TX2(6Cores)Graphics

TX2 ARMv8 rev 3 testing on Ubuntu 16.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1712148-KH-TX26CORES29
Jump To Table - Results

Statistics

Remove Outliers Before Calculating Averages

Graph Settings

Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
TX2 Graphics Test
December 14 2017
 
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):

TX2 Graphics Test

Processor: ARMv8 rev 3 @ 2.04GHz (6 Cores), Memory: 8192MB, Disk: 31GB 032G34, Graphics: NVIDIA Tegra X2 (nvgpu)/

OS: Ubuntu 16.04, Kernel: 4.4.38-tegra (aarch64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 28.1.0, OpenGL: 4.5.0, Compiler: GCC 5.4.0 20160609 + CUDA 8.0, File-System: ext4, Screen Resolution: 1920x1080

Compiler Notes: --build=aarch64-linux-gnu --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new -v
Processor Notes: Scaling Governor: tegra_cpufreq schedutil

TX2(6Cores)GraphicsOpenBenchmarking.orgPhoronix Test SuiteARMv8 rev 3 @ 2.04GHz (6 Cores)8192MB31GB 032G34NVIDIA Tegra X2 (nvgpu)/Ubuntu 16.044.4.38-tegra (aarch64)Unity 7.4.0X Server 1.18.4NVIDIA 28.1.04.5.0GCC 5.4.0 20160609 + CUDA 8.0ext41920x1080ProcessorMemoryDiskGraphicsOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionTX2(6Cores)Graphics PerformanceSystem Logs- --build=aarch64-linux-gnu --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new -v - Scaling Governor: tegra_cpufreq schedutil

TX2(6Cores)Graphicsjxrendermark: 12pt Text Grayscale - 32x32glmark2: 1920 x 1080jxrendermark: Put Composition - 1024x1024shoc: CUDA - Triadshoc: CUDA - FFT SPjxrendermark: Rects Composition - 128x128shoc: CUDA - MD5 Hashjxrendermark: Simple Blit - 1024x1024jxrendermark: Put Composition - 128x128jxrendermark: 12pt Text LCD - 128x128jxrendermark: 12pt Text LCD - 512x512jxrendermark: 12pt Text LCD - 32x32jxrendermark: Simple Blit - 256x256cairo-demos: Fishqgears2: XRender Extension - Image Scalingcuda-mini-nbody: SOA Data Layoutcuda-mini-nbody: Flush Denormals To Zerocuda-mini-nbody: Loop Unrollingcuda-mini-nbody: Cache Blockingcuda-mini-nbody: Originaljxrendermark: Put Composition - 512x512cairo-demos: Chartsshoc: CUDA - Max SP Flopsqgears2: OpenGL - Textjxrendermark: Simple Blit - 32x32qgears2: OpenGL - Gearsjxrendermark: Simple Blit - 128x128qgears2: OpenGL - Image Scalingjxrendermark: Simple Blit - 512x512qgears2: CPU-based Raster - Textjxrendermark: 12pt Text LCD - 256x256qgears2: CPU-based Raster - Gearsjxrendermark: Put Composition - 32x32qgears2: XRender Extension - Textjxrendermark: 12pt Text LCD - 1024x1024qgears2: XRender Extension - Gearsjxrendermark: Put Composition - 256x256qgears2: CPU-based Raster - Image Scalingjxrendermark: Rects Composition - 32x32jxrendermark: Rects Composition - 512x512shoc: CUDA - Bus Speed Downloadjxrendermark: Gradient+Temp Texture - 256x256jxrendermark: Gradient+Temp Texture - 512x512jxrendermark: 12pt Text Grayscale - 1024x1024jxrendermark: Gradient+Temp Texture - 128x128jxrendermark: Radial Gradient Paint - 256x256jxrendermark: Rects Composition - 1024x1024jxrendermark: Gradient+Temp Texture - 32x32jxrendermark: Linear Gradient Blend - 512x512jxrendermark: 12pt Text Grayscale - 256x256jxrendermark: 12pt Text Grayscale - 512x512jxrendermark: 12pt Text Grayscale - 128x128shoc: CUDA - Bus Speed Readbackjxrendermark: Rects Composition - 256x256jxrendermark: Linear Gradient Blend - 128x128jxrendermark: Linear Gradient Blend - 32x32jxrendermark: Linear Gradient Blend - 256x256jxrendermark: Radial Gradient Paint - 32x32jxrendermark: Radial Gradient Paint - 128x128jxrendermark: Radial Gradient Paint - 512x512shoc: CUDA - Texture Read Bandwidthjxrendermark: Transformed Texture Paint - 256x256jxrendermark: Transformed Blit Linear - 1024x1024jxrendermark: Transformed Texture Paint - 128x128jxrendermark: Transformed Blit Bilinear - 256x256jxrendermark: Transformed Blit Bilinear - 512x512jxrendermark: Transformed Texture Paint - 1024x1024jxrendermark: Transformed Blit Linear - 32x32jxrendermark: Transformed Blit Linear - 256x256jxrendermark: Transformed Blit Linear - 512x512jxrendermark: Transformed Blit Bilinear - 32x32jxrendermark: Transformed Blit Linear - 128x128jxrendermark: Radial Gradient Paint - 1024x1024jxrendermark: Linear Gradient Blend - 1024x1024jxrendermark: Gradient+Temp Texture - 1024x1024jxrendermark: Transformed Texture Paint - 512x512jxrendermark: Transformed Texture Paint - 32x32jxrendermark: Transformed Blit Bilinear - 1024x1024jxrendermark: Transformed Blit Bilinear - 128x128askap: Griddingaskap: DegriddingTX2 Graphics Test19793.7813341755.589.4478.7020789.550.868691.077771.3112876.0310154.2420184.9127997.96505.37122.46424.33424.31180.14188.27465.612043.091.31562.9112.0526948.9759.5928375.9159.9114958.6911.2812937.3051.3712924.7319.858417.3650.382455.47123.8619568.8817025.4818.186924.155462.027772.586527.589563.8810902.576747.098507.2211689.3910527.4513716.2218.1119316.129184.008679.549383.848646.179494.428395.2870.4124223.318423.7623898.7921842.3913446.9310783.9927898.9928031.0514488.0421670.6627371.653606.563898.133050.4319121.9522567.658256.7321352.70975.321681.63OpenBenchmarking.org

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text Grayscale - Size: 32x32TX2 Graphics Test4K8K12K16K20KSE +/- 913.58, N = 1219793.781. (CC) gcc options: -lX11 -lXrender

GLmark2

This is a test of Linaro's glmark2 port, currently using the X11 OpenGL 2.0 target. GLmark2 is a basic OpenGL benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterGLmark2 276Resolution: 1920 x 1080TX2 Graphics Test300600900120015001334

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Put Composition - Size: 1024x1024TX2 Graphics Test400800120016002000SE +/- 67.42, N = 121755.581. (CC) gcc options: -lX11 -lXrender

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: CUDA - Benchmark: TriadTX2 Graphics Test3691215SE +/- 0.02, N = 39.441. (CXX) g++ options: -O2 -lSHOCCommon -lcudadevrt -lcudart_static -lrt -lpthread -ldl -lcufft

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: CUDA - Benchmark: FFT SPTX2 Graphics Test20406080100SE +/- 0.05, N = 378.701. (CXX) g++ options: -O2 -lSHOCCommon -lcudadevrt -lcudart_static -lrt -lpthread -ldl -lcufft

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Rects Composition - Size: 128x128TX2 Graphics Test4K8K12K16K20KSE +/- 424.94, N = 1220789.551. (CC) gcc options: -lX11 -lXrender

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: CUDA - Benchmark: MD5 HashTX2 Graphics Test0.19350.3870.58050.7740.9675SE +/- 0.00, N = 30.861. (CXX) g++ options: -O2 -lSHOCCommon -lcudadevrt -lcudart_static -lrt -lpthread -ldl -lcufft

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Simple Blit - Size: 1024x1024TX2 Graphics Test2K4K6K8K10KSE +/- 93.64, N = 108691.071. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Put Composition - Size: 128x128TX2 Graphics Test17003400510068008500SE +/- 40.80, N = 67771.311. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text LCD - Size: 128x128TX2 Graphics Test3K6K9K12K15KSE +/- 998.33, N = 1212876.031. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text LCD - Size: 512x512TX2 Graphics Test2K4K6K8K10KSE +/- 786.25, N = 1210154.241. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text LCD - Size: 32x32TX2 Graphics Test4K8K12K16K20KSE +/- 188.73, N = 620184.911. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Simple Blit - Size: 256x256TX2 Graphics Test6K12K18K24K30KSE +/- 566.28, N = 1227997.961. (CC) gcc options: -lX11 -lXrender

Cairo Performance Demos

The cairo-demos test is a benchmark of Cairo using a collection of Cairo demos created by Chris Wilson at Intel OSTC. This test profile tests Cairo's Xlib back-end (others could be supported upon request) expecting an existing Cairo installation on the test system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterCairo Performance Demos 2012-01-30Demo: FishTX2 Graphics Test110220330440550SE +/- 3.65, N = 3505.37

QGears2

This test runs qgears2, which is a test written by Zack Rusin and is based upon cairogears. This test can see how well different rendering backends perform and uses Qt4. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQGears2Rendering: XRender Extension - Test: Image ScalingTX2 Graphics Test306090120150SE +/- 0.42, N = 3122.461. (CXX) g++ options: -lQtOpenGL -lQtGui -lQtCore -lGL -lpthread

CUDA Mini-Nbody

OpenBenchmarking.orgSeconds, Fewer Is BetterCUDA Mini-Nbody 2015-11-10Test: SOA Data LayoutTX2 Graphics Test90180270360450SE +/- 0.12, N = 3424.33

OpenBenchmarking.orgSeconds, Fewer Is BetterCUDA Mini-Nbody 2015-11-10Test: Flush Denormals To ZeroTX2 Graphics Test90180270360450SE +/- 0.03, N = 3424.31

OpenBenchmarking.orgSeconds, Fewer Is BetterCUDA Mini-Nbody 2015-11-10Test: Loop UnrollingTX2 Graphics Test4080120160200SE +/- 0.44, N = 3180.14

OpenBenchmarking.orgSeconds, Fewer Is BetterCUDA Mini-Nbody 2015-11-10Test: Cache BlockingTX2 Graphics Test4080120160200SE +/- 0.16, N = 3188.27

OpenBenchmarking.orgSeconds, Fewer Is BetterCUDA Mini-Nbody 2015-11-10Test: OriginalTX2 Graphics Test100200300400500SE +/- 0.20, N = 3465.61

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Put Composition - Size: 512x512TX2 Graphics Test400800120016002000SE +/- 62.57, N = 122043.091. (CC) gcc options: -lX11 -lXrender

Cairo Performance Demos

The cairo-demos test is a benchmark of Cairo using a collection of Cairo demos created by Chris Wilson at Intel OSTC. This test profile tests Cairo's Xlib back-end (others could be supported upon request) expecting an existing Cairo installation on the test system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterCairo Performance Demos 2012-01-30Demo: ChartsTX2 Graphics Test0.29480.58960.88441.17921.474SE +/- 0.01, N = 31.31

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: CUDA - Benchmark: Max SP FlopsTX2 Graphics Test120240360480600SE +/- 0.08, N = 3562.911. (CXX) g++ options: -O2 -lSHOCCommon -lcudadevrt -lcudart_static -lrt -lpthread -ldl -lcufft

QGears2

This test runs qgears2, which is a test written by Zack Rusin and is based upon cairogears. This test can see how well different rendering backends perform and uses Qt4. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQGears2Rendering: OpenGL - Test: TextTX2 Graphics Test3691215SE +/- 0.06, N = 312.051. (CXX) g++ options: -lQtOpenGL -lQtGui -lQtCore -lGL -lpthread

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Simple Blit - Size: 32x32TX2 Graphics Test6K12K18K24K30KSE +/- 1304.16, N = 1226948.971. (CC) gcc options: -lX11 -lXrender

QGears2

This test runs qgears2, which is a test written by Zack Rusin and is based upon cairogears. This test can see how well different rendering backends perform and uses Qt4. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQGears2Rendering: OpenGL - Test: GearsTX2 Graphics Test1326395265SE +/- 0.06, N = 359.591. (CXX) g++ options: -lQtOpenGL -lQtGui -lQtCore -lGL -lpthread

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Simple Blit - Size: 128x128TX2 Graphics Test6K12K18K24K30KSE +/- 351.65, N = 628375.911. (CC) gcc options: -lX11 -lXrender

QGears2

This test runs qgears2, which is a test written by Zack Rusin and is based upon cairogears. This test can see how well different rendering backends perform and uses Qt4. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQGears2Rendering: OpenGL - Test: Image ScalingTX2 Graphics Test1326395265SE +/- 0.04, N = 359.911. (CXX) g++ options: -lQtOpenGL -lQtGui -lQtCore -lGL -lpthread

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Simple Blit - Size: 512x512TX2 Graphics Test3K6K9K12K15KSE +/- 128.88, N = 614958.691. (CC) gcc options: -lX11 -lXrender

QGears2

This test runs qgears2, which is a test written by Zack Rusin and is based upon cairogears. This test can see how well different rendering backends perform and uses Qt4. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQGears2Rendering: CPU-based Raster - Test: TextTX2 Graphics Test3691215SE +/- 0.03, N = 311.281. (CXX) g++ options: -lQtOpenGL -lQtGui -lQtCore -lGL -lpthread

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text LCD - Size: 256x256TX2 Graphics Test3K6K9K12K15KSE +/- 735.94, N = 1212937.301. (CC) gcc options: -lX11 -lXrender

QGears2

This test runs qgears2, which is a test written by Zack Rusin and is based upon cairogears. This test can see how well different rendering backends perform and uses Qt4. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQGears2Rendering: CPU-based Raster - Test: GearsTX2 Graphics Test1224364860SE +/- 0.36, N = 351.371. (CXX) g++ options: -lQtOpenGL -lQtGui -lQtCore -lGL -lpthread

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Put Composition - Size: 32x32TX2 Graphics Test3K6K9K12K15KSE +/- 288.95, N = 1212924.731. (CC) gcc options: -lX11 -lXrender

QGears2

This test runs qgears2, which is a test written by Zack Rusin and is based upon cairogears. This test can see how well different rendering backends perform and uses Qt4. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQGears2Rendering: XRender Extension - Test: TextTX2 Graphics Test510152025SE +/- 0.06, N = 319.851. (CXX) g++ options: -lQtOpenGL -lQtGui -lQtCore -lGL -lpthread

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text LCD - Size: 1024x1024TX2 Graphics Test2K4K6K8K10KSE +/- 71.07, N = 68417.361. (CC) gcc options: -lX11 -lXrender

QGears2

This test runs qgears2, which is a test written by Zack Rusin and is based upon cairogears. This test can see how well different rendering backends perform and uses Qt4. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQGears2Rendering: XRender Extension - Test: GearsTX2 Graphics Test1122334455SE +/- 0.75, N = 650.381. (CXX) g++ options: -lQtOpenGL -lQtGui -lQtCore -lGL -lpthread

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Put Composition - Size: 256x256TX2 Graphics Test5001000150020002500SE +/- 49.78, N = 122455.471. (CC) gcc options: -lX11 -lXrender

QGears2

This test runs qgears2, which is a test written by Zack Rusin and is based upon cairogears. This test can see how well different rendering backends perform and uses Qt4. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterQGears2Rendering: CPU-based Raster - Test: Image ScalingTX2 Graphics Test306090120150SE +/- 0.34, N = 3123.861. (CXX) g++ options: -lQtOpenGL -lQtGui -lQtCore -lGL -lpthread

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Rects Composition - Size: 32x32TX2 Graphics Test4K8K12K16K20KSE +/- 1227.82, N = 1219568.881. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Rects Composition - Size: 512x512TX2 Graphics Test4K8K12K16K20KSE +/- 297.81, N = 1217025.481. (CC) gcc options: -lX11 -lXrender

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: CUDA - Benchmark: Bus Speed DownloadTX2 Graphics Test48121620SE +/- 0.02, N = 318.181. (CXX) g++ options: -O2 -lSHOCCommon -lcudadevrt -lcudart_static -lrt -lpthread -ldl -lcufft

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Gradient+Temp Texture - Size: 256x256TX2 Graphics Test15003000450060007500SE +/- 99.44, N = 126924.151. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Gradient+Temp Texture - Size: 512x512TX2 Graphics Test12002400360048006000SE +/- 57.11, N = 105462.021. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text Grayscale - Size: 1024x1024TX2 Graphics Test17003400510068008500SE +/- 362.87, N = 127772.581. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Gradient+Temp Texture - Size: 128x128TX2 Graphics Test14002800420056007000SE +/- 274.54, N = 126527.581. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Radial Gradient Paint - Size: 256x256TX2 Graphics Test2K4K6K8K10KSE +/- 55.94, N = 69563.881. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Rects Composition - Size: 1024x1024TX2 Graphics Test2K4K6K8K10KSE +/- 141.64, N = 1210902.571. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Gradient+Temp Texture - Size: 32x32TX2 Graphics Test14002800420056007000SE +/- 40.34, N = 66747.091. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Linear Gradient Blend - Size: 512x512TX2 Graphics Test2K4K6K8K10KSE +/- 223.01, N = 128507.221. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text Grayscale - Size: 256x256TX2 Graphics Test3K6K9K12K15KSE +/- 861.04, N = 1211689.391. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text Grayscale - Size: 512x512TX2 Graphics Test2K4K6K8K10KSE +/- 429.47, N = 1210527.451. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: 12pt Text Grayscale - Size: 128x128TX2 Graphics Test3K6K9K12K15KSE +/- 749.01, N = 1213716.221. (CC) gcc options: -lX11 -lXrender

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: CUDA - Benchmark: Bus Speed ReadbackTX2 Graphics Test48121620SE +/- 0.06, N = 318.111. (CXX) g++ options: -O2 -lSHOCCommon -lcudadevrt -lcudart_static -lrt -lpthread -ldl -lcufft

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Rects Composition - Size: 256x256TX2 Graphics Test4K8K12K16K20KSE +/- 134.06, N = 619316.121. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Linear Gradient Blend - Size: 128x128TX2 Graphics Test2K4K6K8K10KSE +/- 189.38, N = 129184.001. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Linear Gradient Blend - Size: 32x32TX2 Graphics Test2K4K6K8K10KSE +/- 448.07, N = 128679.541. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Linear Gradient Blend - Size: 256x256TX2 Graphics Test2K4K6K8K10KSE +/- 58.37, N = 69383.841. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Radial Gradient Paint - Size: 32x32TX2 Graphics Test2K4K6K8K10KSE +/- 413.22, N = 128646.171. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Radial Gradient Paint - Size: 128x128TX2 Graphics Test2K4K6K8K10KSE +/- 99.64, N = 69494.421. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Radial Gradient Paint - Size: 512x512TX2 Graphics Test2K4K6K8K10KSE +/- 90.31, N = 68395.281. (CC) gcc options: -lX11 -lXrender

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: CUDA - Benchmark: Texture Read BandwidthTX2 Graphics Test1632486480SE +/- 1.21, N = 470.411. (CXX) g++ options: -O2 -lSHOCCommon -lcudadevrt -lcudart_static -lrt -lpthread -ldl -lcufft

JXRenderMark

JXRenderMark is an open-source benchmark of the X Render extension for X.Org. Currently all tests are rendering to the backbuffer and not directly to the screen. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Texture Paint - Size: 256x256TX2 Graphics Test5K10K15K20K25KSE +/- 148.07, N = 624223.311. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Linear - Size: 1024x1024TX2 Graphics Test2K4K6K8K10KSE +/- 40.61, N = 68423.761. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Texture Paint - Size: 128x128TX2 Graphics Test5K10K15K20K25KSE +/- 278.64, N = 923898.791. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Bilinear - Size: 256x256TX2 Graphics Test5K10K15K20K25KSE +/- 370.30, N = 1221842.391. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Bilinear - Size: 512x512TX2 Graphics Test3K6K9K12K15KSE +/- 70.93, N = 613446.931. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Texture Paint - Size: 1024x1024TX2 Graphics Test2K4K6K8K10KSE +/- 116.05, N = 1010783.991. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Linear - Size: 32x32TX2 Graphics Test6K12K18K24K30KSE +/- 460.13, N = 1227898.991. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Linear - Size: 256x256TX2 Graphics Test6K12K18K24K30KSE +/- 340.04, N = 1228031.051. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Linear - Size: 512x512TX2 Graphics Test3K6K9K12K15KSE +/- 94.98, N = 614488.041. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Bilinear - Size: 32x32TX2 Graphics Test5K10K15K20K25KSE +/- 153.23, N = 621670.661. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Linear - Size: 128x128TX2 Graphics Test6K12K18K24K30KSE +/- 684.36, N = 1227371.651. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Radial Gradient Paint - Size: 1024x1024TX2 Graphics Test8001600240032004000SE +/- 26.43, N = 63606.561. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Linear Gradient Blend - Size: 1024x1024TX2 Graphics Test8001600240032004000SE +/- 45.26, N = 93898.131. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Gradient+Temp Texture - Size: 1024x1024TX2 Graphics Test7001400210028003500SE +/- 24.72, N = 63050.431. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Texture Paint - Size: 512x512TX2 Graphics Test4K8K12K16K20KSE +/- 330.58, N = 1219121.951. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Texture Paint - Size: 32x32TX2 Graphics Test5K10K15K20K25KSE +/- 214.07, N = 622567.651. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Bilinear - Size: 1024x1024TX2 Graphics Test2K4K6K8K10KSE +/- 60.06, N = 68256.731. (CC) gcc options: -lX11 -lXrender

OpenBenchmarking.orgOperations Per Second, More Is BetterJXRenderMark 1.0.1Test: Transformed Blit Bilinear - Size: 128x128TX2 Graphics Test5K10K15K20K25KSE +/- 495.62, N = 1221352.701. (CC) gcc options: -lX11 -lXrender

ASKAP tConvolveCuda

This is a CUDA benchmark of ATNF's ASKAP Benchmark with currently using the tConvolveCuda sub-test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP tConvolveCuda 2015-11-10Processing: GriddingTX2 Graphics Test2004006008001000SE +/- 3.59, N = 3975.321. (CXX) g++ options: -fPIC -O3 -lcudadevrt -lcudart_static -lrt -lpthread -ldl

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP tConvolveCuda 2015-11-10Processing: DegriddingTX2 Graphics Test400800120016002000SE +/- 3.53, N = 31681.631. (CXX) g++ options: -fPIC -O3 -lcudadevrt -lcudart_static -lrt -lpthread -ldl

81 Results Shown

JXRenderMark
GLmark2
JXRenderMark
SHOC Scalable HeterOgeneous Computing:
  CUDA - Triad
  CUDA - FFT SP
JXRenderMark
SHOC Scalable HeterOgeneous Computing
JXRenderMark:
  Simple Blit - 1024x1024
  Put Composition - 128x128
  12pt Text LCD - 128x128
  12pt Text LCD - 512x512
  12pt Text LCD - 32x32
  Simple Blit - 256x256
Cairo Performance Demos
QGears2
CUDA Mini-Nbody:
  SOA Data Layout
  Flush Denormals To Zero
  Loop Unrolling
  Cache Blocking
  Original
JXRenderMark
Cairo Performance Demos
SHOC Scalable HeterOgeneous Computing
QGears2
JXRenderMark
QGears2
JXRenderMark
QGears2
JXRenderMark
QGears2
JXRenderMark
QGears2
JXRenderMark
QGears2
JXRenderMark
QGears2
JXRenderMark
QGears2
JXRenderMark:
  Rects Composition - 32x32
  Rects Composition - 512x512
SHOC Scalable HeterOgeneous Computing
JXRenderMark:
  Gradient+Temp Texture - 256x256
  Gradient+Temp Texture - 512x512
  12pt Text Grayscale - 1024x1024
  Gradient+Temp Texture - 128x128
  Radial Gradient Paint - 256x256
  Rects Composition - 1024x1024
  Gradient+Temp Texture - 32x32
  Linear Gradient Blend - 512x512
  12pt Text Grayscale - 256x256
  12pt Text Grayscale - 512x512
  12pt Text Grayscale - 128x128
SHOC Scalable HeterOgeneous Computing
JXRenderMark:
  Rects Composition - 256x256
  Linear Gradient Blend - 128x128
  Linear Gradient Blend - 32x32
  Linear Gradient Blend - 256x256
  Radial Gradient Paint - 32x32
  Radial Gradient Paint - 128x128
  Radial Gradient Paint - 512x512
SHOC Scalable HeterOgeneous Computing
JXRenderMark:
  Transformed Texture Paint - 256x256
  Transformed Blit Linear - 1024x1024
  Transformed Texture Paint - 128x128
  Transformed Blit Bilinear - 256x256
  Transformed Blit Bilinear - 512x512
  Transformed Texture Paint - 1024x1024
  Transformed Blit Linear - 32x32
  Transformed Blit Linear - 256x256
  Transformed Blit Linear - 512x512
  Transformed Blit Bilinear - 32x32
  Transformed Blit Linear - 128x128
  Radial Gradient Paint - 1024x1024
  Linear Gradient Blend - 1024x1024
  Gradient+Temp Texture - 1024x1024
  Transformed Texture Paint - 512x512
  Transformed Texture Paint - 32x32
  Transformed Blit Bilinear - 1024x1024
  Transformed Blit Bilinear - 128x128
ASKAP tConvolveCuda:
  Gridding
  Degridding

TX2 Graphics Test

Processor: ARMv8 rev 3 @ 2.04GHz (6 Cores), Memory: 8192MB, Disk: 31GB 032G34, Graphics: NVIDIA Tegra X2 (nvgpu)/

OS: Ubuntu 16.04, Kernel: 4.4.38-tegra (aarch64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 28.1.0, OpenGL: 4.5.0, Compiler: GCC 5.4.0 20160609 + CUDA 8.0, File-System: ext4, Screen Resolution: 1920x1080

Compiler Notes: --build=aarch64-linux-gnu --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new -v
Processor Notes: Scaling Governor: tegra_cpufreq schedutil

Testing initiated at 14 December 2017 01:50 by user ubuntu.