MSU Annual Video Codecs Comparison 2019:
Call for codecs

All industry professionals will know about your codec’s success!

compression.ru	in cooperation with	Lomonosov MSU Graphics & Media Lab (Video Group)	Fourteen modern video codecs comparison For real researchers, developers and professional users in field of high-end video compression
		Dubna State University
		Institute for Information Transmission Problems RAS

Important Dates

March, 10

Deadline for receipt of a codec with required presets

Participants are to provide a codec's name (and the name of encoding standard if not H.265/HEVC), a company name, and list of use cases in which they want to participate

March, 1	Deadline for applications
~~February, 25~~ March, 10	Deadline for settling technical problems with codec’s functioning
September	Draft version of report that will be sent to all participants
September	Reception of comments to the draft
September-October	Comparison report release

About Annual MSU Video Codecs Comparisons

MSU team has up to 21 years of experience in video codec analysis, testing and optimization. Some facts about previous MSU Video Codecs Comparisons:

There were more than 400.000 downloads of previous H.264/MPEG-4 AVC and H.265/HEVC video codec comparisons reports

Many codecs' bugs were found and reported to developers

210+ encoders were tested

More than 35 private reports for codec developers (description of codec's weak and strong points) after public report versions

Structure of this page:

New in 2019 MSU Video Codecs Comparison
Task of the Comparison
Typical Scope of Test
- Summary report topics
- Comparison methodology main points
Encoders analysis methodology
- Link for downloading full comparison methodology
- Comparison Rules
- Test Hardware Characteristics
- Encoding speed requirements
- Codec Requirements
- Developers Deliverables
Apply for 2019 MSU Video Codecs Comparison!

Survey: "What would you like to see in comparison reports?"

Special thanks to our sponsors
Contact information

New features in MSU Video Codecs Comparison

Since 2019	Increasing the number of test videos up to 100+, adding UGC (user-generated) videos (vlogs, noise videos, etc.) Open to compare cloud-based solutions Enlarging of video collection Going to use new no-reference metrics
Since 2018	Increased target bitrates for comparison on 4K videos Started to use VMAF objective metric Upgraded testing hardware to Coffee Lake Enlarged video collection (15833 videos)
Since 2017	Started to perform subjective comparisons with hundreds of observers Started to compare AV1 Enlarged video collection (9299 videos)
Since 2016	New methodology of videos selection New video collection (2909 videos)

Task of the Comparison

We perform comparative unbiased analysis for

software implementations
hardware (GPU-based) implementations
cloud-based implementations

of H.265/HEVC video coding standard and compare it to the best implementations of other video coding standards (H.264/MPEG-4 AVC, AV1, VP9, VP8, MPEG-4 and other) using objective metrics (SSIM, VMAF, PSNR and other).

With MSU Codecs Comparison developers can verify the perfomance of their codec. We share test sequences, encoding parameters (presets) and encoder versions so all developers can reproduce the results of the comparison. Participation with publishing of all results is FREE.

Scope of Test

Summary report topics:

Objective measurements + Subjective analysis
Encoding time
Bitrate keeping for evaluating rate-control mechanism
Speed/Quality trade-off analysis
Averaged objective results analysis
Leaders in different use-cases

Comparison methodology main points:

25-30 HD video sequences (main report) + 10-12 4K video sequences (report appendix).
Number of videos may be increased (up to 100+) and depends on the number of participated codecs. This year UGC (user-generated content) videos may be also added to the test set.
SSIM, PSNR, VMAF objective metrics and subjective evaluation
3 color-planes (Y,U,V) and integral metric values
3 various use-cases (Fast, Universal and Ripping) differ by speed/quality trade-off + special use-case for comparison on 4K videos. Also plan to use special Ultra-Ripping use-case (for comparison with AV1).
8-10 different target bitrates (1-12 Mbps for HD and 2-20 Mbps for 4K)
Prosumer-level modern hardware
Fully automatic testing system
7000+ result figures, PDF and HTML reports

Encoders analysis methodology

Full comparison methodology (PDF, 1.1 MB)

Comparison Rules

This year encoder developers send us a bundle of same presets (with different speed/quality characteristics) for all use-cases.
Please pay attention that we will use multi-core CPU for encoding, so you can use multi-threading

Decoding is performed with reference decoder (for H.264/AVC and H.265/HEVC only). For other standards participants should provide decoder
We don't limit GOP size and intra-period
We don’t limit number of passes in the encoding. Total encoding time should fulfill
VBR mode is used during all tests (other modes are discussed individually)
Before results' publishing each developer will receive the results of its codec and competitive free open-source codecs. Developers of each codec can validate the results and write a comment (one paragraph) about the comparison results which will be included in the report
Participation is free with publication of the results
- You can join comparison for free if you agree that your codec's results will be published
- Private participation. Compare your codec with world leaders staying incognito! If your company wants to know results of your codec's testing with possibility to exclude them from publication and information disclosure, you should pay for measurements and report preparing before comparison begins
Enterprise version of comparison report is available for direct participants for free
All participants will receive following deliverables to verify the results for free:
- video sequences used in comparison
- binaries of all free encoders used in comparison to verify the results
- all raw video quality metric and encoding speed data for its encoder and for all of free encoders used in comparison
We are willing to completely or partially delete information about some codec in the public version of comparison report only in exceptional cases (e.g. critical errors in the codec)

Test Hardware Characteristics

Next hardware for codecs testing will be used:

CPU: Intel Socket 1151 Core i7 8700K (Coffee Lake) (3.7Ghz, 6C12T, TDP 95W)
Mainboard: ASRock Z370M Pro4
RAM: Crucial CT16G4DFD824A 2x16GB (totally 32 GB) DIMM DDR4 2400MHz CL15
OS: Windows 10 x64

Encoding speed requirements

For encoder alignment selected presets should provide following encoding speed.
All speed requirements are checked at video sequence encoded at 6Mbps:

Fast/High Density – 1080p@60fps
Universal/Broadcast VQ – 1080p@25fps
Ripping/Pristine VQ – 1080p@1fps and SSIM-RD curve better than x264-veryslow
For subjective comparison: 1080p@1fps
For comparison on 4K videos: 20fps and 1fps
Ultra-Ripping 1080p@0.005fps

Codec Requirements

Presets for different speed requirements should be provided by the developers

Codec should allow to set arbitrary bitrate of resulting stream in VBR mode

Preferable codec interface - console codec version (with batch processing support — bitrate and file names must be possible to assign from the command line)

Encoder should be compatible with reference decoder