Ubuntu 20.10 Devel

Intel Core i9-10980XE testing with a ASRock X299 Steel Legend (P1.30 BIOS) and NVIDIA NV132 11GB on openSUSE Tumbleweed 20200929 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 2010023-FI-2010015FI16
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
Bioinformatics 2 Tests
Chess Test Suite 2 Tests
Timed Code Compilation 2 Tests
C/C++ Compiler Tests 13 Tests
CPU Massive 25 Tests
Creator Workloads 18 Tests
Cryptography 2 Tests
Database Test Suite 3 Tests
Encoding 3 Tests
Fortran Tests 4 Tests
Game Development 5 Tests
HPC - High Performance Computing 20 Tests
Imaging 7 Tests
Java 2 Tests
Common Kernel Benchmarks 4 Tests
Machine Learning 9 Tests
Molecular Dynamics 5 Tests
MPI Benchmarks 3 Tests
Multi-Core 17 Tests
NVIDIA GPU Compute 5 Tests
Intel oneAPI 3 Tests
OpenMPI Tests 4 Tests
Productivity 2 Tests
Programmer / Developer System Benchmarks 8 Tests
Python 2 Tests
Renderers 2 Tests
Scientific Computing 10 Tests
Server 6 Tests
Server CPU Tests 18 Tests
Single-Threaded 7 Tests
Texture Compression 2 Tests
Video Encoding 3 Tests
Common Workstation Benchmarks 4 Tests

Statistics

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

Graph Settings

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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Ubuntu 20.10 20200929
September 29 2020
  11 Hours, 55 Minutes
Fedora Workstation 33 Beta
September 30 2020
  9 Hours, 25 Minutes
Clear Linux 33780
October 01 2020
  11 Hours, 34 Minutes
openSUSE Tumbleweed
October 01 2020
  9 Hours, 3 Minutes
Invert Hiding All Results Option
  10 Hours, 29 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):

Ubuntu 20.10 20200929

Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.30 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB, Graphics: NVIDIA NV132 11GB, Audio: Realtek ALC1220, Monitor: ASUS MG28U, Network: Intel I219-V + Intel I211

OS: Ubuntu 20.10, Kernel: 5.8.0-19-generic (x86_64), Desktop: GNOME Shell 3.38.0, Display Server: X Server 1.20.8, Display Driver: nouveau 1.0.16, OpenGL: 4.3 Mesa 20.1.8, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160

Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-rn7uax/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-rn7uax/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: intel_cpufreq ondemand - CPU Microcode: 0x5002f01
Java Notes: OpenJDK Runtime Environment (build 11.0.8+10-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.8.6
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Fedora Workstation 33 Beta

OS: Fedora 33, Kernel: 5.8.12-300.fc33.x86_64 (x86_64), Desktop: GNOME Shell 3.38.0, Display Server: X Server + Wayland, OpenGL: 4.3 Mesa 20.2.0, Compiler: GCC 10.2.1 20200826 + Clang 11.0.0, File-System: btrfs, Screen Resolution: 3840x2160

Compiler Notes: --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-gcc-major-version-only --with-isl --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver
Processor Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x5002f01
Java Notes: OpenJDK Runtime Environment 18.9 (build 11.0.9-ea+6)
Python Notes: Python 3.9.0rc2
Security Notes: SELinux + itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Clear Linux 33780

Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.30 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB + 31GB UDisk, Graphics: NVIDIA NV132 11GB, Audio: Realtek ALC1220, Monitor: ASUS MG28U, Network: Intel I219-V + Intel I211

OS: Clear Linux OS 33780, Kernel: 5.7.13-975.native (x86_64), Desktop: GNOME Shell 3.38.0, Display Server: X Server 1.20.8, Display Driver: nouveau 1.0.16, OpenGL: 4.3 Mesa 20.2.0-devel, Compiler: GCC 10.2.1 20200928 releases/gcc-10.2.0-302-ga6c47f4ce2 + Clang 10.0.1 + LLVM 10.0.1, File-System: ext4, Screen Resolution: 3840x2160

Environment Notes: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags -Wa,-mbranches-within-32B-boundaries" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries -fvisibility-inlines-hidden -Wl,--enable-new-dtags" MESA_GLSL_CACHE_DISABLE=0 FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,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 -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries" 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-languages=c,c++,fortran,go --enable-ld=default --enable-libstdcxx-pch --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=westmere --with-gcc-major-version-only --with-glibc-version=2.19 --with-gnu-ld --with-isl --with-ppl=yes --with-tune=haswell
Processor Notes: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500012c
Java Notes: OpenJDK Runtime Environment (build 1.8.0-u252-ga-b00)
Python Notes: Python 3.8.6
Security Notes: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

openSUSE Tumbleweed

OS: openSUSE Tumbleweed 20200929, Kernel: 5.8.10-1-default (x86_64), Desktop: KDE Plasma 5.19.5, Display Server: X Server 1.20.9, Display Driver: nouveau 1.0.16, OpenGL: 4.3 Mesa 20.1.8, Compiler: GCC 10.2.1 20200825 [revision c0746a1beb1ba073c7981eb09f55b3d993b32e5c], File-System: btrfs, Screen Resolution: 3840x2160

Compiler Notes: --build=x86_64-suse-linux --disable-libcc1 --disable-libssp --disable-libstdcxx-pch --disable-libvtv --disable-werror --enable-cet=auto --enable-checking=release --enable-gnu-indirect-function --enable-languages=c,c++,objc,fortran,obj-c++,ada,go,d --enable-libphobos --enable-libstdcxx-allocator=new --enable-link-mutex --enable-linux-futex --enable-multilib --enable-offload-targets=nvptx-none=/usr/nvptx-none,amdgcn-amdhsa=/usr/amdgcn-amdhsa, --enable-plugin --enable-ssp --enable-version-specific-runtime-libs --host=x86_64-suse-linux --mandir=/usr/share/man --with-arch-32=x86-64 --with-build-config=bootstrap-lto-lean --with-gcc-major-version-only --with-slibdir=/lib64 --with-tune=generic --without-cuda-driver --without-system-libunwind
Processor Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x5002f01
Java Notes: OpenJDK Runtime Environment (build 11.0.8+10-suse-1.3-x8664)
Python Notes: Python 2.7.18 + Python 3.8.5
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Ubuntu 20.10 DevelProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE TumbleweedIntel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads)ASRock X299 Steel Legend (P1.30 BIOS)Intel Sky Lake-E DMI3 Registers32GBSamsung SSD 970 PRO 512GBNVIDIA NV132 11GBRealtek ALC1220ASUS MG28UIntel I219-V + Intel I211Ubuntu 20.105.8.0-19-generic (x86_64)GNOME Shell 3.38.0X Server 1.20.8nouveau 1.0.164.3 Mesa 20.1.8GCC 10.2.0ext43840x2160Fedora 335.8.12-300.fc33.x86_64 (x86_64)X Server + Wayland4.3 Mesa 20.2.0GCC 10.2.1 20200826 + Clang 11.0.0btrfsSamsung SSD 970 PRO 512GB + 31GB UDiskClear Linux OS 337805.7.13-975.native (x86_64)X Server 1.20.8nouveau 1.0.164.3 Mesa 20.2.0-develGCC 10.2.1 20200928 releases/gcc-10.2.0-302-ga6c47f4ce2 + Clang 10.0.1 + LLVM 10.0.1ext4openSUSE Tumbleweed 202009295.8.10-1-default (x86_64)KDE Plasma 5.19.5X Server 1.20.94.3 Mesa 20.1.8GCC 10.2.1 20200825 [revision c0746a1beb1ba073c7981eb09f55b3d993b32e5c]btrfsOpenBenchmarking.orgCompiler Details- Ubuntu 20.10 20200929: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-rn7uax/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-rn7uax/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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 - Fedora Workstation 33 Beta: --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-gcc-major-version-only --with-isl --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver - Clear Linux 33780: --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-languages=c,c++,fortran,go --enable-ld=default --enable-libstdcxx-pch --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=westmere --with-gcc-major-version-only --with-glibc-version=2.19 --with-gnu-ld --with-isl --with-ppl=yes --with-tune=haswell - openSUSE Tumbleweed: --build=x86_64-suse-linux --disable-libcc1 --disable-libssp --disable-libstdcxx-pch --disable-libvtv --disable-werror --enable-cet=auto --enable-checking=release --enable-gnu-indirect-function --enable-languages=c,c++,objc,fortran,obj-c++,ada,go,d --enable-libphobos --enable-libstdcxx-allocator=new --enable-link-mutex --enable-linux-futex --enable-multilib --enable-offload-targets=nvptx-none=/usr/nvptx-none,amdgcn-amdhsa=/usr/amdgcn-amdhsa, --enable-plugin --enable-ssp --enable-version-specific-runtime-libs --host=x86_64-suse-linux --mandir=/usr/share/man --with-arch-32=x86-64 --with-build-config=bootstrap-lto-lean --with-gcc-major-version-only --with-slibdir=/lib64 --with-tune=generic --without-cuda-driver --without-system-libunwind Processor Details- Ubuntu 20.10 20200929: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x5002f01- Fedora Workstation 33 Beta: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x5002f01- Clear Linux 33780: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500012c- openSUSE Tumbleweed: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x5002f01Java Details- Ubuntu 20.10 20200929: OpenJDK Runtime Environment (build 11.0.8+10-post-Ubuntu-0ubuntu1)- Fedora Workstation 33 Beta: OpenJDK Runtime Environment 18.9 (build 11.0.9-ea+6)- Clear Linux 33780: OpenJDK Runtime Environment (build 1.8.0-u252-ga-b00)- openSUSE Tumbleweed: OpenJDK Runtime Environment (build 11.0.8+10-suse-1.3-x8664)Python Details- Ubuntu 20.10 20200929: Python 3.8.6- Fedora Workstation 33 Beta: Python 3.9.0rc2- Clear Linux 33780: Python 3.8.6- openSUSE Tumbleweed: Python 2.7.18 + Python 3.8.5Security Details- Ubuntu 20.10 20200929: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled - Fedora Workstation 33 Beta: SELinux + itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled - Clear Linux 33780: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled - openSUSE Tumbleweed: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled Environment Details- Clear Linux 33780: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags -Wa,-mbranches-within-32B-boundaries" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries -fvisibility-inlines-hidden -Wl,--enable-new-dtags" MESA_GLSL_CACHE_DISABLE=0 FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,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 -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx""

Ubuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE TumbleweedResult OverviewPhoronix Test Suite100%111%122%134%145%PHPBenchPyPerformanceTimed HMMer Searchlibavif avifencPyBenchDaCapo BenchmarkTimed PHP CompilationMobile Neural NetworkCP2K Molecular DynamicsSVT-AV1SQLite SpeedtestBasis UniversalFFTETimed Linux Kernel CompilationIntel Open Image DenoiseGEGLCoremarkoneDNNLevelDBWebP Image EncodeRawTherapeeGitDeepSpeechDarktableHimeno BenchmarkCraftyasmFishBRL-CADSVT-VP9TNNOpenVKLNCNNASTC EncoderAppleseedBlenderZstd CompressionDolfynHierarchical INTegrationKripkeOpenSSLTensorFlow LiteNAMDlibjpeg-turbo tjbenchCrypto++

Ubuntu 20.10 Develdacapobench: Tradebeanspyperformance: python_startuponednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUgimp: resizeamg: caffe: GoogleNet - CPU - 200node-express-loadtest: caffe: AlexNet - CPU - 200pyperformance: nbodyavifenc: 10avifenc: 8onednn: Recurrent Neural Network Inference - f32 - CPUgimp: rotatephpbench: PHP Benchmark Suitepyperformance: raytraceplaidml: No - Inference - DenseNet 201 - CPUgimp: auto-levelsdacapobench: H2hmmer: Pfam Database Searchpyperformance: chaosgimp: unsharp-maskjava-gradle-perf: Reactorlammps: 20k Atomspyperformance: django_templatepyperformance: crypto_pyaespyperformance: pickle_pure_pythonpybench: Total For Average Test Timespyperformance: floatbasis: ETC1Sgegl: Wavelet Blurbasis: UASTC Level 2 + RDO Post-Processingdacapobench: Tradesoappyperformance: gopyperformance: regex_compilebuild-php: Time To Compiledacapobench: Jythononednn: Deconvolution Batch deconv_3d - u8s8f32 - CPUmnn: inception-v3pyperformance: 2to3onednn: Recurrent Neural Network Training - f32 - CPUcp2k: Fayalite-FIST Datasvt-av1: Enc Mode 8 - 1080ppyperformance: pathlibbasis: UASTC Level 0svt-av1: Enc Mode 4 - 1080ppyperformance: json_loadssqlite-speedtest: Timed Time - Size 1,000webp: Quality 100, Lossless, Highest Compressionmnn: SqueezeNetV1.0onednn: IP Batch 1D - u8s8f32 - CPUrocksdb: Seq Fillffte: N=256, 3D Complex FFT Routinebuild-linux-kernel: Time To Compilerocksdb: Rand Filloidn: Memorialleveldb: Hot Readcoremark: CoreMark Size 666 - Iterations Per Secondwebp: Quality 100, Losslessplaidml: No - Inference - ResNet 50 - CPUdarktable: Masskrug - CPU-onlyleveldb: Rand Readplaidml: No - Inference - VGG16 - CPUgegl: Rotate 90 Degreesonednn: Deconvolution Batch deconv_3d - f32 - CPUgegl: Color Enhancebasis: UASTC Level 2plaidml: No - Inference - VGG19 - CPUdarktable: Server Room - CPU-onlyncnn: CPU - yolov4-tinyavifenc: 2ncnn: CPU - alexnetrocksdb: Read While Writingbasis: UASTC Level 3avifenc: 0onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: IP Batch All - u8s8f32 - CPUleveldb: Seek Randgegl: Antialiasrawtherapee: Total Benchmark Timegegl: Tile Glassncnn: CPU - shufflenet-v2ncnn: CPU - squeezenetncnn: CPU - blazefacegit: Time To Complete Common Git Commandstnn: CPU - MobileNet v2ncnn: CPU - mobilenetdeepspeech: CPUsvt-vp9: Visual Quality Optimized - Bosphorus 1080phimeno: Poisson Pressure Solverhpcg: crafty: Elapsed Timeplaidml: No - Inference - Inception V3 - CPUasmfish: 1024 Hash Memory, 26 Depthwebp: Quality 100, Highest Compressionbrl-cad: VGR Performance Metricdarktable: Server Rack - CPU-onlyncnn: CPU - resnet50ai-benchmark: Device Inference Scoreplaidml: No - Inference - Mobilenet - CPUonednn: IP Batch 1D - f32 - CPUastcenc: Mediumonednn: Deconvolution Batch deconv_1d - f32 - CPUgromacs: Water Benchmarkonednn: IP Batch All - f32 - CPUonednn: Deconvolution Batch deconv_1d - bf16bf16bf16 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUappleseed: Emilyncnn: CPU - googlenetastcenc: Fastopenvkl: vklBenchmarksvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080ponednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPUncnn: CPU - vgg16compress-zstd: 19ncnn: CPU - mnasnetblender: Pabellon Barcelona - CPU-Onlyncnn: CPU - efficientnet-b0blender: Barbershop - CPU-Onlyplaidml: No - Inference - IMDB LSTM - CPUai-benchmark: Device AI Scoredarktable: Boat - CPU-onlyblender: Fishy Cat - CPU-Onlyastcenc: Thoroughappleseed: Disney Materialtnn: CPU - SqueezeNet v1.1rocksdb: Rand Readonednn: IP Batch 1D - bf16bf16bf16 - CPUonednn: Deconvolution Batch deconv_1d - u8s8f32 - CPUcompress-zstd: 3blender: BMW27 - CPU-Onlytensorflow-lite: NASNet Mobileappleseed: Material Testeronednn: Deconvolution Batch deconv_3d - bf16bf16bf16 - CPUblender: Classroom - CPU-Onlyastcenc: Exhaustivedolfyn: Computational Fluid Dynamicstensorflow-lite: Mobilenet Floatonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUtensorflow-lite: SqueezeNettensorflow-lite: Mobilenet Quanthint: FLOATkripke: openssl: RSA 4096-bit Performancenamd: ATPase Simulation - 327,506 Atomstjbench: Decompression Throughputtensorflow-lite: Inception ResNet V2onednn: IP Batch All - bf16bf16bf16 - CPUtensorflow-lite: Inception V4cryptopp: Keyed Algorithmsai-benchmark: Device Training Scorencnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU-v2-v2 - mobilenet-v2mnn: mobilenet-v1-1.0ncnn: CPU - resnet18mnn: MobileNetV2_224mnn: resnet-v2-50lammps: Rhodopsin ProteinUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed747512.10.73323231092.282305426885868071075.0765.26173.58497830843892.654595145.54588.7253.5439.50238.588.634991493.745.96755.016652.153346920513549.15438570.73948636.730278190.422751.9842.56614.67.0625.20522.055.49937.6558.6540.597088156786352659.20669741149.459143379922.1927.366687280.77143416.9615.563.27227.55126.4637.7672.8924652.04820.86522.312.67924.8642.99110.93421910536.43672.0561.589327.6822934.14735.32449.75928.7185.0714.172.1248.632297.19717.16115.53150254.113994.3735247.3853890405566.62532347676.6332075150.12821.79193013.222.320487.011.782641.57532.70019.148029.44985227.09758414.065.09268.67312.2510.009591.7675641.2660.35.25311.577.15389.49787.5134788.353127.1117.93121.510816281.025992441135.581130.4632724886.693.52118100126.62310610.9180283.47141.1218.47687493.18.0185712795690286.2455254464.97370347071535191.90.97719216.386139164234063.86641857207709.204184154812.015.14945.2428.19378549.770.3694218.01331431.111913066228876021265.9056.08353.735512.09877569545813.3674276134.16110517.079275.40811.73944.9104398105610654.64155.076801.561359822915345.97540190.68462334.857292168.749711.85547.04315.17.6645.66422.560.35138.1588.5530.533189180298448693.66660985246.718163657925.1330.617643394.64375017.1973.27730.97137.5952.6410850.44422.5722.68925.7142.07110.76397017939.02770.8901.446577.1772535.86434.93250.66927.8285.1114.052.1447.236286.64117.47112.12810251.984005.4285187.814049501807529413346.6072058460.13021.532.223596.951.742211.52232.00798.958049.31276228.99093914.205.06274.81308.519.808001.7515940.1560.25.34313.837.11392.308.463128.0617.76122.403929279.3591016523155.595280.4662564863.493.49117441126.4651410.9146284.29141.3818.38888258.68.0291112830290878.0454854722.62813344756705175.60.98268216.437167165001063.86441863017711.24424511.325.19541.5339.32028197.180.3810546.46056200.2514036698595614594.03.8363.97348.14949.7869745083803.7311.3933392108.92787.614.198208.72712.49038.388.434685490.243.43744.878665.179289219813439.73432540.60710529.991255156.337725.88350.87113.76.4696.11221.952.67332.7487.4290.516754166436148688.74947123043.237150911525.1628.009728013.37859415.2066.282.90527.62929.7935.5042.6049846.50520.22724.852.76323.1838.9049.92383151736.76465.5841.502107.0597433.98433.86051.01827.3445.1413.162.1845.239276.89616.43108.69367268.784250.7133377.8461893591496.73561651136.2662113560.12521.13203213.902.212866.861.705511.59531.22299.220909.17499219.45105913.615.05280.11321.459.629941.7137240.8862.25.32303.527.19380.51770.7335858.220124.7817.65119.318381275.7231015359615.492180.4605954909.491.73116724124.75868310.8534280.87140.7618.28087810.97.9906812909990284.4455035663.30417344797535212.00.97837217.101844164635363.96961861030712.202296155311.324.41634.27411.87614.90.37027111.7061454.0584.15848.207714.8206733775363.5515.9694452141.86311718.73949.9114446108111445.32856.268642.50524616645.47338450.60308832.667311159.349853.78847.02116.36.6645.82225.761.81037.8248.0440.51324945877.80041825245.25125.0227.030666366.94135217.0286.083.25727.46529.7639.9542.5803051.91820.72024.372.97824.0040.88710.0635.47769.5201.460127.0049332.79536.90154.07529.6685.5014.202.2846.567288.60916.29116.51155255.984023.59530095983276.34544610516.5601998740.13222.2913.352.210726.681.6989631.37618.827939.05209222.76797313.964.88270.17309.819.808821.7015741.6660.05.44303.077.35379.86795.388.351124.6217.45120.837293282.4585.476010.4563414959.592.23116137124.74970510.7797280.85139.8518.29487430.37.9541112796790079.2458498603.13456344342805175.90.97596217.649185164141063.64251854170710.6218805.356.037.04311.884.99340.258OpenBenchmarking.org

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: TradebeansClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta2K4K6K8K10KSE +/- 14.92, N = 4SE +/- 100.02, N = 4SE +/- 97.67, N = 4281974757854
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta14002800420056007000Min: 2791 / Avg: 2819 / Max: 2857Min: 7254 / Avg: 7475 / Max: 7719Min: 7674 / Avg: 7854.25 / Max: 8055

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: python_startupClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed48121620SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.23, N = 127.189.7712.1014.90
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: python_startupClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed48121620Min: 7.17 / Avg: 7.18 / Max: 7.2Min: 9.64 / Avg: 9.77 / Max: 9.84Min: 12 / Avg: 12.13 / Max: 12.2Min: 12.9 / Avg: 14.87 / Max: 15.7

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUFedora Workstation 33 BetaopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 202009290.1650.330.4950.660.825SE +/- 0.001747, N = 3SE +/- 0.003270, N = 3SE +/- 0.004133, N = 3SE +/- 0.009743, N = 120.3694210.3702710.3810540.733232-O2 - MIN: 0.34MIN: 0.35MIN: 0.36MIN: 0.611. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUFedora Workstation 33 BetaopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929246810Min: 0.37 / Avg: 0.37 / Max: 0.37Min: 0.36 / Avg: 0.37 / Max: 0.38Min: 0.38 / Avg: 0.38 / Max: 0.39Min: 0.64 / Avg: 0.73 / Max: 0.771. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

GIMP

GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.20Test: resizeClear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed3691215SE +/- 0.042, N = 3SE +/- 0.047, N = 3SE +/- 0.042, N = 36.4608.01311.706
OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.20Test: resizeClear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed3691215Min: 6.41 / Avg: 6.46 / Max: 6.55Min: 7.93 / Avg: 8.01 / Max: 8.09Min: 11.64 / Avg: 11.71 / Max: 11.78

Algebraic Multi-Grid Benchmark

AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid BenchmarkClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 2020092912K24K36K48K60KSE +/- 14.87, N = 3SE +/- 17.97, N = 3SE +/- 4.04, N = 356200.2531431.1131092.28-O31. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -pthread -lmpi
OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid BenchmarkClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 2020092910K20K30K40K50KMin: 56184.71 / Avg: 56200.25 / Max: 56229.98Min: 31396.47 / Avg: 31431.11 / Max: 31456.71Min: 31084.53 / Avg: 31092.28 / Max: 31098.111. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -pthread -lmpi

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: GoogleNet - Acceleration: CPU - Iterations: 200Clear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 2020092950K100K150K200K250KSE +/- 137.72, N = 3SE +/- 509.79, N = 3SE +/- 106.25, N = 3140366191306230542-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -lboost_system -lboost_thread -lboost_filesystem -lboost_chrono -lboost_date_time -lboost_atomic -lhdf5_cpp -lhdf5 -lz -ldl -lm -lhdf5_hl_cpp -lhdf5_hl-O2 -lboost_system -lboost_thread -lboost_filesystem -lboost_chrono -lboost_date_time -lboost_atomic -lhdf5_cpp -lhdf5 -lhdf5_hl_cpp -lhdf5_hl-O3 -lsz -lz -ldl -lm1. (CXX) g++ options: -fPIC -rdynamic -lglog -lgflags -lprotobuf -lpthread -llmdb -lopenblas
OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: GoogleNet - Acceleration: CPU - Iterations: 200Clear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 2020092940K80K120K160K200KMin: 140130 / Avg: 140366.33 / Max: 140607Min: 190578 / Avg: 191305.67 / Max: 192288Min: 230360 / Avg: 230542.33 / Max: 2307281. (CXX) g++ options: -fPIC -rdynamic -lglog -lgflags -lprotobuf -lpthread -llmdb -lopenblas

Node.js Express HTTP Load Test

A Node.js Express server with a Node-based loadtest client for facilitating HTTP benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterNode.js Express HTTP Load TestClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta2K4K6K8K10KSE +/- 12.29, N = 3SE +/- 11.79, N = 3SE +/- 4.91, N = 39859688562281. Clear Linux 33780: Nodejs2. Ubuntu 20.10 20200929: Nodejs v12.18.23. Fedora Workstation 33 Beta: Nodejs
OpenBenchmarking.orgRequests Per Second, More Is BetterNode.js Express HTTP Load TestClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta2K4K6K8K10KMin: 9836 / Avg: 9859 / Max: 9878Min: 6869 / Avg: 6885 / Max: 6908Min: 6222 / Avg: 6228.33 / Max: 62381. Clear Linux 33780: Nodejs2. Ubuntu 20.10 20200929: Nodejs v12.18.23. Fedora Workstation 33 Beta: Nodejs

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: AlexNet - Acceleration: CPU - Iterations: 200Clear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta20K40K60K80K100KSE +/- 23.71, N = 3SE +/- 47.95, N = 3SE +/- 463.52, N = 3561458680787602-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -lboost_system -lboost_thread -lboost_filesystem -lboost_chrono -lboost_date_time -lboost_atomic -lhdf5_cpp -lhdf5 -lz -ldl -lm -lhdf5_hl_cpp -lhdf5_hl-O3 -lsz -lz -ldl -lm-O2 -lboost_system -lboost_thread -lboost_filesystem -lboost_chrono -lboost_date_time -lboost_atomic -lhdf5_cpp -lhdf5 -lhdf5_hl_cpp -lhdf5_hl1. (CXX) g++ options: -fPIC -rdynamic -lglog -lgflags -lprotobuf -lpthread -llmdb -lopenblas
OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: AlexNet - Acceleration: CPU - Iterations: 200Clear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta15K30K45K60K75KMin: 56103 / Avg: 56145.33 / Max: 56185Min: 86712 / Avg: 86807.33 / Max: 86864Min: 86852 / Avg: 87602.33 / Max: 884491. (CXX) g++ options: -fPIC -rdynamic -lglog -lgflags -lprotobuf -lpthread -llmdb -lopenblas

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 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed306090120150SE +/- 0.06, N = 3SE +/- 0.67, N = 394.0107.0126.0145.0
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: nbodyClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed306090120150Min: 93.9 / Avg: 94 / Max: 94.1Min: 144 / Avg: 144.67 / Max: 146

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 0.7.3Encoder Speed: 10Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta1.32862.65723.98585.31446.643SE +/- 0.012, N = 3SE +/- 0.007, N = 3SE +/- 0.024, N = 3SE +/- 0.011, N = 33.8364.0585.0765.9051. (CXX) g++ options: -O3 -fPIC
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 10Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta246810Min: 3.81 / Avg: 3.84 / Max: 3.85Min: 4.05 / Avg: 4.06 / Max: 4.07Min: 5.04 / Avg: 5.08 / Max: 5.12Min: 5.88 / Avg: 5.91 / Max: 5.921. (CXX) g++ options: -O3 -fPIC

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 8Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta246810SE +/- 0.026, N = 3SE +/- 0.022, N = 3SE +/- 0.014, N = 3SE +/- 0.017, N = 33.9734.1585.2616.0831. (CXX) g++ options: -O3 -fPIC
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 8Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta246810Min: 3.93 / Avg: 3.97 / Max: 4.02Min: 4.13 / Avg: 4.16 / Max: 4.2Min: 5.24 / Avg: 5.26 / Max: 5.29Min: 6.05 / Avg: 6.08 / Max: 6.111. (CXX) g++ options: -O3 -fPIC

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291632486480SE +/- 0.04, N = 3SE +/- 0.55, N = 3SE +/- 0.37, N = 3SE +/- 0.57, N = 348.1548.2153.7473.58MIN: 46.52MIN: 45.8-O2 - MIN: 51.31MIN: 71.211. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291428425670Min: 48.07 / Avg: 48.15 / Max: 48.2Min: 47.2 / Avg: 48.21 / Max: 49.09Min: 53.09 / Avg: 53.74 / Max: 54.37Min: 72.73 / Avg: 73.58 / Max: 74.671. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

GIMP

GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.20Test: rotateClear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed48121620SE +/- 0.003, N = 3SE +/- 0.029, N = 3SE +/- 0.020, N = 39.78612.09814.820
OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.20Test: rotateClear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed48121620Min: 9.78 / Avg: 9.79 / Max: 9.79Min: 12.05 / Avg: 12.1 / Max: 12.15Min: 14.79 / Avg: 14.82 / Max: 14.86

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. The number of iterations used is 1,000,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed200K400K600K800K1000KSE +/- 8801.15, N = 3SE +/- 1520.41, N = 3SE +/- 582.41, N = 3SE +/- 4500.90, N = 3974508783084775695673377
OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed200K400K600K800K1000KMin: 956914 / Avg: 974508.33 / Max: 983764Min: 781408 / Avg: 783083.67 / Max: 786119Min: 774995 / Avg: 775694.67 / Max: 776851Min: 664384 / Avg: 673376.67 / Max: 678224

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: raytraceClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed120240360480600SE +/- 0.33, N = 3SE +/- 0.33, N = 3380389458536
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: raytraceClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed90180270360450Min: 389 / Avg: 389.33 / Max: 390Min: 536 / Avg: 536.33 / Max: 537

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 202009290.83931.67862.51793.35724.1965SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 33.733.552.65
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929246810Min: 3.7 / Avg: 3.73 / Max: 3.76Min: 3.54 / Avg: 3.55 / Max: 3.56Min: 2.64 / Avg: 2.65 / Max: 2.65

GIMP

GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.20Test: auto-levelsClear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed48121620SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 311.3913.3715.97
OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.20Test: auto-levelsClear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed48121620Min: 11.33 / Avg: 11.39 / Max: 11.43Min: 13.31 / Avg: 13.37 / Max: 13.43Min: 15.9 / Avg: 15.97 / Max: 16.02

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 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 2020092910002000300040005000SE +/- 13.92, N = 4SE +/- 43.70, N = 4SE +/- 22.54, N = 4SE +/- 54.09, N = 53392427644524595
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2Clear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 202009298001600240032004000Min: 3368 / Avg: 3391.75 / Max: 3424Min: 4218 / Avg: 4276 / Max: 4404Min: 4425 / Avg: 4451.5 / Max: 4519Min: 4429 / Avg: 4595.2 / Max: 4763

Timed HMMer Search

This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 20200929306090120150SE +/- 0.02, N = 3SE +/- 0.19, N = 3SE +/- 0.28, N = 3SE +/- 0.18, N = 3108.93134.16141.86145.55-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 20200929306090120150Min: 108.9 / Avg: 108.93 / Max: 108.96Min: 133.87 / Avg: 134.16 / Max: 134.51Min: 141.42 / Avg: 141.86 / Max: 142.38Min: 145.21 / Avg: 145.54 / Max: 145.841. (CC) gcc options: -O3 -pthread -lhmmer -leasel -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: chaosClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed306090120150SE +/- 0.06, N = 3SE +/- 0.06, N = 387.688.7105.0117.0
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: chaosClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed20406080100Min: 87.5 / Avg: 87.6 / Max: 87.7Min: 88.6 / Avg: 88.7 / Max: 88.8

GIMP

GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.20Test: unsharp-maskClear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed510152025SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.11, N = 314.2017.0818.74
OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.20Test: unsharp-maskClear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed510152025Min: 14.15 / Avg: 14.2 / Max: 14.22Min: 17.07 / Avg: 17.08 / Max: 17.1Min: 18.53 / Avg: 18.74 / Max: 18.91

Java Gradle Build

This test runs Java software project builds using the Gradle build system. It is intended to give developers an idea as to the build performance for development activities and build servers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterJava Gradle BuildGradle Build: ReactorClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta60120180240300SE +/- 3.33, N = 3SE +/- 3.64, N = 3SE +/- 2.98, N = 7208.73253.54275.41
OpenBenchmarking.orgSeconds, Fewer Is BetterJava Gradle BuildGradle Build: ReactorClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta50100150200250Min: 204.96 / Avg: 208.73 / Max: 215.36Min: 247.28 / Avg: 253.54 / Max: 259.89Min: 265.82 / Avg: 275.41 / Max: 290.31

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 24Aug2020Model: 20k AtomsClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215SE +/- 0.011, N = 3SE +/- 0.055, N = 3SE +/- 0.045, N = 312.49011.7399.502-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O31. (CXX) g++ options: -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 24Aug2020Model: 20k AtomsClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 2020092948121620Min: 12.48 / Avg: 12.49 / Max: 12.51Min: 11.65 / Avg: 11.74 / Max: 11.84Min: 9.42 / Avg: 9.5 / Max: 9.571. (CXX) g++ options: -pthread -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: django_templateClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed1122334455SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 338.338.544.949.9
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: django_templateClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed1020304050Min: 38.3 / Avg: 38.33 / Max: 38.4Min: 38.5 / Avg: 38.53 / Max: 38.6Min: 44.8 / Avg: 44.9 / Max: 45Min: 49.9 / Avg: 49.93 / Max: 50

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: crypto_pyaesClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed306090120150SE +/- 0.03, N = 3SE +/- 0.06, N = 388.488.6104.0114.0
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: crypto_pyaesClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed20406080100Min: 88.4 / Avg: 88.43 / Max: 88.5Min: 88.5 / Avg: 88.6 / Max: 88.7

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pickle_pure_pythonClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed100200300400500SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 1.00, N = 3346349398446
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pickle_pure_pythonClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed80160240320400Min: 349 / Avg: 349.33 / Max: 350Min: 398 / Avg: 398.33 / Max: 399Min: 444 / Avg: 446 / Max: 447

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 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed2004006008001000SE +/- 1.53, N = 3SE +/- 1.00, N = 385491410561081
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed2004006008001000Min: 851 / Avg: 854 / Max: 856Min: 913 / Avg: 914 / Max: 916

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 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed306090120150SE +/- 0.03, N = 3SE +/- 0.07, N = 390.293.7106.0114.0
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: floatClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed20406080100Min: 90.2 / Avg: 90.23 / Max: 90.3Min: 93.6 / Avg: 93.67 / Max: 93.8

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: ETC1SClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta1224364860SE +/- 0.04, N = 3SE +/- 0.66, N = 3SE +/- 0.42, N = 3SE +/- 0.17, N = 343.4445.3345.9754.64-O3-O3-O3-O21. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: ETC1SClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta1122334455Min: 43.37 / Avg: 43.44 / Max: 43.5Min: 44.06 / Avg: 45.33 / Max: 46.25Min: 45.22 / Avg: 45.97 / Max: 46.67Min: 54.46 / Avg: 54.64 / Max: 54.971. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Wavelet BlurClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed1326395265SE +/- 0.05, N = 3SE +/- 0.20, N = 3SE +/- 0.11, N = 3SE +/- 0.02, N = 344.8855.0255.0856.27
OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Wavelet BlurClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed1122334455Min: 44.78 / Avg: 44.88 / Max: 44.96Min: 54.7 / Avg: 55.02 / Max: 55.37Min: 54.9 / Avg: 55.08 / Max: 55.29Min: 56.24 / Avg: 56.27 / Max: 56.3

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 2 + RDO Post-ProcessingopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta2004006008001000SE +/- 0.28, N = 3SE +/- 0.19, N = 3SE +/- 1.11, N = 3SE +/- 0.29, N = 3642.51652.15665.18801.56-O3-O3-O3-O21. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 2 + RDO Post-ProcessingopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta140280420560700Min: 642.19 / Avg: 642.51 / Max: 643.07Min: 651.78 / Avg: 652.15 / Max: 652.4Min: 662.98 / Avg: 665.18 / Max: 666.54Min: 801 / Avg: 801.56 / Max: 801.921. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread

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: TradesoapClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta8001600240032004000SE +/- 25.26, N = 4SE +/- 30.86, N = 4SE +/- 21.24, N = 4289234693598
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradesoapClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta6001200180024003000Min: 2829 / Avg: 2891.75 / Max: 2941Min: 3400 / Avg: 3468.75 / Max: 3535Min: 3537 / Avg: 3598 / Max: 3628

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: goClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed50100150200250SE +/- 0.33, N = 3198205229246
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: goClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed4080120160200Min: 246 / Avg: 246.33 / Max: 247

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: regex_compileClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed4080120160200134135153166

Timed PHP Compilation

This test times how long it takes to build PHP 7. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 7.4.2Time To CompileClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291122334455SE +/- 0.05, N = 3SE +/- 0.11, N = 3SE +/- 0.09, N = 3SE +/- 0.16, N = 339.7345.4745.9849.15
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 7.4.2Time To CompileClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291020304050Min: 39.63 / Avg: 39.73 / Max: 39.79Min: 45.33 / Avg: 45.47 / Max: 45.68Min: 45.89 / Avg: 45.98 / Max: 46.15Min: 48.9 / Avg: 49.15 / Max: 49.44

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 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta9001800270036004500SE +/- 9.25, N = 4SE +/- 14.70, N = 4SE +/- 24.35, N = 4SE +/- 23.83, N = 43254384538574019
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta7001400210028003500Min: 3239 / Avg: 3254.25 / Max: 3281Min: 3805 / Avg: 3844.5 / Max: 3874Min: 3821 / Avg: 3857.25 / Max: 3928Min: 3970 / Avg: 4018.5 / Max: 4067

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009290.16640.33280.49920.66560.832SE +/- 0.000671, N = 3SE +/- 0.000320, N = 3SE +/- 0.007849, N = 15SE +/- 0.007689, N = 30.6030880.6071050.6846230.739486MIN: 0.59MIN: 0.6-O2 - MIN: 0.62MIN: 0.71. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810Min: 0.6 / Avg: 0.6 / Max: 0.6Min: 0.61 / Avg: 0.61 / Max: 0.61Min: 0.65 / Avg: 0.68 / Max: 0.75Min: 0.73 / Avg: 0.74 / Max: 0.751. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Mobile Neural Network

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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: inception-v3Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929816243240SE +/- 0.21, N = 15SE +/- 0.39, N = 3SE +/- 0.15, N = 3SE +/- 0.17, N = 929.9932.6734.8636.73-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 27.78 / MAX: 31.53MIN: 31.87 / MAX: 33.64MIN: 32.13 / MAX: 36.98MIN: 33.59 / MAX: 45.41. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: inception-v3Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929816243240Min: 28.33 / Avg: 29.99 / Max: 31.15Min: 32.04 / Avg: 32.67 / Max: 33.37Min: 34.61 / Avg: 34.86 / Max: 35.14Min: 35.72 / Avg: 36.73 / Max: 37.611. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

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 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed70140210280350255278292311

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009294080120160200SE +/- 0.49, N = 3SE +/- 1.85, N = 3SE +/- 0.21, N = 3SE +/- 0.83, N = 3156.34159.35168.75190.42MIN: 154.1MIN: 155.5-O2 - MIN: 166.69MIN: 187.051. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929306090120150Min: 155.45 / Avg: 156.34 / Max: 157.14Min: 156.51 / Avg: 159.35 / Max: 162.82Min: 168.35 / Avg: 168.75 / Max: 169.05Min: 189.19 / Avg: 190.42 / Max: 1921. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

CP2K Molecular Dynamics

CP2K is an open-source molecular dynamics software package focused on quantum chemistry and solid-state physics. This test profile currently makes use of the OpenMP implementation and using the Fayalite-FIST molecular dynamics run and measures the total time to complete. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 6.1Fayalite-FIST DataFedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 20200929openSUSE Tumbleweed2004006008001000711.86725.88751.98853.79

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 8 - Input: 1080pClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 202009291122334455SE +/- 0.20, N = 3SE +/- 0.08, N = 3SE +/- 0.16, N = 3SE +/- 0.27, N = 350.8747.0447.0242.571. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 8 - Input: 1080pClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 202009291020304050Min: 50.47 / Avg: 50.87 / Max: 51.11Min: 46.93 / Avg: 47.04 / Max: 47.2Min: 46.77 / Avg: 47.02 / Max: 47.32Min: 42.07 / Avg: 42.57 / Max: 43.011. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

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 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed48121620SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 313.714.615.116.3
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: pathlibClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed48121620Min: 13.6 / Avg: 13.67 / Max: 13.7Min: 14.5 / Avg: 14.57 / Max: 14.6Min: 15.1 / Avg: 15.1 / Max: 15.1Min: 16.3 / Avg: 16.33 / Max: 16.4

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 0Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta246810SE +/- 0.008, N = 3SE +/- 0.036, N = 3SE +/- 0.003, N = 3SE +/- 0.005, N = 36.4696.6647.0627.664-O3-O3-O3-O21. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 0Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta3691215Min: 6.45 / Avg: 6.47 / Max: 6.48Min: 6.62 / Avg: 6.66 / Max: 6.73Min: 7.06 / Avg: 7.06 / Max: 7.07Min: 7.66 / Avg: 7.66 / Max: 7.671. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 4 - Input: 1080pClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929246810SE +/- 0.005, N = 3SE +/- 0.037, N = 3SE +/- 0.051, N = 3SE +/- 0.043, N = 36.1125.8225.6645.2051. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 4 - Input: 1080pClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929246810Min: 6.1 / Avg: 6.11 / Max: 6.12Min: 5.75 / Avg: 5.82 / Max: 5.87Min: 5.6 / Avg: 5.66 / Max: 5.77Min: 5.13 / Avg: 5.2 / Max: 5.281. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

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 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed612182430SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 321.922.022.525.7
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: json_loadsClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed612182430Min: 21.8 / Avg: 21.93 / Max: 22.1Min: 21.9 / Avg: 22.03 / Max: 22.2Min: 22.5 / Avg: 22.5 / Max: 22.5Min: 25.6 / Avg: 25.67 / Max: 25.7

SQLite Speedtest

This is a benchmark of SQLite's speedtest1 benchmark program with an increased problem size of 1,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Clear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed1428425670SE +/- 0.12, N = 3SE +/- 0.00, N = 3SE +/- 0.21, N = 3SE +/- 0.54, N = 352.6755.5060.3561.81-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O2-O21. (CC) gcc options: -ldl -lz -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Clear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed1224364860Min: 52.44 / Avg: 52.67 / Max: 52.81Min: 55.5 / Avg: 55.5 / Max: 55.5Min: 60.03 / Avg: 60.35 / Max: 60.75Min: 60.88 / Avg: 61.81 / Max: 62.731. (CC) gcc options: -ldl -lz -lpthread

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest CompressionClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta918273645SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 332.7537.6637.8238.16-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -ltiff-O2 -ltiff-O2-O2 -ltiff1. (CC) gcc options: -fvisibility=hidden -pthread -lm -ljpeg -lpng16
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest CompressionClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta816243240Min: 32.67 / Avg: 32.75 / Max: 32.84Min: 37.63 / Avg: 37.66 / Max: 37.69Min: 37.62 / Avg: 37.82 / Max: 37.95Min: 37.94 / Avg: 38.16 / Max: 38.271. (CC) gcc options: -fvisibility=hidden -pthread -lm -ljpeg -lpng16

Mobile Neural Network

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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: SqueezeNetV1.0Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929246810SE +/- 0.074, N = 15SE +/- 0.025, N = 3SE +/- 0.074, N = 3SE +/- 0.085, N = 97.4298.0448.5538.654-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 6.58 / MAX: 8.09MIN: 7.7 / MAX: 8.8MIN: 8.07 / MAX: 9.2MIN: 7.84 / MAX: 9.731. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: SqueezeNetV1.0Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009293691215Min: 6.94 / Avg: 7.43 / Max: 7.84Min: 8 / Avg: 8.04 / Max: 8.08Min: 8.45 / Avg: 8.55 / Max: 8.7Min: 8.19 / Avg: 8.65 / Max: 8.951. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009290.13430.26860.40290.53720.6715SE +/- 0.000391, N = 3SE +/- 0.000503, N = 3SE +/- 0.007438, N = 3SE +/- 0.005982, N = 30.5132490.5167540.5331890.597088MIN: 0.5MIN: 0.5-O2 - MIN: 0.49MIN: 0.561. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810Min: 0.51 / Avg: 0.51 / Max: 0.51Min: 0.52 / Avg: 0.52 / Max: 0.52Min: 0.53 / Avg: 0.53 / Max: 0.55Min: 0.59 / Avg: 0.6 / Max: 0.611. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Facebook RocksDB

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

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Sequential FillFedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 20200929400K800K1200K1600K2000KSE +/- 20906.40, N = 3SE +/- 10674.35, N = 3SE +/- 7849.59, N = 3180298416643611567863-O21. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Sequential FillFedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 20200929300K600K900K1200K1500KMin: 1764671 / Avg: 1802983.67 / Max: 1836643Min: 1644581 / Avg: 1664361.33 / Max: 1681207Min: 1556058 / Avg: 1567862.67 / Max: 15827281. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

FFTE

FFTE is a package by Daisuke Takahashi to compute Discrete Fourier Transforms of 1-, 2- and 3- dimensional sequences of length (2^p)*(3^q)*(5^r). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 7.0N=256, 3D Complex FFT RoutineUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed11K22K33K44K55KSE +/- 185.91, N = 3SE +/- 141.79, N = 3SE +/- 331.24, N = 3SE +/- 753.53, N = 352659.2148693.6748688.7545877.801. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 7.0N=256, 3D Complex FFT RoutineUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed9K18K27K36K45KMin: 52311.83 / Avg: 52659.21 / Max: 52947.72Min: 48507.16 / Avg: 48693.67 / Max: 48971.92Min: 48315.39 / Avg: 48688.75 / Max: 49349.36Min: 44571.64 / Avg: 45877.8 / Max: 47181.951. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.4Time To CompileClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291122334455SE +/- 0.60, N = 3SE +/- 0.64, N = 4SE +/- 0.67, N = 3SE +/- 0.85, N = 343.2445.2546.7249.46
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.4Time To CompileClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291020304050Min: 42.41 / Avg: 43.24 / Max: 44.4Min: 44.46 / Avg: 45.25 / Max: 47.17Min: 45.99 / Avg: 46.72 / Max: 48.06Min: 48.36 / Avg: 49.46 / Max: 51.13

Facebook RocksDB

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

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random FillFedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 20200929400K800K1200K1600K2000KSE +/- 6254.49, N = 3SE +/- 3895.19, N = 3SE +/- 16959.02, N = 6163657915091151433799-O21. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random FillFedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 20200929300K600K900K1200K1500KMin: 1625885 / Avg: 1636578.67 / Max: 1647546Min: 1501325 / Avg: 1509115 / Max: 1513076Min: 1401384 / Avg: 1433799.17 / Max: 15030511. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

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 1.2.0Scene: MemorialClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 20200929612182430SE +/- 0.07, N = 3SE +/- 0.34, N = 3SE +/- 0.17, N = 3SE +/- 0.21, N = 1425.1625.1325.0222.19
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.2.0Scene: MemorialClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 20200929612182430Min: 25.04 / Avg: 25.16 / Max: 25.29Min: 24.44 / Avg: 25.13 / Max: 25.52Min: 24.73 / Avg: 25.02 / Max: 25.33Min: 20.06 / Avg: 22.19 / Max: 23.2

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta714212835SE +/- 0.23, N = 3SE +/- 0.05, N = 3SE +/- 0.26, N = 10SE +/- 0.38, N = 1227.0327.3728.0130.62-O3-O3-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -lsqlite3-O21. (CXX) g++ options: -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta714212835Min: 26.7 / Avg: 27.03 / Max: 27.49Min: 27.26 / Avg: 27.37 / Max: 27.43Min: 27.5 / Avg: 28.01 / Max: 30.19Min: 26.46 / Avg: 30.62 / Max: 31.431. (CXX) g++ options: -lsnappy -lpthread

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta160K320K480K640K800KSE +/- 4785.64, N = 3SE +/- 2918.93, N = 3SE +/- 1709.93, N = 3SE +/- 656.29, N = 3728013.38687280.77666366.94643394.64-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O2 -lrt" -lrt
OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta130K260K390K520K650KMin: 722190.64 / Avg: 728013.38 / Max: 737503.41Min: 683479.42 / Avg: 687280.77 / Max: 693018.48Min: 663451.79 / Avg: 666366.94 / Max: 669373.08Min: 642130.92 / Avg: 643394.64 / Max: 644333.741. (CC) gcc options: -O2 -lrt" -lrt

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta48121620SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 315.2116.9617.0317.20-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -ltiff-O2 -ltiff-O2-O2 -ltiff1. (CC) gcc options: -fvisibility=hidden -pthread -lm -ljpeg -lpng16
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta48121620Min: 15.12 / Avg: 15.21 / Max: 15.26Min: 16.94 / Avg: 16.96 / Max: 16.99Min: 16.97 / Avg: 17.03 / Max: 17.14Min: 17.08 / Avg: 17.2 / Max: 17.261. (CC) gcc options: -fvisibility=hidden -pthread -lm -ljpeg -lpng16

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929246810SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 36.286.085.56
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 202009293691215Min: 6.27 / Avg: 6.28 / Max: 6.31Min: 6.05 / Avg: 6.08 / Max: 6.1Min: 5.53 / Avg: 5.56 / Max: 5.62

Darktable

Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Masskrug - Acceleration: CPU-onlyClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta0.73731.47462.21192.94923.6865SE +/- 0.000, N = 3SE +/- 0.018, N = 3SE +/- 0.013, N = 3SE +/- 0.005, N = 32.9053.2573.2723.277
OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Masskrug - Acceleration: CPU-onlyClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta246810Min: 2.9 / Avg: 2.9 / Max: 2.91Min: 3.24 / Avg: 3.26 / Max: 3.29Min: 3.25 / Avg: 3.27 / Max: 3.3Min: 3.27 / Avg: 3.28 / Max: 3.29

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta714212835SE +/- 0.21, N = 3SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.34, N = 327.4727.5527.6330.97-O3-O3-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -lsqlite3-O21. (CXX) g++ options: -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta714212835Min: 27.1 / Avg: 27.46 / Max: 27.83Min: 27.37 / Avg: 27.55 / Max: 27.68Min: 27.44 / Avg: 27.63 / Max: 27.74Min: 30.52 / Avg: 30.97 / Max: 31.631. (CXX) g++ options: -lsnappy -lpthread

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929714212835SE +/- 0.17, N = 3SE +/- 0.21, N = 3SE +/- 0.06, N = 329.7929.7626.46
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929714212835Min: 29.49 / Avg: 29.79 / Max: 30.08Min: 29.34 / Avg: 29.76 / Max: 29.97Min: 26.39 / Avg: 26.46 / Max: 26.58

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Rotate 90 DegreesClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed918273645SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 335.5037.6037.7739.95
OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Rotate 90 DegreesClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed816243240Min: 35.48 / Avg: 35.5 / Max: 35.52Min: 37.57 / Avg: 37.59 / Max: 37.64Min: 37.67 / Avg: 37.77 / Max: 37.92Min: 39.83 / Avg: 39.95 / Max: 40.08

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009290.65081.30161.95242.60323.254SE +/- 0.00447, N = 3SE +/- 0.00552, N = 3SE +/- 0.01296, N = 3SE +/- 0.03490, N = 32.580302.604982.641082.89246MIN: 2.55MIN: 2.58-O2 - MIN: 2.58MIN: 2.771. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810Min: 2.58 / Avg: 2.58 / Max: 2.59Min: 2.6 / Avg: 2.6 / Max: 2.62Min: 2.62 / Avg: 2.64 / Max: 2.67Min: 2.84 / Avg: 2.89 / Max: 2.961. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Color EnhanceClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 202009291224364860SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 346.5150.4451.9252.05
OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Color EnhanceClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 202009291020304050Min: 46.48 / Avg: 46.51 / Max: 46.52Min: 50.36 / Avg: 50.44 / Max: 50.55Min: 51.86 / Avg: 51.92 / Max: 51.96Min: 51.98 / Avg: 52.05 / Max: 52.18

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 2Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta510152025SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 320.2320.7220.8722.57-O3-O3-O3-O21. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 2Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta510152025Min: 20.21 / Avg: 20.23 / Max: 20.26Min: 20.66 / Avg: 20.72 / Max: 20.83Min: 20.82 / Avg: 20.87 / Max: 20.89Min: 22.57 / Avg: 22.57 / Max: 22.571. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929612182430SE +/- 0.23, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 324.8524.3722.31
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929612182430Min: 24.49 / Avg: 24.85 / Max: 25.29Min: 24.25 / Avg: 24.37 / Max: 24.52Min: 22.13 / Avg: 22.31 / Max: 22.41

Darktable

Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Server Room - Acceleration: CPU-onlyUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed0.67011.34022.01032.68043.3505SE +/- 0.006, N = 3SE +/- 0.008, N = 3SE +/- 0.007, N = 3SE +/- 0.001, N = 32.6792.6892.7632.978
OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Server Room - Acceleration: CPU-onlyUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed246810Min: 2.67 / Avg: 2.68 / Max: 2.69Min: 2.68 / Avg: 2.69 / Max: 2.7Min: 2.76 / Avg: 2.76 / Max: 2.78Min: 2.98 / Avg: 2.98 / Max: 2.98

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: yolov4-tinyClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta612182430SE +/- 0.02, N = 3SE +/- 0.09, N = 5SE +/- 0.14, N = 3SE +/- 0.03, N = 523.1824.0024.8625.71-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 22.74 / MAX: 28-O3 - MIN: 23.21 / MAX: 34.2-O3 - MIN: 24.03 / MAX: 30.04-O2 - MIN: 24.55 / MAX: 32.721. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: yolov4-tinyClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta612182430Min: 23.14 / Avg: 23.18 / Max: 23.22Min: 23.75 / Avg: 24 / Max: 24.31Min: 24.69 / Avg: 24.86 / Max: 25.13Min: 25.64 / Avg: 25.71 / Max: 25.831. (CXX) g++ options: -rdynamic -lgomp -lpthread

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 2Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291020304050SE +/- 0.22, N = 3SE +/- 0.38, N = 3SE +/- 0.16, N = 3SE +/- 0.29, N = 338.9040.8942.0742.991. (CXX) g++ options: -O3 -fPIC
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 2Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929918273645Min: 38.52 / Avg: 38.9 / Max: 39.27Min: 40.33 / Avg: 40.89 / Max: 41.61Min: 41.78 / Avg: 42.07 / Max: 42.33Min: 42.41 / Avg: 42.99 / Max: 43.331. (CXX) g++ options: -O3 -fPIC

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: alexnetClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009293691215SE +/- 0.07, N = 3SE +/- 0.16, N = 5SE +/- 0.13, N = 5SE +/- 0.01, N = 39.9210.0610.7610.93-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 9.73 / MAX: 10.06-O3 - MIN: 9.72 / MAX: 10.97-O2 - MIN: 10.13 / MAX: 29.35-O3 - MIN: 10.8 / MAX: 11.871. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: alexnetClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009293691215Min: 9.79 / Avg: 9.92 / Max: 10.01Min: 9.77 / Avg: 10.06 / Max: 10.49Min: 10.27 / Avg: 10.76 / Max: 10.95Min: 10.92 / Avg: 10.93 / Max: 10.941. (CXX) g++ options: -rdynamic -lgomp -lpthread

Facebook RocksDB

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

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Read While WritingUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780900K1800K2700K3600K4500KSE +/- 31411.63, N = 3SE +/- 7677.96, N = 3SE +/- 4720.39, N = 3421910539701793831517-O21. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Read While WritingUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780700K1400K2100K2800K3500KMin: 4187459 / Avg: 4219105.33 / Max: 4281928Min: 3956807 / Avg: 3970179 / Max: 3983403Min: 3826639 / Avg: 3831517 / Max: 38409561. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 3openSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta918273645SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 335.4836.4436.7639.03-O3-O3-O3-O21. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 3openSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta816243240Min: 35.44 / Avg: 35.48 / Max: 35.53Min: 36.39 / Avg: 36.44 / Max: 36.49Min: 36.74 / Avg: 36.76 / Max: 36.8Min: 39 / Avg: 39.03 / Max: 39.041. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -rdynamic -lm -lpthread

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 0Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291632486480SE +/- 0.07, N = 3SE +/- 0.19, N = 3SE +/- 0.30, N = 3SE +/- 0.23, N = 365.5869.5270.8972.061. (CXX) g++ options: -O3 -fPIC
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 0Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291428425670Min: 65.46 / Avg: 65.58 / Max: 65.72Min: 69.17 / Avg: 69.52 / Max: 69.84Min: 70.39 / Avg: 70.89 / Max: 71.43Min: 71.6 / Avg: 72.06 / Max: 72.351. (CXX) g++ options: -O3 -fPIC

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUFedora Workstation 33 BetaopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 202009290.35760.71521.07281.43041.788SE +/- 0.00837, N = 3SE +/- 0.00293, N = 3SE +/- 0.01165, N = 3SE +/- 0.00344, N = 31.446571.460121.502101.58932-O2 - MIN: 1.36MIN: 1.39MIN: 1.42MIN: 1.531. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUFedora Workstation 33 BetaopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929246810Min: 1.43 / Avg: 1.45 / Max: 1.46Min: 1.45 / Avg: 1.46 / Max: 1.47Min: 1.49 / Avg: 1.5 / Max: 1.52Min: 1.59 / Avg: 1.59 / Max: 1.61. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810SE +/- 0.03276, N = 3SE +/- 0.04482, N = 3SE +/- 0.00859, N = 3SE +/- 0.03569, N = 37.004937.059747.177257.68229MIN: 6.7MIN: 6.79-O2 - MIN: 6.9MIN: 7.281. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215Min: 6.95 / Avg: 7 / Max: 7.07Min: 6.99 / Avg: 7.06 / Max: 7.14Min: 7.17 / Avg: 7.18 / Max: 7.19Min: 7.61 / Avg: 7.68 / Max: 7.731. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta816243240SE +/- 0.17, N = 3SE +/- 0.05, N = 3SE +/- 0.20, N = 3SE +/- 0.15, N = 332.8033.9834.1535.86-O3-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -lsqlite3-O3-O21. (CXX) g++ options: -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta816243240Min: 32.47 / Avg: 32.8 / Max: 33.03Min: 33.92 / Avg: 33.98 / Max: 34.09Min: 33.75 / Avg: 34.15 / Max: 34.38Min: 35.64 / Avg: 35.86 / Max: 36.161. (CXX) g++ options: -lsnappy -lpthread

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: AntialiasClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed816243240SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 333.8634.9335.3236.90
OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: AntialiasClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed816243240Min: 33.79 / Avg: 33.86 / Max: 33.99Min: 34.92 / Avg: 34.93 / Max: 34.95Min: 35.29 / Avg: 35.32 / Max: 35.36Min: 36.84 / Avg: 36.9 / Max: 36.96

RawTherapee

RawTherapee is a cross-platform, open-source multi-threaded RAW image processing program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRawTherapeeTotal Benchmark TimeUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed1224364860SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 349.7650.6751.0254.081. Ubuntu 20.10 20200929: RawTherapee, version 5.8, command line.2. Fedora Workstation 33 Beta: RawTherapee, version 5.8, command line.3. Clear Linux 33780: RawTherapee, version , command line.4. openSUSE Tumbleweed: RawTherapee, version 5.8, command line.
OpenBenchmarking.orgSeconds, Fewer Is BetterRawTherapeeTotal Benchmark TimeUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed1122334455Min: 49.74 / Avg: 49.76 / Max: 49.78Min: 50.58 / Avg: 50.67 / Max: 50.73Min: 50.88 / Avg: 51.02 / Max: 51.09Min: 54.04 / Avg: 54.08 / Max: 54.151. Ubuntu 20.10 20200929: RawTherapee, version 5.8, command line.2. Fedora Workstation 33 Beta: RawTherapee, version 5.8, command line.3. Clear Linux 33780: RawTherapee, version , command line.4. openSUSE Tumbleweed: RawTherapee, version 5.8, command line.

GEGL

GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Tile GlassClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed714212835SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 327.3427.8328.7229.67
OpenBenchmarking.orgSeconds, Fewer Is BetterGEGLOperation: Tile GlassClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed714212835Min: 27.32 / Avg: 27.34 / Max: 27.36Min: 27.8 / Avg: 27.83 / Max: 27.85Min: 28.66 / Avg: 28.72 / Max: 28.77Min: 29.58 / Avg: 29.67 / Max: 29.77

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: shufflenet-v2Ubuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed1.23752.4753.71254.956.1875SE +/- 0.05, N = 2SE +/- 0.05, N = 5SE +/- 0.10, N = 3SE +/- 0.09, N = 55.075.115.145.50-O3 - MIN: 4.75 / MAX: 8.65-O2 - MIN: 4.7 / MAX: 20.31-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 4.81 / MAX: 16.62-O3 - MIN: 5.07 / MAX: 6.51. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: shufflenet-v2Ubuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed246810Min: 5.02 / Avg: 5.07 / Max: 5.11Min: 4.94 / Avg: 5.11 / Max: 5.24Min: 4.93 / Avg: 5.14 / Max: 5.24Min: 5.16 / Avg: 5.5 / Max: 5.671. (CXX) g++ options: -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: squeezenetClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed48121620SE +/- 0.03, N = 3SE +/- 0.17, N = 5SE +/- 0.06, N = 3SE +/- 0.17, N = 513.1614.0514.1714.20-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 13 / MAX: 13.52-O2 - MIN: 13.46 / MAX: 27.02-O3 - MIN: 13.89 / MAX: 26.2-O3 - MIN: 13.71 / MAX: 15.541. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: squeezenetClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed48121620Min: 13.1 / Avg: 13.16 / Max: 13.19Min: 13.63 / Avg: 14.05 / Max: 14.62Min: 14.1 / Avg: 14.17 / Max: 14.28Min: 13.82 / Avg: 14.2 / Max: 14.751. (CXX) g++ options: -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: blazefaceUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed0.5131.0261.5392.0522.565SE +/- 0.01, N = 3SE +/- 0.01, N = 5SE +/- 0.02, N = 3SE +/- 0.03, N = 52.122.142.182.28-O3 - MIN: 2.07 / MAX: 2.18-O2 - MIN: 2.07 / MAX: 2.26-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 2.1 / MAX: 14.99-O3 - MIN: 2.18 / MAX: 2.471. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: blazefaceUbuntu 20.10 20200929Fedora Workstation 33 BetaClear Linux 33780openSUSE Tumbleweed246810Min: 2.11 / Avg: 2.12 / Max: 2.13Min: 2.12 / Avg: 2.14 / Max: 2.16Min: 2.14 / Avg: 2.18 / Max: 2.22Min: 2.23 / Avg: 2.28 / Max: 2.411. (CXX) g++ options: -rdynamic -lgomp -lpthread

Git

This test measures the time needed to carry out some sample Git operations on an example, static repository that happens to be a copy of the GNOME GTK tool-kit repository. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGitTime To Complete Common Git CommandsClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291122334455SE +/- 0.10, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 345.2446.5747.2448.631. Clear Linux 33780: git version 2.28.02. openSUSE Tumbleweed: git version 2.28.03. Fedora Workstation 33 Beta: git version 2.28.04. Ubuntu 20.10 20200929: git version 2.27.0
OpenBenchmarking.orgSeconds, Fewer Is BetterGitTime To Complete Common Git CommandsClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291020304050Min: 45.1 / Avg: 45.24 / Max: 45.43Min: 46.4 / Avg: 46.57 / Max: 46.65Min: 47.19 / Avg: 47.24 / Max: 47.27Min: 48.51 / Avg: 48.63 / Max: 48.781. Clear Linux 33780: git version 2.28.02. openSUSE Tumbleweed: git version 2.28.03. Fedora Workstation 33 Beta: git version 2.28.04. Ubuntu 20.10 20200929: git version 2.27.0

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v2Clear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 2020092960120180240300SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.29, N = 3SE +/- 0.27, N = 3276.90286.64288.61297.20-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 275.91 / MAX: 285.12-O2 - MIN: 285.14 / MAX: 294.79-O3 - MIN: 286.94 / MAX: 301.68-O3 - MIN: 294.96 / MAX: 312.431. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v2Clear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 2020092950100150200250Min: 276.8 / Avg: 276.9 / Max: 277Min: 286.59 / Avg: 286.64 / Max: 286.72Min: 288.07 / Avg: 288.61 / Max: 289.06Min: 296.87 / Avg: 297.2 / Max: 297.731. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -rdynamic -ldl

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mobilenetopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta48121620SE +/- 0.08, N = 5SE +/- 0.00, N = 3SE +/- 0.16, N = 3SE +/- 0.25, N = 516.2916.4317.1617.47-O3 - MIN: 15.99 / MAX: 17.28-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 15.94 / MAX: 20.27-O3 - MIN: 16.76 / MAX: 17.48-O2 - MIN: 16.65 / MAX: 46.981. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mobilenetopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta48121620Min: 16.1 / Avg: 16.29 / Max: 16.56Min: 16.43 / Avg: 16.43 / Max: 16.44Min: 16.84 / Avg: 17.16 / Max: 17.39Min: 16.89 / Avg: 17.47 / Max: 18.141. (CXX) g++ options: -rdynamic -lgomp -lpthread

DeepSpeech

Mozilla DeepSpeech is a speech-to-text engine powered by TensorFlow for machine learning and derived from Baidu's Deep Speech research paper. This test profile times the speech-to-text process for a roughly three minute audio recording. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDeepSpeech 0.6Acceleration: CPUClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed306090120150SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.11, N = 3108.69112.13115.53116.51
OpenBenchmarking.orgSeconds, Fewer Is BetterDeepSpeech 0.6Acceleration: CPUClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed20406080100Min: 108.57 / Avg: 108.69 / Max: 108.76Min: 112.01 / Avg: 112.13 / Max: 112.36Min: 115.39 / Avg: 115.53 / Max: 115.74Min: 116.3 / Avg: 116.51 / Max: 116.64

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: Visual Quality Optimized - Input: Bosphorus 1080pClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta60120180240300SE +/- 0.77, N = 3SE +/- 1.16, N = 3SE +/- 1.24, N = 3SE +/- 1.39, N = 3268.78255.98254.11251.98-O21. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: Visual Quality Optimized - Input: Bosphorus 1080pClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta50100150200250Min: 267.38 / Avg: 268.78 / Max: 270.03Min: 253.7 / Avg: 255.98 / Max: 257.51Min: 252.42 / Avg: 254.11 / Max: 256.52Min: 249.38 / Avg: 251.98 / Max: 254.131. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

Himeno Benchmark

The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009299001800270036004500SE +/- 1.81, N = 3SE +/- 2.23, N = 3SE +/- 1.02, N = 3SE +/- 10.55, N = 34250.714023.604005.433994.37-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -mavx2
OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009297001400210028003500Min: 4247.14 / Avg: 4250.71 / Max: 4253.06Min: 4020.14 / Avg: 4023.6 / Max: 4027.76Min: 4003.48 / Avg: 4005.43 / Max: 4006.94Min: 3982.26 / Avg: 3994.37 / Max: 4015.381. (CC) gcc options: -O3 -mavx2

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 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810SE +/- 0.00431, N = 3SE +/- 0.00386, N = 3SE +/- 0.06631, N = 37.846187.814047.38538-lmpi_cxx-lmpi_cxx1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Clear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215Min: 7.84 / Avg: 7.85 / Max: 7.85Min: 7.81 / Avg: 7.81 / Max: 7.82Min: 7.25 / Avg: 7.39 / Max: 7.461. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi

Crafty

This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed TimeopenSUSE TumbleweedFedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 202009292M4M6M8M10MSE +/- 39588.48, N = 3SE +/- 15600.26, N = 3SE +/- 10545.32, N = 3SE +/- 56119.05, N = 395983279501807935914990405561. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed TimeopenSUSE TumbleweedFedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 202009291.7M3.4M5.1M6.8M8.5MMin: 9521797 / Avg: 9598327.33 / Max: 9654173Min: 9471597 / Avg: 9501806.67 / Max: 9523667Min: 9346634 / Avg: 9359148.67 / Max: 9380108Min: 8928397 / Avg: 9040556.33 / Max: 91002771. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Inception V3 - Device: CPUClear Linux 33780Ubuntu 20.10 20200929openSUSE Tumbleweed246810SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 36.736.626.34
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Inception V3 - Device: CPUClear Linux 33780Ubuntu 20.10 20200929openSUSE Tumbleweed3691215Min: 6.71 / Avg: 6.73 / Max: 6.76Min: 6.6 / Avg: 6.62 / Max: 6.64Min: 6.33 / Avg: 6.34 / Max: 6.36

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta12M24M36M48M60MSE +/- 421420.95, N = 3SE +/- 405576.86, N = 3SE +/- 704473.66, N = 3SE +/- 648467.99, N = 356165113544610515323476752941334
OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta10M20M30M40M50MMin: 55500059 / Avg: 56165113.33 / Max: 56946042Min: 53737000 / Avg: 54461051.33 / Max: 55139763Min: 51850643 / Avg: 53234767.33 / Max: 54154863Min: 52283958 / Avg: 52941334 / Max: 54238229

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest CompressionClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929246810SE +/- 0.007, N = 3SE +/- 0.004, N = 3SE +/- 0.018, N = 3SE +/- 0.003, N = 36.2666.5606.6076.633-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -ltiff-O2-O2 -ltiff-O2 -ltiff1. (CC) gcc options: -fvisibility=hidden -pthread -lm -ljpeg -lpng16
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest CompressionClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009293691215Min: 6.26 / Avg: 6.27 / Max: 6.28Min: 6.55 / Avg: 6.56 / Max: 6.57Min: 6.58 / Avg: 6.61 / Max: 6.64Min: 6.63 / Avg: 6.63 / Max: 6.641. (CC) gcc options: -fvisibility=hidden -pthread -lm -ljpeg -lpng16

BRL-CAD

BRL-CAD 7.28.0 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 7.30.8VGR Performance MetricClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed50K100K150K200K250K211356207515205846199874-lSM -lICE -lXext -luuid-lSM -lICE -lXext1. (CXX) g++ options: -std=c++11 -pipe -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -rdynamic -lGLU -lGL -lGLdispatch -lX11 -lpthread -ldl -lm

Darktable

Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Server Rack - Acceleration: CPU-onlyClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed0.02970.05940.08910.11880.1485SE +/- 0.001, N = 3SE +/- 0.002, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 30.1250.1280.1300.132
OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Server Rack - Acceleration: CPU-onlyClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed12345Min: 0.12 / Avg: 0.12 / Max: 0.13Min: 0.13 / Avg: 0.13 / Max: 0.13Min: 0.13 / Avg: 0.13 / Max: 0.13Min: 0.13 / Avg: 0.13 / Max: 0.14

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet50Clear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed510152025SE +/- 0.09, N = 3SE +/- 0.31, N = 5SE +/- 0.01, N = 3SE +/- 0.43, N = 521.1321.5321.7922.29-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 20.86 / MAX: 21.4-O2 - MIN: 20.05 / MAX: 23.48-O3 - MIN: 21.52 / MAX: 35.09-O3 - MIN: 21.02 / MAX: 23.981. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet50Clear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed510152025Min: 20.96 / Avg: 21.13 / Max: 21.22Min: 20.29 / Avg: 21.53 / Max: 21.88Min: 21.77 / Avg: 21.79 / Max: 21.81Min: 21.18 / Avg: 22.29 / Max: 23.081. (CXX) g++ options: -rdynamic -lgomp -lpthread

AI Benchmark Alpha

AI Benchmark Alpha is a Python library for evaluating artificial intelligence (AI) performance on diverse hardware platforms and relies upon the TensorFlow machine learning library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device Inference ScoreClear Linux 33780Ubuntu 20.10 2020092940080012001600200020321930

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 2020092948121620SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 313.9013.3513.22
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: CPUClear Linux 33780openSUSE TumbleweedUbuntu 20.10 2020092948121620Min: 13.75 / Avg: 13.9 / Max: 14Min: 13.22 / Avg: 13.35 / Max: 13.51Min: 13.12 / Avg: 13.22 / Max: 13.34

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch 1D - Data Type: f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009290.52211.04421.56632.08842.6105SE +/- 0.00089, N = 3SE +/- 0.00286, N = 3SE +/- 0.00643, N = 3SE +/- 0.03226, N = 42.210722.212862.223592.32048MIN: 2.13MIN: 2.14-O2 - MIN: 2.04MIN: 2.131. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch 1D - Data Type: f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810Min: 2.21 / Avg: 2.21 / Max: 2.21Min: 2.21 / Avg: 2.21 / Max: 2.22Min: 2.21 / Avg: 2.22 / Max: 2.23Min: 2.23 / Avg: 2.32 / Max: 2.381. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

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.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: MediumopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 36.686.866.957.011. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: MediumopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215Min: 6.67 / Avg: 6.68 / Max: 6.7Min: 6.85 / Avg: 6.86 / Max: 6.86Min: 6.94 / Avg: 6.95 / Max: 6.95Min: 6.99 / Avg: 7.01 / Max: 7.031. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009290.40110.80221.20331.60442.0055SE +/- 0.00188, N = 3SE +/- 0.00301, N = 3SE +/- 0.00624, N = 3SE +/- 0.02002, N = 31.698961.705511.742211.78264MIN: 1.68MIN: 1.67-O2 - MIN: 1.69MIN: 1.711. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810Min: 1.7 / Avg: 1.7 / Max: 1.7Min: 1.7 / Avg: 1.71 / Max: 1.71Min: 1.73 / Avg: 1.74 / Max: 1.75Min: 1.75 / Avg: 1.78 / Max: 1.821. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing on the CPU with the water_GMX50 data. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water BenchmarkClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta0.35890.71781.07671.43561.7945SE +/- 0.002, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 31.5951.5751.522-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -ldl-O3-O2 -ldl1. (CXX) g++ options: -pthread -lrt -lpthread -lm
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water BenchmarkClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta246810Min: 1.59 / Avg: 1.59 / Max: 1.6Min: 1.57 / Avg: 1.58 / Max: 1.58Min: 1.52 / Avg: 1.52 / Max: 1.521. (CXX) g++ options: -pthread -lrt -lpthread -lm

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch All - Data Type: f32 - Engine: CPUClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929816243240SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.12, N = 3SE +/- 0.10, N = 331.2231.3832.0132.70MIN: 29.81MIN: 29.81-O2 - MIN: 30.67MIN: 31.081. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch All - Data Type: f32 - Engine: CPUClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929714212835Min: 31.1 / Avg: 31.22 / Max: 31.33Min: 31.35 / Avg: 31.38 / Max: 31.41Min: 31.81 / Avg: 32.01 / Max: 32.22Min: 32.55 / Avg: 32.7 / Max: 32.891. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_1d - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929Clear Linux 337803691215SE +/- 0.00341, N = 3SE +/- 0.01309, N = 3SE +/- 0.00075, N = 3SE +/- 0.00147, N = 38.827938.958049.148029.22090MIN: 8.74-O2 - MIN: 8.67MIN: 8.92MIN: 9.081. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_1d - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929Clear Linux 337803691215Min: 8.82 / Avg: 8.83 / Max: 8.83Min: 8.94 / Avg: 8.96 / Max: 8.98Min: 9.15 / Avg: 9.15 / Max: 9.15Min: 9.22 / Avg: 9.22 / Max: 9.221. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215SE +/- 0.01124, N = 3SE +/- 0.04894, N = 3SE +/- 0.01577, N = 3SE +/- 0.02565, N = 39.052099.174999.312769.44985MIN: 8.98MIN: 9.05-O2 - MIN: 9.21MIN: 9.331. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215Min: 9.03 / Avg: 9.05 / Max: 9.07Min: 9.12 / Avg: 9.17 / Max: 9.27Min: 9.29 / Avg: 9.31 / Max: 9.34Min: 9.4 / Avg: 9.45 / Max: 9.481. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: EmilyClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta50100150200250219.45222.77227.10228.99

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: googlenetClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta48121620SE +/- 0.04, N = 3SE +/- 0.35, N = 5SE +/- 0.45, N = 3SE +/- 0.35, N = 513.6113.9614.0614.20-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 13.49 / MAX: 20.39-O3 - MIN: 13.36 / MAX: 16.65-O3 - MIN: 13.52 / MAX: 15.76-O2 - MIN: 13.26 / MAX: 40.261. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: googlenetClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta48121620Min: 13.54 / Avg: 13.61 / Max: 13.65Min: 13.44 / Avg: 13.96 / Max: 15.32Min: 13.59 / Avg: 14.06 / Max: 14.96Min: 13.56 / Avg: 14.2 / Max: 15.141. (CXX) g++ options: -rdynamic -lgomp -lpthread

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.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: FastopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009291.14532.29063.43594.58125.7265SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 34.885.055.065.091. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: FastopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810Min: 4.85 / Avg: 4.88 / Max: 4.93Min: 5.03 / Avg: 5.05 / Max: 5.08Min: 5.05 / Avg: 5.06 / Max: 5.07Min: 5.07 / Avg: 5.09 / Max: 5.11. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

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 0.9Benchmark: vklBenchmarkClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 2020092960120180240300SE +/- 0.86, N = 3SE +/- 0.47, N = 3SE +/- 1.42, N = 3SE +/- 0.22, N = 3280.11274.81270.17268.67MIN: 1 / MAX: 1134MIN: 1 / MAX: 1115MIN: 1 / MAX: 1098MIN: 1 / MAX: 1084
OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 2020092950100150200250Min: 278.83 / Avg: 280.11 / Max: 281.75Min: 274.33 / Avg: 274.81 / Max: 275.75Min: 268 / Avg: 270.17 / Max: 272.83Min: 268.25 / Avg: 268.67 / Max: 269

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta70140210280350SE +/- 3.43, N = 3SE +/- 4.02, N = 4SE +/- 2.74, N = 3SE +/- 4.41, N = 3321.45312.25309.81308.51-O21. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta60120180240300Min: 314.63 / Avg: 321.45 / Max: 325.56Min: 300.6 / Avg: 312.25 / Max: 318.98Min: 304.57 / Avg: 309.81 / Max: 313.81Min: 299.7 / Avg: 308.51 / Max: 313.321. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 202009293691215SE +/- 0.00277, N = 3SE +/- 0.02520, N = 3SE +/- 0.05240, N = 3SE +/- 0.03370, N = 39.629949.808009.8088210.00959MIN: 9.56-O2 - MIN: 9.68MIN: 9.59MIN: 9.861. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 202009293691215Min: 9.63 / Avg: 9.63 / Max: 9.64Min: 9.77 / Avg: 9.81 / Max: 9.86Min: 9.73 / Avg: 9.81 / Max: 9.91Min: 9.94 / Avg: 10.01 / Max: 10.051. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009290.39770.79541.19311.59081.9885SE +/- 0.00363, N = 3SE +/- 0.00724, N = 3SE +/- 0.01563, N = 3SE +/- 0.00967, N = 31.701571.713721.751591.76756MIN: 1.63MIN: 1.62-O2 - MIN: 1.64MIN: 1.651. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929246810Min: 1.7 / Avg: 1.7 / Max: 1.71Min: 1.71 / Avg: 1.71 / Max: 1.73Min: 1.73 / Avg: 1.75 / Max: 1.78Min: 1.75 / Avg: 1.77 / Max: 1.781. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: vgg16Fedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 20200929openSUSE Tumbleweed1020304050SE +/- 0.25, N = 5SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.33, N = 540.1540.8841.2641.66-O2 - MIN: 39.37 / MAX: 52.59-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 40.7 / MAX: 42.66-O3 - MIN: 41.12 / MAX: 44.72-O3 - MIN: 40.85 / MAX: 45.071. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: vgg16Fedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 20200929openSUSE Tumbleweed918273645Min: 39.56 / Avg: 40.15 / Max: 40.96Min: 40.81 / Avg: 40.88 / Max: 41.02Min: 41.23 / Avg: 41.26 / Max: 41.28Min: 41.01 / Avg: 41.66 / Max: 42.691. (CXX) g++ options: -rdynamic -lgomp -lpthread

Zstd Compression

This test measures the time needed to compress a sample file (an Ubuntu ISO) using Zstd compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 19Clear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed1428425670SE +/- 0.07, N = 3SE +/- 0.23, N = 3SE +/- 0.20, N = 3SE +/- 0.15, N = 362.260.360.260.0-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 19Clear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 BetaopenSUSE Tumbleweed1224364860Min: 62.1 / Avg: 62.23 / Max: 62.3Min: 59.8 / Avg: 60.27 / Max: 60.5Min: 59.8 / Avg: 60.2 / Max: 60.4Min: 59.7 / Avg: 59.97 / Max: 60.21. (CC) gcc options: -O3 -pthread -lz -llzma

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mnasnetUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed1.2242.4483.6724.8966.12SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 55.255.325.345.44-O3 - MIN: 4.93 / MAX: 11.38-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 5.1 / MAX: 5.47-O2 - MIN: 4.81 / MAX: 46.89-O3 - MIN: 5.14 / MAX: 5.841. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mnasnetUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed246810Min: 5.14 / Avg: 5.25 / Max: 5.42Min: 5.27 / Avg: 5.32 / Max: 5.35Min: 5.28 / Avg: 5.34 / Max: 5.42Min: 5.33 / Avg: 5.44 / Max: 5.751. (CXX) g++ options: -rdynamic -lgomp -lpthread

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Pabellon Barcelona - Compute: CPU-OnlyopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta70140210280350SE +/- 0.49, N = 3SE +/- 1.16, N = 3SE +/- 0.28, N = 3SE +/- 0.35, N = 3303.07303.52311.57313.83
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Pabellon Barcelona - Compute: CPU-OnlyopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta60120180240300Min: 302.18 / Avg: 303.07 / Max: 303.87Min: 302.27 / Avg: 303.52 / Max: 305.83Min: 311.15 / Avg: 311.57 / Max: 312.09Min: 313.28 / Avg: 313.83 / Max: 314.47

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: efficientnet-b0Fedora Workstation 33 BetaUbuntu 20.10 20200929Clear Linux 33780openSUSE Tumbleweed246810SE +/- 0.12, N = 5SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.14, N = 57.117.157.197.35-O2 - MIN: 6.51 / MAX: 8.85-O3 - MIN: 6.78 / MAX: 11.03-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 6.9 / MAX: 7.52-O3 - MIN: 6.87 / MAX: 8.681. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: efficientnet-b0Fedora Workstation 33 BetaUbuntu 20.10 20200929Clear Linux 33780openSUSE Tumbleweed3691215Min: 6.8 / Avg: 7.11 / Max: 7.4Min: 7.02 / Avg: 7.15 / Max: 7.22Min: 7.07 / Avg: 7.19 / Max: 7.36Min: 7.09 / Avg: 7.35 / Max: 7.861. (CXX) g++ options: -rdynamic -lgomp -lpthread

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: CPU-OnlyopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta90180270360450SE +/- 0.58, N = 3SE +/- 0.32, N = 3SE +/- 0.41, N = 3SE +/- 0.40, N = 3379.86380.51389.49392.30
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: CPU-OnlyopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta70140210280350Min: 379.1 / Avg: 379.86 / Max: 380.99Min: 380.13 / Avg: 380.51 / Max: 381.15Min: 388.91 / Avg: 389.49 / Max: 390.29Min: 391.57 / Avg: 392.3 / Max: 392.95

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPUopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 337802004006008001000SE +/- 4.35, N = 3SE +/- 1.65, N = 3SE +/- 3.02, N = 3795.38787.51770.73
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPUopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780140280420560700Min: 787.6 / Avg: 795.38 / Max: 802.63Min: 784.26 / Avg: 787.51 / Max: 789.64Min: 765.73 / Avg: 770.73 / Max: 776.17

AI Benchmark Alpha

AI Benchmark Alpha is a Python library for evaluating artificial intelligence (AI) performance on diverse hardware platforms and relies upon the TensorFlow machine learning library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device AI ScoreClear Linux 33780Ubuntu 20.10 20200929800160024003200400035853478

Darktable

Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Boat - Acceleration: CPU-onlyClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta246810SE +/- 0.014, N = 3SE +/- 0.017, N = 3SE +/- 0.022, N = 3SE +/- 0.011, N = 38.2208.3518.3538.463
OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.2.1Test: Boat - Acceleration: CPU-onlyClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta3691215Min: 8.2 / Avg: 8.22 / Max: 8.25Min: 8.32 / Avg: 8.35 / Max: 8.37Min: 8.31 / Avg: 8.35 / Max: 8.39Min: 8.44 / Avg: 8.46 / Max: 8.48

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Fishy Cat - Compute: CPU-OnlyopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta306090120150SE +/- 0.17, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3124.62124.78127.11128.06
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Fishy Cat - Compute: CPU-OnlyopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta20406080100Min: 124.31 / Avg: 124.62 / Max: 124.91Min: 124.7 / Avg: 124.78 / Max: 124.91Min: 126.98 / Avg: 127.11 / Max: 127.18Min: 127.93 / Avg: 128.06 / Max: 128.17

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.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: ThoroughopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 2020092948121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 317.4517.6517.7617.931. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: ThoroughopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929510152025Min: 17.44 / Avg: 17.45 / Max: 17.46Min: 17.64 / Avg: 17.65 / Max: 17.66Min: 17.75 / Avg: 17.76 / Max: 17.76Min: 17.91 / Avg: 17.93 / Max: 17.951. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Disney MaterialClear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta306090120150119.32120.84121.51122.40

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.1Clear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed60120180240300SE +/- 0.07, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 3.12, N = 3275.72279.36281.03282.46-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 275.44 / MAX: 276.54-O2 - MIN: 278.84 / MAX: 280.84-O3 - MIN: 280.52 / MAX: 282.72-O3 - MIN: 278.62 / MAX: 297.81. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.1Clear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929openSUSE Tumbleweed50100150200250Min: 275.64 / Avg: 275.72 / Max: 275.87Min: 279.17 / Avg: 279.36 / Max: 279.49Min: 280.98 / Avg: 281.03 / Max: 281.1Min: 279.15 / Avg: 282.46 / Max: 288.691. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -rdynamic -ldl

Facebook RocksDB

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

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random ReadFedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 2020092920M40M60M80M100MSE +/- 87613.91, N = 3SE +/- 27355.59, N = 3SE +/- 116666.65, N = 310165231510153596199244113-O21. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random ReadFedora Workstation 33 BetaClear Linux 33780Ubuntu 20.10 2020092920M40M60M80M100MMin: 101479523 / Avg: 101652314.67 / Max: 101763927Min: 101481315 / Avg: 101535961 / Max: 101565596Min: 99114924 / Avg: 99244113 / Max: 994769811. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch 1D - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta1.25892.51783.77675.03566.2945SE +/- 0.00504, N = 3SE +/- 0.01117, N = 3SE +/- 0.03159, N = 3SE +/- 0.03682, N = 35.476015.492185.581135.59528MIN: 5.38MIN: 5.43MIN: 5.43-O2 - MIN: 5.411. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch 1D - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta246810Min: 5.47 / Avg: 5.48 / Max: 5.48Min: 5.48 / Avg: 5.49 / Max: 5.51Min: 5.52 / Avg: 5.58 / Max: 5.63Min: 5.52 / Avg: 5.6 / Max: 5.641. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta0.10490.20980.31470.41960.5245SE +/- 0.000445, N = 3SE +/- 0.000033, N = 3SE +/- 0.001807, N = 3SE +/- 0.001959, N = 30.4563410.4605950.4632720.466256MIN: 0.44MIN: 0.45MIN: 0.45-O2 - MIN: 0.451. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta12345Min: 0.46 / Avg: 0.46 / Max: 0.46Min: 0.46 / Avg: 0.46 / Max: 0.46Min: 0.46 / Avg: 0.46 / Max: 0.47Min: 0.46 / Avg: 0.47 / Max: 0.471. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Zstd Compression

This test measures the time needed to compress a sample file (an Ubuntu ISO) using Zstd compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 3openSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta11002200330044005500SE +/- 15.85, N = 3SE +/- 12.17, N = 3SE +/- 36.88, N = 3SE +/- 19.66, N = 34959.54909.44886.64863.4-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 3openSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta9001800270036004500Min: 4928.9 / Avg: 4959.5 / Max: 4982Min: 4887.7 / Avg: 4909.4 / Max: 4929.8Min: 4817.1 / Avg: 4886.57 / Max: 4942.8Min: 4827.9 / Avg: 4863.43 / Max: 4895.81. (CC) gcc options: -O3 -pthread -lz -llzma

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: BMW27 - Compute: CPU-OnlyClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 2020092920406080100SE +/- 0.33, N = 3SE +/- 0.15, N = 3SE +/- 0.06, N = 3SE +/- 0.14, N = 391.7392.2393.4993.52
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: BMW27 - Compute: CPU-OnlyClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 2020092920406080100Min: 91.28 / Avg: 91.73 / Max: 92.38Min: 92.05 / Avg: 92.23 / Max: 92.52Min: 93.38 / Avg: 93.49 / Max: 93.57Min: 93.24 / Avg: 93.52 / Max: 93.71

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 2020092930K60K90K120K150KSE +/- 259.29, N = 3SE +/- 181.79, N = 3SE +/- 195.25, N = 3SE +/- 340.53, N = 3116137116724117441118100
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 2020092920K40K60K80K100KMin: 115774 / Avg: 116136.67 / Max: 116639Min: 116506 / Avg: 116724 / Max: 117085Min: 117053 / Avg: 117441.33 / Max: 117671Min: 117655 / Avg: 118100 / Max: 118769

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Material TesteropenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 20200929306090120150124.75124.76126.47126.62

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_3d - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 310.7810.8510.9110.92MIN: 10.67MIN: 10.72-O2 - MIN: 10.71MIN: 10.731. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_3d - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215Min: 10.78 / Avg: 10.78 / Max: 10.78Min: 10.82 / Avg: 10.85 / Max: 10.88Min: 10.88 / Avg: 10.91 / Max: 10.94Min: 10.87 / Avg: 10.92 / Max: 10.961. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Classroom - Compute: CPU-OnlyopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta60120180240300SE +/- 0.16, N = 3SE +/- 0.88, N = 3SE +/- 0.51, N = 3SE +/- 0.17, N = 3280.85280.87283.47284.29
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Classroom - Compute: CPU-OnlyopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta50100150200250Min: 280.57 / Avg: 280.85 / Max: 281.11Min: 279.11 / Avg: 280.87 / Max: 281.89Min: 282.51 / Avg: 283.47 / Max: 284.25Min: 283.98 / Avg: 284.29 / Max: 284.57

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.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: ExhaustiveopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta306090120150SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3139.85140.76141.12141.381. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: ExhaustiveopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta306090120150Min: 139.8 / Avg: 139.85 / Max: 139.9Min: 140.7 / Avg: 140.76 / Max: 140.79Min: 141.08 / Avg: 141.12 / Max: 141.16Min: 141.37 / Avg: 141.38 / Max: 141.391. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

Dolfyn

Dolfyn is a Computational Fluid Dynamics (CFD) code of modern numerical simulation techniques. The Dolfyn test profile measures the execution time of the bundled computational fluid dynamics demos that are bundled with Dolfyn. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDolfyn 0.527Computational Fluid DynamicsClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929510152025SE +/- 0.14, N = 3SE +/- 0.10, N = 3SE +/- 0.10, N = 3SE +/- 0.05, N = 318.2818.2918.3918.48
OpenBenchmarking.orgSeconds, Fewer Is BetterDolfyn 0.527Computational Fluid DynamicsClear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929510152025Min: 18.03 / Avg: 18.28 / Max: 18.49Min: 18.1 / Avg: 18.29 / Max: 18.44Min: 18.21 / Avg: 18.39 / Max: 18.54Min: 18.38 / Avg: 18.48 / Max: 18.55

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta20K40K60K80K100KSE +/- 41.21, N = 3SE +/- 59.33, N = 3SE +/- 83.96, N = 3SE +/- 36.06, N = 387430.387493.187810.988258.6
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta15K30K45K60K75KMin: 87349.3 / Avg: 87430.3 / Max: 87484Min: 87414.6 / Avg: 87493.1 / Max: 87609.4Min: 87723.3 / Avg: 87810.93 / Max: 87978.8Min: 88214.5 / Avg: 88258.63 / Max: 88330.1

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta246810SE +/- 0.00284, N = 3SE +/- 0.00386, N = 3SE +/- 0.01495, N = 3SE +/- 0.04760, N = 37.954117.990688.018578.02911MIN: 7.65MIN: 7.71MIN: 7.6-O2 - MIN: 7.671. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta3691215Min: 7.95 / Avg: 7.95 / Max: 7.96Min: 7.98 / Avg: 7.99 / Max: 8Min: 7.99 / Avg: 8.02 / Max: 8.04Min: 7.93 / Avg: 8.03 / Max: 8.081. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetUbuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 BetaClear Linux 3378030K60K90K120K150KSE +/- 20.08, N = 3SE +/- 21.14, N = 3SE +/- 67.98, N = 3SE +/- 69.22, N = 3127956127967128302129099
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetUbuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 BetaClear Linux 3378020K40K60K80K100KMin: 127917 / Avg: 127956.33 / Max: 127983Min: 127937 / Avg: 127967.33 / Max: 128008Min: 128214 / Avg: 128302.33 / Max: 128436Min: 128961 / Avg: 129099.33 / Max: 129173

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta20K40K60K80K100KSE +/- 19.54, N = 3SE +/- 44.75, N = 3SE +/- 67.67, N = 3SE +/- 73.74, N = 390079.290284.490286.290878.0
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantopenSUSE TumbleweedClear Linux 33780Ubuntu 20.10 20200929Fedora Workstation 33 Beta16K32K48K64K80KMin: 90045.6 / Avg: 90079.2 / Max: 90113.3Min: 90195.2 / Avg: 90284.37 / Max: 90335.6Min: 90181.1 / Avg: 90286.23 / Max: 90412.6Min: 90752.9 / Avg: 90878.03 / Max: 91008.2

Hierarchical INTegration

This test runs the U.S. Department of Energy's Ames Laboratory Hierarchical INTegration (HINT) benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQUIPs, More Is BetterHierarchical INTegration 1.0Test: FLOATopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta100M200M300M400M500MSE +/- 472141.24, N = 3SE +/- 885405.39, N = 3SE +/- 750713.54, N = 3SE +/- 311710.44, N = 3458498603.13455254464.97455035663.30454854722.63-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -march=native -lm
OpenBenchmarking.orgQUIPs, More Is BetterHierarchical INTegration 1.0Test: FLOATopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta80M160M240M320M400MMin: 457603302.96 / Avg: 458498603.13 / Max: 459206215.25Min: 453602751.21 / Avg: 455254464.97 / Max: 456633231.73Min: 453675780.68 / Avg: 455035663.3 / Max: 456266741.55Min: 454245023.76 / Avg: 454854722.63 / Max: 455272224.891. (CC) gcc options: -O3 -march=native -lm

Kripke

Kripke is a simple, scalable, 3D Sn deterministic particle transport code. Its primary purpose is to research how data layout, programming paradigms and architectures effect the implementation and performance of Sn transport. Kripke is developed by LLNL. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.4Ubuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed7M14M21M28M35MSE +/- 21100.15, N = 3SE +/- 74000.40, N = 3SE +/- 12013.89, N = 3SE +/- 64733.28, N = 334707153344797533447567034434280-O3-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O31. (CXX) g++ options: -fopenmp
OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.4Ubuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 BetaopenSUSE Tumbleweed6M12M18M24M30MMin: 34672280 / Avg: 34707153.33 / Max: 34745170Min: 34353140 / Avg: 34479753.33 / Max: 34609430Min: 34452540 / Avg: 34475670 / Max: 34492870Min: 34310840 / Avg: 34434280 / Max: 345298101. (CXX) g++ options: -fopenmp

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test measures the RSA 4096-bit performance of OpenSSL. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.1RSA 4096-bit PerformanceClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta11002200330044005500SE +/- 4.95, N = 3SE +/- 5.55, N = 3SE +/- 13.63, N = 3SE +/- 5.10, N = 35212.05191.95175.95175.6-pipe -fexceptions -fstack-protector -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.1RSA 4096-bit PerformanceClear Linux 33780Ubuntu 20.10 20200929openSUSE TumbleweedFedora Workstation 33 Beta9001800270036004500Min: 5204.9 / Avg: 5211.97 / Max: 5221.5Min: 5183.2 / Avg: 5191.87 / Max: 5202.2Min: 5154.6 / Avg: 5175.93 / Max: 5201.3Min: 5166.4 / Avg: 5175.63 / Max: 51841. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

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 AtomsopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta0.22110.44220.66330.88441.1055SE +/- 0.00123, N = 3SE +/- 0.00123, N = 3SE +/- 0.00202, N = 3SE +/- 0.00086, N = 30.975960.977190.978370.98268
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsopenSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta246810Min: 0.97 / Avg: 0.98 / Max: 0.98Min: 0.98 / Avg: 0.98 / Max: 0.98Min: 0.97 / Avg: 0.98 / Max: 0.98Min: 0.98 / Avg: 0.98 / Max: 0.98

libjpeg-turbo tjbench

tjbench is a JPEG decompression/compression benchmark part of libjpeg-turbo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.0.2Test: Decompression ThroughputopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 2020092950100150200250SE +/- 0.12, N = 3SE +/- 0.42, N = 3SE +/- 0.32, N = 3SE +/- 0.18, N = 3217.65217.10216.44216.39-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -rdynamic
OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.0.2Test: Decompression ThroughputopenSUSE TumbleweedClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009294080120160200Min: 217.43 / Avg: 217.65 / Max: 217.84Min: 216.29 / Avg: 217.1 / Max: 217.68Min: 215.85 / Avg: 216.44 / Max: 216.95Min: 216.03 / Avg: 216.39 / Max: 216.571. (CC) gcc options: -O3 -rdynamic

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2openSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta400K800K1200K1600K2000KSE +/- 229.13, N = 3SE +/- 425.79, N = 3SE +/- 412.81, N = 3SE +/- 581.84, N = 31641410164234016463531650010
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2openSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta300K600K900K1200K1500KMin: 1641110 / Avg: 1641410 / Max: 1641860Min: 1641760 / Avg: 1642340 / Max: 1643170Min: 1645550 / Avg: 1646353.33 / Max: 1646920Min: 1649350 / Avg: 1650010 / Max: 1651170

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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch All - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929Clear Linux 337801428425670SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 363.6463.8663.8763.97MIN: 63.15-O2 - MIN: 63.2MIN: 63.01MIN: 63.431. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch All - Data Type: bf16bf16bf16 - Engine: CPUopenSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929Clear Linux 337801326395265Min: 63.62 / Avg: 63.64 / Max: 63.66Min: 63.74 / Avg: 63.86 / Max: 63.96Min: 63.76 / Avg: 63.87 / Max: 64Min: 63.83 / Avg: 63.97 / Max: 64.091. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4openSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta400K800K1200K1600K2000KSE +/- 655.01, N = 3SE +/- 309.86, N = 3SE +/- 399.62, N = 3SE +/- 222.29, N = 31854170185720718610301863017
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4openSUSE TumbleweedUbuntu 20.10 20200929Clear Linux 33780Fedora Workstation 33 Beta300K600K900K1200K1500KMin: 1853380 / Avg: 1854170 / Max: 1855470Min: 1856610 / Avg: 1857206.67 / Max: 1857650Min: 1860530 / Avg: 1861030 / Max: 1861820Min: 1862630 / Avg: 1863016.67 / Max: 1863400

Crypto++

Crypto++ is a C++ class library of cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/second, More Is BetterCrypto++ 8.2Test: Keyed AlgorithmsClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 20200929150300450600750SE +/- 0.18, N = 3SE +/- 0.35, N = 3SE +/- 0.05, N = 3SE +/- 0.34, N = 3712.20711.24710.62709.20-fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-g2-g2-g21. (CXX) g++ options: -O3 -pipe -fPIC -pthread
OpenBenchmarking.orgMiB/second, More Is BetterCrypto++ 8.2Test: Keyed AlgorithmsClear Linux 33780Fedora Workstation 33 BetaopenSUSE TumbleweedUbuntu 20.10 20200929130260390520650Min: 711.88 / Avg: 712.2 / Max: 712.52Min: 710.56 / Avg: 711.24 / Max: 711.68Min: 710.55 / Avg: 710.62 / Max: 710.71Min: 708.56 / Avg: 709.2 / Max: 709.731. (CXX) g++ options: -O3 -pipe -fPIC -pthread

AI Benchmark Alpha

AI Benchmark Alpha is a Python library for evaluating artificial intelligence (AI) performance on diverse hardware platforms and relies upon the TensorFlow machine learning library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device Training ScoreClear Linux 33780Ubuntu 20.10 202009293006009001200150015531548

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU-v3-v3 - Model: mobilenet-v3openSUSE Tumbleweed1.20382.40763.61144.81526.019SE +/- 0.10, N = 55.35MIN: 5.02 / MAX: 6.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU-v2-v2 - Model: mobilenet-v2openSUSE Tumbleweed246810SE +/- 0.14, N = 56.03MIN: 5.48 / MAX: 8.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Mobile Neural Network

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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: mobilenet-v1-1.0openSUSE Tumbleweed246810SE +/- 0.018, N = 37.043MIN: 6.91 / MAX: 7.51. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet18Fedora Workstation 33 BetaClear Linux 33780openSUSE TumbleweedUbuntu 20.10 202009293691215SE +/- 0.32, N = 5SE +/- 0.03, N = 3SE +/- 0.28, N = 5SE +/- 0.02, N = 311.3211.3211.8812.01-O2 - MIN: 10.72 / MAX: 12.22-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 11.22 / MAX: 15.55-O3 - MIN: 11.5 / MAX: 13.11-O3 - MIN: 11.92 / MAX: 12.121. (CXX) g++ options: -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet18Fedora Workstation 33 BetaClear Linux 33780openSUSE TumbleweedUbuntu 20.10 202009293691215Min: 10.8 / Avg: 11.32 / Max: 12.11Min: 11.27 / Avg: 11.32 / Max: 11.38Min: 11.58 / Avg: 11.88 / Max: 12.98Min: 11.98 / Avg: 12.01 / Max: 12.031. (CXX) g++ options: -rdynamic -lgomp -lpthread

Mobile Neural Network

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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: MobileNetV2_224Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta1.16892.33783.50674.67565.8445SE +/- 0.080, N = 15SE +/- 0.021, N = 3SE +/- 0.068, N = 9SE +/- 0.138, N = 34.4164.9935.1495.195-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 3.51 / MAX: 5.35MIN: 4.64 / MAX: 5.35MIN: 4.51 / MAX: 6.34MIN: 4.72 / MAX: 5.581. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: MobileNetV2_224Clear Linux 33780openSUSE TumbleweedUbuntu 20.10 20200929Fedora Workstation 33 Beta246810Min: 3.79 / Avg: 4.42 / Max: 4.95Min: 4.96 / Avg: 4.99 / Max: 5.03Min: 4.68 / Avg: 5.15 / Max: 5.38Min: 4.92 / Avg: 5.2 / Max: 5.341. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: resnet-v2-50Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 202009291020304050SE +/- 0.67, N = 15SE +/- 0.26, N = 3SE +/- 2.22, N = 3SE +/- 0.69, N = 934.2740.2641.5345.24-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 28.55 / MAX: 36.93MIN: 34.58 / MAX: 40.92MIN: 36.84 / MAX: 44.93MIN: 39.62 / MAX: 54.481. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: resnet-v2-50Clear Linux 33780openSUSE TumbleweedFedora Workstation 33 BetaUbuntu 20.10 20200929918273645Min: 28.79 / Avg: 34.27 / Max: 36.54Min: 39.81 / Avg: 40.26 / Max: 40.72Min: 37.11 / Avg: 41.53 / Max: 44.01Min: 41.61 / Avg: 45.24 / Max: 47.161. (CXX) g++ options: -O3 -std=c++11 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 24Aug2020Model: Rhodopsin ProteinClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215SE +/- 0.126, N = 3SE +/- 0.259, N = 15SE +/- 0.061, N = 311.8769.3208.193-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O31. (CXX) g++ options: -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 24Aug2020Model: Rhodopsin ProteinClear Linux 33780Fedora Workstation 33 BetaUbuntu 20.10 202009293691215Min: 11.73 / Avg: 11.88 / Max: 12.13Min: 7.9 / Avg: 9.32 / Max: 10.89Min: 8.08 / Avg: 8.19 / Max: 8.291. (CXX) g++ options: -pthread -lm

158 Results Shown

DaCapo Benchmark
PyPerformance
oneDNN
GIMP
Algebraic Multi-Grid Benchmark
Caffe
Node.js Express HTTP Load Test
Caffe
PyPerformance
libavif avifenc:
  10
  8
oneDNN
GIMP
PHPBench
PyPerformance
PlaidML
GIMP
DaCapo Benchmark
Timed HMMer Search
PyPerformance
GIMP
Java Gradle Build
LAMMPS Molecular Dynamics Simulator
PyPerformance:
  django_template
  crypto_pyaes
  pickle_pure_python
PyBench
PyPerformance
Basis Universal
GEGL
Basis Universal
DaCapo Benchmark
PyPerformance:
  go
  regex_compile
Timed PHP Compilation
DaCapo Benchmark
oneDNN
Mobile Neural Network
PyPerformance
oneDNN
CP2K Molecular Dynamics
SVT-AV1
PyPerformance
Basis Universal
SVT-AV1
PyPerformance
SQLite Speedtest
WebP Image Encode
Mobile Neural Network
oneDNN
Facebook RocksDB
FFTE
Timed Linux Kernel Compilation
Facebook RocksDB
Intel Open Image Denoise
LevelDB
Coremark
WebP Image Encode
PlaidML
Darktable
LevelDB
PlaidML
GEGL
oneDNN
GEGL
Basis Universal
PlaidML
Darktable
NCNN
libavif avifenc
NCNN
Facebook RocksDB
Basis Universal
libavif avifenc
oneDNN:
  Matrix Multiply Batch Shapes Transformer - f32 - CPU
  IP Batch All - u8s8f32 - CPU
LevelDB
GEGL
RawTherapee
GEGL
NCNN:
  CPU - shufflenet-v2
  CPU - squeezenet
  CPU - blazeface
Git
TNN
NCNN
DeepSpeech
SVT-VP9
Himeno Benchmark
High Performance Conjugate Gradient
Crafty
PlaidML
asmFish
WebP Image Encode
BRL-CAD
Darktable
NCNN
AI Benchmark Alpha
PlaidML
oneDNN
ASTC Encoder
oneDNN
GROMACS
oneDNN:
  IP Batch All - f32 - CPU
  Deconvolution Batch deconv_1d - bf16bf16bf16 - CPU
  Convolution Batch Shapes Auto - u8s8f32 - CPU
Appleseed
NCNN
ASTC Encoder
OpenVKL
SVT-VP9
oneDNN:
  Convolution Batch Shapes Auto - f32 - CPU
  Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPU
NCNN
Zstd Compression
NCNN
Blender
NCNN
Blender
PlaidML
AI Benchmark Alpha
Darktable
Blender
ASTC Encoder
Appleseed
TNN
Facebook RocksDB
oneDNN:
  IP Batch 1D - bf16bf16bf16 - CPU
  Deconvolution Batch deconv_1d - u8s8f32 - CPU
Zstd Compression
Blender
TensorFlow Lite
Appleseed
oneDNN
Blender
ASTC Encoder
Dolfyn
TensorFlow Lite
oneDNN
TensorFlow Lite:
  SqueezeNet
  Mobilenet Quant
Hierarchical INTegration
Kripke
OpenSSL
NAMD
libjpeg-turbo tjbench
TensorFlow Lite
oneDNN
TensorFlow Lite
Crypto++
AI Benchmark Alpha
NCNN:
  CPU-v3-v3 - mobilenet-v3
  CPU-v2-v2 - mobilenet-v2
Mobile Neural Network
NCNN
Mobile Neural Network:
  MobileNetV2_224
  resnet-v2-50
LAMMPS Molecular Dynamics Simulator

Ubuntu 20.10 20200929

Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.30 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB, Graphics: NVIDIA NV132 11GB, Audio: Realtek ALC1220, Monitor: ASUS MG28U, Network: Intel I219-V + Intel I211

OS: Ubuntu 20.10, Kernel: 5.8.0-19-generic (x86_64), Desktop: GNOME Shell 3.38.0, Display Server: X Server 1.20.8, Display Driver: nouveau 1.0.16, OpenGL: 4.3 Mesa 20.1.8, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160

Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-rn7uax/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-rn7uax/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: intel_cpufreq ondemand - CPU Microcode: 0x5002f01
Java Notes: OpenJDK Runtime Environment (build 11.0.8+10-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.8.6
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 29 September 2020 10:52 by user pts.

Fedora Workstation 33 Beta

Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.30 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB, Graphics: NVIDIA NV132 11GB, Audio: Realtek ALC1220, Monitor: ASUS MG28U, Network: Intel I219-V + Intel I211

OS: Fedora 33, Kernel: 5.8.12-300.fc33.x86_64 (x86_64), Desktop: GNOME Shell 3.38.0, Display Server: X Server + Wayland, OpenGL: 4.3 Mesa 20.2.0, Compiler: GCC 10.2.1 20200826 + Clang 11.0.0, File-System: btrfs, Screen Resolution: 3840x2160

Compiler Notes: --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-gcc-major-version-only --with-isl --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver
Processor Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x5002f01
Java Notes: OpenJDK Runtime Environment 18.9 (build 11.0.9-ea+6)
Python Notes: Python 3.9.0rc2
Security Notes: SELinux + itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 30 September 2020 06:00 by user phoronix.

Clear Linux 33780

Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.30 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB + 31GB UDisk, Graphics: NVIDIA NV132 11GB, Audio: Realtek ALC1220, Monitor: ASUS MG28U, Network: Intel I219-V + Intel I211

OS: Clear Linux OS 33780, Kernel: 5.7.13-975.native (x86_64), Desktop: GNOME Shell 3.38.0, Display Server: X Server 1.20.8, Display Driver: nouveau 1.0.16, OpenGL: 4.3 Mesa 20.2.0-devel, Compiler: GCC 10.2.1 20200928 releases/gcc-10.2.0-302-ga6c47f4ce2 + Clang 10.0.1 + LLVM 10.0.1, File-System: ext4, Screen Resolution: 3840x2160

Environment Notes: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags -Wa,-mbranches-within-32B-boundaries" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries -fvisibility-inlines-hidden -Wl,--enable-new-dtags" MESA_GLSL_CACHE_DISABLE=0 FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,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 -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries" 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-languages=c,c++,fortran,go --enable-ld=default --enable-libstdcxx-pch --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=westmere --with-gcc-major-version-only --with-glibc-version=2.19 --with-gnu-ld --with-isl --with-ppl=yes --with-tune=haswell
Processor Notes: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500012c
Java Notes: OpenJDK Runtime Environment (build 1.8.0-u252-ga-b00)
Python Notes: Python 3.8.6
Security Notes: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 1 October 2020 01:00 by user phoronix.

openSUSE Tumbleweed

Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.30 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB + 31GB UDisk, Graphics: NVIDIA NV132 11GB, Audio: Realtek ALC1220, Monitor: ASUS MG28U, Network: Intel I219-V + Intel I211

OS: openSUSE Tumbleweed 20200929, Kernel: 5.8.10-1-default (x86_64), Desktop: KDE Plasma 5.19.5, Display Server: X Server 1.20.9, Display Driver: nouveau 1.0.16, OpenGL: 4.3 Mesa 20.1.8, Compiler: GCC 10.2.1 20200825 [revision c0746a1beb1ba073c7981eb09f55b3d993b32e5c], File-System: btrfs, Screen Resolution: 3840x2160

Compiler Notes: --build=x86_64-suse-linux --disable-libcc1 --disable-libssp --disable-libstdcxx-pch --disable-libvtv --disable-werror --enable-cet=auto --enable-checking=release --enable-gnu-indirect-function --enable-languages=c,c++,objc,fortran,obj-c++,ada,go,d --enable-libphobos --enable-libstdcxx-allocator=new --enable-link-mutex --enable-linux-futex --enable-multilib --enable-offload-targets=nvptx-none=/usr/nvptx-none,amdgcn-amdhsa=/usr/amdgcn-amdhsa, --enable-plugin --enable-ssp --enable-version-specific-runtime-libs --host=x86_64-suse-linux --mandir=/usr/share/man --with-arch-32=x86-64 --with-build-config=bootstrap-lto-lean --with-gcc-major-version-only --with-slibdir=/lib64 --with-tune=generic --without-cuda-driver --without-system-libunwind
Processor Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x5002f01
Java Notes: OpenJDK Runtime Environment (build 11.0.8+10-suse-1.3-x8664)
Python Notes: Python 2.7.18 + Python 3.8.5
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 1 October 2020 17:44 by user .