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.

To run this test with the Phoronix Test Suite, the basic command is: phoronix-test-suite benchmark vkpeak.

Project Site

github.com

Source Repository

github.com

Test Created

24 April 2021

Last Updated

1 August 2023

Test Maintainer

Michael Larabel

Test Type

Graphics

Average Install Time

2 Seconds

Average Run Time

11 Minutes, 9 Seconds

Test Dependencies

Vulkan

Accolades

10k+ Downloads

Supported Platforms

Supported Architectures

x86_64

* Uploading of benchmark result data to OpenBenchmarking.org is always optional (opt-in) via the Phoronix Test Suite for users wishing to share their results publicly.
** Data based on those opting to upload their test results to OpenBenchmarking.org and users enabling the opt-in anonymous statistics reporting while running benchmarks from an Internet-connected platform.
Data updated weekly as of 27 September 2023.

Revision History

pts/vkpeak-1.1.0 [View Source] Tue, 01 Aug 2023 11:42:09 GMT
Update against latest upstream.

pts/vkpeak-1.0.2 [View Source] Sat, 01 May 2021 06:41:01 GMT
Update per https://github.com/phoronix-test-suite/test-profiles/pull/194

pts/vkpeak-1.0.1 [View Source] Sat, 24 Apr 2021 11:55:54 GMT
Update hashes, add macOS support while at it.

pts/vkpeak-1.0.0 [View Source] Sat, 24 Apr 2021 11:34:45 GMT
Initial commit of vkpeak test profile, per https://github.com/phoronix-test-suite/test-profiles/issues/192

Suites Using This Test

NVIDIA GPU Compute

Vulkan Compute

Performance Metrics

Analyze Test Configuration:

vkpeak 20230730

fp32-scalar

OpenBenchmarking.org metrics for this test profile configuration based on 137 public results since 1 August 2023 with the latest data as of 20 September 2023.

Below is an overview of the generalized performance for components where there is sufficient statistically significant data based upon user-uploaded results. It is important to keep in mind particularly in the Linux/open-source space there can be vastly different OS configurations, with this overview intended to offer just general guidance as to the performance expectations.

Component

Percentile Rank

# Compatible Public Results

GFLOPS (Average)

NVIDIA GeForce RTX 4090

100th

44816 ^{+/- 61}

AMD Radeon RX 7900 XTX

93rd

25498 ^{+/- 168}

AMD Radeon RX 6800 XT

87th

21661 ^{+/- 653}

NVIDIA GeForce RTX 3090

85th

20968 ^{+/- 253}

NVIDIA GeForce RTX 3080 Ti

79th

19563 ^{+/- 37}

Mid-Tier

75th

< 17506

NVIDIA Quadro RTX 6000

72nd

16637 ^{+/- 41}

AMD Radeon RX 6700 XT

61st

12987 ^{+/- 193}

NVIDIA GeForce RTX 3070

56th

11521 ^{+/- 4}

Median

50th

8532

MSI NVIDIA GeForce RTX 4060

48th

8526 ^{+/- 8}

AMD Radeon RX 5700

44th

7853 ^{+/- 15}

eVGA NVIDIA GeForce RTX 3060

37th

6824 ^{+/- 13}

ASUS AMD Phoenix1

32nd

3287 ^{+/- 19}

AMD Radeon 680M

29th

2991 ^{+/- 209}

Low-Tier

25th

< 2143

AMD Custom GPU 0405

21st

1639 ^{+/- 1}

Intel Xe TGL GT2

19th

935

Intel UHD 620 WHL GT2

15th

268

GFX1103

14th

192

NVIDIA GeForce 920M

8th

187

AMD Radeon HD 8200

8th

115

Based on OpenBenchmarking.org data, the selected test / test configuration (vkpeak 20230730 - fp32-scalar) has an average run-time of 9 minutes. By default this test profile is set to run at least 3 times but may increase if the standard deviation exceeds pre-defined defaults or other calculations deem additional runs necessary for greater statistical accuracy of the result.

Based on public OpenBenchmarking.org results, the selected test / test configuration has an average standard deviation of 0.1%.

Notable Instruction Set Usage

Notable instruction set extensions supported by this test, based on an automatic analysis by the Phoronix Test Suite / OpenBenchmarking.org analytics engine.

Instruction Set

Support

Instructions Detected

SSE2 (SSE2)

Used by default on supported hardware.

MOVDQA MOVDQU MOVAPD SUBSD COMISD MULSD ADDSD DIVSD MAXSD CVTSI2SD PUNPCKLQDQ PSRLDQ MOVD CVTTSD2SI PSHUFD SQRTSD PADDQ PSUBQ SHUFPD CVTSD2SS ANDPD UCOMISD CVTSS2SD MOVUPD MOVHPD MULPD UNPCKHPD CMPNLESD ANDNPD ORPD XORPD PUNPCKHQDQ MINSD SUBPD

Advanced Vector Extensions (AVX)

Used by default on supported hardware.
Found on Intel processors since Sandy Bridge (2011).
Found on AMD processors since Bulldozer (2011).

VZEROUPPER VINSERTF128 VEXTRACTF128 VPERM2F128

Advanced Vector Extensions 512 (AVX512)

Used by default on supported hardware.

(ZMM REGISTER USE)

Last automated analysis: 18 September 2023

This test profile binary relies on the shared libraries libvulkan.so.1, libpthread.so.0, libm.so.6, libc.so.6.

Tested CPU Architectures

This benchmark has been successfully tested on the below mentioned architectures. The CPU architectures listed is where successful OpenBenchmarking.org result uploads occurred, namely for helping to determine if a given test is compatible with various alternative CPU architectures.

CPU Architecture

Kernel Identifier

Verified On

Intel / AMD x86 64-bit