OpenCL Intel Arc Graphics Linux

Tests 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 2303075-PTS-OPENCLIN52
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:

CPU Massive 3 Tests
Creator Workloads 2 Tests
HPC - High Performance Computing 2 Tests
Machine Learning 2 Tests
NVIDIA GPU Compute 5 Tests
OpenCL 4 Tests

Statistics

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

Graph Settings

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

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
Intel Arc A770
March 06 2023
  1 Hour, 5 Minutes
Intel Arc A750
March 06 2023
  1 Hour, 28 Minutes
Intel Arc A380
March 07 2023
  43 Minutes
A380
March 07 2023
  42 Minutes
A380 two
March 07 2023
  2 Hours, 9 Minutes
Invert Hiding All Results Option
  1 Hour, 13 Minutes

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

Intel Arc A770

Processor: AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (9922 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 2000GB, Graphics: Intel Arc A770 DG2 16GB (2400MHz), Audio: Intel DG2 Audio, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Ubuntu 23.04, Kernel: 6.2.2-060202-generic (x86_64), Desktop: GNOME Shell 43.2, Display Server: X Server 1.21.1.6 + Wayland, OpenGL: 4.6 Mesa 23.1.0-devel (git-0d8a54f 2023-03-06 lunar-oibaf-ppa), OpenCL: OpenCL 3.0, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Intel Arc A750

Changed Graphics to Intel Arc A750 DG2 8GB (2400MHz).

Intel Arc A380

A380

A380 two

Changed Graphics to Intel Arc A380 DG2 6GB (2450MHz).

OpenCL Intel Arc Graphics LinuxOpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR X670E HERO (9922 BIOS)AMD Device 14d832GBWestern Digital WD_BLACK SN850X 1000GB + 2000GBIntel Arc A770 DG2 16GB (2400MHz)Intel Arc A750 DG2 8GB (2400MHz)Intel Arc A380 DG2 6GB (2450MHz)Intel DG2 AudioASUS MG28UIntel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411Ubuntu 23.046.2.2-060202-generic (x86_64)GNOME Shell 43.2X Server 1.21.1.6 + Wayland4.6 Mesa 23.1.0-devel (git-0d8a54f 2023-03-06 lunar-oibaf-ppa)OpenCL 3.0GCC 12.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLOpenCLCompilerFile-SystemScreen ResolutionOpenCL Intel Arc Graphics Linux BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203 - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Intel Arc A770Intel Arc A750Intel Arc A380A380A380 twoResult OverviewPhoronix Test Suite100%166%232%298%364%FluidX3DBlenderLeelaChessZeroSHOC Scalable HeterOgeneous ComputingDarktableclpeakHashcat

Intel Arc A770Intel Arc A750Intel Arc A380A380A380 twoPer Watt Result OverviewPhoronix Test Suite100%155%209%264%318%Meta Performance Per WattsSHOC Scalable HeterOgeneous ComputingclpeakHashcatLeelaChessZeroFluidX3DP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.M

OpenCL Intel Arc Graphics Linuxshoc: OpenCL - Bus Speed Readbackshoc: OpenCL - Bus Speed Downloadshoc: OpenCL - Triadclpeak: Transfer Bandwidth enqueueWriteBuffershoc: OpenCL - GEMM SGEMM_Nclpeak: Single-Precision Computeshoc: OpenCL - MD5 Hashfluidx3d: FP32-FP16Chashcat: SHA1hashcat: SHA-512fluidx3d: FP32-FP16Sblender: Pabellon Barcelona - Intel oneAPIfluidx3d: FP32-FP32hashcat: TrueCrypt RIPEMD160 + XTSblender: BMW27 - Intel oneAPIblender: Classroom - Intel oneAPIlczero: OpenCLhashcat: MD5blender: Fishy Cat - Intel oneAPIclpeak: Global Memory Bandwidthblender: Barbershop - Intel oneAPIclpeak: Integer Computeclpeak: Integer 24-bit Computeclpeak: Kernel Latencydarktable: Server Room - OpenCLshoc: OpenCL - S3Dclpeak: Transfer Bandwidth enqueueReadBufferdarktable: Boat - OpenCLdarktable: Server Rack - OpenCLshoc: OpenCL - Texture Read Bandwidthshoc: OpenCL - Reductiondarktable: Masskrug - OpenCLhashcat: 7-Zipshoc: OpenCL - Max SP FlopsIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two22.485422.957920.074617.202459.8713007.2224.1784421415178616667973616667434882.77262335710020.7856.1132203345098333329.54397.26191.344442.404435.6035.700.779225.9115.691.8580.101980.579150.5491.44824792535749072.84362.96602.91692.802100.1011381.5721.1627396613380116667863450000396385.18251231368821.0658.60101012952321666731.35396.57198.813914.623911.9196.670.874209.9382.402.1660.207882.97878.50842.07624690075876511.269111.745210.61210.02608.4053320.416.173112040666000002683000001203296.7673910180072.02191.863909998810000098.11120.71611.991453.221454.5535.12.26878.7545.544.0820.217451.805113.2012.476154678011.2711.744410.60619.92608.3673320.426.1776112140656000002679000001203296.7573910240072.07192.033975998670000098.29120.75612.091453.681454.9132.62.28678.77045.674.0930.218451.488113.7632.423161060011.271111.744810.61159.87608.5883320.696.1738112140696000002684000001203296.6773910250072.20191.593952999483333398.25120.74610.951454.291454.0432.552.28078.72355.654.0940.221450.397113.2492.4391525558OpenBenchmarking.org

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed ReadbackIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two510152025SE +/- 0.0004, N = 13SE +/- 0.0000, N = 8SE +/- 0.0006, N = 1222.48542.843611.269111.270011.27111. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed ReadbackIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.14110.28220.42330.56440.70550.5700.0560.6260.6250.627
MinAvgMaxIntel Arc A77034.439.556.2Intel Arc A75035.651.166.3A380 two16.618.022.5OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor204060801001. Intel Arc A770: Approximate power consumption of 97 Joules per run.2. Intel Arc A750: Approximate power consumption of 224 Joules per run.3. A380 two: Approximate power consumption of 48 Joules per run.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed DownloadIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two612182430SE +/- 0.0026, N = 14SE +/- 0.0001, N = 8SE +/- 0.0001, N = 1222.95792.966011.745211.744411.74481. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed DownloadIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.14810.29620.44430.59240.74050.5850.0580.6300.6300.658
MinAvgMaxIntel Arc A77034.739.356.0Intel Arc A75036.050.965.1A380 two16.617.922.1OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor204060801001. Intel Arc A770: Approximate power consumption of 93 Joules per run.2. Intel Arc A750: Approximate power consumption of 210 Joules per run.3. A380 two: Approximate power consumption of 46 Joules per run.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: TriadIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two510152025SE +/- 0.0106, N = 12SE +/- 0.0004, N = 10SE +/- 0.0007, N = 1120.07462.916910.612010.606110.61151. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: TriadIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.14150.2830.42450.5660.70750.4810.0570.6270.6290.575
MinAvgMaxIntel Arc A77034.241.764.9Intel Arc A75035.751.175.6A380 two16.618.524.0OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor204060801001. Intel Arc A770: Approximate power consumption of 115 Joules per run.2. Intel Arc A750: Approximate power consumption of 174 Joules per run.3. A380 two: Approximate power consumption of 52 Joules per run.

clpeak

Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGBPS, More Is Betterclpeak 1.1.2OpenCL Test: Transfer Bandwidth enqueueWriteBufferIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two48121620SE +/- 0.16, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 317.202.8010.029.929.871. (CXX) g++ options: -O3
OpenBenchmarking.orgGBPS Per Watt, More Is Betterclpeak 1.1.2OpenCL Test: Transfer Bandwidth enqueueWriteBufferIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.10760.21520.32280.43040.5380.3120.0450.4780.4750.474
MinAvgMaxIntel Arc A77034.255.268.5Intel Arc A75035.662.365.9Intel Arc A38016.621.022.4A38016.720.922.3A380 two16.620.822.4OpenBenchmarking.orgWatts, Fewer Is Betterclpeak 1.1.2GPU Power Consumption Monitor204060801001. Intel Arc A770: Approximate power consumption of 1270 Joules per run.2. Intel Arc A750: Approximate power consumption of 5045 Joules per run.3. A380 two: Approximate power consumption of 465 Joules per run.

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_NIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two5001000150020002500SE +/- 16.77, N = 15SE +/- 14.19, N = 9SE +/- 0.15, N = 72459.872100.10608.41608.37608.591. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_NIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two91827364538.8831.7520.8920.9620.50
MinAvgMaxIntel Arc A77034.463.3160.8Intel Arc A75036.066.1108.6A380 two16.629.742.3OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 186 Joules per run.2. Intel Arc A750: Approximate power consumption of 257 Joules per run.3. A380 two: Approximate power consumption of 136 Joules per run.

clpeak

Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is Betterclpeak 1.1.2OpenCL Test: Single-Precision ComputeIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two3K6K9K12K15KSE +/- 2.82, N = 3SE +/- 4.43, N = 3SE +/- 0.48, N = 313007.2211381.573320.413320.423320.691. (CXX) g++ options: -O3
OpenBenchmarking.orgGFLOPS Per Watt, More Is Betterclpeak 1.1.2OpenCL Test: Single-Precision ComputeIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two306090120150111.75103.1897.0196.9695.92
MinAvgMaxIntel Arc A77034.8116.4144.5Intel Arc A75036.2110.3138.3Intel Arc A38016.734.241.0A38016.734.241.0A380 two16.534.640.8OpenBenchmarking.orgWatts, Fewer Is Betterclpeak 1.1.2GPU Power Consumption Monitor40801201602001. Intel Arc A770: Approximate power consumption of 1242 Joules per run.2. Intel Arc A750: Approximate power consumption of 1213 Joules per run.3. A380 two: Approximate power consumption of 369 Joules per run.

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 HashIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two612182430SE +/- 0.0113, N = 15SE +/- 0.0060, N = 15SE +/- 0.0012, N = 924.178421.16276.17306.17766.17381. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGHash/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 HashIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.09540.19080.28620.38160.4770.4240.3330.1980.1990.182
MinAvgMaxIntel Arc A77034.557.0151.0Intel Arc A75035.763.6152.4A380 two16.634.060.0OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor40801201602001. Intel Arc A770: Approximate power consumption of 141 Joules per run.2. Intel Arc A750: Approximate power consumption of 161 Joules per run.3. A380 two: Approximate power consumption of 117 Joules per run.

FluidX3D

FluidX3D is a speedy and memory efficient Boltzmann CFD (Computational Fluid Dynamics) software package implemented using OpenCL and intended for GPU acceleration. FluidX3D is developed by Moritz Lehmann and written free for non-commercial use. This is a test profile measuring the system OpenCL performance using the FluidX3D benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMLUPs/s, More Is BetterFluidX3D 2.3Test: FP32-FP16CIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two9001800270036004500SE +/- 0.88, N = 3SE +/- 2.03, N = 3SE +/- 0.33, N = 342143966112011211121
OpenBenchmarking.orgMLUPs/s Per Watt, More Is BetterFluidX3D 2.3Test: FP32-FP16CIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two61218243024.4022.8122.6822.8422.65
MinAvgMaxIntel Arc A77035.3172.7190.1Intel Arc A75036.0173.9190.2Intel Arc A38016.649.450.5A38016.649.150.6A380 two16.549.550.7OpenBenchmarking.orgWatts, Fewer Is BetterFluidX3D 2.3GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 3915 Joules per run.2. Intel Arc A750: Approximate power consumption of 4174 Joules per run.3. A380 two: Approximate power consumption of 3828 Joules per run.

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA1Intel Arc A770Intel Arc A750Intel Arc A380A380A380 two3000M6000M9000M12000M15000MSE +/- 10398667.12, N = 6SE +/- 1699297.24, N = 6SE +/- 3098601.84, N = 61517861666713380116667406660000040656000004069600000
OpenBenchmarking.orgH/s Per Watt, More Is BetterHashcat 6.2.4Benchmark: SHA1Intel Arc A770Intel Arc A750Intel Arc A380A380A380 two30M60M90M120M150M124309909.67118364467.38129326501.77104693562.23102289051.53
MinAvgMaxIntel Arc A77034.7122.1191.1Intel Arc A75035.8113.0190.5Intel Arc A38016.631.460.1A380 two16.639.860.1OpenBenchmarking.orgWatts, Fewer Is BetterHashcat 6.2.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 672 Joules per run.2. Intel Arc A750: Approximate power consumption of 678 Joules per run.3. A380 two: Approximate power consumption of 225 Joules per run.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA-512Intel Arc A770Intel Arc A750Intel Arc A380A380A380 two200M400M600M800M1000MSE +/- 611782.46, N = 6SE +/- 930859.82, N = 6SE +/- 129099.44, N = 6973616667863450000268300000267900000268400000
OpenBenchmarking.orgH/s Per Watt, More Is BetterHashcat 6.2.4Benchmark: SHA-512Intel Arc A770Intel Arc A750Intel Arc A380A380A380 two2M4M6M8M10M8080212.767327052.197914454.286994778.076767724.71
MinAvgMaxIntel Arc A77034.7120.5190.3Intel Arc A75035.8117.8190.2A380 two16.539.760.1OpenBenchmarking.orgWatts, Fewer Is BetterHashcat 6.2.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 663 Joules per run.2. Intel Arc A750: Approximate power consumption of 668 Joules per run.3. A380 two: Approximate power consumption of 225 Joules per run.

FluidX3D

FluidX3D is a speedy and memory efficient Boltzmann CFD (Computational Fluid Dynamics) software package implemented using OpenCL and intended for GPU acceleration. FluidX3D is developed by Moritz Lehmann and written free for non-commercial use. This is a test profile measuring the system OpenCL performance using the FluidX3D benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMLUPs/s, More Is BetterFluidX3D 2.3Test: FP32-FP16SIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two9001800270036004500SE +/- 2.96, N = 3SE +/- 9.94, N = 3SE +/- 0.00, N = 343483963120312031203
OpenBenchmarking.orgMLUPs/s Per Watt, More Is BetterFluidX3D 2.3Test: FP32-FP16SIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two71421283528.7928.6631.2431.4231.28
MinAvgMaxIntel Arc A77034.7151.0164.7Intel Arc A75037.4138.3152.6Intel Arc A38016.638.539.3A38016.738.339.3A380 two16.538.539.3OpenBenchmarking.orgWatts, Fewer Is BetterFluidX3D 2.3GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 3273 Joules per run.2. Intel Arc A750: Approximate power consumption of 3319 Joules per run.3. A380 two: Approximate power consumption of 2782 Joules per run.

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Pabellon Barcelona - Compute: Intel oneAPIIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two60120180240300SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 382.7785.18296.76296.75296.67
MinAvgMaxIntel Arc A77035.2180.5193.0Intel Arc A75036.0179.8192.8Intel Arc A38016.751.455.2A38016.851.155.1A380 two16.551.355.8OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 7882 Joules per run.2. Intel Arc A750: Approximate power consumption of 8031 Joules per run.3. A380 two: Approximate power consumption of 7731 Joules per run.

FluidX3D

FluidX3D is a speedy and memory efficient Boltzmann CFD (Computational Fluid Dynamics) software package implemented using OpenCL and intended for GPU acceleration. FluidX3D is developed by Moritz Lehmann and written free for non-commercial use. This is a test profile measuring the system OpenCL performance using the FluidX3D benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMLUPs/s, More Is BetterFluidX3D 2.3Test: FP32-FP32Intel Arc A770Intel Arc A750Intel Arc A380A380A380 two6001200180024003000SE +/- 2.52, N = 3SE +/- 2.03, N = 3SE +/- 0.00, N = 326232512739739739
OpenBenchmarking.orgMLUPs/s Per Watt, More Is BetterFluidX3D 2.3Test: FP32-FP32Intel Arc A770Intel Arc A750Intel Arc A380A380A380 two51015202518.9019.1421.0821.0121.12
MinAvgMaxIntel Arc A77034.5138.8148.1Intel Arc A75036.5131.3138.7Intel Arc A38016.735.135.5A38016.635.235.6A380 two16.735.035.5OpenBenchmarking.orgWatts, Fewer Is BetterFluidX3D 2.3GPU Power Consumption Monitor40801201602001. Intel Arc A770: Approximate power consumption of 4810 Joules per run.2. Intel Arc A750: Approximate power consumption of 4769 Joules per run.3. A380 two: Approximate power consumption of 4072 Joules per run.

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: TrueCrypt RIPEMD160 + XTSIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two80K160K240K320K400KSE +/- 819.84, N = 8SE +/- 722.21, N = 8SE +/- 265.83, N = 6357100313688101800102400102500
OpenBenchmarking.orgH/s Per Watt, More Is BetterHashcat 6.2.4Benchmark: TrueCrypt RIPEMD160 + XTSIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two80016002400320040003909.273623.823448.252614.912724.90
MinAvgMaxIntel Arc A77034.691.3190.1Intel Arc A75035.686.6190.8Intel Arc A38016.629.560.0A380 two16.537.660.4OpenBenchmarking.orgWatts, Fewer Is BetterHashcat 6.2.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 365 Joules per run.2. Intel Arc A750: Approximate power consumption of 379 Joules per run.3. A380 two: Approximate power consumption of 194 Joules per run.

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: BMW27 - Compute: Intel oneAPIIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two1632486480SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 320.7821.0672.0272.0772.20
MinAvgMaxIntel Arc A77034.1153.4191.1Intel Arc A75034.8144.3179.4Intel Arc A38016.644.849.1A38016.646.150.1A380 two16.645.950.1OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 1892 Joules per run.2. Intel Arc A750: Approximate power consumption of 1828 Joules per run.3. A380 two: Approximate power consumption of 1743 Joules per run.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Classroom - Compute: Intel oneAPIIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two4080120160200SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.12, N = 356.1058.60191.86192.03191.59
MinAvgMaxIntel Arc A77034.6162.8178.0Intel Arc A75035.8152.6165.9Intel Arc A38016.644.646.2A38016.844.946.2A380 two16.545.146.4OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 4885 Joules per run.2. Intel Arc A750: Approximate power consumption of 4781 Joules per run.3. A380 two: Approximate power consumption of 4404 Joules per run.

LeelaChessZero

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

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: OpenCLIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two3K6K9K12K15KSE +/- 95.50, N = 3SE +/- 118.76, N = 4SE +/- 12.25, N = 313220101013909397539521. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.28Backend: OpenCLIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two2040608010092.0778.3090.8981.7781.09
MinAvgMaxIntel Arc A77034.8143.6154.2Intel Arc A75036.0129.0142.1Intel Arc A38016.743.050.5A38016.748.650.7A380 two16.648.750.7OpenBenchmarking.orgWatts, Fewer Is BetterLeelaChessZero 0.28GPU Power Consumption Monitor40801201602001. Intel Arc A770: Approximate power consumption of 25319 Joules per run.2. Intel Arc A750: Approximate power consumption of 24059 Joules per run.3. A380 two: Approximate power consumption of 8691 Joules per run.

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: MD5Intel Arc A770Intel Arc A750Intel Arc A380A380A380 two7000M14000M21000M28000M35000MSE +/- 13663123.52, N = 6SE +/- 6369794.17, N = 6SE +/- 1102018.35, N = 63345098333329523216667998810000099867000009994833333
OpenBenchmarking.orgH/s Per Watt, More Is BetterHashcat 6.2.4Benchmark: MD5Intel Arc A770Intel Arc A750Intel Arc A380A380A380 two70M140M210M280M350M279704669.84259062473.66305914241.96256837548.22256656680.41
MinAvgMaxIntel Arc A77034.4119.6191.0Intel Arc A75035.8114.0191.1Intel Arc A38016.732.760.0A380 two16.638.960.1OpenBenchmarking.orgWatts, Fewer Is BetterHashcat 6.2.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 658 Joules per run.2. Intel Arc A750: Approximate power consumption of 646 Joules per run.3. A380 two: Approximate power consumption of 214 Joules per run.

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Fishy Cat - Compute: Intel oneAPIIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two20406080100SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 329.5431.3598.1198.2998.25
MinAvgMaxIntel Arc A77034.9154.9190.7Intel Arc A75036.2156.1190.7Intel Arc A38016.652.157.0A38016.652.056.8A380 two16.652.256.9OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 2632 Joules per run.2. Intel Arc A750: Approximate power consumption of 2809 Joules per run.3. A380 two: Approximate power consumption of 2678 Joules per run.

clpeak

Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGBPS, More Is Betterclpeak 1.1.2OpenCL Test: Global Memory BandwidthIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two90180270360450SE +/- 0.09, N = 8SE +/- 0.20, N = 8SE +/- 0.02, N = 6397.26396.57120.71120.75120.741. (CXX) g++ options: -O3
OpenBenchmarking.orgGBPS Per Watt, More Is Betterclpeak 1.1.2OpenCL Test: Global Memory BandwidthIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two1.3012.6023.9035.2046.5055.4095.7824.5724.4914.388
MinAvgMaxIntel Arc A77034.573.4147.9Intel Arc A75036.068.6142.2A38016.726.935.8A380 two16.727.536.9OpenBenchmarking.orgWatts, Fewer Is Betterclpeak 1.1.2GPU Power Consumption Monitor40801201602001. Intel Arc A770: Approximate power consumption of 312 Joules per run.2. Intel Arc A750: Approximate power consumption of 309 Joules per run.3. A380 two: Approximate power consumption of 170 Joules per run.

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Barbershop - Compute: Intel oneAPIIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two130260390520650SE +/- 0.06, N = 3SE +/- 0.37, N = 3SE +/- 1.21, N = 3191.34198.81611.99612.09610.95
MinAvgMaxIntel Arc A77034.9176.4192.2Intel Arc A75035.9166.9181.5Intel Arc A38016.650.452.4A38016.750.552.7A380 two16.550.452.6OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 17351 Joules per run.2. Intel Arc A750: Approximate power consumption of 17026 Joules per run.3. A380 two: Approximate power consumption of 15525 Joules per run.

Meta Performance Per Watts

OpenBenchmarking.orgPerformance Per Watts, More Is BetterMeta Performance Per WattsPerformance Per WattsIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two2K4K6K8K10K8502.244998.162673.912683.312675.01

clpeak

Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGIOPS, More Is Betterclpeak 1.1.2OpenCL Test: Integer ComputeIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two10002000300040005000SE +/- 7.85, N = 3SE +/- 1.75, N = 3SE +/- 0.28, N = 34442.403914.621453.221453.681454.291. (CXX) g++ options: -O3
OpenBenchmarking.orgGIOPS Per Watt, More Is Betterclpeak 1.1.2OpenCL Test: Integer ComputeIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two71421283528.5524.5630.1830.1829.29
MinAvgMaxIntel Arc A77034.8155.6192.1Intel Arc A75036.2159.4191.4Intel Arc A38016.648.260.0A38016.748.260.0A380 two16.749.660.0OpenBenchmarking.orgWatts, Fewer Is Betterclpeak 1.1.2GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 1919 Joules per run.2. Intel Arc A750: Approximate power consumption of 1966 Joules per run.3. A380 two: Approximate power consumption of 513 Joules per run.

OpenBenchmarking.orgGIOPS, More Is Betterclpeak 1.1.2OpenCL Test: Integer 24-bit ComputeIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two10002000300040005000SE +/- 3.59, N = 3SE +/- 1.12, N = 3SE +/- 0.08, N = 34435.603911.911454.551454.911454.041. (CXX) g++ options: -O3
OpenBenchmarking.orgGIOPS Per Watt, More Is Betterclpeak 1.1.2OpenCL Test: Integer 24-bit ComputeIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two71421283528.4224.9928.5929.8229.41
MinAvgMaxIntel Arc A77034.9156.1190.4Intel Arc A75035.8156.5190.2Intel Arc A38016.650.960.0A38016.648.860.0A380 two16.649.460.1OpenBenchmarking.orgWatts, Fewer Is Betterclpeak 1.1.2GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 1925 Joules per run.2. Intel Arc A750: Approximate power consumption of 1931 Joules per run.3. A380 two: Approximate power consumption of 527 Joules per run.

OpenBenchmarking.orgus, Fewer Is Betterclpeak 1.1.2OpenCL Test: Kernel LatencyIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two20406080100SE +/- 0.23, N = 10SE +/- 1.08, N = 15SE +/- 0.02, N = 1035.7096.6735.1032.6032.551. (CXX) g++ options: -O3
MinAvgMaxIntel Arc A77034.048.0100.7Intel Arc A75035.657.797.9A380 two16.619.429.6OpenBenchmarking.orgWatts, Fewer Is Betterclpeak 1.1.2GPU Power Consumption Monitor204060801001. Intel Arc A770: Approximate power consumption of 154 Joules per run.2. Intel Arc A750: Approximate power consumption of 231 Joules per run.3. A380 two: Approximate power consumption of 62 Joules per run.

Darktable

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

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 4.2.1Test: Server Room - Acceleration: OpenCLIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.51441.02881.54322.05762.572SE +/- 0.001, N = 12SE +/- 0.001, N = 11SE +/- 0.002, N = 90.7790.8742.2682.2862.280
MinAvgMaxIntel Arc A75035.945.079.9A380 two16.620.532.4OpenBenchmarking.orgWatts, Fewer Is BetterDarktable 4.2.1GPU Power Consumption Monitor204060801001. Intel Arc A750: Approximate power consumption of 131 Joules per run.2. A380 two: Approximate power consumption of 73 Joules per run.

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: S3DIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two50100150200250SE +/- 0.40, N = 3SE +/- 0.42, N = 3SE +/- 0.03, N = 3225.91209.9478.7578.7778.721. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: S3DIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two2468106.5705.7414.6994.7234.678
MinAvgMaxIntel Arc A77033.934.435.1Intel Arc A75035.636.637.5Intel Arc A38016.616.816.9A38016.516.716.9A380 two16.616.817.0OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor11223344551. Intel Arc A770: Approximate power consumption of 745 Joules per run.2. Intel Arc A750: Approximate power consumption of 817 Joules per run.3. A380 two: Approximate power consumption of 370 Joules per run.

clpeak

Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGBPS, More Is Betterclpeak 1.1.2OpenCL Test: Transfer Bandwidth enqueueReadBufferIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two1.28032.56063.84095.12126.4015SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 35.692.405.545.675.651. (CXX) g++ options: -O3
OpenBenchmarking.orgGBPS Per Watt, More Is Betterclpeak 1.1.2OpenCL Test: Transfer Bandwidth enqueueReadBufferIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.0610.1220.1830.2440.3050.1050.0390.2650.2700.271
MinAvgMaxIntel Arc A77034.454.466.9Intel Arc A75036.062.265.6Intel Arc A38016.620.922.4A38016.621.022.4A380 two16.520.822.4OpenBenchmarking.orgWatts, Fewer Is Betterclpeak 1.1.2GPU Power Consumption Monitor204060801001. Intel Arc A770: Approximate power consumption of 1268 Joules per run.2. Intel Arc A750: Approximate power consumption of 5042 Joules per run.3. A380 two: Approximate power consumption of 465 Joules per run.

Darktable

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

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 4.2.1Test: Boat - Acceleration: OpenCLIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.92121.84242.76363.68484.606SE +/- 0.003, N = 9SE +/- 0.003, N = 9SE +/- 0.002, N = 81.8582.1664.0824.0934.094
MinAvgMaxIntel Arc A75035.656.5146.6Intel Arc A38016.722.443.3A380 two16.626.143.3OpenBenchmarking.orgWatts, Fewer Is BetterDarktable 4.2.1GPU Power Consumption Monitor40801201602001. Intel Arc A750: Approximate power consumption of 213 Joules per run.2. A380 two: Approximate power consumption of 111 Joules per run.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 4.2.1Test: Server Rack - Acceleration: OpenCLIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.04970.09940.14910.19880.2485SE +/- 0.000, N = 14SE +/- 0.000, N = 14SE +/- 0.000, N = 140.1010.2070.2170.2180.221
MinAvgMaxIntel Arc A75035.738.548.2A380 two16.617.419.9OpenBenchmarking.orgWatts, Fewer Is BetterDarktable 4.2.1GPU Power Consumption Monitor14284256701. Intel Arc A750: Approximate power consumption of 91 Joules per run.2. A380 two: Approximate power consumption of 41 Joules per run.

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read BandwidthIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two2004006008001000SE +/- 1.38, N = 3SE +/- 0.24, N = 3SE +/- 0.16, N = 3980.58882.98451.81451.49450.401. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read BandwidthIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two36912156.4876.42412.00311.99111.934
MinAvgMaxIntel Arc A77034.6151.2191.5Intel Arc A75036.2137.5171.0Intel Arc A38016.737.651.0A38016.737.750.9A380 two16.637.750.0OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 5039 Joules per run.2. Intel Arc A750: Approximate power consumption of 4811 Joules per run.3. A380 two: Approximate power consumption of 1271 Joules per run.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: ReductionIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two306090120150SE +/- 0.15, N = 12SE +/- 0.03, N = 10SE +/- 0.04, N = 15150.5578.51113.20113.76113.251. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: ReductionIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two1.19142.38283.57424.76565.9572.9331.3555.1385.1635.295
MinAvgMaxIntel Arc A77034.351.3101.3Intel Arc A75035.757.997.4A380 two16.621.432.6OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor204060801001. Intel Arc A770: Approximate power consumption of 141 Joules per run.2. Intel Arc A750: Approximate power consumption of 191 Joules per run.3. A380 two: Approximate power consumption of 63 Joules per run.

Darktable

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

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 4.2.1Test: Masskrug - Acceleration: OpenCLIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two0.55711.11421.67132.22842.7855SE +/- 0.005, N = 11SE +/- 0.003, N = 10SE +/- 0.005, N = 91.4482.0762.4762.4232.439
MinAvgMaxIntel Arc A75035.643.871.9A380 two16.619.632.2OpenBenchmarking.orgWatts, Fewer Is BetterDarktable 4.2.1GPU Power Consumption Monitor204060801001. Intel Arc A750: Approximate power consumption of 149 Joules per run.2. A380 two: Approximate power consumption of 70 Joules per run.

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: 7-ZipIntel Arc A770Intel Arc A75050K100K150K200K250KSE +/- 311.92, N = 4SE +/- 302.77, N = 4247925246900
OpenBenchmarking.orgH/s Per Watt, More Is BetterHashcat 6.2.4Benchmark: 7-ZipIntel Arc A770Intel Arc A7504008001200160020001663.701786.18
MinAvgMaxIntel Arc A77035.1149.0191.1Intel Arc A75036.0138.2184.4OpenBenchmarking.orgWatts, Fewer Is BetterHashcat 6.2.4GPU Power Consumption Monitor501001502002501. Intel Arc A770: Approximate power consumption of 1453 Joules per run.2. Intel Arc A750: Approximate power consumption of 1244 Joules per run.

Benchmark: 7-Zip

Intel Arc A380: The test run did not produce a result.

A380: The test run did not produce a result.

A380 two: The test run did not produce a result.

GPU Power Consumption Monitor

OpenBenchmarking.orgWattsGPU Power Consumption MonitorPhoronix Test Suite System MonitoringIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two4080120160200Min: 33.9 / Avg: 117.73 / Max: 193Min: 34.8 / Avg: 108.15 / Max: 192.8Min: 16.5 / Avg: 39.59 / Max: 60.1Min: 16.5 / Avg: 40.45 / Max: 60.2Min: 16.5 / Avg: 41.62 / Max: 60.4

SHOC Scalable HeterOgeneous Computing

MinAvgMaxIntel Arc A77034.381.0147.3Intel Arc A75035.968.5120.0Intel Arc A38016.732.752.8A38016.733.152.9A380 two16.633.552.5OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor4080120160200

OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Max SP FlopsIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two10K20K30K40K50K44147.4811084.1047302.1448642.2845530.28

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Max SP FlopsIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two800K1600K2400K3200K4000KSE +/- 29520.59, N = 6SE +/- 87205.31, N = 15SE +/- 16360.41, N = 435749077587651546780161060015255581. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Max SP FlopsIntel Arc A770Intel Arc A750Intel Arc A380A380A380 two10K20K30K40K50K44147.4811084.1047302.1448642.2845530.28
MinAvgMaxIntel Arc A77034.381.0147.3Intel Arc A75035.968.5120.0Intel Arc A38016.732.752.8A38016.733.152.9A380 two16.633.552.5OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17GPU Power Consumption Monitor40801201602001. Intel Arc A770: Approximate power consumption of 459 Joules per run.2. Intel Arc A750: Approximate power consumption of 347 Joules per run.3. A380 two: Approximate power consumption of 260 Joules per run.

Intel Arc A770

Processor: AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (9922 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 2000GB, Graphics: Intel Arc A770 DG2 16GB (2400MHz), Audio: Intel DG2 Audio, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Ubuntu 23.04, Kernel: 6.2.2-060202-generic (x86_64), Desktop: GNOME Shell 43.2, Display Server: X Server 1.21.1.6 + Wayland, OpenGL: 4.6 Mesa 23.1.0-devel (git-0d8a54f 2023-03-06 lunar-oibaf-ppa), OpenCL: OpenCL 3.0, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 6 March 2023 15:49 by user pts.

Intel Arc A750

Processor: AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (9922 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 2000GB, Graphics: Intel Arc A750 DG2 8GB (2400MHz), Audio: Intel DG2 Audio, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Ubuntu 23.04, Kernel: 6.2.2-060202-generic (x86_64), Desktop: GNOME Shell 43.2, Display Server: X Server 1.21.1.6 + Wayland, OpenGL: 4.6 Mesa 23.1.0-devel (git-0d8a54f 2023-03-06 lunar-oibaf-ppa), OpenCL: OpenCL 3.0, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 6 March 2023 18:30 by user pts.

Intel Arc A380

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 7 March 2023 06:24 by user pts.

A380

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 7 March 2023 07:20 by user pts.

A380 two

Processor: AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (9922 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 2000GB, Graphics: Intel Arc A380 DG2 6GB (2450MHz), Audio: Intel DG2 Audio, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Ubuntu 23.04, Kernel: 6.2.2-060202-generic (x86_64), Desktop: GNOME Shell 43.2, Display Server: X Server 1.21.1.6 + Wayland, OpenGL: 4.6 Mesa 23.1.0-devel (git-0d8a54f 2023-03-06 lunar-oibaf-ppa), OpenCL: OpenCL 3.0, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-PG9DkL/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 7 March 2023 08:23 by user pts.