Hi all,

there is on doom9 forum the first result about best codec for new HD format:

http://forum.doom9.org/showthread.php?p=1028887#post1028887




Golgot13

So, is that little chart saying that VC-1 isn't as good as MPEG-2 or H264?

Even if it is, those results look almost 100% the same.

And regardless, I think we can all agree VC-1 is far superior to MPEG-2 in side-by-side tests (as seen in a lot of Blu-ray to HD-DVD tests since Blu-ray titles are often MPEG-2).

But is H264 better in picture quality? I'd doubt it. I'd bet it's on par with VC-1.

I honestly think MPEG-2 can look just as good as any codec as long as it's not bit starved.

No. It tells us that both OPSNR and SSIM 2 objective review/benchmark indicators show that "Mpeg2<VC-1<H.264/AVC" (to put it very simple...).

OPSNR and SSIM are probably the most objective and widely accepted encoding quality tests/indicators available today... higher is better = less deviation from the source.

With h.264/AVC (using x164 encoding, while VC-1 was tested with PEP (=application used for most commercial HD-DVD and Blu-Ray VC-1 encodings) coming out on top in each bitrate-szenario test.


Please ask Amir and Ben in the insider thread why OPSNR and SSIM tests are not so objective after all and these results (as well as many independent benchmark tests before them - do a search on Doom9) showing/demonstrating/"proofing" AVCs superiority mean nothing really. ( :rolleyes: )

I bet you will get the usual highly political, professional and utterly convincing answer. They do their jobs really well around here.

I honestly think MPEG-2 can look just as good as any codec as long as it's not bit starved.
By bit-starved, you mean any more than 20% compression?

MPEG-2 suffers from DCT artifacts, whereas the advanced MPEG-4 variants (AVC and VC-1) do not.

DCT artifacts are most evident in fast-action scenes (e.g. fire, smoke) and are very visible even at 30Mbps.

By bit-starved, you mean any more than 20% compression?
MPEG-2 suffers from DCT artifacts, whereas the advanced MPEG-4 variants (AVC and VC-1) do not.

If VC1 and AVC are bit starved they filter the blocking away. That is just another kind of artifact that softens the picture.

If VC1 and AVC are bit starved they filter the blocking away. That is just another kind of artifact that softens the picture.

What I mean is, even if MPEG-2 is not really bit-starved it still has DCT artifacts.

I.e. DCT artifacts are not simply caused by bit-starving - they are a fundamental problem with how MPEG-2 is encoded.

AVCs superiority mean nothing really. ( :rolleyes: )

I bet you will get the usual highly political, professional and utterly convincing answer. They do their jobs really well around here.

AVC superiority mean:
At same bitrate = a better video quality
At same quality = more video space for more movie or video bonus


All HD hardware will use H264 (camcorder, IPTV, HD broadcast like Voom in USA,...).
And for video embedded (like video on mobil phone,...) it use H264 ( 3GPP2,...).
I don't see mobile phone which can recieve a video stream in VC1 by HSPA/CDMA/UMTS (but it exist in H264)


Golgot13

There is a nice answer from Ben (thank you about Swisscom info) about VC1.
http://forum.doom9.org/showthread.php?p=1029329#post1029329

I hope Amir will answer too (if he is not in holiday). :)





Golgot13

After a reply from Ben about a movie in 720P at 2Mbps optimized with VC1 codec(with best solution from MS ?):

http://forum.doom9.org/showthread.php?p=1029298#post1029298


There is a answer with the same movie in 720P at 1.863Mbps with H264 codec:

http://forum.doom9.org/showthread.php?p=1029477#post1029477


So, score = 2 for H264 and 0 for VC1 (First result of codec challenge and 720P movie encoded).


I hope Amir and other MS team (there is Zambelli and Ben on doom9 only) can answer to show the possiblities of VC1.
X264 is a free software, all people can use it (GPL software).




Regards,



Golgot13

New thread start in doom9 forum about this test (challenge between MPEG2, VC1 and H264):


http://forum.doom9.org/showthread.php?t=128498


All encoded file with different codec are fully compliant with HD DVD specification :D



Somebody (like me) wait answer (better encoded file) from MS team.





Regards,




Golgot13

MPEG-2 suffers from DCT artifacts, whereas the advanced MPEG-4 variants (AVC and VC-1) do not.

DCT artifacts are most evident in fast-action scenes (e.g. fire, smoke) and are very visible even at 30Mbps.

you sir are dead wrong and obviously never saw D-Theater releases

??? > Mpeg2 is nothing but format war fanboy spin ;)

-Gary

as this is a "format neutral" topic continue discussion here:

http://www.avsforum.com/avs-vb/showthread.php?t=883667

DCT artifacts are very visible even at 30Mbps.


It is not true because VC1 and H264 are equivalent at MPEG2 for a bitrate higher than ~ 28Mbps.

You can see the first conclusion about the codec challenge:
http://forum.doom9.org/showthread.php?p=1029513#post1029513


If somebody can give a movie hard to encode which can use for this challenge, it will be wellcome.





Golgot13

Nice to see someone finally scientifically challenge the FUD being spread about AVC around here.

VC1's slogan should be "We are better because MS said so!" :p

This challenge will end up being just as useless as ANY other challenge in the past that had Sagittaire as one of the contributors.

Reason:

he constantly keeps changing the rules to these challenges whenever the results are not as favorable to his position as he anticipated.

Just witness that he did that again today (just a couple of hours ago).

This challenge will end up being just as useless as ANY other challenge in the past that had Sagittaire as one of the contributors.

Reason:

he constantly keeps changing the rules to these challenges whenever the results are not as favorable to his position as he anticipated.

Just witness that he did that again today (just a couple of hours ago).
You inadvertently answered my question without knowing it. :p
I was getting ready to post differently until I refreshed and saw this.