K360K480K600KSE +/- 3472.31, N = 3SE +/- 1033.82, N = 3SE +/- 1758.24, N = 35567615708835725161. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector Floating Pointv6.6.6 Stockv6.7v6.8 20 Jan50K100K150K200K250KSE +/- 217.62, N = 3SE +/- 394.44, N = 3SE +/- 1244.36, N = 15228832.13234027.64235304.731. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Matrix Mathv6.6.6 Stockv6.7v6.8 20 Jan80K160K240K320K400KSE +/- 637.54, N = 3SE +/- 272.90, N = 3SE +/- 266.39, N = 3385258.18396041.26395555.261. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

Quicksilver

Quicksilver is a proxy application that represents some elements of the Mercury workload by solving a simplified dynamic Monte Carlo particle transport problem. Quicksilver is developed by Lawrence Livermore National Laboratory (LLNL) and this test profile currently makes use of the OpenMP CPU threaded code path. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterQuicksilver 20230818Input: CORAL2 P1v6.6.6 Stockv6.7v6.8 20 Jan6M12M18M24M30MSE +/- 211266.03, N = 3SE +/- 55075.71, N = 3SE +/- 8819.17, N = 32728000028040000279133331. (CXX) g++ options: -fopenmp -O3 -march=native

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Mallocv6.6.6 Stockv6.7v6.8 20 Jan30M60M90M120M150MSE +/- 917737.04, N = 3SE +/- 1238064.72, N = 3SE +/- 123392.16, N = 3150760521.22154494959.66150399862.321. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 19.8.1Time To Compilev6.6.6 Stockv6.7v6.8 20 Jan306090120150SE +/- 0.10, N = 3SE +/- 0.15, N = 3SE +/- 0.24, N = 3121.21118.05118.90

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan3K6K9K12K15KSE +/- 10.92, N = 3SE +/- 19.49, N = 3SE +/- 20.70, N = 313960.2214310.2514329.961. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Fused Multiply-Addv6.6.6 Stockv6.7v6.8 20 Jan15M30M45M60M75MSE +/- 21862.72, N = 3SE +/- 42666.45, N = 3SE +/- 471689.87, N = 370244822.7572100478.1471381983.241. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan0.08780.17560.26340.35120.439SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.390.380.38MIN: 0.18 / MAX: 26.28MIN: 0.18 / MAX: 19.31MIN: 0.18 / MAX: 23.041. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan0.9811.9622.9433.9244.905SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 34.364.254.25MIN: 2.15 / MAX: 27MIN: 1.95 / MAX: 30.92MIN: 2.02 / MAX: 29.341. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan246810SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 36.376.216.22MIN: 3.34 / MAX: 29.67MIN: 3.17 / MAX: 23.34MIN: 3.17 / MAX: 35.721. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 37.697.507.50MIN: 3.31 / MAX: 27.53MIN: 3.43 / MAX: 27.91MIN: 3.39 / MAX: 29.371. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan2K4K6K8K10KSE +/- 4.62, N = 3SE +/- 9.48, N = 3SE +/- 14.31, N = 37959.978161.268146.271. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan11002200330044005500SE +/- 7.01, N = 3SE +/- 2.97, N = 3SE +/- 5.72, N = 34926.835045.515040.811. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan20406080100SE +/- 0.20, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 383.2685.2285.251. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection FP16-INT8 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan80160240320400SE +/- 0.84, N = 3SE +/- 0.25, N = 3SE +/- 0.18, N = 3383.15374.51374.31MIN: 303.22 / MAX: 402.09MIN: 321.16 / MAX: 396.63MIN: 196.32 / MAX: 397.311. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

Quicksilver

Quicksilver is a proxy application that represents some elements of the Mercury workload by solving a simplified dynamic Monte Carlo particle transport problem. Quicksilver is developed by Lawrence Livermore National Laboratory (LLNL) and this test profile currently makes use of the OpenMP CPU threaded code path. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterQuicksilver 20230818Input: CTS2v6.6.6 Stockv6.7v6.8 20 Jan5M10M15M20M25MSE +/- 32829.53, N = 3SE +/- 38441.88, N = 3SE +/- 14529.66, N = 32332666723876667236633331. (CXX) g++ options: -fopenmp -O3 -march=native

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Zlibv6.6.6 Stockv6.7v6.8 20 Jan2K4K6K8K10KSE +/- 15.44, N = 3SE +/- 39.53, N = 3SE +/- 14.00, N = 37881.108041.128060.141. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

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 4.0Blend File: BMW27 - Compute: CPU-Onlyv6.6.6 Stockv6.7v6.8 20 Jan48121620SE +/- 0.08, N = 3SE +/- 0.13, N = 3SE +/- 0.13, N = 317.7617.3817.40

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.8Encoder Mode: Preset 13 - Input: Bosphorus 4Kv6.6.6 Stockv6.7v6.8 20 Jan4080120160200SE +/- 0.41, N = 3SE +/- 0.83, N = 3SE +/- 0.78, N = 3188.72192.31188.231. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

TensorFlow

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

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 1 - Model: ResNet-50v6.6.6 Stockv6.7v6.8 20 Jan246810SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 37.197.127.27

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Glibc C String Functionsv6.6.6 Stockv6.7v6.8 20 Jan16M32M48M64M80MSE +/- 346470.68, N = 3SE +/- 469274.41, N = 3SE +/- 224384.57, N = 373865789.5675382089.0474823547.611. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression Ratingv6.6.6 Stockv6.7v6.8 20 Jan80K160K240K320K400KSE +/- 1642.20, N = 3SE +/- 967.86, N = 3SE +/- 716.12, N = 33563953633973590391. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

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 0.30Backend: Eigenv6.6.6 Stockv6.7v6.8 20 Jan80160240320400SE +/- 2.67, N = 3SE +/- 4.26, N = 4SE +/- 3.25, N = 73653583631. (CXX) g++ options: -flto -pthread

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.8Encoder Mode: Preset 8 - Input: Bosphorus 4Kv6.6.6 Stockv6.7v6.8 20 Jan20406080100SE +/- 0.57, N = 3SE +/- 0.69, N = 3SE +/- 0.77, N = 388.8390.3688.961. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

QuantLib

QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.32Configuration: Multi-Threadedv6.6.6 Stockv6.7v6.8 20 Jan60K120K180K240K300KSE +/- 1629.93, N = 3SE +/- 922.38, N = 3SE +/- 704.18, N = 3266288.4269955.5270826.11. (CXX) g++ options: -O3 -march=native -fPIE -pie

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: AVX-512 VNNIv6.6.6 Stockv6.7v6.8 20 Jan2M4M6M8M10MSE +/- 86949.56, N = 5SE +/- 3726.82, N = 3SE +/- 10648.54, N = 38158371.948282501.928295943.771. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

Timed PHP Compilation

This test times how long it takes to build PHP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 8.1.9Time To Compilev6.6.6 Stockv6.7v6.8 20 Jan612182430SE +/- 0.02, N = 3SE +/- 0.13, N = 3SE +/- 0.08, N = 327.1226.6926.92

QuantLib

QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.32Configuration: Single-Threadedv6.6.6 Stockv6.7v6.8 20 Jan9001800270036004500SE +/- 23.44, N = 3SE +/- 17.53, N = 3SE +/- 38.02, N = 34354.94410.44422.21. (CXX) g++ options: -O3 -march=native -fPIE -pie

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.8Encoder Mode: Preset 4 - Input: Bosphorus 4Kv6.6.6 Stockv6.7v6.8 20 Jan3691215SE +/- 0.052, N = 3SE +/- 0.044, N = 3SE +/- 0.023, N = 39.0819.2089.1251. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

rav1e

Xiph rav1e is a Rust-written AV1 video encoder that claims to be the fastest and safest AV1 encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.7Speed: 6v6.6.6 Stockv6.7v6.8 20 Jan246810SE +/- 0.022, N = 3SE +/- 0.031, N = 3SE +/- 0.030, N = 37.3707.3987.465

FLAC Audio Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.4WAV To FLACv6.6.6 Stockv6.7v6.8 20 Jan3691215SE +/- 0.03, N = 5SE +/- 0.02, N = 5SE +/- 0.03, N = 512.5212.3712.381. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Pthreadv6.6.6 Stockv6.7v6.8 20 Jan30K60K90K120K150KSE +/- 452.99, N = 3SE +/- 142.99, N = 3SE +/- 80.91, N = 3125130.59125537.87124066.291. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

rav1e

Xiph rav1e is a Rust-written AV1 video encoder that claims to be the fastest and safest AV1 encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.7Speed: 1v6.6.6 Stockv6.7v6.8 20 Jan0.31140.62280.93421.24561.557SE +/- 0.004, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 31.3691.3841.380

Timed LLVM Compilation

This test times how long it takes to compile/build the LLVM compiler stack. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Unix Makefilesv6.6.6 Stockv6.7v6.8 20 Jan4080120160200SE +/- 0.10, N = 3SE +/- 0.32, N = 3SE +/- 0.48, N = 3181.65180.19179.85

rav1e

Xiph rav1e is a Rust-written AV1 video encoder that claims to be the fastest and safest AV1 encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.7Speed: 5v6.6.6 Stockv6.7v6.8 20 Jan1.27622.55243.82865.10486.381SE +/- 0.008, N = 3SE +/- 0.009, N = 3SE +/- 0.015, N = 35.6205.6725.666

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.7Speed: 10v6.6.6 Stockv6.7v6.8 20 Jan48121620SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 317.6017.5617.72

Y-Cruncher

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

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.3Pi Digits To Calculate: 500Mv6.6.6 Stockv6.7v6.8 20 Jan1.2332.4663.6994.9326.165SE +/- 0.006, N = 3SE +/- 0.004, N = 3SE +/- 0.009, N = 35.4635.4805.444

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.3Pi Digits To Calculate: 1Bv6.6.6 Stockv6.7v6.8 20 Jan3691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 311.0211.0611.02

Quicksilver

Quicksilver is a proxy application that represents some elements of the Mercury workload by solving a simplified dynamic Monte Carlo particle transport problem. Quicksilver is developed by Lawrence Livermore National Laboratory (LLNL) and this test profile currently makes use of the OpenMP CPU threaded code path. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterQuicksilver 20230818Input: CORAL2 P2v6.6.6 Stockv6.7v6.8 20 Jan5M10M15M20M25MSE +/- 34641.02, N = 3SE +/- 14529.66, N = 3SE +/- 32829.53, N = 32207000022036667220866671. (CXX) g++ options: -fopenmp -O3 -march=native

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Atomicv6.6.6 Stockv6.7v6.8 20 Jan70140210280350SE +/- 0.15, N = 3SE +/- 0.17, N = 3SE +/- 0.19, N = 3319.94319.64319.561. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: x86_64 RdRandv6.6.6 Stockv6.7v6.8 20 Jan2K4K6K8K10KSE +/- 0.34, N = 3SE +/- 0.18, N = 3SE +/- 0.33, N = 311327.0611326.1211328.361. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Person Detection FP16 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan50100150200250SE +/- 6.32, N = 15SE +/- 0.26, N = 3SE +/- 0.30, N = 3249.98169.15162.79MIN: 40.51 / MAX: 811.65MIN: 69.93 / MAX: 221.3MIN: 79.12 / MAX: 219.931. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Detection FP16 - Device: CPUv6.6.6 Stockv6.7v6.8 20 Jan4080120160200SE +/- 2.69, N = 15SE +/- 0.29, N = 3SE +/- 0.37, N = 3128.73188.83196.251. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Matrix 3D Mathv6.6.6 Stockv6.7v6.8 20 Jan9001800270036004500SE +/- 161.28, N = 15SE +/- 46.41, N = 13SE +/- 1.56, N = 32308.984379.784331.881. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Semaphoresv6.6.6 Stockv6.7v6.8 20 Jan20M40M60M80M100MSE +/- 315601.64, N = 3SE +/- 673729.70, N = 3SE +/- 6780651.21, N = 1593013794.86100143123.74106517080.671. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: CPU Cachev6.6.6 Stockv6.7v6.8 20 Jan300K600K900K1200K1500KSE +/- 16798.31, N = 3SE +/- 23675.00, N = 12SE +/- 21363.01, N = 151268835.491271089.941289645.531. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: IO_uringv6.6.6 Stockv6.7v6.8 20 Jan500K1000K1500K2000K2500KSE +/- 109628.32, N = 12SE +/- 27934.77, N = 3SE +/- 103521.52, N = 121486513.952249186.661779056.721. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: AVL Treev6.6.6 Stockv6.7v6.8 20 Jan140280420560700SE +/- 9.75, N = 14SE +/- 0.67, N = 3SE +/- 0.39, N = 3432.83635.15638.581. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Futexv6.6.6 Stockv6.7v6.8 20 Jan700K1400K2100K2800K3500KSE +/- 38705.18, N = 12SE +/- 51913.47, N = 15SE +/- 17686.67, N = 33008388.983014245.493115686.211. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lz

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latencyv6.6.6 Stockv6.7v6.8 20 Jan816243240SE +/- 1.47, N = 12SE +/- 1.45, N = 12SE +/- 0.03, N = 334.6931.9728.961. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Writev6.6.6 Stockv6.7v6.8 20 Jan7K14K21K28K35KSE +/- 1131.34, N = 12SE +/- 1238.01, N = 12SE +/- 37.82, N = 32935631899345261. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

98 Results Shown

TensorFlow
OpenVINO:
  Machine Translation EN To DE FP16 - CPU:
    FPS
    ms
Stress-NG
Timed LLVM Compilation
OpenVINO:
  Road Segmentation ADAS FP16-INT8 - CPU:
    FPS
    ms
Stress-NG:
  SENDFILE
  MEMFD
Timed Linux Kernel Compilation
Stress-NG
OpenVINO:
  Person Vehicle Bike Detection FP16 - CPU:
    ms
    FPS
Stress-NG:
  Cloning
  System V Message Passing
PyTorch
Speedb
PyTorch:
  CPU - 16 - ResNet-50
  CPU - 512 - ResNet-50
Speedb
Stress-NG:
  MMAP
  Pipe
  Glibc Qsort Data Sorting
  Poll
  CPU Stress
  Context Switching
OpenVINO:
  Handwritten English Recognition FP16-INT8 - CPU:
    ms
    FPS
PostgreSQL:
  100 - 1000 - Read Only
  100 - 1000 - Read Only - Average Latency
Stress-NG:
  NUMA
  Mixed Scheduler
  Function Call
Blender
SVT-AV1
Timed Linux Kernel Compilation
Speedb
OpenVINO
Speedb
Stress-NG:
  Floating Point
  Crypto
  Hash
  Vector Math
  Vector Shuffle
  Wide Vector Math
  Memory Copying
7-Zip Compression
Stress-NG:
  Vector Floating Point
  Matrix Math
Quicksilver
Stress-NG
Timed Node.js Compilation
OpenVINO
Stress-NG
OpenVINO:
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU
  Face Detection Retail FP16-INT8 - CPU
  Vehicle Detection FP16-INT8 - CPU
  Weld Porosity Detection FP16-INT8 - CPU
  Weld Porosity Detection FP16-INT8 - CPU
  Vehicle Detection FP16-INT8 - CPU
  Face Detection FP16-INT8 - CPU
  Face Detection FP16-INT8 - CPU
Quicksilver
Stress-NG
Blender
SVT-AV1
TensorFlow
Stress-NG
7-Zip Compression
LeelaChessZero
SVT-AV1
QuantLib
Stress-NG
Timed PHP Compilation
QuantLib
SVT-AV1
rav1e
FLAC Audio Encoding
Stress-NG
rav1e
Timed LLVM Compilation
rav1e:
  5
  10
Y-Cruncher:
  500M
  1B
Quicksilver
Stress-NG:
  Atomic
  x86_64 RdRand
OpenVINO:
  Person Detection FP16 - CPU:
    ms
    FPS
Stress-NG:
  Matrix 3D Math
  Semaphores
  CPU Cache
  IO_uring
  AVL Tree
  Futex
PostgreSQL:
  100 - 1000 - Read Write - Average Latency
  100 - 1000 - Read Write

v6.6.6 Stock

Processor: AMD Ryzen Threadripper 7980X 64-Cores @ 7.79GHz (64 Cores / 128 Threads), Motherboard: System76 Thelio Major (FA Z5 BIOS), Chipset: AMD Device 14a4, Memory: 4 x 32 GB DRAM-4800MT/s Micron MTC20F1045S1RC48BA2, Disk: 1000GB CT1000T700SSD5, Graphics: NAVI31 45GB (1760/1124MHz), Audio: AMD Device 14cc, Monitor: ASUS VP28U, Network: Aquantia Device 14c0 + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Pop 22.04, Kernel: 6.6.6-76060606-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, OpenGL: 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.54), Vulkan: 1.3.267, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105
Python Notes: Python 3.10.12
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + 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 / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 18 January 2024 11:40 by user system76.

v6.7

Processor: AMD Ryzen Threadripper 7980X 64-Cores @ 7.79GHz (64 Cores / 128 Threads), Motherboard: System76 Thelio Major (FA Z5 BIOS), Chipset: AMD Device 14a4, Memory: 4 x 32 GB DRAM-4800MT/s Micron MTC20F1045S1RC48BA2, Disk: 1000GB CT1000T700SSD5, Graphics: NAVI31 45GB (1760/1124MHz), Audio: AMD Device 14cc, Monitor: ASUS VP28U, Network: Aquantia Device 14c0 + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Pop 22.04, Kernel: 6.7.0-060700-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, OpenGL: 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.56), Vulkan: 1.3.267, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105
Python Notes: Python 3.10.12
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + 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 / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 19 January 2024 04:47 by user system76.

v6.8 20 Jan

Processor: AMD Ryzen Threadripper 7980X 64-Cores @ 7.79GHz (64 Cores / 128 Threads), Motherboard: System76 Thelio Major (FA Z5 BIOS), Chipset: AMD Device 14a4, Memory: 4 x 32 GB DRAM-4800MT/s Micron MTC20F1045S1RC48BA2, Disk: 1000GB CT1000T700SSD5, Graphics: NAVI31 45GB (1760/1124MHz), Audio: AMD Device 14cc, Monitor: ASUS VP28U, Network: Aquantia Device 14c0 + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Pop 22.04, Kernel: 6.7.0-060700daily20240120-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, OpenGL: 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.57), Vulkan: 1.3.267, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105
Python Notes: Python 3.10.12
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + 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 / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 20 January 2024 04:31 by user pts.