Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit.

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

Source Repository

github.com

Test Created

30 September 2019

Last Updated

18 May 2021

Test Maintainer

Michael Larabel 

Test Type

Processor

Average Install Time

1 Second

Average Run Time

43 Seconds

Accolades

20k+ Downloads

Supported Platforms

Supported Architectures

x86_64

Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page Views ***OpenBenchmarking.orgEventsIntel Open Image Denoise Popularity Statisticspts/oidn2019.092019.102019.112019.122020.012020.022020.032020.042020.052020.062020.072020.082020.092020.102020.112020.122021.012021.022021.032021.042021.052021.062021.072021.082021.092021.102021.112021.122022.012022.022022.032022.042022.052022.062022.072022.082022.092022.102022.112022.122023.012023.022K4K6K8K10K
* 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 2 February 2023.
RT.hdr_alb_nrm.3840x216036.1%RTLightmap.hdr.4096x409628.2%RT.ldr_alb_nrm.3840x216035.7%Run Option PopularityOpenBenchmarking.org

Revision History

pts/oidn-1.4.0   [View Source]   Tue, 18 May 2021 05:41:11 GMT
Update to upstream OIDn 1.4, switch to using its new built in oidnBenchmark.

pts/oidn-1.2.0   [View Source]   Thu, 09 Apr 2020 15:36:38 GMT
Update against Intel Open Image Denoise 1.2.

pts/oidn-1.1.1   [View Source]   Tue, 15 Oct 2019 17:30:30 GMT
Add macOS support.

pts/oidn-1.1.0   [View Source]   Mon, 30 Sep 2019 11:23:44 GMT
Use img/sec as unit, set image bench count to 100.

pts/oidn-1.0.0   [View Source]   Mon, 30 Sep 2019 10:57:17 GMT
Initial commit of Intel Open Image Denoise library benchmark. Simple PFM input file until finding better option(s).

Suites Using This Test

Multi-Core

Intel oneAPI

Creator Workloads

Game Development


Performance Metrics

Analyze Test Configuration:

Intel Open Image Denoise 1.4.0

Run: RT.hdr_alb_nrm.3840x2160

OpenBenchmarking.org metrics for this test profile configuration based on 1,012 public results since 18 May 2021 with the latest data as of 31 January 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
Images / Sec (Average)
100th
12
4.08 +/- 0.39
100th
9
3.87 +/- 0.04
97th
37
3.36 +/- 0.25
97th
9
3.17 +/- 0.15
92nd
47
2.85 +/- 0.14
90th
9
2.70 +/- 0.03
90th
5
2.69 +/- 0.04
87th
22
2.53 +/- 0.03
86th
13
2.25 +/- 0.01
84th
3
2.06 +/- 0.02
84th
4
2.00 +/- 0.04
83rd
17
1.94 +/- 0.02
81st
15
1.81 +/- 0.03
78th
16
1.66 +/- 0.20
78th
14
1.65 +/- 0.01
77th
11
1.64 +/- 0.02
Mid-Tier
75th
< 1.61
75th
4
1.59 +/- 0.03
75th
7
1.56 +/- 0.03
74th
3
1.48 +/- 0.01
72nd
11
1.41 +/- 0.01
69th
6
1.14 +/- 0.01
68th
10
1.10 +/- 0.01
64th
26
0.94 +/- 0.03
64th
3
0.90 +/- 0.01
63rd
4
0.89 +/- 0.02
63rd
18
0.89 +/- 0.01
60th
3
0.87 +/- 0.05
59th
3
0.80 +/- 0.01
56th
37
0.58 +/- 0.03
54th
13
0.56 +/- 0.03
52nd
7
0.53 +/- 0.01
51st
14
0.50 +/- 0.03
51st
9
0.50 +/- 0.01
Median
50th
0.50
49th
12
0.49 +/- 0.01
48th
6
0.48 +/- 0.01
43rd
4
0.41 +/- 0.02
43rd
9
0.38 +/- 0.01
38th
5
0.34 +/- 0.01
34th
8
0.32 +/- 0.02
34th
9
0.31 +/- 0.02
34th
20
0.31 +/- 0.01
31st
7
0.30 +/- 0.01
29th
6
0.28 +/- 0.01
29th
4
0.27 +/- 0.01
27th
4
0.25 +/- 0.01
Low-Tier
25th
< 0.25
25th
4
0.24 +/- 0.01
25th
3
0.24 +/- 0.01
24th
5
0.22 +/- 0.01
22nd
6
0.21 +/- 0.01
20th
4
0.19 +/- 0.01
18th
5
0.18 +/- 0.01
18th
3
0.18 +/- 0.01
17th
4
0.17 +/- 0.01
13th
6
0.15 +/- 0.01
8th
6
0.11 +/- 0.01
8th
6
0.11 +/- 0.01
5th
5
0.08
OpenBenchmarking.orgDistribution Of Public Results - Run: RT.hdr_alb_nrm.3840x21601004 Results Range From 0 To 147537 Images / Sec2951590288531180414755177062065723608265592951032461354123836341314442654721650167531185606959020619716492267873708247377576726796778262885579885309148194432973831003341032851062361091871121381150891180401209911239421268931298441327951357461386971416481445991475502004006008001000

Based on OpenBenchmarking.org data, the selected test / test configuration (Intel Open Image Denoise 1.4.0 - Run: RT.hdr_alb_nrm.3840x2160) has an average run-time of 4 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 BenchmarkRun: RT.hdr_alb_nrm.3840x2160Run-Time612182430Min: 1 / Avg: 3.98 / Max: 28

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 RunsRun: RT.hdr_alb_nrm.3840x2160Deviation246810Min: 0 / Avg: 0.15 / Max: 2

Does It Scale Well With Increasing Cores?

Yes, based on the automated analysis of the collected public benchmark data, this test / test settings does 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 BaseIntel Open Image Denoise CPU Core ScalingRun: RT.hdr_alb_nrm.3840x2160468101216243264128612182430

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 CVTSI2SD DIVSD MULSD CVTTSD2SI MOVHPD ADDSD SUBSD CMPNLESD CMPLTSD MOVD CMPEQSD MOVDQA PSRLDQ PSLLDQ MOVNTDQ MOVNTI PUNPCKLQDQ CVTSS2SD CVTSD2SS COMISD MINSD PSHUFD UCOMISD PADDQ PMULUDQ PSUBQ CVTDQ2PS CVTTPS2DQ UNPCKLPD CVTSD2SI MOVAPD MULPD ORPD ADDPD UNPCKHPD XORPD ANDPD CVTPD2PS
SSE3 (SSE3)
Used by default on supported hardware.
 
LDDQU HADDPS MOVSHDUP MOVDDUP
SSSE3 (SSSE3)
Used by default on supported hardware.
 
PALIGNR
Used by default on supported hardware.
Found on Intel processors since Sandy Bridge (2011).
Found on AMD processors since Bulldozer (2011).

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

 
VPBROADCASTD VEXTRACTI128 VPBROADCASTQ VGATHERQPS
Advanced Vector Extensions 512 (AVX512)
Used by default on supported hardware.
 
(ZMM REGISTER USE)
FMA (FMA)
Used by default on supported hardware.
Found on Intel processors since Haswell (2013).
Found on AMD processors since Bulldozer (2011).

 
VFMADD213PS VFMADD231PS VFMSUB231PS VFNMSUB231PS VFMSUB213PS VFMADD132PS VFMSUB213SD VFMADD213SD VFMADD231SD VFMSUB213SS
WAITPKG / UMWAIT / TPAUSE (WAITPKG)
Used by default on supported hardware.
 
TPAUSE
Last automated analysis: 18 January 2022

This test profile binary relies on the shared libraries libOpenImageDenoise.so.1, libpthread.so.0, libtbb.so.12, libm.so.6, libc.so.6, libdl.so.2.

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)
ARMv8 64-bit
arm64
Apple M1