dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content.

To run this test with the Phoronix Test Suite, the basic command is: phoronix-test-suite benchmark dav1d.

Project Site

videolan.org

Source Repository

code.videolan.org

Test Created

18 December 2018

Last Updated

18 March 2022

Test Maintainer

Michael Larabel 

Test Type

Processor

Average Install Time

3 Minutes

Average Run Time

10 Minutes, 8 Seconds

Test Dependencies

C/C++ Compiler Toolchain + Yasm Assembler + Meson Build System

Accolades

80k+ Downloads

Supported Platforms


Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page Views ***OpenBenchmarking.orgEventsdav1d Popularity Statisticspts/dav1d2018.122019.022019.042019.062019.082019.102019.122020.022020.042020.062020.082020.102020.122021.022021.042021.062021.082021.102021.122022.022022.042022.062022.085K10K15K20K25K
* 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.
*** Test profile page view reporting began March 2021.
Data current as of 14 August 2022.
Summer Nature 4K24.8%Chimera 1080p 10-bit23.5%Chimera 1080p25.8%Summer Nature 1080p25.9%Video Input Option PopularityOpenBenchmarking.org

Revision History

pts/dav1d-1.12.0   [View Source]   Fri, 18 Mar 2022 12:15:53 GMT
Update against dav1d 1.0 upstream release.

pts/dav1d-1.11.2   [View Source]   Sun, 06 Feb 2022 13:24:36 GMT
Update download mirrors, closes: https://github.com/phoronix-test-suite/phoronix-test-suite/issues/597

pts/dav1d-1.11.1   [View Source]   Sun, 17 Oct 2021 11:42:30 GMT
Update Windows FFmpeg build to newer location, newer version. Closes: https://github.com/phoronix-test-suite/phoronix-test-suite/pull/553

pts/dav1d-1.11.0   [View Source]   Sat, 04 Sep 2021 06:04:51 GMT
Update against dav1d 0.9.2 upstream.

pts/dav1d-1.10.0   [View Source]   Sun, 01 Aug 2021 14:55:34 GMT
Update against dav1d 0.9.1 upstream.

pts/dav1d-1.9.0   [View Source]   Sun, 16 May 2021 13:39:23 GMT
Update against dav1d 0.9 upstream.

pts/dav1d-1.8.0   [View Source]   Mon, 22 Feb 2021 06:31:01 GMT
Update against new upstream dav1d 0.8.2.

pts/dav1d-1.7.1   [View Source]   Sat, 09 Jan 2021 16:54:35 GMT
Update to latest dav1d hash.

pts/dav1d-1.7.0   [View Source]   Sun, 03 Jan 2021 08:43:19 GMT
Update against new upstream dav1d 0.8.1.

pts/dav1d-1.6.0   [View Source]   Thu, 21 May 2020 14:58:06 GMT
Update against upstream dav1d 0.7.

pts/dav1d-1.5.0   [View Source]   Tue, 10 Mar 2020 06:45:56 GMT
Update against dav1d 0.6.0 upstream.

pts/dav1d-1.4.1   [View Source]   Fri, 22 Nov 2019 19:26:24 GMT
Update download links per https://github.com/phoronix-test-suite/test-profiles/issues/113

pts/dav1d-1.4.0   [View Source]   Fri, 11 Oct 2019 07:04:25 GMT
Update against dav1d 0.5 upstream, add Windows support.

pts/dav1d-1.3.0   [View Source]   Tue, 24 Sep 2019 13:53:19 GMT
Update against dav1d 0.4, make use of new FPS reporting, add Chimera video options. https://github.com/phoronix-test-suite/test-profiles/pull/103

pts/dav1d-1.2.0   [View Source]   Fri, 10 May 2019 10:52:01 GMT
Update against dav1d 0.3 https://github.com/phoronix-test-suite/test-profiles/pull/79

pts/dav1d-1.1.0   [View Source]   Tue, 15 Jan 2019 11:23:31 GMT
Use null mixer for video output.

pts/dav1d-1.0.0   [View Source]   Tue, 18 Dec 2018 14:52:01 GMT
Initial commit of dav1d AV1 video decoder benchmark.

Suites Using This Test

C/C++ Compiler Tests

Video Encoding

Encoding

Multi-Core

CPU Massive

AV1

Server CPU Tests

Creator Workloads


Performance Metrics

Analyze Test Configuration:

dav1d 1.0

Video Input: Summer Nature 1080p

OpenBenchmarking.org metrics for this test profile configuration based on 267 public results since 18 March 2022 with the latest data as of 13 August 2022.

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
FPS (Average)
98th
17
1098 +/- 78
88th
4
958 +/- 3
85th
4
902 +/- 5
85th
3
899 +/- 4
83rd
15
877 +/- 33
78th
5
834 +/- 10
Mid-Tier
75th
< 796
73rd
3
767 +/- 50
72nd
3
758 +/- 16
68th
3
683 +/- 4
66th
3
659 +/- 5
63rd
4
623 +/- 2
63rd
3
622 +/- 1
58th
6
595 +/- 14
55th
4
525 +/- 4
54th
3
521 +/- 4
Median
50th
492
39th
3
425 +/- 5
38th
3
408 +/- 5
36th
3
385 +/- 6
33rd
3
370 +/- 1
30th
3
349 +/- 4
27th
3
338 +/- 1
Low-Tier
25th
< 333
24th
3
300 +/- 13
19th
3
236 +/- 1
16th
3
215 +/- 1
OpenBenchmarking.orgDistribution Of Public Results - Video Input: Summer Nature 1080p260 Results Range From 49 To 1177 FPS4983117151185219253287321355389423457491525559593627661695729763797831865899933967100110351069110311371171120548121620

Based on OpenBenchmarking.org data, the selected test / test configuration (dav1d 1.0 - Video Input: Summer Nature 1080p) has an average run-time of 2 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.

OpenBenchmarking.orgMinutesTime Required To Complete BenchmarkVideo Input: Summer Nature 1080pRun-Time246810Min: 1 / Avg: 1.21 / Max: 4

Based on public OpenBenchmarking.org results, the selected test / test configuration has an average standard deviation of 0.3%.

OpenBenchmarking.orgPercent, Fewer Is BetterAverage Deviation Between RunsVideo Input: Summer Nature 1080pDeviation246810Min: 0 / Avg: 0.32 / Max: 3

Does It Scale Well With Increasing Cores?

No, based on the automated analysis of the collected public benchmark data, this test / test settings does not generally scale well with increasing CPU core counts. Data based on publicly available results for this test / test settings, separated by vendor, result divided by the reference CPU clock speed, grouped by matching physical CPU core count, and normalized against the smallest core count tested from each vendor for each CPU having a sufficient number of test samples and statistically significant data.

IntelAMDOpenBenchmarking.orgRelative Core Scaling To Basedav1d CPU Core ScalingVideo Input: Summer Nature 1080p468320.48540.97081.45621.94162.427

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.
 
PUNPCKLQDQ CVTSI2SD DIVSD COMISD CVTTSD2SI SUBSD MOVAPD MULSD ADDSD MOVDQU PSUBQ ANDPD ORPD MOVD MOVDQA PSHUFLW PSHUFD PSLLDQ PSRLDQ PUNPCKHQDQ SHUFPD PSHUFHW PADDQ PMULUDQ
SSE3 (SSE3)
Used by default on supported hardware.
 
MOVDDUP
SSSE3 (SSSE3)
Used by default on supported hardware.
 
PSHUFB PABSB PSIGNB PMADDUBSW PMULHRSW PHADDD PHADDW PALIGNR PABSW PSIGNW PHADDSW
SSE 4.2 (SSE4_2)
Used by default on supported hardware.
Found on Intel processors since at least 2010.
Found on AMD processors since Bulldozer (2011).

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

 
VZEROUPPER VINSERTF128 VEXTRACTF128 VBROADCASTSD VBROADCASTSS
Used by default on supported hardware.
Found on Intel processors since Haswell (2013).
Found on AMD processors since Excavator (2016).

 
VPBROADCASTW VINSERTI128 VPBROADCASTB VPBROADCASTD VEXTRACTI128 VPBLENDD VPERMQ VPBROADCASTQ VPERM2I128 VPERMD VPGATHERDD VPGATHERDQ VPSRLVD VPSLLVD
Advanced Vector Extensions 512 (AVX512)
Used by default on supported hardware.
 
(ZMM REGISTER USE)
AVX Vector Neural Network Instructions (AVX-VNNI)
Used by default on supported hardware.
 
VPDPBUSD VPDPWSSD
The test / benchmark does honor compiler flag changes.
Last automated analysis: 18 March 2022

This test profile binary relies on the shared libraries libdav1d.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
x86_64
(Many Processors)
Loongson LoongArch 64-bit
loongarch64
Loongson-3C5000L
ARMv8 64-bit
aarch64
ARMv8 Cortex-A72 4-Core, ARMv8 Cortex-N2, ARMv8 Neoverse-N1 4-Core, ARMv8 Neoverse-V1 4-Core, Ampere ARMv8 Neoverse-N1 256-Core, Apple M2