Vulkan Compute

Intel Core i7-2600K testing with a Gigabyte P67A-UD7-B3 (F7 BIOS) and NVIDIA GeForce RTX 3060 12GB on Microsoft Windows 7 Professional Build 7601 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 2111054-TJ-2110305TJ16
Jump To Table - Results

View

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

Limit displaying results to tests within:

NVIDIA GPU Compute 5 Tests
Vulkan Compute 5 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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
RX 480 4GB
October 30 2021
  1 Hour, 27 Minutes
RTX 3060
November 05 2021
  2 Hours, 40 Minutes
Invert Hiding All Results Option
  2 Hours, 3 Minutes
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):

RX 480 4GB

Processor: Intel Core i7-2600K (8 Cores), Motherboard: Gigabyte P67A-UD7-B3 (F7 BIOS), Memory: 0 x 4096 MB, Disk: 224GB INTEL SSDSC2CW240A3 + 1863GB WDC WD20EARX-008FB0, Graphics: AMD Radeon RX 480, Audio: Realtek HD Audio + AMD HD Audio Device, Monitor: S7A950D

OS: Microsoft Windows 7 Professional Build 7601, Kernel: 6.1 (x86_64), Display Driver: 27.20.14501.18003, OpenCL: OpenCL 2.1 AMD-APP (3188.4), File-System: NTFS, Screen Resolution: 1920x1080

Environment Notes: windows_tracing_flags=3
Security Notes: __user pointer sanitization: Disabled

RTX 3060

Processor: Intel Core i7-2600K (8 Cores), Motherboard: Gigabyte P67A-UD7-B3 (F7 BIOS), Memory: 0 x 4096 MB, Disk: 224GB INTEL SSDSC2CW240A3 + 1863GB WDC WD20EARX-008FB0, Graphics: NVIDIA GeForce RTX 3060 12GB, Audio: NVIDIA HD Audio + Realtek HD Audio, Monitor: S7A950D

OS: Microsoft Windows 7 Professional Build 7601, Kernel: 6.1 (x86_64), Display Driver: 472.12 (30.0.14.7212), OpenCL: OpenCL 3.0 CUDA 11.4.136 + OpenCL 2.1 AMD-APP (3188.4), File-System: NTFS, Screen Resolution: 1920x1080

Environment Notes: windows_tracing_flags=3
Security Notes: __user pointer sanitization: Disabled + KPTI Enabled: Yes + PTE Inversion: Yes

Vulkan ComputeProcessorMotherboardMemoryDiskGraphicsAudioMonitorOSKernelDisplay DriverOpenCLFile-SystemScreen ResolutionRX 480 4GBRTX 3060Intel Core i7-2600K (8 Cores)Gigabyte P67A-UD7-B3 (F7 BIOS)0 x 4096 MB224GB INTEL SSDSC2CW240A3 + 1863GB WDC WD20EARX-008FB0AMD Radeon RX 480Realtek HD Audio + AMD HD Audio DeviceS7A950DMicrosoft Windows 7 Professional Build 76016.1 (x86_64)27.20.14501.18003OpenCL 2.1 AMD-APP (3188.4)NTFS1920x1080NVIDIA GeForce RTX 3060 12GBNVIDIA HD Audio + Realtek HD Audio472.12 (30.0.14.7212)OpenCL 3.0 CUDA 11.4.136 + OpenCL 2.1 AMD-APP (3188.4)OpenBenchmarking.orgEnvironment Details- windows_tracing_flags=3Security Details- RX 480 4GB: __user pointer sanitization: Disabled- RTX 3060: __user pointer sanitization: Disabled + KPTI Enabled: Yes + PTE Inversion: Yes

RX 480 4GB vs. RTX 3060 ComparisonPhoronix Test SuiteBaseline+121.5%+121.5%+243%+243%+364.5%+364.5%+486%+486%486.1%485.2%163.2%116%100.6%67.7%24.1%int32-scalarint32-vec44x - Yes4x - No2x - 3 - Yesfp64-vec470.8%fp64-scalar70.1%fp32-vec4fp32-scalar2x - 3 - No6.7%vkpeakvkpeakRealSR-NCNNRealSR-NCNNWaifu2x-NCNN VulkanvkpeakvkpeakvkpeakvkpeakWaifu2x-NCNN VulkanRX 480 4GBRTX 3060

Vulkan Computevkpeak: fp32-scalarvkpeak: fp32-vec4vkpeak: fp64-scalarvkpeak: fp64-vec4vkpeak: int32-scalarvkpeak: int32-vec4realsr-ncnn: 4x - Norealsr-ncnn: 4x - Yeswaifu2x-ncnn: 2x - 3 - Nowaifu2x-ncnn: 2x - 3 - Yesvkfft: vkresample: 2x - Doublevkresample: 2x - Singlevkpeak: fp16-scalarvkpeak: fp16-vec4vkpeak: int16-scalarvkpeak: int16-vec4RX 480 4GBRTX 30605621.785522.92372.65372.671192.331192.4825.359185.0792.07911.3526977.639262.87219.08218.166988.836978.9211.74270.3302.2185.6582779247.08127.1867004.1813614.374628.096162.13OpenBenchmarking.org

vkpeak

Vkpeak is a Vulkan compute benchmark inspired by OpenCL's clpeak. Vkpeak provides Vulkan compute performance measurements for FP16 / FP32 / FP64 / INT16 / INT32 scalar and vec4 performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarRX 480 4GBRTX 306015003000450060007500SE +/- 10.76, N = 3SE +/- 20.38, N = 35621.786977.63
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarRX 480 4GBRTX 306012002400360048006000Min: 5607.48 / Avg: 5621.78 / Max: 5642.85Min: 6955.24 / Avg: 6977.63 / Max: 7018.31

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4RX 480 4GBRTX 30602K4K6K8K10KSE +/- 6.80, N = 3SE +/- 27.73, N = 35522.929262.87
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4RX 480 4GBRTX 306016003200480064008000Min: 5510.26 / Avg: 5522.92 / Max: 5533.57Min: 9234.88 / Avg: 9262.87 / Max: 9318.34

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarRX 480 4GBRTX 306080160240320400SE +/- 0.01, N = 3SE +/- 0.69, N = 3372.65219.08
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarRX 480 4GBRTX 306070140210280350Min: 372.64 / Avg: 372.65 / Max: 372.67Min: 218.38 / Avg: 219.08 / Max: 220.46

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4RX 480 4GBRTX 306080160240320400SE +/- 0.00, N = 2SE +/- 0.18, N = 3372.67218.16
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4RX 480 4GBRTX 306070140210280350Min: 372.67 / Avg: 372.67 / Max: 372.67Min: 217.8 / Avg: 218.16 / Max: 218.34

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarRX 480 4GBRTX 306015003000450060007500SE +/- 0.01, N = 3SE +/- 19.64, N = 31192.336988.83
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarRX 480 4GBRTX 306012002400360048006000Min: 1192.3 / Avg: 1192.33 / Max: 1192.35Min: 6949.55 / Avg: 6988.83 / Max: 7008.53

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4RX 480 4GBRTX 306015003000450060007500SE +/- 0.02, N = 3SE +/- 0.07, N = 31192.486978.92
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4RX 480 4GBRTX 306012002400360048006000Min: 1192.45 / Avg: 1192.48 / Max: 1192.52Min: 6978.83 / Avg: 6978.92 / Max: 6979.05

RealSR-NCNN

RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoRX 480 4GBRTX 3060612182430SE +/- 0.23, N = 7SE +/- 0.09, N = 1225.3611.74
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoRX 480 4GBRTX 3060612182430Min: 25.07 / Avg: 25.36 / Max: 26.72Min: 11.56 / Avg: 11.74 / Max: 12.68

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRX 480 4GBRTX 30604080120160200SE +/- 0.21, N = 3SE +/- 0.06, N = 3185.0870.33
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRX 480 4GBRTX 3060306090120150Min: 184.83 / Avg: 185.08 / Max: 185.5Min: 70.22 / Avg: 70.33 / Max: 70.43

Waifu2x-NCNN Vulkan

Waifu2x-NCNN is an NCNN neural network implementation of the Waifu2x converter project and accelerated using the Vulkan API. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: NoRX 480 4GBRTX 30600.49910.99821.49731.99642.4955SE +/- 0.054, N = 15SE +/- 0.051, N = 152.0792.218
OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: NoRX 480 4GBRTX 3060246810Min: 2.01 / Avg: 2.08 / Max: 2.84Min: 2.08 / Avg: 2.22 / Max: 2.92

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesRX 480 4GBRTX 30603691215SE +/- 0.051, N = 3SE +/- 0.060, N = 311.3525.658
OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesRX 480 4GBRTX 30603691215Min: 11.28 / Avg: 11.35 / Max: 11.45Min: 5.54 / Avg: 5.66 / Max: 5.73

VkFFT

VkFFT is a Fast Fourier Transform (FFT) Library that is GPU accelerated by means of the Vulkan API. The VkFFT benchmark runs FFT performance differences of many different sizes before returning an overall benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1RTX 30606K12K18K24K30KSE +/- 297.81, N = 427792
OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1RTX 30605K10K15K20K25KMin: 26907 / Avg: 27792.25 / Max: 28201

RX 480 4GB: The test quit with a non-zero exit status.

VkResample

VkResample is a Vulkan-based image upscaling library based on VkFFT. The sample input file is upscaling a 4K image to 8K using Vulkan-based GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: DoubleRTX 30601122334455SE +/- 0.38, N = 1547.08
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: DoubleRTX 30601020304050Min: 44.15 / Avg: 47.08 / Max: 49.3

Upscale: 2x - Precision: Double

RX 480 4GB: The test quit with a non-zero exit status.

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleRTX 3060612182430SE +/- 2.24, N = 1527.19
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleRTX 3060612182430Min: 22.93 / Avg: 27.19 / Max: 44.46

Upscale: 2x - Precision: Single

RX 480 4GB: The test quit with a non-zero exit status.

vkpeak

Vkpeak is a Vulkan compute benchmark inspired by OpenCL's clpeak. Vkpeak provides Vulkan compute performance measurements for FP16 / FP32 / FP64 / INT16 / INT32 scalar and vec4 performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarRTX 306015003000450060007500SE +/- 21.16, N = 37004.18

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4RTX 30603K6K9K12K15KSE +/- 39.61, N = 313614.37

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarRTX 306010002000300040005000SE +/- 3.82, N = 34628.09

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4RTX 306013002600390052006500SE +/- 0.07, N = 36162.13

17 Results Shown

vkpeak:
  fp32-scalar
  fp32-vec4
  fp64-scalar
  fp64-vec4
  int32-scalar
  int32-vec4
RealSR-NCNN:
  4x - No
  4x - Yes
Waifu2x-NCNN Vulkan:
  2x - 3 - No
  2x - 3 - Yes
VkFFT
VkResample:
  2x - Double
  2x - Single
vkpeak:
  fp16-scalar
  fp16-vec4
  int16-scalar
  int16-vec4

RX 480 4GB

Processor: Intel Core i7-2600K (8 Cores), Motherboard: Gigabyte P67A-UD7-B3 (F7 BIOS), Memory: 0 x 4096 MB, Disk: 224GB INTEL SSDSC2CW240A3 + 1863GB WDC WD20EARX-008FB0, Graphics: AMD Radeon RX 480, Audio: Realtek HD Audio + AMD HD Audio Device, Monitor: S7A950D

OS: Microsoft Windows 7 Professional Build 7601, Kernel: 6.1 (x86_64), Display Driver: 27.20.14501.18003, OpenCL: OpenCL 2.1 AMD-APP (3188.4), File-System: NTFS, Screen Resolution: 1920x1080

Environment Notes: windows_tracing_flags=3
Security Notes: __user pointer sanitization: Disabled

Testing initiated at 30 October 2021 08:46 by user Amaterasu.

RTX 3060

Processor: Intel Core i7-2600K (8 Cores), Motherboard: Gigabyte P67A-UD7-B3 (F7 BIOS), Memory: 0 x 4096 MB, Disk: 224GB INTEL SSDSC2CW240A3 + 1863GB WDC WD20EARX-008FB0, Graphics: NVIDIA GeForce RTX 3060 12GB, Audio: NVIDIA HD Audio + Realtek HD Audio, Monitor: S7A950D

OS: Microsoft Windows 7 Professional Build 7601, Kernel: 6.1 (x86_64), Display Driver: 472.12 (30.0.14.7212), OpenCL: OpenCL 3.0 CUDA 11.4.136 + OpenCL 2.1 AMD-APP (3188.4), File-System: NTFS, Screen Resolution: 1920x1080

Environment Notes: windows_tracing_flags=3
Security Notes: __user pointer sanitization: Disabled + KPTI Enabled: Yes + PTE Inversion: Yes

Testing initiated at 5 November 2021 04:22 by user Amaterasu.