Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > Video Encoding > MPEG-4 AVC / H.264

Reply
 
Thread Tools Search this Thread Display Modes
Old 11th May 2011, 10:09   #21  |  Link
Sagittaire
Testeur de codecs
 
Sagittaire's Avatar
 
Join Date: May 2003
Location: France
Posts: 2,484
In this case desactive simply AQ ...
__________________
Le Sagittaire ... ;-)

1- Ateme AVC or x264
2- VP7 or RV10 only for anime
3- XviD, DivX or WMV9
Sagittaire is offline   Reply With Quote
Old 12th May 2011, 18:55   #22  |  Link
kolak
Registered User
 
Join Date: Nov 2004
Location: Poland
Posts: 2,843
Quote:
Originally Posted by Blue_MiSfit View Post
Kolak, as usual you spew nonsense.

"x264 is not very good at high bitrates due its low bitrates optimisations"

"when it comes to source with gradients x264 is not the best choice"

"it was not designed for high bitrates"
Why we don't we have all studios (at leas small ones) using x264- it has been long time since it's BD complaint?
Answer- it does not give real advantage in case of BD production and even if it's free it does not change fact that it misses many features (+ it's relatively slow), which are essential for authoring studios, so there is no that big interest in using it (before even judging quality).

If x264 would have special high bitrate mode these artefacts would not be there- it's way good enough to avoid them, but it needs some optimisation for high bitrates.
Yes- gradients, low level details areas are week points of x264.


Andrew

Last edited by kolak; 12th May 2011 at 19:07.
kolak is offline   Reply With Quote
Old 17th May 2011, 20:16   #23  |  Link
kolak
Registered User
 
Join Date: Nov 2004
Location: Poland
Posts: 2,843
I've done yet another try: quite grainy film source encoded at 35Mbit (38Mbit max), 2 pass BD pressets, with very slow first pass. It was done at about 5fps (on 12 core machine with HT at 99% of all cores) and does look very good, except exactly the same problem as mp3dom described.
Whole frame will look very good, but it has places where details gets flatten and changed into blocks- there is really no reason for this at 35Mbs. There is also still big gap in quality between B and rest of frames (I assume this can be tweaked). This was not anime, but grainy film source.


Andrew

Last edited by kolak; 18th May 2011 at 16:36.
kolak is offline   Reply With Quote
Old 18th May 2011, 13:59   #24  |  Link
mp3dom
Registered User
 
Join Date: Jul 2003
Location: Italy
Posts: 1,135
Quote:
Originally Posted by Blue_MiSfit View Post
Best of luck, mp3dom. This looks like a very difficult case to optimize for, and I'm sure you're doing all this work for a very good reason.
The reason is... to have the maximum quality available.

Quote:
You are quite certain that the artifacts you describe are visible in motion, on a standard uncalibrated display, and are harder to swallow than the softness / inferior handling of complex scenes offered by a professional encoder... right?
No, I need to clarify. The weakest point of an encoder like Blu-Code (have you never tried/seen it in action or seen its output quality?) is in very complex area. Something really uncommon on real-life sources or live films, but can be 'common' in japanese animation where some long-scenes are completely unrelated each others, so motion estimation is really huge and a difficult task. In those cases, the smoothing or even blocking is visible compared to the source. The smooth fortunately is not too strong, because the encoder have an 'adaptive' automatic deblocking but I like anyway to not have smoothness or blocking at all . In all the other cases (normal or quite complex scenes, gradation, fades, grain retention etc) I find the output of Blu-Code the same or more visually pleasant (it depends by scenes) than the x264 output (especially on fades and gradients in both static and dynamic scenes, I find the output of Blu-Code better than any other AVC encoder). I'm always referring to a high bitrate output.

Quote:
x264 may not be the best tool for the job, but I'm glad you're giving it a try! Free is nice, huh?
To make long story short and simple, actually I work in a (small) company that doesn't made BD as a 'service' (for any other company) but made and sell its own BD (like, I think, Criterion do, just to give the idea). This means that it's hard to have access to a 50K$ encoder even if it's the best in the world. The fact that I'm still giving a try to x264 is because I see the 'potentials' of the encoder in the BD field (at low datarate is already known and personally tested to be the best and I'm already using it) and can probably match, with some modifications, what actually I think is the best encoder - but that I can't affort right now - for BD at high bitrate (CC-HDe). The screenshots that I've posted in the first thread comes from CC-HDe (I know it because every encoder has its own 'pattern' easy to spot). Even CC-HDe is not perfect anyway. Gradients and fades IMHO are better managed by Blu-Code but the output of CC-HDe is anyway acceptable enough (it doesn't produce aberrations/blocking during fades or gradients) and very complex scenes have the same quality as x264.

Last edited by mp3dom; 18th May 2011 at 14:13.
mp3dom is offline   Reply With Quote
Old 18th May 2011, 14:43   #25  |  Link
CruNcher
Registered User
 
CruNcher's Avatar
 
Join Date: Apr 2002
Location: Germany
Posts: 4,926
Quote:
(I know it because every encoder has its own 'pattern' easy to spot).
That is something wow i mean ok i believe it's possible between different codec types but between the same i wouldn't buy it that you can detect from which specific H.264 Encoder a output frame comes from no way, i guess you rather meant you can see which one it came from by deep analyzing it visually against the other, that i would believe again

Scaling from 0->100 and somewhere not losing efficiency inside the same codebase is nothing easy and seeing that x264 was mostly optimized visually with low bitrate and the hell with freedom from all restrictions in mind over the years i can follow your and kolaks problems on the blu-ray situation somewhat, and you really tried every possible combination of x264 settings that have todo with compression for quality and you don't get happy @ least with how it looks @ high bitrate i mean their are a lot of options and ranges for possible tweaking high bitrate from the defaults visually ?
__________________
all my compares are riddles so please try to decipher them yourselves :)

It is about Time

Join the Revolution NOW before it is to Late !

http://forum.doom9.org/showthread.php?t=168004

Last edited by CruNcher; 18th May 2011 at 15:09.
CruNcher is offline   Reply With Quote
Old 18th May 2011, 16:24   #26  |  Link
mp3dom
Registered User
 
Join Date: Jul 2003
Location: Italy
Posts: 1,135
Yes, I mean you can detect the encoder based on analyze (StreamEye is enough), not by single screenshot or while watching the movie.
mp3dom is offline   Reply With Quote
Old 18th May 2011, 16:28   #27  |  Link
kolak
Registered User
 
Join Date: Nov 2004
Location: Poland
Posts: 2,843
mp3dom- there are some improvements coming to CC-HDe.
There is quite a lot to improve, but it was all suspended due to MVC encoder.

The best is to have Blu-code, CC-HDe and x264
Very stable grain and overall image look is very strong point of Blu-code.

I wish x264 gets tweaked for high bitrates encodes.


Andrew
kolak is offline   Reply With Quote
Old 18th May 2011, 16:38   #28  |  Link
kolak
Registered User
 
Join Date: Nov 2004
Location: Poland
Posts: 2,843
Quote:
Originally Posted by mp3dom View Post
Yes, I mean you can detect the encoder based on analyze (StreamEye is enough), not by single screenshot or while watching the movie.
You can tell even by the look quite often- if you use encoder for long you can see its patterns


Andrew
kolak is offline   Reply With Quote
Old 19th May 2011, 19:03   #29  |  Link
posdnya
Registered User
 
Join Date: Dec 2001
Posts: 21
Quote:
Originally Posted by mp3dom View Post
Yes, I mean you can detect the encoder based on analyze (StreamEye is enough), not by single screenshot or while watching the movie.
Could you make three screenshots from StreamEye - macroblock types, quantizers, and macroblock size. And frame itself, of course.
Or better just give a few meg of the encoded stream containing problem frames. No more then 200 :-) - then evaluation version could be used to check it.

If there is problem with encoding, it will be evident, doesn't matter what's the nature - too high quantizers (not sure) or some problems with intra predicted blocks in P-frame (more likely)
posdnya is offline   Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 13:21.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.