[Ffmpeg-devel] Re: VC-1

g. the_ether
Sat Apr 8 18:26:15 CEST 2006


M?ns Rullg?rd <mru <at> inprovide.com> writes:

> > I was interested to see in an article by TI that it requires much less 
> > processing than h.264 (40% less to encode and similar for decode)
> 
> And correspondingly lower quality.

I don't want to enter a religious discussion but "correspondingly" is a 
misleading adjective to use.

If there is any difference in quality, it would appear to be negligible. 
Admitedly, the article is by M$ employees, but if you read, Srinivasan, S.; 
(John) Hsu, P.; Holcomb, T.; Mukerjee, K.; Regunathan, S.L.; Lin, B.; Liang, 
J.; Lee, M.-C.; Ribas-Corbera, J., ?Windows Media Video 9: overview and 
applications?, Signal Processing: Image Communication, Volume 19, Issue 9, 1 
October 2004, Pages 851-875, you will see:

"When it comes to subjective quality, however,
WMV-9 has equaled or outperformed optimized
implementations of H.264/AVC. There have been
a variety of studies that have independently
evaluated the compression efficiency of WMV-9
and H.264/AVC. For example, Tandberg Television
evaluated WMV-9 and the H.264/AVC
(baseline and main profile) implementations (version
6.0a), and compared them to their optimized
MPEG-2 codec and MPEG-4 ASP codecs [1,2].
They concluded that the visual quality achieved by
H.264/AVC main profile and WMV-9 was comparable,
and that these two codecs provide the best
quality among the competing codecs in their
comparison."

Furthermore:

"WMV-9 also achieved the highest
perceptual quality in each of the clips tested
at the latest independent test performed by the
DVD Forum, which also evaluated optimized
implementations of H.264/AVC main profile,
MPEG-4 ASP, and MPEG-2."

g.








More information about the ffmpeg-devel mailing list