Second Annual MSU MPEG-4 AVC/H.264 Video Codec Comparison

For professional users and codec authors

Second H.264/AVC comparison home page!

Part 3. “Best speed” preset

Preset “Best speed”

Main codecs’ objective in this mode is the fastest video compression with satisfactory quality. Therefore main graphs in this section are graphs that show compression speed.

Y-PSNR

However, let’s consider Y-PSNR graphs at first.

Graph type Total number of graph Inserted in this document
Average metric 196 7 (3.5%)

Picture 1. Y-PSNR. Sequence "foreman"

Picture 2. Y-PSNR. Sequence "susi"

Picture 3. Y-PSNR. Sequence "bbc"

Picture 4. Y-PSNR. Sequence "battle"

Picture 5. Y-PSNR. Sequence "simpsons"

Picture 6. Y-PSNR. Sequence "matrix"

Picture 7. Y-PSNR. Sequence "concert"

Speed/Quality Tradeoff

Here are the most important graphs for this preset. They show codec's work time (abscissa axis) and quality of sequence, compressed with it (axis of ordinates). If codec A is to the left of codec B and above it then codec A is better than codec B; if to the right and below it then on the contrary, codec B is better than codec A. In other cases nothing can be said about definite advantage of one's codec compared to another's.

Bit rates in cutlines inside this section are target bit rates (not real codec bit rates).

Graph type

Total number of graph

Inserted in this document

Speed/quality tradeoff

1960

8 (0.4%)


Picture 8. Codec comparison with at speed/quality tradeoff graphs

Picture 9. Legend for Speed/quality tradeoff graphs

Sequence “susi”, Y-PSNR

Picture 12. Bitrate 938 kbps Picture 13. Bitrate 2340 kbps

Sequence “battle”, Y-PSNR

Picture 14. Bitrate 100 kbps Picture 15. Bitrate 340 kbps
Picture 16. Bitrate 938 kbps Picture 17. Bitrate 2340 kbps

Bitrate handling

Consider graphs showing how accurately every codec kept requested bit rate. Grades on x-axis show target bitrates (point 0 is 100 kbps, point 9 - 2340 kbps). Grades on y-axis show how many times codec exceeded declared bit rate (real bit rate to predetermined bit rate ratio).


Picture 18. Bitrate handling. Preset "Best speed". Sequence "foreman"

Picture 19. Bitrate handling. Preset "Best speed". Sequence "susi"

Picture 20. Bitrate handling. Preset "Best speed". Sequence "bbc"

Picture 21. Bitrate handling. Preset "Best speed". Sequence "battle"

Picture 22. Bitrate handling. Preset "Best speed". Sequence "simpsons"

Picture 23. Bitrate handling. Preset "Best speed". Sequence "matrix"

Picture 24. Bitrate handling. Preset "Best speed". Sequence "concert"

Conclusions:

Time

Now let’s consider graphs that show time spent for each video sequence encoding in “Best Speed” preset and average normalized time for this preset.


Picture 25. Encoding time. Preset "Best speed". Sequence "foreman"

Picture 26. Encoding time. Preset "Best speed". Sequence "susi"

Picture 27. Encoding time. Preset "Best speed". Sequence "bbc"

Picture 28. Encoding time. Preset "Best speed". Sequence "battle"

Picture 29. Encoding time. Preset "Best speed". Sequence "simpsons"

Picture 30. Encoding time. Preset "Best speed". Sequence "matrix "

Picture 31. Encoding time. Preset "Best speed". Sequence "concert"

Picture 32. Normalized average encoding time. Preset "Best speed"

Conclusions:

10 Mar 2011
See Also
Automatic local color correction in S3D video
Stereo video may contain a huge color discrepancy. Most of the problems are hard to eliminate because of possible different distortions in each area in the frame.
Call for HEVC codecs 2019
Fourteen modern video codec comparison
HEVC Video Codecs Comparison 2018 (Thirteen MSU Video Codec Comparison)
13th MSU video codecs comparison
HEVC Video Codecs Comparison 2017 (Twelfth MSU Video Codec Comparison)
12th MSU video codecs comparison
MSU Video Codec Comparisons (6 test of lossless, MPEG-4 and MPEG-4 AVC)
Call for HEVC codecs 2018
Site structure