NVIDIA GeForce RTX Benchmarks With 470 Linux Driver

NVIDIA 470 Linux driver benchmarks by Michael Larabel.

HTML result view exported from: https://openbenchmarking.org/result/2106232-PTS-MAYEO20228&grt.

ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionNVIDIA 465NVIDIA 470 RTX 3060 Ti RTX 3090 RTX 3060 Ti RTX 3090AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads)ASUS ROG CROSSHAIR VIII HERO (3402 BIOS)AMD Starship/Matisse16GB1000GB Sabrent Rocket 4.0 Plus + 2000GBNVIDIA GeForce RTX 3060 Ti 8GBNVIDIA GA104 HD AudioASUS VP28URealtek RTL8125 2.5GbE + Intel I211Ubuntu 21.045.11.0-17-generic (x86_64)GNOME Shell 3.38.4X Server 1.20.11NVIDIA 465.314.6.0OpenCL 3.0 CUDA 11.3.1161.2.168GCC 10.3.0 + CUDA 11.3ext43840x2160NVIDIA GeForce RTX 3090 24GBNVIDIA GA102 HD AudioASUS ROG CROSSHAIR VIII HERO (3501 BIOS)NVIDIA GeForce RTX 3060 Ti 8GBNVIDIA GA104 HD Audio5.11.0-18-generic (x86_64)NVIDIA 470.42.01OpenCL 3.0 CUDA 11.4.561.2.175NVIDIA GeForce RTX 3090 24GBNVIDIA GA102 HD AudioOpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa201009Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

basemark: Vulkan - 3840 x 2160 - Highbatman-knight: 3840 x 2160 - Highdirt-rally2: 2880 x 1620 - Ultradirt-rally2: 3840 x 1600 - Ultraf12020: 3840 x 2160 - Ultrahitman2: 2560 x 1440 - Ultrahitman2: 3840 x 2160 - Ultrarealsr-ncnn: 4x - Yesshadowofthetombraider: 2560 x 1440 - Highest - Offshadowofthetombraider: 3840 x 2160 - Highest - Offstrange-brigade: 3840 x 2160 - Ultraunigine-heaven: 2560 x 1440 - Fullscreen - OpenGLunigine-heaven: 3840 x 2160 - Fullscreen - OpenGLunigine-super: 2560 x 1440 - Fullscreen - High - OpenGLvkfft: vkpeak: fp32-scalarvkpeak: fp32-vec4vkpeak: fp16-scalarvkpeak: fp16-vec4vkpeak: fp64-scalarvkpeak: fp64-vec4vkpeak: int32-scalarvkpeak: int32-vec4vkpeak: int16-scalarvkpeak: int16-vec4vkresample: 2x - SingleNVIDIA 465NVIDIA 470 RTX 3060 Ti RTX 3090 RTX 3060 Ti RTX 3090103.4590.5584.69611672.652890111100.2058.6351.561107.159.9119.3131.96656.510367.4333679671.4312805.209691.4519431.27302.96304.849689.789649.656405.878471.8717.975190.86156.79142.150375123.940160169140.7495.8228.233173.6103.8213.0233.371108.451117.84367720703.7927391.4420640.7541421.96647.11648.6820613.6320516.9913607.8116889.779.30497.7785.3780.08728368.79708910594.9355.6854.835101.256.7109.1126.50353.707764.7314739672.2812823.869704.5418869.90303.65304.909698.089653.716409.958483.6819.073190.87158.34143.549087124.714999172140.6096.4528.462175.5104.5208.9234.223108.427117.94474220757.0127390.3020655.2940258.60648.70650.8520686.7520598.8513653.9416881.679.290OpenBenchmarking.org

Basemark GPU

Renderer: Vulkan - Resolution: 3840 x 2160 - Graphics Preset: High

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFPS, More Is BetterBasemark GPU 1.2Renderer: Vulkan - Resolution: 3840 x 2160 - Graphics Preset: HighRTX 3060 TiRTX 30904080120160200SE +/- 0.08, N = 3SE +/- 0.17, N = 3SE +/- 0.04, N = 3SE +/- 0.15, N = 3103.45190.8697.77190.87

Basemark GPU

Renderer: Vulkan - Resolution: 3840 x 2160 - Graphics Preset: High

OpenBenchmarking.orgFPS Per Watt, More Is BetterBasemark GPU 1.2Renderer: Vulkan - Resolution: 3840 x 2160 - Graphics Preset: HighRTX 3060 TiRTX 30900.15570.31140.46710.62280.77850.6040.692

Basemark GPU

GPU Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterBasemark GPU 1.2GPU Power Consumption MonitorRTX 3060 TiRTX 309060120180240300Min: 22.01 / Avg: 171.34 / Max: 199.2Min: 10.81 / Avg: 275.66 / Max: 350.12

Batman: Arkham Knight

Resolution: 3840 x 2160 - Quality: High

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterBatman: Arkham KnightResolution: 3840 x 2160 - Quality: HighRTX 3060 TiRTX 3090306090120150SE +/- 0.14, N = 3SE +/- 0.87, N = 3SE +/- 0.36, N = 3SE +/- 0.42, N = 390.55156.7985.37158.34

Batman: Arkham Knight

Resolution: 3840 x 2160 - Quality: High

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterBatman: Arkham KnightResolution: 3840 x 2160 - Quality: HighRTX 3060 TiRTX 30900.13280.26560.39840.53120.6640.5860.590

DiRT Rally 2.0

Resolution: 2880 x 1620 - Graphics Preset: Ultra

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterDiRT Rally 2.0Resolution: 2880 x 1620 - Graphics Preset: UltraRTX 3060 TiRTX 3090306090120150SE +/- 0.06, N = 3SE +/- 0.33, N = 3SE +/- 0.01, N = 3SE +/- 0.15, N = 384.70142.1580.09143.55

DiRT Rally 2.0

Resolution: 3840 x 1600 - Graphics Preset: Ultra

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterDiRT Rally 2.0Resolution: 3840 x 1600 - Graphics Preset: UltraRTX 3060 TiRTX 3090306090120150SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 372.65123.9468.80124.71

DiRT Rally 2.0

Resolution: 3840 x 1600 - Graphics Preset: Ultra

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterDiRT Rally 2.0Resolution: 3840 x 1600 - Graphics Preset: UltraRTX 3060 TiRTX 30900.090.180.270.360.450.400.39

DiRT Rally 2.0

GPU Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterDiRT Rally 2.0GPU Power Consumption MonitorRTX 3060 TiRTX 309060120180240300Min: 21.94 / Avg: 181.66 / Max: 199.56Min: 11.91 / Avg: 317.55 / Max: 348.95

F1 2020

Resolution: 3840 x 2160 - Graphics Preset: Ultra

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterF1 2020Resolution: 3840 x 2160 - Graphics Preset: UltraRTX 3060 TiRTX 30904080120160200SE +/- 0.67, N = 3SE +/- 0.67, N = 3111169105172

HITMAN 2

Resolution: 2560 x 1440 - Quality Level: Ultra

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterHITMAN 2Resolution: 2560 x 1440 - Quality Level: UltraRTX 3060 TiRTX 3090306090120150SE +/- 0.13, N = 3SE +/- 0.85, N = 3SE +/- 0.13, N = 3SE +/- 0.32, N = 3100.20140.7494.93140.60

HITMAN 2

Resolution: 3840 x 2160 - Quality Level: Ultra

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterHITMAN 2Resolution: 3840 x 2160 - Quality Level: UltraRTX 3060 TiRTX 309020406080100SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.26, N = 358.6395.8255.6896.45

RealSR-NCNN

Scale: 4x - TAA: Yes

NVIDIA 465NVIDIA 470OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRTX 3060 TiRTX 30901224364860SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 351.5628.2354.8428.46

Shadow of the Tomb Raider

Resolution: 2560 x 1440 - Graphics Preset: Highest - Anti Aliasing: Off

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterShadow of the Tomb RaiderResolution: 2560 x 1440 - Graphics Preset: Highest - Anti Aliasing: OffRTX 3060 TiRTX 30904080120160200SE +/- 0.07, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 3SE +/- 0.27, N = 3107.1173.6101.2175.5

Shadow of the Tomb Raider

Resolution: 2560 x 1440 - Graphics Preset: Highest - Anti Aliasing: Off

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterShadow of the Tomb RaiderResolution: 2560 x 1440 - Graphics Preset: Highest - Anti Aliasing: OffRTX 3060 TiRTX 30900.13640.27280.40920.54560.6820.6060.565

Shadow of the Tomb Raider

Resolution: 3840 x 2160 - Graphics Preset: Highest - Anti Aliasing: Off

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterShadow of the Tomb RaiderResolution: 3840 x 2160 - Graphics Preset: Highest - Anti Aliasing: OffRTX 3060 TiRTX 309020406080100SE +/- 0.00, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 359.9103.856.7104.5

Shadow of the Tomb Raider

Resolution: 3840 x 2160 - Graphics Preset: Highest - Anti Aliasing: Off

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterShadow of the Tomb RaiderResolution: 3840 x 2160 - Graphics Preset: Highest - Anti Aliasing: OffRTX 3060 TiRTX 30900.07380.14760.22140.29520.3690.3260.328

Shadow of the Tomb Raider

GPU Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterShadow of the Tomb RaiderGPU Power Consumption MonitorRTX 3060 TiRTX 309060120180240300Min: 22.36 / Avg: 183.73 / Max: 199.81Min: 14.65 / Avg: 316.49 / Max: 347.82

Strange Brigade

Resolution: 3840 x 2160 - Graphics Setting: Ultra

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterStrange BrigadeResolution: 3840 x 2160 - Graphics Setting: UltraRTX 3060 TiRTX 309050100150200250SE +/- 0.28, N = 3SE +/- 0.85, N = 3SE +/- 0.38, N = 3SE +/- 0.23, N = 3119.3213.0109.1208.9

Unigine Heaven

Resolution: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGL

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Heaven 4.0Resolution: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGLRTX 3060 TiRTX 309050100150200250SE +/- 0.07, N = 3SE +/- 0.18, N = 3SE +/- 0.05, N = 3SE +/- 0.25, N = 3131.97233.37126.50234.22

Unigine Heaven

Resolution: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGL

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterUnigine Heaven 4.0Resolution: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGLRTX 3060 TiRTX 30900.15750.3150.47250.630.78750.6960.700

Unigine Heaven

Resolution: 3840 x 2160 - Mode: Fullscreen - Renderer: OpenGL

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Heaven 4.0Resolution: 3840 x 2160 - Mode: Fullscreen - Renderer: OpenGLRTX 3060 TiRTX 309020406080100SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 356.51108.4553.71108.43

Unigine Heaven

Resolution: 3840 x 2160 - Mode: Fullscreen - Renderer: OpenGL

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterUnigine Heaven 4.0Resolution: 3840 x 2160 - Mode: Fullscreen - Renderer: OpenGLRTX 3060 TiRTX 30900.07290.14580.21870.29160.36450.2970.324

Unigine Heaven

GPU Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterUnigine Heaven 4.0GPU Power Consumption MonitorRTX 3060 TiRTX 309060120180240300Min: 22.1 / Avg: 190.54 / Max: 199.85Min: 19.8 / Avg: 334.45 / Max: 350.33

Unigine Superposition

Resolution: 2560 x 1440 - Mode: Fullscreen - Quality: High - Renderer: OpenGL

NVIDIA 465NVIDIA 470OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 2560 x 1440 - Mode: Fullscreen - Quality: High - Renderer: OpenGLRTX 3060 TiRTX 3090306090120150SE +/- 0.03, N = 3SE +/- 0.12, N = 3SE +/- 0.00, N = 3SE +/- 0.15, N = 367.4117.864.7117.9

Unigine Superposition

Resolution: 2560 x 1440 - Mode: Fullscreen - Quality: High - Renderer: OpenGL

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterUnigine Superposition 1.0Resolution: 2560 x 1440 - Mode: Fullscreen - Quality: High - Renderer: OpenGLRTX 3060 TiRTX 30900.08460.16920.25380.33840.4230.3750.376

Unigine Superposition

GPU Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterUnigine Superposition 1.0GPU Power Consumption MonitorRTX 3060 TiRTX 309060120180240300Min: 21.49 / Avg: 179.81 / Max: 199.52Min: 13.13 / Avg: 313.54 / Max: 348.38

VkFFT

NVIDIA 465NVIDIA 470OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1RTX 3060 TiRTX 309010K20K30K40K50KSE +/- 459.48, N = 3SE +/- 531.03, N = 4SE +/- 126.89, N = 3SE +/- 426.11, N = 9333674367731473447421. (CXX) g++ options: -O3 -pthread

VkFFT

OpenBenchmarking.orgBenchmark Score Per Watt, More Is BetterVkFFT 1.1.1RTX 3060 TiRTX 309090180270360450399.01291.23

VkFFT

GPU Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterVkFFT 1.1.1GPU Power Consumption MonitorRTX 3060 TiRTX 30904080120160200Min: 21.03 / Avg: 83.62 / Max: 130.36Min: 9.9 / Avg: 149.98 / Max: 224.48

vkpeak

fp32-scalar

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarRTX 3060 TiRTX 30904K8K12K16K20KSE +/- 14.79, N = 3SE +/- 49.55, N = 3SE +/- 7.07, N = 3SE +/- 30.07, N = 39671.4320703.799672.2820757.01

vkpeak

fp32-vec4

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4RTX 3060 TiRTX 30906K12K18K24K30KSE +/- 1.54, N = 3SE +/- 1.45, N = 3SE +/- 11.74, N = 3SE +/- 1.02, N = 312805.2027391.4412823.8627390.30

vkpeak

fp16-scalar

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarRTX 3060 TiRTX 30904K8K12K16K20KSE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 8.80, N = 3SE +/- 29.01, N = 39691.4520640.759704.5420655.29

vkpeak

fp16-vec4

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4RTX 3060 TiRTX 30909K18K27K36K45KSE +/- 0.12, N = 3SE +/- 46.65, N = 3SE +/- 21.47, N = 3SE +/- 1.13, N = 319431.2741421.9618869.9040258.60

vkpeak

fp64-scalar

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarRTX 3060 TiRTX 3090140280420560700SE +/- 0.03, N = 3SE +/- 0.82, N = 3SE +/- 0.61, N = 3SE +/- 0.01, N = 3302.96647.11303.65648.70

vkpeak

fp64-vec4

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4RTX 3060 TiRTX 3090140280420560700SE +/- 0.00, N = 3SE +/- 0.06, N = 3SE +/- 0.00, N = 3SE +/- 0.12, N = 3304.84648.68304.90650.85

vkpeak

int32-scalar

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarRTX 3060 TiRTX 30904K8K12K16K20KSE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 1.09, N = 39689.7820613.639698.0820686.75

vkpeak

int32-vec4

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4RTX 3060 TiRTX 30904K8K12K16K20KSE +/- 0.07, N = 3SE +/- 0.78, N = 3SE +/- 0.05, N = 3SE +/- 1.54, N = 39649.6520516.999653.7120598.85

vkpeak

int16-scalar

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarRTX 3060 TiRTX 30903K6K9K12K15KSE +/- 0.18, N = 3SE +/- 0.22, N = 3SE +/- 0.08, N = 3SE +/- 1.35, N = 36405.8713607.816409.9513653.94

vkpeak

int16-vec4

NVIDIA 465NVIDIA 470OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4RTX 3060 TiRTX 30904K8K12K16K20KSE +/- 0.07, N = 3SE +/- 8.04, N = 3SE +/- 8.53, N = 3SE +/- 0.18, N = 38471.8716889.778483.6816881.67

VkResample

Upscale: 2x - Precision: Single

NVIDIA 465NVIDIA 470OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleRTX 3060 TiRTX 3090510152025SE +/- 0.011, N = 5SE +/- 0.006, N = 5SE +/- 0.014, N = 3SE +/- 0.008, N = 317.9759.30419.0739.2901. (CXX) g++ options: -O3 -pthread

Phoronix Test Suite v10.8.5