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 7th December 2004, 22:24   #201  |  Link
karl_lillevold
Moderator
 
karl_lillevold's Avatar
 
Join Date: Oct 2002
Location: Seattle
Posts: 1,584
Re: Re: Nero (Ateme) H.264 in Avisynth duration not found

Quote:
Originally posted by Sagittaire
try with trim:
a = DirectShowSource("output.mp4", fps=23.976, audio=false)
a = trim(a, 0, 1500)
return a

perhabs bobololo or babagaya could post in PM beta parser and beta decoder for your test ... lol
Trim did not make any difference.

Actually, Ateme was kind enough to let me participate in the Beta test, so I have version beta-4a. With this, I can kind of get Avisynth to work without seek. However, normal DirectShow playback, which the Nero versions of the filters handle well, is not working well in Ateme's versions, with frozen video, failed seeks, player will not open etc. (on the Nero encoded content)
__________________
This information is provided "AS IS" with no warranties, grants no rights, and reflects my personal opinion.

Last edited by karl_lillevold; 7th December 2004 at 22:27.
karl_lillevold is offline   Reply With Quote
Old 7th December 2004, 22:27   #202  |  Link
Andrey
Registered User
 
Join Date: May 2003
Location: Riga, Latvia
Posts: 186
Hi, guys !
>>Sharktooth: Finally recode works.
You are lucky. Still no chance for me.
But I'll try your method step by step when I have spare time, may be it will help...
>>Soulhunter: If I add "ConvertToRGB32" to the script, recode simply disappears !?!
My current installation (show mpeg4 enc) disappears EVERY time I tried to use avs source, regardless of what is inside
And
>>Noise can be safely filtered out and then added during playback.
>>Film grain cant...
Hm ? Is that bolobolo, who showed here an example of analysing film grain before encoding and adding it after ?
And said that it is one of the tools they gonna to implement in the future...
Or I'm wrong ? (It was in that beta test chain on this forum)
Andrey is offline   Reply With Quote
Old 7th December 2004, 22:41   #203  |  Link
Soulhunter
Bored...
 
Soulhunter's Avatar
 
Join Date: Apr 2003
Location: Unknown
Posts: 2,812
Quote:
Originally posted by Andrey

My current installation disappears EVERY time I tried to use avs source...
Have you tried to delete the WarpSharp.dll from AviSyths plugin folder ???

Maybe moving all plugins in a different folder works...


Quote:
Originally posted by Andrey

Hm ? Is that bolobolo, who showed here an example of analysing film grain before encoding and adding it after ?
Nope, that was me...

But you need ffdshow to do this !!!


Bye
__________________

Visit my IRC channel

Last edited by Soulhunter; 7th December 2004 at 22:45.
Soulhunter is offline   Reply With Quote
Old 7th December 2004, 23:00   #204  |  Link
babayaga
Registered User
 
Join Date: Dec 2003
Location: Paris
Posts: 90
Quote:
Originally posted by Soulhunter
Well, for me the noise and film-grain is part of the image detail !!!

Btw, I think most ppl use denoising to raise the compressibility...

Thanks, the "Samples.zip" is on the way...

EDIT: Ok, its uploaded now !!!

Bye
I disagree about the film grain "removal" in the shot you sent

There is indeed a loss in grain for both encodes (Xvid and ND) but it appears to me that it's closely related to the B-frames sequence :
- ND B-frames are a little bit more blurry than XviD's but less noisy
- ND P-frames are a little bit more precise than Xvid's and less noisy
- The P-B sequences are not the same so the alternance of noisy/blurry frames is not the same

As a result, a part of the grain you think Xvid preserves is noise added (mostly in B-frames) and the overall impression is closely related to the P-B frame sequence.

As a matter of taste, in this very shot I prefer our way of handling the grain (it looks more "smooth" or less "rusty" to my eyes).
babayaga is offline   Reply With Quote
Old 7th December 2004, 23:15   #205  |  Link
bond
Registered User
 
Join Date: Nov 2001
Posts: 9,770
Quote:
Originally posted by Selur
did anyone get it working in Graphedit?
gives me a
"The filter could not be created. Resource used by this filter may already be in use. Class nor registered (return code: 0x80040154)"
error,...
hm, any news on that? did anyone manage to encode using neros avc encoder directly in graphedit as it was possible with neros asp encoder?
__________________
Between the weak and the strong one it is the freedom which oppresses and the law that liberates (Jean Jacques Rousseau)
I know, that I know nothing (Socrates)

MPEG-4 ASP FAQ | AVC/H.264 FAQ | AAC FAQ | MP4 FAQ | MP4Menu stores DVD Menus in MP4 (guide)
Ogg Theora | Ogg Vorbis
use WM9 today and get Micro$oft controlling the A/V market tomorrow for free
bond is offline   Reply With Quote
Old 7th December 2004, 23:17   #206  |  Link
Soulhunter
Bored...
 
Soulhunter's Avatar
 
Join Date: Apr 2003
Location: Unknown
Posts: 2,812
Quote:
Originally posted by babayaga

As a result, a part of the grain you think Xvid preserves is noise added (mostly in B-frames) and the overall impression is closely related to the P-B frame sequence.

As a matter of taste, in this very shot I prefer our way of handling the grain (it looks more "smooth" or less "rusty" to my eyes).
Well, imo its not really a "matter of taste" thing...

I think a codec should be able to reproduce the source as exact as possible !?!

For me this means, if the source looks "rusty" the encode should look rusty as well...

If you prefer a clean "not rusty" look, you can use one of the numerous denoisers out there !!!


Bye
__________________

Visit my IRC channel

Last edited by Soulhunter; 7th December 2004 at 23:31.
Soulhunter is offline   Reply With Quote
Old 7th December 2004, 23:22   #207  |  Link
Andrey
Registered User
 
Join Date: May 2003
Location: Riga, Latvia
Posts: 186
>>Maybe moving all plugins in a different folder works...
It works. Thanks a lot !!!
BTW, why the hell unused(loaded automatically?) plugins (and plugins at all) affects this ?
Andrey is offline   Reply With Quote
Old 7th December 2004, 23:26   #208  |  Link
Manao
Registered User
 
Join Date: Jan 2002
Location: France
Posts: 2,856
Because they are loaded when avisynth dll is loaded ( in order to know which filters are available )
Manao is offline   Reply With Quote
Old 8th December 2004, 01:16   #209  |  Link
nicco
TeacH Me!! I'm so n00b!
 
nicco's Avatar
 
Join Date: Nov 2003
Location: Italy
Posts: 316
Quote:
originally posted by Teegedeck
Perhaps that Gladiators don't really have such a smooth skin?
it depends, I think, on their make-up!


@soulhunter
Premised that quality is subjective (and personally I don't like source-noise), I think you are right with this:
Quote:
I think a codec should be able to reproduce the source as exact as possible !?!

For me this means, if the source looks "rusty" the encode should look rusty as well...

If you prefer a clean "not rusty" look, you can use one of the numerous denoisers out there !!!
But I really like avc denoising in your shot!
nicco is offline   Reply With Quote
Old 8th December 2004, 01:48   #210  |  Link
babayaga
Registered User
 
Join Date: Dec 2003
Location: Paris
Posts: 90
Quote:
Originally posted by Soulhunter
Well, imo its not really a "matter of taste" thing...
I think a codec should be able to reproduce the source as exact as possible !?!
Well that's exactly what it does

Quote:

For me this means, if the source looks "rusty" the encode should look rusty as well...

If you prefer a clean "not rusty" look, you can use one of the numerous denoisers !!!
I was refering at the look of the noise *not* the video itself.

The source noise is by far not as rusty as the Xvid result especially in B-frames : the added noise in B-frames compensates too much the compression loss.

So in my opinion, there is also a compression loss in ND. This one is smaller than with Xvid (since we have a higher coding efficiency) but is not compensated by an artificial encoder noise like in Xvid.

For those who want to make their own mind, the clips are in ftp://mood.ateme.com/Samples.zip
babayaga is offline   Reply With Quote
Old 8th December 2004, 04:38   #211  |  Link
Sharktooth
Mr. Sandman
 
Sharktooth's Avatar
 
Join Date: Sep 2003
Location: Haddonfield, IL
Posts: 11,768
I prefer the xvid encode. It's sharper and noise/grain is more like the source. AVC encode b-frames are too much "denoised/smoothed".
BTW i spotted some "imperceptible blocks" in the background in both the encodes.
Sharktooth is offline   Reply With Quote
Old 8th December 2004, 13:06   #212  |  Link
nicco
TeacH Me!! I'm so n00b!
 
nicco's Avatar
 
Join Date: Nov 2003
Location: Italy
Posts: 316
Quote:
@soulhunter
Btw, I think most ppl use denoising to raise the compressibility...
If you look at my DV-movies you'll understand why I use denoising filters..

Personally, for my sources (almost only DV) I prefer avc "denoised picture", becouse details are preserved, and it's hard to me to notice lousy bloks
nicco is offline   Reply With Quote
Old 8th December 2004, 13:54   #213  |  Link
temporance
Registered User
 
Join Date: Mar 2002
Posts: 486
Quote:
Originally posted by nicco
Personally, for my sources (almost only DV) I prefer avc "denoised picture", becouse details are preserved, and it's hard to me to notice lousy bloks
For me, it's the opposite. We're not talking about denoising, we're talking about "de-texturing" too. I would rather have a few blocking artifacts if it means I can actually see some detail in my movie. AVC is as good at ASP at making sharp edges and strong details look sharp. It is not good though at preserving fine textures, and I don't just mean film grain. Things that make the movie look 'real' (e.g. rough paintwork, actor's crowsfeet or stubble, lichen growing on a wall, leaves on a tree and blades of grass) are better represented, IMHO, by ASP than AVC. Anyway, it's early days and I'm sure this argument will be fought, won and lost over the next year or so.
temporance is offline   Reply With Quote
Old 8th December 2004, 14:13   #214  |  Link
Tommy Carrot
Registered User
 
Tommy Carrot's Avatar
 
Join Date: Mar 2002
Posts: 863
Well, the smoothing away of the fine textures is an unfortunate side-effect of the loop-filter, similarly to the post-filter of mpeg4 asp, and this is the reason most of us don't use post-filtering at playback. The problem is, disabling the loop-filter won't solve the issue, but it creates other, because IMO the blocking effect of h.264 is much more annoying than in the case of mpeg4 asp. You can try to decrease the loop-filter strength, perhaps you can find a more optimal setting, where the smoothness is not so obvious, and the blockyness is still tolerable.
Tommy Carrot is offline   Reply With Quote
Old 8th December 2004, 14:18   #215  |  Link
Soulhunter
Bored...
 
Soulhunter's Avatar
 
Join Date: Apr 2003
Location: Unknown
Posts: 2,812
@ temporance

True, much ppl dislike WMV/RV9 for its "smooths out too much fine details" behavior...

Maybe lowering the high frequency cut-off in exchange for some ringing could prevent this !?!


Bye
__________________

Visit my IRC channel

Last edited by Soulhunter; 8th December 2004 at 14:23.
Soulhunter is offline   Reply With Quote
Old 8th December 2004, 14:36   #216  |  Link
nicco
TeacH Me!! I'm so n00b!
 
nicco's Avatar
 
Join Date: Nov 2003
Location: Italy
Posts: 316
Quote:
originally posted by Tommy carrot
The problem is, disabling the loop-filter won't solve the issue, but it creates other, because IMO the blocking effect of h.264 is much more annoying than in the case of mpeg4 asp
True!
Quote:
originally posted by Tommy carrot
You can try to decrease the loop-filter strength, perhaps you can find a more optimal setting, where the smoothness is not so obvious, and the blockyness is still tolerable.
Is what I've done!, a deblocking value of 1 (no adaptive) is very good for me
Quote:
originally posted by temporance
We're not talking about denoising, we're talking about "de-texturing" too.
I was talking to de-texturing too.
Quote:
originally posted by temporance
I would rather have a few blocking artifacts if it means I can actually see some detail in my movie
I agree, but these details should be more noticeable than blocking artifacts, If I can see these details only with a lens I prefer a bit smoother picture rather than a "party of blocks".
However these are only my own opinions, you have yours!

my shots (postprocessing disabled):
[EDIT]updated

Last edited by nicco; 9th December 2004 at 13:08.
nicco is offline   Reply With Quote
Old 8th December 2004, 15:40   #217  |  Link
temporance
Registered User
 
Join Date: Mar 2002
Posts: 486
@ babayaga

I downloaded your samples, wow that's a high bitrate you use! I prefer the xvid version: the film grain / noise is more alive. The AVC encode has a slight 'dirty window' effect - there seems to be some artificial temporal correlation or "stickiness" of the noise (and the image appears very slightly softer).

@ Tommy

I don't have a problem with ASP deblocking. It doesn't hit texture in flattish areas of the image in the same way that AVC does. Perhaps it's partly helped by the larger block size in ASP. Also ASP deblocking is completely in the hands of the person and the codec doing the decoding - in-loop filtering is decided at encode time.

I have more experimentation to do. Trying AVC with no and minimum deblocking next....
temporance is offline   Reply With Quote
Old 8th December 2004, 15:44   #218  |  Link
temporance
Registered User
 
Join Date: Mar 2002
Posts: 486
Quote:
Originally posted by nicco
[EDIT] my shots (postprocessing disabled):
[/B]
Interesting. But is it fair to compare DivX and xvid without PP against a deblocked AVC image? Btw, how does the moving picture look - is the blockiness of ASP more/less apparent?
temporance is offline   Reply With Quote
Old 8th December 2004, 16:54   #219  |  Link
skal
Registered User
 
Join Date: Jun 2003
Posts: 121
Quote:
Originally posted by temporance
Interesting. But is it fair to compare DivX and xvid without PP against a deblocked AVC image?
Also: how comes "AVC DEBL=1" invented the two sharp skin spots
on the bottom-right corner of the very last picture?

(is it what is refered to as "film grain generation" ? hehe )

More seriously: did you check all the pics are P-frames, e.g.?
(or b-frames, for that matter)

-Skal
skal is offline   Reply With Quote
Old 8th December 2004, 19:02   #220  |  Link
nicco
TeacH Me!! I'm so n00b!
 
nicco's Avatar
 
Join Date: Nov 2003
Location: Italy
Posts: 316
Quote:
originally posted by temporance
But is it fair to compare DivX and xvid without PP against a deblocked AVC image?
In fact, there is also a non-deblocked avc image! Postprocessing is disabled both in asp and avc.
Quote:
originally posted by temporance
how does the moving picture look - is the blockiness of ASP more/less apparent
The blockiness is less apparent in avc-deblocked movie(the last one), even compared with postprocessed asp.
Quote:
originally posted by skal
how comes "AVC DEBL=1" invented the two sharp skin spots
on the bottom-right corner of the very last picture?
Don't ask me!
nicco 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 23:52.


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