Raspberry Pi 400 vs. Orange Pi 5 Benchmarks

Orange Pi 5 benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2303239-NE-ORANGEPI565
Jump To Table - Results

View

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

Limit displaying results to tests within:

Audio Encoding 2 Tests
AV1 3 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 18 Tests
Compression Tests 2 Tests
CPU Massive 18 Tests
Creator Workloads 17 Tests
Cryptography 2 Tests
Database Test Suite 4 Tests
Encoding 8 Tests
Fortran Tests 4 Tests
HPC - High Performance Computing 10 Tests
Imaging 6 Tests
Common Kernel Benchmarks 4 Tests
Machine Learning 5 Tests
Molecular Dynamics 3 Tests
Multi-Core 15 Tests
OpenMPI Tests 3 Tests
Programmer / Developer System Benchmarks 7 Tests
Python 3 Tests
Scientific Computing 4 Tests
Server 6 Tests
Server CPU Tests 8 Tests
Single-Threaded 6 Tests
Speech 2 Tests
Telephony 2 Tests
Video Encoding 6 Tests
Common Workstation Benchmarks 2 Tests

Statistics

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

Graph Settings

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

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Raspberry Pi 400
March 13 2023
  4 Days, 13 Hours, 4 Minutes
Orange Pi 5
March 20 2023
  2 Days, 20 Hours, 58 Minutes
Invert Hiding All Results Option
  3 Days, 17 Hours, 1 Minute
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):


Raspberry Pi 400 vs. Orange Pi 5 BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsMonitorOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionOpenGLOpenCLRaspberry Pi 400Orange Pi 5ARMv8 Cortex-A72 @ 1.80GHz (4 Cores)BCM2835 Raspberry Pi 400 Rev 1.0Broadcom BCM27114096MB128GB SF128vc4drmfbDELL P2210HDebian 115.15.84-v8+ (aarch64)LXDEX Server 1.20.11GCC 10.2.1 20210110ext41920x1080Rockchip ARMv8 Cortex-A76 @ 1.80GHz (4 Cores / 8 Threads)Orange Pi 58GB128GB SR128llvmpipeDELL S2409WUbuntu 22.045.10.110-rockchip-rk3588 (aarch64)Xfce 4.16X Server 1.21.1.34.5 Mesa 22.2.5 (LLVM 15.0.6 128 bits)OpenCL 2.1 v1.g6p0-01eac0.efb75e2978d783a80fe78be1bfb0efc1GCC 11.3.0OpenBenchmarking.orgKernel Details- Raspberry Pi 400: snd_bcm2835.enable_compat_alsa=0 snd_bcm2835.enable_hdmi=1Compiler Details- Raspberry Pi 400: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - Orange Pi 5: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v Processor Details- Scaling Governor: cpufreq-dt ondemandPython Details- Raspberry Pi 400: Python 3.9.2- Orange Pi 5: Python 3.10.6Security Details- Raspberry Pi 400: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable + srbds: Not affected + tsx_async_abort: Not affected - Orange Pi 5: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable: Unprivileged eBPF enabled + srbds: Not affected + tsx_async_abort: Not affected Environment Details- Orange Pi 5: GST_DEBUG_NO_COLOR=1

Raspberry Pi 400 vs. Orange Pi 5 ComparisonPhoronix Test SuiteBaseline+848.6%+848.6%+1697.2%+1697.2%+2545.8%+2545.8%755.2%721.9%561.6%510%510%492.9%490%417.7%401.1%391.2%382.3%374.6%369.6%360.7%354.7%350.6%349.8%335%330.6%3394.3%3308.4%281.1%278.6%277.5%273.6%270.2%261%261%258.3%256.6%251.5%250.3%250.2%247.4%243.9%243.8%239.1%239%237.3%227.9%227.1%219.4%216%211.3%210.8%208.3%208.1%201.6%1953%1950.1%1902.7%196.7%194.3%189.3%188.6%187.1%186.9%186.2%177.9%177.4%174.8%173.5%170.9%169.4%168.6%167.6%167%166.1%164.7%163.4%162.6%160.8%160.5%160.5%160.1%159.3%159.2%158.3%158%157.8%157.4%156.4%156.3%148.9%147.2%143.7%142.7%134.7%134.5%134.2%133.5%133.3%131.8%130.5%130.3%130.3%130.2%130%129.9%129.2%128.4%127.6%127.6%126.3%126.3%124%123%123%122.8%122.6%122.5%122.2%122%116.7%116.2%114.3%114.3%114.2%113.7%113.4%109.7%106.5%106.2%104.8%104.6%104.5%101.9%1899.4%1876.1%1843.6%98.8%95.7%94.4%92.7%90.4%85.9%85%83.5%82.8%82.7%82%78%77.9%75.6%73.3%72.8%72.8%71.2%71.1%68.2%65.1%60.1%56.5%56.4%53.3%53.1%51.8%48.5%36.4%36.3%36.2%36.1%34.4%34.3%26.3%26.3%12.7%3 - Compression Speed3, Long Mode - Compression SpeedCPU - resnet18SHA2568 - Compression Speed8, Long Mode - Compression Speed12 - Compression SpeedCPU - vgg16CPU-v2-v2 - mobilenet-v2CPU - resnet50CPU - mnasnetCPU - yolov4-tinyCPU - mobilenet1:100CPU-v3-v3 - mobilenet-v31:10CPU - squeezenet_ssdCPU - efficientnet-b0CPU - regnety_400mAES-256-GCMAES-128-GCMC.u.1.0.3.s.i.i.C.L.9ResizingCPU - FastestDetdjango_templateOpenMP - BM1OpenMP - BM11 - 100 - Read Only1 - 100 - Read Only - Average LatencyRotate1 - 250 - Read Only1 - 250 - Read Only - Average LatencypathlibTime To CompileTimed Time - Size 1,000Speed 0 - Bosphorus 4KTime To Compiledefconfig1 - 500 - Read Only1 - 500 - Read Only - Average LatencyUpdate RandCPU - googlenetSpeed 10 Realtime - Bosphorus 1080p19 - Compression SpeedSpeed 9 Realtime - Bosphorus 1080pBosphorus 1080p - FasterSpeed 8 Realtime - Bosphorus 1080p2010004000Layered HalfspacechaosP.P.SCPU - shufflenet-v2regex_compileRhodopsin ProteinAllL.E.HSummer Nature 4KCPU - vision_transformer61PNG - 80PNG - 90S.N.1CPU - alexnetW.l.HH.HT.T.S.SfloatChaCha20Tomographic ModelSpeed 5 - Bosphorus 4K1 - 50 - Read OnlyC.F.D1 - 50 - Read Only - Average LatencyP.P.BJPEG - 90CPU - DenseNetJPEG - 80Mount St. Helens12 - D.S8 - D.S8, Long Mode - D.Scrypto_pyaesV.P.M6, LosslessQ.1.LCPU - SqueezeNet v1.1ChaCha20-Poly1305R.R.W.Rjson_loads19, Long Mode - Compression Speedlibx264 - Video On Demandlibx264 - Video On Demand19, Long Mode - D.Slibx264 - Platformlibx264 - Platform3 - D.S19 - D.Slibx264 - Uploadlibx264 - Uploadlibx264 - Livelibx264 - Livelibx265 - Uploadlibx265 - Video On Demandlibx265 - Platformlibx265 - Video On Demandlibx265 - Platformlibx265 - Upload3, Long Mode - D.SSwirlCPU - MobileNet v2mobilenet-v1-1.0EnhancedSharpenCoreMark Size 666 - I.P.SWPA PSKRSA4096nbodyNoise-GaussianHWB Color Spacelibx265 - Livelibx265 - LiveRand Read10, Lossless100200500Church FacadeWAV To MP3T.F.A.T.TSHA512CPU - SqueezeNet v2RSA4096A.C.PRead While WritingDefaultQuality 100WAV To FLACSqueezeNetV1.0LionCPU - blazefacebcryptBlowfishsqueezenetv1.1Speed 8 Realtime - Bosphorus 4KQ.1.H.Cresnet-v2-50MD5HMAC-SHA512MobileNetV2_224Speed 10 Realtime - Bosphorus 4KmobilenetV3Speed 9 Realtime - Bosphorus 4Kinception-v31 - 500 - Read Write - Average Latency1 - 50 - Read Write - Average Latency1 - 500 - Read Write1 - 50 - Read Write1 - 250 - Read Write1 - 250 - Read Write - Average Latency1 - 100 - Read Write1 - 100 - Read Write - Average LatencynasnetZstd CompressionZstd CompressionNCNNOpenSSLZstd CompressionZstd CompressionZstd CompressionNCNNNCNNNCNNNCNNNCNNNCNNMemcachedNCNNMemcachedNCNNNCNNNCNNOpenSSLOpenSSLXZ CompressionGraphicsMagickNCNNNumpy BenchmarkPyPerformanceminiBUDEminiBUDEPostgreSQLPostgreSQLGraphicsMagickPostgreSQLPostgreSQLPyPerformanceTimed MPlayer CompilationSQLite SpeedtestVP9 libvpx EncodingTimed FFmpeg CompilationTimed Linux Kernel CompilationPostgreSQLPostgreSQLRocksDBNCNNAOM AV1Zstd CompressionAOM AV1VVenCAOM AV1nginxnginxnginxSPECFEM3DPyPerformanceHimeno BenchmarkNCNNPyPerformanceLAMMPS Molecular Dynamics SimulatorJPEG XL Decoding libjxlCloverLeafdav1dNCNNlibavif avifencJPEG XL Decoding libjxlJPEG XL libjxlJPEG XL libjxldav1dNCNNSPECFEM3DSPECFEM3DeSpeak-NG Speech EnginePyPerformanceOpenSSLSPECFEM3DVP9 libvpx EncodingPostgreSQLDolfynPostgreSQLLibRawJPEG XL libjxlTNNJPEG XL libjxlSPECFEM3DZstd CompressionZstd CompressionZstd CompressionPyPerformanceBRL-CADlibavif avifencWebP Image EncodeTNNOpenSSLRocksDBPyPerformanceZstd CompressionFFmpegFFmpegZstd CompressionFFmpegFFmpegZstd CompressionZstd CompressionFFmpegFFmpegFFmpegFFmpegFFmpegFFmpegFFmpegFFmpegFFmpegFFmpegZstd CompressionGraphicsMagickTNNMobile Neural NetworkGraphicsMagickGraphicsMagickCoremarkJohn The RipperOpenSSLPyPerformanceGraphicsMagickGraphicsMagickFFmpegFFmpegRocksDBlibavif avifencnginxnginxnginxGoogle DracoLAME MP3 EncodingRNNoisePyBenchOpenSSLTNNOpenSSLTSCPRocksDBWebP Image EncodeWebP Image EncodeFLAC Audio EncodingMobile Neural NetworkGoogle DracoNCNNJohn The RipperJohn The RipperMobile Neural NetworkAOM AV1WebP Image EncodeMobile Neural NetworkJohn The RipperJohn The RipperMobile Neural NetworkAOM AV1Mobile Neural NetworkAOM AV1Mobile Neural NetworkPostgreSQLPostgreSQLPostgreSQLPostgreSQLPostgreSQLPostgreSQLPostgreSQLPostgreSQLMobile Neural NetworkRaspberry Pi 400Orange Pi 5

Raspberry Pi 400 vs. Orange Pi 5 Benchmarkscompress-zstd: 3 - Compression Speedcompress-zstd: 3, Long Mode - Compression Speedncnn: CPU - resnet18openssl: SHA256compress-zstd: 8 - Compression Speedcompress-zstd: 8, Long Mode - Compression Speedcompress-zstd: 12 - Compression Speedncnn: CPU - vgg16ncnn: CPU - resnet50ncnn: CPU - yolov4-tinyncnn: CPU - mobilenetmemcached: 1:100memcached: 1:10ncnn: CPU - squeezenet_ssdncnn: CPU - efficientnet-b0ncnn: CPU - regnety_400mopenssl: AES-256-GCMopenssl: AES-128-GCMcompress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9graphics-magick: Resizingncnn: CPU - FastestDetnumpy: pyperformance: django_templateminibude: OpenMP - BM1minibude: OpenMP - BM1graphics-magick: Rotatepgbench: 1 - 250 - Read Onlypgbench: 1 - 250 - Read Only - Average Latencypyperformance: pathlibbuild-mplayer: Time To Compilevpxenc: Speed 0 - Bosphorus 4Kbuild-ffmpeg: Time To Compilebuild-linux-kernel: defconfigpgbench: 1 - 500 - Read Onlypgbench: 1 - 500 - Read Only - Average Latencyrocksdb: Update Randncnn: CPU - googlenetaom-av1: Speed 10 Realtime - Bosphorus 1080pcompress-zstd: 19 - Compression Speedaom-av1: Speed 9 Realtime - Bosphorus 1080pvvenc: Bosphorus 1080p - Fasternginx: 20nginx: 1000nginx: 4000specfem3d: Layered Halfspacepyperformance: chaoshimeno: Poisson Pressure Solverpyperformance: regex_compilelammps: Rhodopsin Proteinjpegxl-decode: Allcloverleaf: Lagrangian-Eulerian Hydrodynamicsdav1d: Summer Nature 4Kncnn: CPU - vision_transformeravifenc: 6jpegxl-decode: 1jpegxl: PNG - 80jpegxl: PNG - 90dav1d: Summer Nature 1080pncnn: CPU - alexnetspecfem3d: Water-layered Halfspacespecfem3d: Homogeneous Halfspacepyperformance: floatopenssl: ChaCha20specfem3d: Tomographic Modelvpxenc: Speed 5 - Bosphorus 4Kdolfyn: Computational Fluid Dynamicslibraw: Post-Processing Benchmarkjpegxl: JPEG - 90tnn: CPU - DenseNetjpegxl: JPEG - 80specfem3d: Mount St. Helenscompress-zstd: 12 - Decompression Speedcompress-zstd: 8 - Decompression Speedcompress-zstd: 8, Long Mode - Decompression Speedpyperformance: crypto_pyaesbrl-cad: VGR Performance Metricavifenc: 6, Losslesswebp: Quality 100, Losslesstnn: CPU - SqueezeNet v1.1openssl: ChaCha20-Poly1305rocksdb: Read Rand Write Randpyperformance: json_loadscompress-zstd: 19, Long Mode - Compression Speedffmpeg: libx264 - Video On Demandffmpeg: libx264 - Video On Demandcompress-zstd: 19, Long Mode - Decompression Speedffmpeg: libx264 - Platformffmpeg: libx264 - Platformcompress-zstd: 3 - Decompression Speedcompress-zstd: 19 - Decompression Speedffmpeg: libx264 - Uploadffmpeg: libx264 - Uploadffmpeg: libx264 - Liveffmpeg: libx264 - Liveffmpeg: libx265 - Uploadffmpeg: libx265 - Video On Demandffmpeg: libx265 - Platformffmpeg: libx265 - Video On Demandffmpeg: libx265 - Platformffmpeg: libx265 - Uploadcompress-zstd: 3, Long Mode - Decompression Speedgraphics-magick: Swirltnn: CPU - MobileNet v2graphics-magick: Enhancedgraphics-magick: Sharpencoremark: CoreMark Size 666 - Iterations Per Secondjohn-the-ripper: WPA PSKopenssl: RSA4096pyperformance: nbodygraphics-magick: Noise-Gaussiangraphics-magick: HWB Color Spaceffmpeg: libx265 - Liveffmpeg: libx265 - Liverocksdb: Rand Readavifenc: 10, Losslessnginx: 100nginx: 200nginx: 500draco: Church Facadeencode-mp3: WAV To MP3pybench: Total For Average Test Timesopenssl: SHA512tnn: CPU - SqueezeNet v2openssl: RSA4096tscp: AI Chess Performancewebp: Defaultwebp: Quality 100encode-flac: WAV To FLACdraco: Lionncnn: CPU - blazefacejohn-the-ripper: bcryptjohn-the-ripper: Blowfishaom-av1: Speed 8 Realtime - Bosphorus 4Kwebp: Quality 100, Highest Compressionjohn-the-ripper: MD5john-the-ripper: HMAC-SHA512aom-av1: Speed 10 Realtime - Bosphorus 4Kaom-av1: Speed 9 Realtime - Bosphorus 4Kpgbench: 1 - 50 - Read Write - Average Latencypgbench: 1 - 50 - Read Writepgbench: 1 - 250 - Read Writepgbench: 1 - 250 - Read Write - Average Latencypgbench: 1 - 100 - Read Writepgbench: 1 - 100 - Read Write - Average Latencyrocksdb: Read While Writingncnn: CPU - mnasnetncnn: CPU - shufflenet-v2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU-v2-v2 - mobilenet-v2mnn: inception-v3mnn: mobilenet-v1-1.0mnn: MobileNetV2_224mnn: SqueezeNetV1.0mnn: resnet-v2-50mnn: squeezenetv1.1mnn: mobilenetV3mnn: nasnetsqlite-speedtest: Timed Time - Size 1,000pgbench: 1 - 500 - Read Write - Average Latencypgbench: 1 - 500 - Read Writepgbench: 1 - 100 - Read Only - Average Latencypgbench: 1 - 100 - Read Onlypgbench: 1 - 50 - Read Only - Average Latencypgbench: 1 - 50 - Read Onlyrnnoise: espeak: Text-To-Speech Synthesisaom-av1: Speed 8 Realtime - Bosphorus 1080pRaspberry Pi 400Orange Pi 528.827.8146.0944486444014.014.05.00619.53292.97247.31193.8475136.5978927.15156.5076.3972.82237227850281296987396.0735641.7952.182580.40510.1281011701214.696107633.1900.23852.9104866.4621659430.14038090122.7414.201.8614.181.148477.26415.23407.142807.202416241465796.4856666690.59018.341396.0613.512853.1093.7265.951.211.1861.7385.662669.9743068281441.58097184747816295087671139.7213754470.8689.5816.571.1918882.9901.221126.883080076255.1300.3302.542414400107.4800.29899.071129040129016470987.41.64924.758.19236.9924.238.20314.0224.01122.8766472642.2534.63145.813571.591.391.395451.185447.520.71331.0501092.702212139918.9568841672115.5453311464.331165.298798049597740538.538503.90500.94465.052707023.8683672515639207230.9708930.34365023.462.4493.4681754210.45157915813.571.103751821583334.674.63140.2913572471010.643350285.84126987152.7226.1247.0261.64356.04457.58733.04061.657284.61031.16610.86779.823469.7233300.8941525.559180132.0222477258.261154.54111.75246.3228.522.08271386598385.483.029.5119.6759.6452.1141.28346191.70355623.6734.7917.5616.9182894768289587671153103.92621211.07194.9569.71.46236.560355595904.19730.8184.1190.78251.5681442.791544109.21412166638.8444.215.7843.713.5379798.148512.548153.60946.2933873631582304.0591062331.69352.48502.3337.471038.1634.27216.123.263.17165.2032.081003.470066713544.7075018101824250201837437.5227245262.2434.55016.973.077324.8673.14439.539479170653.8747.3747.81743495245.7930.68383.971301358066338431537.73.78401.5618.86545.3401.8218.85719.7511.7493.265.1278.3764.441594.6348456193.103.102446.7571329192447.4719754761.58735.4111504.170454585500.3074203573246.5216643018.87569.611222226719.08310074.789899.329038.601361512.1971906981851067124.25216520.88009006.324.4452.49699886.03272927326.111.856004933770007.167.03102.926486332752.775442226.36649327110.939.0510.3412.30239.72826.63921.12434.666172.36418.2047.09970.801136.6302420.5412071.559645350.7806444229.97258.67435.44OpenBenchmarking.org

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3 - Compression SpeedOrange Pi 5Raspberry Pi 40050100150200250SE +/- 0.84, N = 3SE +/- 0.13, N = 3246.328.81. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3, Long Mode - Compression SpeedOrange Pi 5Raspberry Pi 40050100150200250SE +/- 1.33, N = 3SE +/- 0.24, N = 3228.527.81. (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 20220729Target: CPU - Model: resnet18Orange Pi 5Raspberry Pi 400306090120150SE +/- 0.71, N = 3SE +/- 0.26, N = 322.08146.09MIN: 18.01 / MAX: 66.6-pthread - MIN: 144.46 / MAX: 177.051. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: SHA256Orange Pi 5Raspberry Pi 400600M1200M1800M2400M3000MSE +/- 5122655.63, N = 3SE +/- 71645.67, N = 327138659834448644401. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8 - Compression SpeedOrange Pi 5Raspberry Pi 40020406080100SE +/- 0.90, N = 3SE +/- 0.03, N = 385.414.01. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8, Long Mode - Compression SpeedOrange Pi 5Raspberry Pi 40020406080100SE +/- 1.04, N = 3SE +/- 0.07, N = 383.014.01. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 12 - Compression SpeedOrange Pi 5Raspberry Pi 400714212835SE +/- 0.28, N = 6SE +/- 0.00, N = 329.505.001. (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 20220729Target: CPU - Model: vgg16Orange Pi 5Raspberry Pi 400130260390520650SE +/- 0.17, N = 3SE +/- 0.80, N = 3119.67619.53MIN: 110.53 / MAX: 164.37-pthread - MIN: 605 / MAX: 680.751. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: resnet50Orange Pi 5Raspberry Pi 40060120180240300SE +/- 0.76, N = 3SE +/- 0.52, N = 359.64292.97MIN: 49.04 / MAX: 119.91-pthread - MIN: 289.07 / MAX: 314.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: yolov4-tinyOrange Pi 5Raspberry Pi 40050100150200250SE +/- 0.38, N = 3SE +/- 1.06, N = 352.11247.31MIN: 44 / MAX: 114.26-pthread - MIN: 241.09 / MAX: 265.51. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: mobilenetOrange Pi 5Raspberry Pi 4004080120160200SE +/- 0.04, N = 3SE +/- 1.08, N = 341.28193.84MIN: 28.32 / MAX: 88.5-pthread - MIN: 189.61 / MAX: 210.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.18Set To Get Ratio: 1:100Orange Pi 5Raspberry Pi 40070K140K210K280K350KSE +/- 2298.49, N = 3SE +/- 741.56, N = 3346191.7075136.591. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.18Set To Get Ratio: 1:10Orange Pi 5Raspberry Pi 40080K160K240K320K400KSE +/- 2142.19, N = 15SE +/- 819.29, N = 5355623.6778927.151. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

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 20220729Target: CPU - Model: squeezenet_ssdOrange Pi 5Raspberry Pi 400306090120150SE +/- 0.24, N = 3SE +/- 0.39, N = 334.79156.50MIN: 24.81 / MAX: 84.31-pthread - MIN: 155 / MAX: 168.751. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: efficientnet-b0Orange Pi 5Raspberry Pi 40020406080100SE +/- 0.43, N = 3SE +/- 0.53, N = 317.5676.39MIN: 11.41 / MAX: 69.99-pthread - MIN: 74.66 / MAX: 87.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: regnety_400mOrange Pi 5Raspberry Pi 4001632486480SE +/- 0.33, N = 3SE +/- 0.23, N = 316.9172.82MIN: 12.06 / MAX: 73.54-pthread - MIN: 71.69 / MAX: 82.511. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: AES-256-GCMOrange Pi 5Raspberry Pi 4002000M4000M6000M8000M10000MSE +/- 85355221.83, N = 5SE +/- 76741.38, N = 382894768282372278501. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: AES-128-GCMOrange Pi 5Raspberry Pi 4002000M4000M6000M8000M10000MSE +/- 101628515.09, N = 12SE +/- 202998.46, N = 395876711532812969871. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

XZ Compression

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

OpenBenchmarking.orgSeconds, Fewer Is BetterXZ Compression 5.2.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9Orange Pi 5Raspberry Pi 40090180270360450SE +/- 1.25, N = 3SE +/- 2.33, N = 3103.93396.07-pthread1. (CC) gcc options: -fvisibility=hidden -O2

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: ResizingOrange Pi 5Raspberry Pi 40050100150200250SE +/- 1.86, N = 321256-pthread -lwebp -lwebpmux -lSM -lICE1. (CC) gcc options: -fopenmp -O2 -ljbig -ltiff -lfreetype -ljpeg -lXext -lX11 -llzma -lz -lm -lpthread

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 20220729Target: CPU - Model: FastestDetOrange Pi 5Raspberry Pi 4001020304050SE +/- 0.26, N = 3SE +/- 0.15, N = 311.0741.79MIN: 4.94 / MAX: 64.99-pthread - MIN: 41.15 / MAX: 59.321. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkOrange Pi 5Raspberry Pi 4004080120160200SE +/- 0.79, N = 3SE +/- 0.03, N = 3194.9552.18

PyPerformance

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

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: django_templateOrange Pi 5Raspberry Pi 40060120180240300SE +/- 0.07, N = 369.7258.0

miniBUDE

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

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Orange Pi 5Raspberry Pi 4000.3290.6580.9871.3161.645SE +/- 0.006, N = 3SE +/- 0.000, N = 31.4620.4051. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Orange Pi 5Raspberry Pi 400816243240SE +/- 0.16, N = 3SE +/- 0.00, N = 336.5610.131. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -mcpu=native -lm

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: RotateOrange Pi 5Raspberry Pi 40080160240320400SE +/- 0.33, N = 3SE +/- 0.67, N = 3355101-pthread -lwebp -lwebpmux -lSM -lICE1. (CC) gcc options: -fopenmp -O2 -ljbig -ltiff -lfreetype -ljpeg -lXext -lX11 -llzma -lz -lm -lpthread

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read OnlyOrange Pi 5Raspberry Pi 40013K26K39K52K65KSE +/- 667.25, N = 4SE +/- 104.90, N = 35959017012-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Only - Average LatencyOrange Pi 5Raspberry Pi 40048121620SE +/- 0.048, N = 4SE +/- 0.091, N = 34.19714.696-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -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: pathlibOrange Pi 5Raspberry Pi 40020406080100SE +/- 0.03, N = 330.8107.0

Timed MPlayer Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MPlayer Compilation 1.5Time To CompileOrange Pi 5Raspberry Pi 400140280420560700SE +/- 1.80, N = 3SE +/- 0.11, N = 3184.12633.19

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 0 - Input: Bosphorus 4KOrange Pi 5Raspberry Pi 4000.17550.3510.52650.7020.8775SE +/- 0.00, N = 3SE +/- 0.00, N = 30.780.231. (CXX) g++ options: -lm -lpthread -march=armv8-a -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

Timed FFmpeg Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 6.0Time To CompileOrange Pi 5Raspberry Pi 4002004006008001000SE +/- 2.32, N = 7SE +/- 0.70, N = 3251.57852.91

Timed Linux Kernel Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: defconfigOrange Pi 5Raspberry Pi 40010002000300040005000SE +/- 3.56, N = 3SE +/- 16.95, N = 31442.794866.46

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read OnlyOrange Pi 5Raspberry Pi 40012K24K36K48K60KSE +/- 845.64, N = 12SE +/- 190.50, N = 35441016594-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Only - Average LatencyOrange Pi 5Raspberry Pi 400714212835SE +/- 0.145, N = 12SE +/- 0.347, N = 39.21430.140-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

RocksDB

This is a benchmark of Meta/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 BetterRocksDB 7.9.2Test: Update RandomOrange Pi 5Raspberry Pi 40030K60K90K120K150KSE +/- 1196.67, N = 5SE +/- 405.92, N = 3121666380901. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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 20220729Target: CPU - Model: googlenetOrange Pi 5Raspberry Pi 400306090120150SE +/- 0.53, N = 3SE +/- 0.22, N = 338.84122.74MIN: 27.86 / MAX: 87.11-pthread - MIN: 121.57 / MAX: 139.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080pOrange Pi 5Raspberry Pi 4001020304050SE +/- 0.62, N = 15SE +/- 0.06, N = 344.2114.20-lpthread1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Compression SpeedOrange Pi 5Raspberry Pi 4001.30052.6013.90155.2026.5025SE +/- 0.05, N = 15SE +/- 0.00, N = 35.781.861. (CC) gcc options: -O3 -pthread -lz -llzma

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080pOrange Pi 5Raspberry Pi 4001020304050SE +/- 0.61, N = 15SE +/- 0.20, N = 343.7114.18-lpthread1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under The Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 1080p - Video Preset: FasterOrange Pi 5Raspberry Pi 4000.79581.59162.38743.18323.979SE +/- 0.029, N = 8SE +/- 0.005, N = 33.5371.148-lpthread1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

nginx

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

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 20Orange Pi 5Raspberry Pi 4002K4K6K8K10KSE +/- 83.72, N = 15SE +/- 3.93, N = 159798.14477.261. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 1000Orange Pi 5Raspberry Pi 4002K4K6K8K10KSE +/- 94.01, N = 5SE +/- 5.72, N = 38512.54415.231. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 4000Orange Pi 5Raspberry Pi 4002K4K6K8K10KSE +/- 68.29, N = 3SE +/- 0.60, N = 38153.60407.141. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Layered HalfspaceOrange Pi 5Raspberry Pi 4006001200180024003000SE +/- 4.22, N = 3SE +/- 5.92, N = 3946.292807.20-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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: chaosOrange Pi 5Raspberry Pi 400100200300400500SE +/- 0.00, N = 3158465

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 SolverOrange Pi 5Raspberry Pi 4005001000150020002500SE +/- 9.43, N = 3SE +/- 6.58, N = 32304.06796.491. (CC) gcc options: -O3

PyPerformance

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

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: regex_compileOrange Pi 5Raspberry Pi 400140280420560700SE +/- 0.33, N = 3SE +/- 0.58, N = 3233669

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 23Jun2022Model: Rhodopsin ProteinOrange Pi 5Raspberry Pi 4000.38090.76181.14271.52361.9045SE +/- 0.011, N = 15SE +/- 0.002, N = 31.6930.590-pthread1. (CXX) g++ options: -O3 -ldl

JPEG XL Decoding libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: AllOrange Pi 5Raspberry Pi 4001224364860SE +/- 0.45, N = 8SE +/- 0.01, N = 352.4818.34

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm.in input file (Problem 5). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsOrange Pi 5Raspberry Pi 40030060090012001500SE +/- 0.23, N = 3SE +/- 0.84, N = 3502.331396.061. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.1Video Input: Summer Nature 4KOrange Pi 5Raspberry Pi 400918273645SE +/- 0.35, N = 3SE +/- 0.02, N = 337.4713.511. (CC) gcc options: -pthread -lm

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 20220729Target: CPU - Model: vision_transformerOrange Pi 5Raspberry Pi 4006001200180024003000SE +/- 0.78, N = 3SE +/- 4.73, N = 31038.162853.10MIN: 966.15 / MAX: 1128.84-pthread - MIN: 2776.33 / MAX: 2992.571. (CXX) g++ options: -O3 -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 0.11Encoder Speed: 6Orange Pi 5Raspberry Pi 40020406080100SE +/- 0.28, N = 15SE +/- 0.25, N = 334.2793.731. (CXX) g++ options: -O3 -fPIC -lm

JPEG XL Decoding libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: 1Orange Pi 5Raspberry Pi 40048121620SE +/- 0.22, N = 3SE +/- 0.01, N = 316.125.95

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 80Orange Pi 5Raspberry Pi 4000.73351.4672.20052.9343.6675SE +/- 0.01, N = 3SE +/- 0.00, N = 33.261.21-pthread1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 90Orange Pi 5Raspberry Pi 4000.71331.42662.13992.85323.5665SE +/- 0.01, N = 3SE +/- 0.00, N = 33.171.18-pthread1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.1Video Input: Summer Nature 1080pOrange Pi 5Raspberry Pi 4004080120160200SE +/- 0.78, N = 3SE +/- 0.18, N = 3165.2061.731. (CC) gcc options: -pthread -lm

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 20220729Target: CPU - Model: alexnetOrange Pi 5Raspberry Pi 40020406080100SE +/- 0.41, N = 3SE +/- 0.53, N = 332.0885.66MIN: 24.52 / MAX: 83.92-pthread - MIN: 83.58 / MAX: 98.841. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Water-layered HalfspaceOrange Pi 5Raspberry Pi 4006001200180024003000SE +/- 11.11, N = 4SE +/- 32.56, N = 31003.472669.97-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Homogeneous HalfspaceOrange Pi 5Raspberry Pi 40030060090012001500SE +/- 4.54, N = 3SE +/- 4.48, N = 3544.711441.58-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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: floatOrange Pi 5Raspberry Pi 400100200300400500SE +/- 0.33, N = 3SE +/- 1.73, N = 3182478

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: ChaCha20Orange Pi 5Raspberry Pi 400900M1800M2700M3600M4500MSE +/- 17103865.91, N = 3SE +/- 376917.52, N = 3425020183716295087671. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Tomographic ModelOrange Pi 5Raspberry Pi 4002004006008001000SE +/- 3.65, N = 3SE +/- 2.27, N = 3437.521139.72-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 5 - Input: Bosphorus 4KOrange Pi 5Raspberry Pi 4000.5041.0081.5122.0162.52SE +/- 0.01, N = 3SE +/- 0.00, N = 32.240.861. (CXX) g++ options: -lm -lpthread -march=armv8-a -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

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 DynamicsOrange Pi 5Raspberry Pi 40020406080100SE +/- 0.07, N = 3SE +/- 0.59, N = 334.5589.58

LibRaw

LibRaw is a RAW image decoder for digital camera photos. This test profile runs LibRaw's post-processing benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing BenchmarkOrange Pi 5Raspberry Pi 40048121620SE +/- 0.15, N = 3SE +/- 0.01, N = 316.976.571. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 90Orange Pi 5Raspberry Pi 4000.69081.38162.07242.76323.454SE +/- 0.00, N = 3SE +/- 0.00, N = 33.071.19-pthread1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

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 0.3Target: CPU - Model: DenseNetOrange Pi 5Raspberry Pi 4004K8K12K16K20KSE +/- 8.00, N = 3SE +/- 32.53, N = 37324.8718882.99MIN: 7093.42 / MAX: 7544.4MIN: 18781.08 / MAX: 19205.871. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 80Orange Pi 5Raspberry Pi 4000.70651.4132.11952.8263.5325SE +/- 0.01, N = 3SE +/- 0.00, N = 33.141.22-pthread1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Mount St. HelensOrange Pi 5Raspberry Pi 4002004006008001000SE +/- 6.15, N = 3SE +/- 4.41, N = 3439.541126.88-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 12 - Decompression SpeedOrange Pi 5Raspberry Pi 400140280420560700SE +/- 2.39, N = 6SE +/- 0.69, N = 3653.8255.11. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8 - Decompression SpeedOrange Pi 5Raspberry Pi 400160320480640800SE +/- 0.95, N = 3SE +/- 0.86, N = 3747.3300.31. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8, Long Mode - Decompression SpeedOrange Pi 5Raspberry Pi 400160320480640800SE +/- 0.41, N = 3SE +/- 1.27, N = 3747.8302.51. (CC) gcc options: -O3 -pthread -lz -llzma

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: crypto_pyaesOrange Pi 5Raspberry Pi 40090180270360450SE +/- 0.00, N = 3174424

BRL-CAD

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

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.34VGR Performance MetricOrange Pi 5Raspberry Pi 4007K14K21K28K35K3495214400-pthread1. (CXX) g++ options: -std=c++14 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lregex_brl -lz_brl -lnetpbm -ldl -lm -ltk8.6

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6, LosslessOrange Pi 5Raspberry Pi 40020406080100SE +/- 0.28, N = 3SE +/- 0.22, N = 345.79107.481. (CXX) g++ options: -O3 -fPIC -lm

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.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, LosslessOrange Pi 5Raspberry Pi 4000.1530.3060.4590.6120.765SE +/- 0.00, N = 3SE +/- 0.00, N = 30.680.29-lpng16 -lgif -ltiff -pthread1. (CC) gcc options: -fvisibility=hidden -O2 -lm

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 0.3Target: CPU - Model: SqueezeNet v1.1Orange Pi 5Raspberry Pi 4002004006008001000SE +/- 0.22, N = 3SE +/- 6.59, N = 3383.97899.07MIN: 383.04 / MAX: 384.46MIN: 883.84 / MAX: 912.191. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: ChaCha20-Poly1305Orange Pi 5Raspberry Pi 400600M1200M1800M2400M3000MSE +/- 34736811.53, N = 3SE +/- 1146742.49, N = 3301358066312904012901. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

RocksDB

This is a benchmark of Meta/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 BetterRocksDB 7.9.2Test: Read Random Write RandomOrange Pi 5Raspberry Pi 40080K160K240K320K400KSE +/- 3825.62, N = 6SE +/- 1408.45, N = 33843151647091. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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_loadsOrange Pi 5Raspberry Pi 40020406080100SE +/- 0.00, N = 3SE +/- 0.12, N = 337.787.4

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Compression SpeedOrange Pi 5Raspberry Pi 4000.85051.7012.55153.4024.2525SE +/- 0.00, N = 3SE +/- 0.00, N = 33.781.641. (CC) gcc options: -O3 -pthread -lz -llzma

FFmpeg

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Video On DemandOrange Pi 5Raspberry Pi 4002004006008001000SE +/- 0.13, N = 3SE +/- 0.87, N = 3401.56924.751. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Video On DemandOrange Pi 5Raspberry Pi 400510152025SE +/- 0.01, N = 3SE +/- 0.01, N = 318.868.191. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Decompression SpeedOrange Pi 5Raspberry Pi 400120240360480600SE +/- 0.74, N = 3SE +/- 0.52, N = 3545.3236.91. (CC) gcc options: -O3 -pthread -lz -llzma

FFmpeg

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: PlatformOrange Pi 5Raspberry Pi 4002004006008001000SE +/- 0.19, N = 3SE +/- 1.80, N = 3401.82924.231. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: PlatformOrange Pi 5Raspberry Pi 400510152025SE +/- 0.01, N = 3SE +/- 0.02, N = 318.858.201. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3 - Decompression SpeedOrange Pi 5Raspberry Pi 400160320480640800SE +/- 2.29, N = 3SE +/- 0.41, N = 3719.7314.01. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Decompression SpeedOrange Pi 5Raspberry Pi 400110220330440550SE +/- 0.26, N = 15SE +/- 1.17, N = 3511.7224.01. (CC) gcc options: -O3 -pthread -lz -llzma

FFmpeg

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: UploadOrange Pi 5Raspberry Pi 4002004006008001000SE +/- 0.27, N = 3SE +/- 0.63, N = 3493.261122.881. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: UploadOrange Pi 5Raspberry Pi 4001.1522.3043.4564.6085.76SE +/- 0.00, N = 3SE +/- 0.00, N = 35.122.251. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: LiveOrange Pi 5Raspberry Pi 40020406080100SE +/- 0.02, N = 3SE +/- 0.04, N = 378.3734.631. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: LiveOrange Pi 5Raspberry Pi 400306090120150SE +/- 0.02, N = 3SE +/- 0.16, N = 364.44145.811. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: UploadOrange Pi 5Raspberry Pi 4008001600240032004000SE +/- 0.79, N = 3SE +/- 0.57, N = 31594.633571.591. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On DemandOrange Pi 5Raspberry Pi 4000.69751.3952.09252.793.4875SE +/- 0.00, N = 3SE +/- 0.00, N = 33.101.391. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: PlatformOrange Pi 5Raspberry Pi 4000.69751.3952.09252.793.4875SE +/- 0.00, N = 3SE +/- 0.00, N = 33.101.391. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On DemandOrange Pi 5Raspberry Pi 40012002400360048006000SE +/- 0.92, N = 3SE +/- 0.58, N = 32446.765451.181. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: PlatformOrange Pi 5Raspberry Pi 40012002400360048006000SE +/- 1.35, N = 3SE +/- 2.80, N = 32447.475447.521. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: UploadOrange Pi 5Raspberry Pi 4000.35550.7111.06651.4221.7775SE +/- 0.00, N = 3SE +/- 0.00, N = 31.580.711. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3, Long Mode - Decompression SpeedOrange Pi 5Raspberry Pi 400160320480640800SE +/- 0.52, N = 3SE +/- 1.53, N = 3735.4331.01. (CC) gcc options: -O3 -pthread -lz -llzma

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: SwirlOrange Pi 5Raspberry Pi 40020406080100SE +/- 0.58, N = 3SE +/- 0.33, N = 311150-pthread -lwebp -lwebpmux -lSM -lICE1. (CC) gcc options: -fopenmp -O2 -ljbig -ltiff -lfreetype -ljpeg -lXext -lX11 -llzma -lz -lm -lpthread

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 0.3Target: CPU - Model: MobileNet v2Orange Pi 5Raspberry Pi 4002004006008001000SE +/- 0.95, N = 3SE +/- 10.76, N = 3504.171092.70MIN: 479.3 / MAX: 551.57MIN: 1069.72 / MAX: 1115.241. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: EnhancedOrange Pi 5Raspberry Pi 4001020304050SE +/- 0.33, N = 34521-pthread -lwebp -lwebpmux -lSM -lICE1. (CC) gcc options: -fopenmp -O2 -ljbig -ltiff -lfreetype -ljpeg -lXext -lX11 -llzma -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: SharpenOrange Pi 5Raspberry Pi 4001020304050SE +/- 0.29, N = 154521-pthread -lwebp -lwebpmux -lSM -lICE1. (CC) gcc options: -fopenmp -O2 -ljbig -ltiff -lfreetype -ljpeg -lXext -lX11 -llzma -lz -lm -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 SecondOrange Pi 5Raspberry Pi 40020K40K60K80K100KSE +/- 183.76, N = 3SE +/- 9.43, N = 385500.3139918.961. (CC) gcc options: -O2 -lrt" -lrt

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: WPA PSKOrange Pi 5Raspberry Pi 4008001600240032004000SE +/- 37.92, N = 3SE +/- 0.33, N = 335731672-lssl -lcrypto -lm -lrt -lz -ldl -lcrypt1. (CC) gcc options: -fopenmp

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.1Algorithm: RSA4096Orange Pi 5Raspberry Pi 40050100150200250SE +/- 0.09, N = 3SE +/- 0.06, N = 3246.5115.51. (CC) gcc options: -pthread -O3 -lssl -lcrypto -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: nbodyOrange Pi 5Raspberry Pi 400100200300400500SE +/- 0.33, N = 3SE +/- 1.15, N = 3216453

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: Noise-GaussianOrange Pi 5Raspberry Pi 4001428425670SE +/- 0.33, N = 36431-pthread -lwebp -lwebpmux -lSM -lICE1. (CC) gcc options: -fopenmp -O2 -ljbig -ltiff -lfreetype -ljpeg -lXext -lX11 -llzma -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: HWB Color SpaceOrange Pi 5Raspberry Pi 40070140210280350SE +/- 0.67, N = 3301146-pthread -lwebp -lwebpmux -lSM -lICE1. (CC) gcc options: -fopenmp -O2 -ljbig -ltiff -lfreetype -ljpeg -lXext -lX11 -llzma -lz -lm -lpthread

FFmpeg

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

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: LiveOrange Pi 5Raspberry Pi 400246810SE +/- 0.02, N = 3SE +/- 0.00, N = 38.874.331. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: LiveOrange Pi 5Raspberry Pi 40030060090012001500SE +/- 1.24, N = 3SE +/- 0.15, N = 3569.611165.301. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

RocksDB

This is a benchmark of Meta/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 BetterRocksDB 7.9.2Test: Random ReadOrange Pi 5Raspberry Pi 4003M6M9M12M15MSE +/- 116757.70, N = 3SE +/- 50304.99, N = 31222226759774051. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 10, LosslessOrange Pi 5Raspberry Pi 400918273645SE +/- 0.23, N = 3SE +/- 0.20, N = 319.0838.541. (CXX) g++ options: -O3 -fPIC -lm

nginx

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

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 100Orange Pi 5Raspberry Pi 4002K4K6K8K10KSE +/- 76.61, N = 3SE +/- 2.01, N = 310074.78503.901. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 200Orange Pi 5Raspberry Pi 4002K4K6K8K10KSE +/- 114.71, N = 4SE +/- 3.02, N = 39899.32500.941. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500Orange Pi 5Raspberry Pi 4002K4K6K8K10KSE +/- 101.37, N = 4SE +/- 2.23, N = 39038.60465.051. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Church FacadeOrange Pi 5Raspberry Pi 4006K12K18K24K30KSE +/- 17.91, N = 3SE +/- 48.60, N = 313615270701. (CXX) g++ options: -O3

LAME MP3 Encoding

LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3Orange Pi 5Raspberry Pi 400612182430SE +/- 0.00, N = 4SE +/- 0.04, N = 312.2023.871. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lncurses -lm

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 TimesOrange Pi 5Raspberry Pi 4008001600240032004000SE +/- 3.51, N = 3SE +/- 7.22, N = 319063672

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: SHA512Orange Pi 5Raspberry Pi 400200M400M600M800M1000MSE +/- 8390014.49, N = 3SE +/- 213656.63, N = 39818510675156392071. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

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 0.3Target: CPU - Model: SqueezeNet v2Orange Pi 5Raspberry Pi 40050100150200250SE +/- 0.29, N = 5SE +/- 0.53, N = 3124.25230.97MIN: 123.64 / MAX: 125.49MIN: 229.07 / MAX: 233.381. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.1Algorithm: RSA4096Orange Pi 5Raspberry Pi 4004K8K12K16K20KSE +/- 0.20, N = 3SE +/- 0.79, N = 316520.88930.31. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

TSCP

This is a performance test of TSCP, Tom Kerrigan's Simple Chess Program, which has a built-in performance benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess PerformanceOrange Pi 5Raspberry Pi 400200K400K600K800K1000KSE +/- 390.11, N = 10SE +/- 254.47, N = 58009004365021. (CC) gcc options: -O3 -march=native

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.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: DefaultOrange Pi 5Raspberry Pi 400246810SE +/- 0.01, N = 8SE +/- 0.01, N = 36.323.46-lpng16 -lgif -ltiff -pthread1. (CC) gcc options: -fvisibility=hidden -O2 -lm

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100Orange Pi 5Raspberry Pi 4000.9991.9982.9973.9964.995SE +/- 0.01, N = 7SE +/- 0.03, N = 34.442.44-lpng16 -lgif -ltiff -pthread1. (CC) gcc options: -fvisibility=hidden -O2 -lm

FLAC Audio Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.4WAV To FLACOrange Pi 5Raspberry Pi 40020406080100SE +/- 0.08, N = 5SE +/- 0.71, N = 2552.5093.471. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: LionOrange Pi 5Raspberry Pi 4004K8K12K16K20KSE +/- 3.68, N = 4SE +/- 8.11, N = 39988175421. (CXX) g++ options: -O3

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 20220729Target: CPU - Model: blazefaceOrange Pi 5Raspberry Pi 4003691215SE +/- 0.09, N = 3SE +/- 0.03, N = 36.0310.45MIN: 3.02 / MAX: 54.1-pthread - MIN: 10.21 / MAX: 11.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcryptOrange Pi 5Raspberry Pi 4006001200180024003000SE +/- 2.40, N = 3SE +/- 19.08, N = 1527291579-lssl -lcrypto -lm -lrt -lz -ldl -lcrypt1. (CC) gcc options: -fopenmp

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishOrange Pi 5Raspberry Pi 4006001200180024003000SE +/- 3.48, N = 3SE +/- 18.75, N = 1527321581-lssl -lcrypto -lm -lrt -lz -ldl -lcrypt1. (CC) gcc options: -fopenmp

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4KOrange Pi 5Raspberry Pi 400246810SE +/- 0.04, N = 3SE +/- 0.00, N = 36.113.57-lpthread1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

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.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Highest CompressionOrange Pi 5Raspberry Pi 4000.41630.83261.24891.66522.0815SE +/- 0.00, N = 4SE +/- 0.00, N = 31.851.10-lpng16 -lgif -ltiff -pthread1. (CC) gcc options: -fvisibility=hidden -O2 -lm

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: MD5Orange Pi 5Raspberry Pi 40013K26K39K52K65KSE +/- 50.83, N = 3SE +/- 60.95, N = 36004937518-lssl -lcrypto -lm -lrt -lz -ldl -lcrypt1. (CC) gcc options: -fopenmp

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: HMAC-SHA512Orange Pi 5Raspberry Pi 400700K1400K2100K2800K3500KSE +/- 1000.00, N = 3SE +/- 1333.33, N = 333770002158333-lssl -lcrypto -lm -lrt -lz -ldl -lcrypt1. (CC) gcc options: -fopenmp

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4KOrange Pi 5Raspberry Pi 400246810SE +/- 0.05, N = 3SE +/- 0.06, N = 37.164.67-lpthread1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KOrange Pi 5Raspberry Pi 400246810SE +/- 0.09, N = 3SE +/- 0.01, N = 37.034.63-lpthread1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average LatencyOrange Pi 5Raspberry Pi 400306090120150SE +/- 0.85, N = 3SE +/- 1.72, N = 12102.93140.29-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 50 - Mode: Read WriteOrange Pi 5Raspberry Pi 400110220330440550SE +/- 4.01, N = 3SE +/- 4.16, N = 12486357-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read WriteOrange Pi 5Raspberry Pi 40070140210280350SE +/- 2.84, N = 12SE +/- 2.03, N = 3332247-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Write - Average LatencyOrange Pi 5Raspberry Pi 4002004006008001000SE +/- 6.64, N = 12SE +/- 8.26, N = 3752.781010.64-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read WriteOrange Pi 5Raspberry Pi 400100200300400500SE +/- 1.68, N = 3SE +/- 2.49, N = 3442350-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Write - Average LatencyOrange Pi 5Raspberry Pi 40060120180240300SE +/- 0.86, N = 3SE +/- 2.03, N = 3226.37285.84-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

CPU Temperature Monitor

OpenBenchmarking.orgCelsiusCPU Temperature MonitorPhoronix Test Suite System MonitoringOrange Pi 51632486480Min: 38.85 / Avg: 71.98 / Max: 85.89

RocksDB

MinAvgMaxOrange Pi 560.183.385.0OpenBenchmarking.orgCelsius, Fewer Is BetterRocksDB 7.9.2CPU Temperature Monitor20406080100

Mobile Neural Network

OpenBenchmarking.orgCelsius, Fewer Is BetterMobile Neural Network 2.1CPU Temperature MonitorOrange Pi 51632486480Min: 38.85 / Avg: 84.27 / Max: 85.89

SQLite Speedtest

MinAvgMaxOrange Pi 546.258.160.1OpenBenchmarking.orgCelsius, Fewer Is BetterSQLite Speedtest 3.30CPU Temperature Monitor1632486480

PostgreSQL

OpenBenchmarking.orgCelsius, Fewer Is BetterPostgreSQL 15CPU Temperature MonitorOrange Pi 51428425670Min: 44.38 / Avg: 56.98 / Max: 73.92

OpenBenchmarking.orgCelsius, Fewer Is BetterPostgreSQL 15CPU Temperature MonitorOrange Pi 51632486480Min: 55.46 / Avg: 81.03 / Max: 85

OpenBenchmarking.orgCelsius, Fewer Is BetterPostgreSQL 15CPU Temperature MonitorOrange Pi 51632486480Min: 47.15 / Avg: 80.5 / Max: 85

RNNoise

MinAvgMaxOrange Pi 545.351.353.6OpenBenchmarking.orgCelsius, Fewer Is BetterRNNoise 2020-06-28CPU Temperature Monitor1530456075

eSpeak-NG Speech Engine

MinAvgMaxOrange Pi 544.451.953.6OpenBenchmarking.orgCelsius, Fewer Is BettereSpeak-NG Speech Engine 20200907CPU Temperature Monitor1530456075

AOM AV1

MinAvgMaxOrange Pi 544.468.281.3OpenBenchmarking.orgCelsius, Fewer Is BetterAOM AV1 3.6CPU Temperature Monitor20406080100

RocksDB

This is a benchmark of Meta/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 BetterRocksDB 7.9.2Test: Read While WritingOrange Pi 5Raspberry Pi 400110K220K330K440K550KSE +/- 49414.01, N = 12SE +/- 2368.64, N = 34932712698711. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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 20220729Target: CPU - Model: mnasnetOrange Pi 5Raspberry Pi 4001224364860SE +/- 0.48, N = 3SE +/- 0.47, N = 310.9352.72MIN: 6.92 / MAX: 59.61-pthread - MIN: 51.76 / MAX: 56.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: shufflenet-v2Orange Pi 5Raspberry Pi 400612182430SE +/- 0.99, N = 3SE +/- 0.09, N = 39.0526.12MIN: 5.45 / MAX: 49.39-pthread - MIN: 25.4 / MAX: 26.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU-v3-v3 - Model: mobilenet-v3Orange Pi 5Raspberry Pi 4001122334455SE +/- 0.94, N = 3SE +/- 0.38, N = 310.3447.02MIN: 6.73 / MAX: 65.96-pthread - MIN: 45.66 / MAX: 58.511. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU-v2-v2 - Model: mobilenet-v2Orange Pi 5Raspberry Pi 4001428425670SE +/- 1.63, N = 3SE +/- 0.75, N = 312.3061.64MIN: 6.61 / MAX: 60.26-pthread - MIN: 59.79 / MAX: 74.491. (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. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: inception-v3Orange Pi 5Raspberry Pi 40080160240320400SE +/- 6.52, N = 9SE +/- 2.11, N = 3239.73356.04MIN: 202.89 / MAX: 320.75MIN: 348.64 / MAX: 461.281. (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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: mobilenet-v1-1.0Orange Pi 5Raspberry Pi 4001326395265SE +/- 0.94, N = 9SE +/- 0.25, N = 326.6457.59MIN: 22.14 / MAX: 49.04MIN: 56.73 / MAX: 139.151. (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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: MobileNetV2_224Orange Pi 5Raspberry Pi 400816243240SE +/- 1.18, N = 9SE +/- 0.06, N = 321.1233.04MIN: 15.36 / MAX: 77.33MIN: 32.6 / MAX: 44.321. (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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: SqueezeNetV1.0Orange Pi 5Raspberry Pi 4001428425670SE +/- 1.22, N = 9SE +/- 0.18, N = 334.6761.66MIN: 28.45 / MAX: 51.62MIN: 60.88 / MAX: 74.421. (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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: resnet-v2-50Orange Pi 5Raspberry Pi 40060120180240300SE +/- 4.53, N = 9SE +/- 1.42, N = 3172.36284.61MIN: 118.57 / MAX: 231.7MIN: 280.49 / MAX: 314.731. (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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: squeezenetv1.1Orange Pi 5Raspberry Pi 400714212835SE +/- 0.49, N = 9SE +/- 0.26, N = 318.2031.17MIN: 14.65 / MAX: 30.58MIN: 30.41 / MAX: 60.271. (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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: mobilenetV3Orange Pi 5Raspberry Pi 4003691215SE +/- 0.224, N = 9SE +/- 0.014, N = 37.09910.867MIN: 5.66 / MAX: 16.33MIN: 10.71 / MAX: 22.021. (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

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: nasnetOrange Pi 5Raspberry Pi 40020406080100SE +/- 2.84, N = 9SE +/- 0.38, N = 370.8079.82MIN: 46.62 / MAX: 98.95MIN: 78.91 / MAX: 136.971. (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

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 1,000Orange Pi 5Raspberry Pi 400100200300400500SE +/- 1.59, N = 4SE +/- 10.37, N = 9136.63469.72-ldl -lpthread1. (CC) gcc options: -O2 -lz

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Write - Average LatencyOrange Pi 5Raspberry Pi 4007001400210028003500SE +/- 47.80, N = 12SE +/- 80.85, N = 92420.543300.89-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read WriteOrange Pi 5Raspberry Pi 40050100150200250SE +/- 4.09, N = 12SE +/- 3.56, N = 9207152-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Only - Average LatencyOrange Pi 5Raspberry Pi 4001.25082.50163.75245.00326.254SE +/- 0.042, N = 9SE +/- 0.059, N = 121.5595.559-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read OnlyOrange Pi 5Raspberry Pi 40014K28K42K56K70KSE +/- 1716.24, N = 9SE +/- 194.44, N = 126453518013-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average LatencyOrange Pi 5Raspberry Pi 4000.4550.911.3651.822.275SE +/- 0.016, N = 12SE +/- 0.026, N = 120.7802.022-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 50 - Mode: Read OnlyOrange Pi 5Raspberry Pi 40014K28K42K56K70KSE +/- 1483.60, N = 12SE +/- 332.97, N = 126444224772-lpthread -lrt -ldl1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

RNNoise

RNNoise is a recurrent neural network for audio noise reduction developed by Mozilla and Xiph.Org. This test profile is a single-threaded test measuring the time to denoise a sample 26 minute long 16-bit RAW audio file using this recurrent neural network noise suppression library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 2020-06-28Orange Pi 5Raspberry Pi 4001326395265SE +/- 0.19, N = 15SE +/- 1.81, N = 1429.9758.261. (CC) gcc options: -O2 -pedantic -fvisibility=hidden

eSpeak-NG Speech Engine

This test times how long it takes the eSpeak speech synthesizer to read Project Gutenberg's The Outline of Science and output to a WAV file. This test profile is now tracking the eSpeak-NG version of eSpeak. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech SynthesisOrange Pi 5Raspberry Pi 400306090120150SE +/- 0.55, N = 4SE +/- 5.45, N = 1658.67154.541. (CC) gcc options: -O2 -std=c99

AOM AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080pOrange Pi 5Raspberry Pi 400816243240SE +/- 1.30, N = 12SE +/- 0.07, N = 335.4411.75-lpthread1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

177 Results Shown

Zstd Compression:
  3 - Compression Speed
  3, Long Mode - Compression Speed
NCNN
OpenSSL
Zstd Compression:
  8 - Compression Speed
  8, Long Mode - Compression Speed
  12 - Compression Speed
NCNN:
  CPU - vgg16
  CPU - resnet50
  CPU - yolov4-tiny
  CPU - mobilenet
Memcached:
  1:100
  1:10
NCNN:
  CPU - squeezenet_ssd
  CPU - efficientnet-b0
  CPU - regnety_400m
OpenSSL:
  AES-256-GCM
  AES-128-GCM
XZ Compression
GraphicsMagick
NCNN
Numpy Benchmark
PyPerformance
miniBUDE:
  OpenMP - BM1:
    Billion Interactions/s
    GFInst/s
GraphicsMagick
PostgreSQL:
  1 - 250 - Read Only
  1 - 250 - Read Only - Average Latency
PyPerformance
Timed MPlayer Compilation
VP9 libvpx Encoding
Timed FFmpeg Compilation
Timed Linux Kernel Compilation
PostgreSQL:
  1 - 500 - Read Only
  1 - 500 - Read Only - Average Latency
RocksDB
NCNN
AOM AV1
Zstd Compression
AOM AV1
VVenC
nginx:
  20
  1000
  4000
SPECFEM3D
PyPerformance
Himeno Benchmark
PyPerformance
LAMMPS Molecular Dynamics Simulator
JPEG XL Decoding libjxl
CloverLeaf
dav1d
NCNN
libavif avifenc
JPEG XL Decoding libjxl
JPEG XL libjxl:
  PNG - 80
  PNG - 90
dav1d
NCNN
SPECFEM3D:
  Water-layered Halfspace
  Homogeneous Halfspace
PyPerformance
OpenSSL
SPECFEM3D
VP9 libvpx Encoding
Dolfyn
LibRaw
JPEG XL libjxl
TNN
JPEG XL libjxl
SPECFEM3D
Zstd Compression:
  12 - Decompression Speed
  8 - Decompression Speed
  8, Long Mode - Decompression Speed
PyPerformance
BRL-CAD
libavif avifenc
WebP Image Encode
TNN
OpenSSL
RocksDB
PyPerformance
Zstd Compression
FFmpeg:
  libx264 - Video On Demand:
    Seconds
    FPS
Zstd Compression
FFmpeg:
  libx264 - Platform:
    Seconds
    FPS
Zstd Compression:
  3 - Decompression Speed
  19 - Decompression Speed
FFmpeg:
  libx264 - Upload:
    Seconds
    FPS
  libx264 - Live:
    FPS
    Seconds
  libx265 - Upload:
    Seconds
  libx265 - Video On Demand:
    FPS
  libx265 - Platform:
    FPS
  libx265 - Video On Demand:
    Seconds
  libx265 - Platform:
    Seconds
  libx265 - Upload:
    FPS
Zstd Compression
GraphicsMagick
TNN
GraphicsMagick:
  Enhanced
  Sharpen
Coremark
John The Ripper
OpenSSL
PyPerformance
GraphicsMagick:
  Noise-Gaussian
  HWB Color Space
FFmpeg:
  libx265 - Live:
    FPS
    Seconds
RocksDB
libavif avifenc
nginx:
  100
  200
  500
Google Draco
LAME MP3 Encoding
PyBench
OpenSSL
TNN
OpenSSL
TSCP
WebP Image Encode:
  Default
  Quality 100
FLAC Audio Encoding
Google Draco
NCNN
John The Ripper:
  bcrypt
  Blowfish
AOM AV1
WebP Image Encode
John The Ripper:
  MD5
  HMAC-SHA512
AOM AV1:
  Speed 10 Realtime - Bosphorus 4K
  Speed 9 Realtime - Bosphorus 4K
PostgreSQL:
  1 - 50 - Read Write - Average Latency
  1 - 50 - Read Write
  1 - 250 - Read Write
  1 - 250 - Read Write - Average Latency
  1 - 100 - Read Write
  1 - 100 - Read Write - Average Latency
CPU Temperature Monitor:
  Phoronix Test Suite System Monitoring
  CPU Temp Monitor
  CPU Temp Monitor
  CPU Temp Monitor
  CPU Temp Monitor
  CPU Temp Monitor
  CPU Temp Monitor
  CPU Temp Monitor
  CPU Temp Monitor
  CPU Temp Monitor
RocksDB
NCNN:
  CPU - mnasnet
  CPU - shufflenet-v2
  CPU-v3-v3 - mobilenet-v3
  CPU-v2-v2 - mobilenet-v2
Mobile Neural Network:
  inception-v3
  mobilenet-v1-1.0
  MobileNetV2_224
  SqueezeNetV1.0
  resnet-v2-50
  squeezenetv1.1
  mobilenetV3
  nasnet
SQLite Speedtest
PostgreSQL:
  1 - 500 - Read Write - Average Latency
  1 - 500 - Read Write
  1 - 100 - Read Only - Average Latency
  1 - 100 - Read Only
  1 - 50 - Read Only - Average Latency
  1 - 50 - Read Only
RNNoise
eSpeak-NG Speech Engine
AOM AV1