AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library.

To run this test with the Phoronix Test Suite, the basic command is: phoronix-test-suite benchmark aom-av1.

Source Repository

aomedia.googlesource.com

Test Created

30 March 2018

Last Updated

3 September 2023

Test Maintainer

Michael Larabel 

Test Type

Processor

Average Install Time

2 Minutes, 49 Seconds

Average Run Time

3 Minutes, 30 Seconds

Test Dependencies

C/C++ Compiler Toolchain + 7-Zip / p7zip + CMake + PERL + Yasm Assembler

Accolades

100k+ Downloads

Supported Platforms


Reported Installs **Public Result Uploads *Reported Test Completions **Test Profile Page Views ***OpenBenchmarking.orgEventsAOM AV1 Popularity Statisticspts/aom-av12018.042018.062018.082018.102018.122019.022019.042019.062019.082019.102019.122020.022020.042020.062020.082020.102020.122021.022021.042021.062021.082021.102021.122022.022022.042022.062022.082022.102022.122023.022023.042023.062023.082023.102023.122024.0220K40K60K80K100K
* 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 updated weekly as of 25 March 2024.
Speed 0 Two-Pass11.1%Speed 6 Realtime12.9%Speed 9 Realtime13.2%Speed 11 Realtime13.2%Speed 6 Two-Pass12.6%Speed 8 Realtime13.0%Speed 4 Two-Pass10.9%Speed 10 Realtime13.2%Encoder Mode Option PopularityOpenBenchmarking.org
Bosphorus 4K50.5%Bosphorus 1080p49.5%Input Option PopularityOpenBenchmarking.org

Revision History

pts/aom-av1-3.7.0   [View Source]   Sun, 03 Sep 2023 17:00:19 GMT
Update against libaom 3.7 upstream.

pts/aom-av1-3.6.0   [View Source]   Tue, 07 Feb 2023 18:34:15 GMT
Update against upstream AOM v3.6 release.

pts/aom-av1-3.5.0   [View Source]   Thu, 22 Sep 2022 09:57:34 GMT
Update against AOM-AV1 3.5, tune parameters too.

pts/aom-av1-3.4.0   [View Source]   Sat, 18 Jun 2022 05:22:19 GMT
Update against AOM v3.4 upstream.

pts/aom-av1-3.3.0   [View Source]   Thu, 17 Feb 2022 14:17:26 GMT
Update against AOM-AV1 3.3 upstream.

pts/aom-av1-3.2.0   [View Source]   Thu, 14 Oct 2021 19:32:32 GMT
Update against upstream libaom v3.2.0 release, drop Windows support for now since the Appveyor builds stopped, also unify versioning.

pts/aom-av1-2.4.0   [View Source]   Thu, 06 May 2021 12:26:26 GMT
Update against libaom 3.1 upstream.

pts/aom-av1-2.3.0   [View Source]   Wed, 24 Mar 2021 09:19:16 GMT
Update against v3.0 upstream, add 4K video option too now that libaom is faster than before...

pts/aom-av1-2.2.0   [View Source]   Fri, 12 Mar 2021 08:39:44 GMT
Update against libaom 2.1-rc1 Git state, increase tile settings.

pts/aom-av1-2.1.2   [View Source]   Sat, 12 Sep 2020 12:42:09 GMT
Speedup installation per https://github.com/phoronix-test-suite/test-profiles/issues/158

pts/aom-av1-2.1.1   [View Source]   Tue, 19 May 2020 14:02:41 GMT
Update argument options.

pts/aom-av1-2.1.0   [View Source]   Tue, 19 May 2020 09:05:19 GMT
Update against libaom 2.0.0.

pts/aom-av1-2.0.2   [View Source]   Fri, 10 Jan 2020 14:03:26 GMT
Cap number of threads for AOM encoding to 64 threads, otherwise the program errors out due to its MAX_NUM_THREADS.

pts/aom-av1-2.0.1   [View Source]   Fri, 10 Jan 2020 13:52:11 GMT
Add Windows support.

pts/aom-av1-2.0.0   [View Source]   Fri, 10 Jan 2020 13:27:51 GMT
Updates per https://github.com/phoronix-test-suite/test-profiles/pull/112 plus also rebase to latest AOM Git.

pts/aom-av1-1.2.0   [View Source]   Mon, 16 Sep 2019 14:50:04 GMT
Update per https://github.com/phoronix-test-suite/test-profiles/pull/101 re-base to AOM Git as of 2019-09-16.

pts/aom-av1-1.1.0   [View Source]   Mon, 11 Feb 2019 15:06:58 GMT
Updates per https://github.com/phoronix-test-suite/test-profiles/pull/58 plus fixing FPS reporting, switch to using AOM AV1 Git as of today, multi-threading is used by default by upstream aomenc already, and other changes.

pts/aom-av1-1.0.0   [View Source]   Fri, 30 Mar 2018 10:41:20 GMT
Initial work on the AOM AV1 video encode test profile, but not too practical yet given the current extremely slow speed of the encoder, etc. So mostly a placeholder until that code is in better shape.

Suites Using This Test

Video Encoding

C/C++ Compiler Tests

Encoding

Multi-Core

AV1

Creator Workloads


Performance Metrics

Analyze Test Configuration:

AOM AV1 2.0

Encoder Mode: Speed 4 Two-Pass

OpenBenchmarking.org metrics for this test profile configuration based on 657 public results since 19 May 2020 with the latest data as of 22 February 2021.

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
Frames Per Second (Average)
100th
18
3.43 +/- 0.10
100th
17
3.41 +/- 0.11
96th
6
3.29 +/- 0.05
95th
8
3.19 +/- 0.02
92nd
29
2.93 +/- 0.04
89th
4
2.81 +/- 0.03
84th
14
2.70 +/- 0.04
84th
14
2.70 +/- 0.09
80th
13
2.66 +/- 0.05
80th
10
2.63 +/- 0.07
78th
11
2.61 +/- 0.02
76th
3
2.55 +/- 0.02
Mid-Tier
75th
< 2.55
75th
8
2.53 +/- 0.12
73rd
4
2.52 +/- 0.01
73rd
7
2.51 +/- 0.03
71st
10
2.43 +/- 0.03
69th
5
2.39 +/- 0.02
68th
11
2.32 +/- 0.06
68th
10
2.31 +/- 0.03
64th
13
2.27 +/- 0.04
63rd
3
2.23 +/- 0.01
61st
4
2.19 +/- 0.03
60th
5
2.18 +/- 0.02
60th
6
2.18 +/- 0.01
59th
3
2.17 +/- 0.11
56th
5
2.13 +/- 0.05
56th
8
2.12 +/- 0.05
55th
4
2.11 +/- 0.02
55th
8
2.11 +/- 0.05
54th
6
2.10 +/- 0.08
53rd
5
2.08 +/- 0.03
53rd
8
2.08 +/- 0.03
Median
50th
2.07
50th
6
2.06 +/- 0.05
47th
6
2.05 +/- 0.04
47th
6
2.05 +/- 0.05
47th
6
2.05 +/- 0.05
47th
6
2.05 +/- 0.05
47th
4
2.04 +/- 0.01
47th
6
2.04 +/- 0.06
42nd
6
2.02 +/- 0.03
41st
3
2.01 +/- 0.17
40th
4
1.98 +/- 0.05
38th
6
1.96 +/- 0.01
35th
12
1.94 +/- 0.09
35th
3
1.93 +/- 0.05
33rd
3
1.92 +/- 0.01
33rd
8
1.91 +/- 0.04
30th
3
1.88 +/- 0.05
30th
6
1.88 +/- 0.03
30th
3
1.88 +/- 0.01
Low-Tier
25th
< 1.81
25th
4
1.80 +/- 0.02
23rd
4
1.70 +/- 0.03
22nd
4
1.68 +/- 0.01
22nd
3
1.67 +/- 0.03
21st
3
1.66 +/- 0.01
20th
5
1.65 +/- 0.04
19th
4
1.64 +/- 0.01
17th
4
1.57 +/- 0.01
15th
4
1.53 +/- 0.01
15th
3
1.53 +/- 0.05
14th
5
1.47 +/- 0.01
13th
4
1.45 +/- 0.04
13th
3
1.45 +/- 0.01
12th
3
1.41 +/- 0.01
11th
8
1.40 +/- 0.02
9th
4
1.15 +/- 0.01
9th
4
1.10 +/- 0.01
8th
3
1.08 +/- 0.01
7th
6
0.98 +/- 0.04
6th
3
0.95 +/- 0.02
OpenBenchmarking.orgDistribution Of Public Results - Encoder Mode: Speed 4 Two-Pass657 Results Range From 0 To 4 Frames Per Second0.3330.6660.9991.3321.6651.9982.3312.6642.9973.333.6633.9964080120160200

Based on OpenBenchmarking.org data, the selected test / test configuration (AOM AV1 2.0 - Encoder Mode: Speed 4 Two-Pass) 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 BenchmarkEncoder Mode: Speed 4 Two-PassRun-Time246810Min: 1 / Avg: 1.25 / Max: 6

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

OpenBenchmarking.orgPercent, Fewer Is BetterAverage Deviation Between RunsEncoder Mode: Speed 4 Two-PassDeviation246810Min: 0 / Avg: 0.11 / Max: 1

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.

AMDIntelOpenBenchmarking.orgRelative Core Scaling To BaseAOM AV1 CPU Core ScalingEncoder Mode: Speed 4 Two-Pass4681216202432640.71671.43342.15012.86683.5835

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.
 
MOVDQU MOVDQA
Last automated analysis: 18 September 2023

This test profile binary relies on the shared libraries libacl.so.1, libselinux.so.1, libc.so.6, libpcre2-8.so.0.

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)
IBM POWER (PowerPC) 64-bit
ppc64le
POWER9 4-Core, POWER9 44-Core
Intel / AMD x86 32-bit
i686
(Many Processors)
ARMv8 64-bit
aarch64
Ampere Altra ARMv8 Neoverse-N1 160-Core, Ampere eMAG ARMv8 32-Core