Re: Questions / Bug fixes / Patches

Carl,

I have applied the frame_drop_ratio patch to CVS. When frame_drop_ratio is
set to "0", no N_VOPs will now be produced anymore by xvidcore. Hopefully,
this will workaround the issues you observed with MP4Box muxed .mp4 files.

I have to admit however that I'm still confused: If I recall correctly
we concluded last time when this topic was discussed that MP4Box appears
to rewrite N_VOPs as P_VOPs with all Skip-MBs upon muxing a xvid encoded
stream into .mp4 files using the -nodrop option. Your most recent post
however seems to indicate that the N_VOPs are preserved by MP4Box for
-nodrop. Has MP4Box behavior changed?

Best regards,
Michael


Quoting Carl Eric Codere <[email protected]>:

> Greetings,
>           The issue of N-VOPs has been discussed again and again,   
> such as in this thread:
>
> http://forum.doom9.org/archive/index.php/t-111622.html
>
> Would it not be possible to add an option, such as my early patch to  
>  not use N-VOPs for compatibility reasons. And yes, it is a bug in   
> most players and codecs out there, and not a bug in Xvid... :)
>
> Best regards,
> Carl
>
> --- On Mon, 11/22/10, Carl Eric Codere <[email protected]> wrote:
>
>> From: Carl Eric Codere <[email protected]>
>> Subject: Re: [XviD-devel] Questions / Bug fixes / Patches
>> To: [email protected]
>> Received: Monday, November 22, 2010, 10:30 PM
>> Greetings,
>>         I once again reproduced the
>> problem with the frame drop ratio. I
>> did as follows with the following mencoder options:
>>
>> -xvidencopts pass=1
>> bitrate=768
>> me_quality=6
>> trellis
>> vhq=1:min_iquant=2
>> min_pquant=2
>> min_bquant=2
>> frame_drop_ratio=0
>> nopacked
>> par=vga11
>> profile=sp4a
>>
>> Same thing for pass 2. When converted to a MP4 file using
>> Mp4Box with the following options:
>> ..\tools\mp4box\mp4box -fps %fps%  -nodrop -new -add
>> %2_%bitRateHigh%_MP4@SP_Xvid.avi
>> %2_MP4@SP_%bitRateHigh%_Xvid.mp4
>>
>> -> 27 frames are defined as being N-VOP, G-Spot
>> specifies them as non packed bitstream N-VOPs.
>>
>> Now, when I use graphedit and replay this video with the
>> ffmpeg video decoder and frame grab all these frames, the 27
>> frames seem to be missing at constant bitrate. Of course,
>> this seems to be a bug with the FFMPEG decoder (which does
>> not occur when i use the 3ivx decoder), but the drop ratio
>> workaround could possibly solve this problem, as well as
>> other numerous compatibility problem, no?
>>
>> Best regards,
>> Carl
>>
>>
>>
>> --- On Mon, 8/30/10, Marcin Jaworski <[email protected]>
>> wrote:
>>
>> > From: Marcin Jaworski <[email protected]>
>> > Subject: Re: [XviD-devel] Questions / Bug fixes /
>> Patches
>> > To: [email protected]
>> > Received: Monday, August 30, 2010, 9:30 AM
>> > Quoting Carl Eric Codere:
>> > > Greetings,
>> > >          But i don't think my
>> > patch was applied, should it be applied? Have you
>> confirmed
>> > that it causes problems in some players or not? On my
>> side,
>> > I thoughr it would cause problems (I am converting the
>> Xvid
>> > streams to MP4 streams using mp4box, but it seems
>> after
>> > testing that i could not reproduce the problem...
>> > >
>> > > Best regards,
>> > > Carl
>> > >   
>> > It is sufficient to use frame_drop_ratio == -1, that
>> will
>> > take care of
>> > the "<=" in line 1887 (line 1178 uses the same
>> method).
>> >
>> > There have been some reports (on Doom9, I think)
>> about
>> > problems with
>> > N-VOPs (the real ones, not Packed Bitstream) on
>> standalone
>> > players. But
>> > these were AVI files... If a player is smart enough to
>> play
>> > MP4 (like
>> > those few players with Nero Digital certification), it
>> may
>> > be smart
>> > enough to cope with N-VOPs.
>> >
>> > Some patches that disable N-VOPs with frame_drop_ratio
>> == 0
>> > have been
>> > floating around before, but I thought it would be
>> better to
>> > retain the
>> > original behavior of 0 and use -1 to turn off N-VOPs.
>> It
>> > seemed
>> > reasonable to keep the official API as it is instead
>> of
>> > redefining it. I
>> > must admit I haven't tested the behavior of my own
>> DVD
>> > player with
>> > N-VOPs, but I think users should have the possibility
>> of
>> > disabling them
>> > if they wish it (flexibility has always been a part of
>> the
>> > open source
>> > philosophy).
>> >
>> > Best regards,
>> > Marcin Jaworski
>> > _______________________________________________
>> > Xvid-devel mailing list
>> > [email protected]
>> > http://list.xvid.org/mailman/listinfo/xvid-devel
>> >
>> _______________________________________________
>> Xvid-devel mailing list
>> [email protected]
>> http://list.xvid.org/mailman/listinfo/xvid-devel
>>
> _______________________________________________
> Xvid-devel mailing list
> [email protected]
> http://list.xvid.org/mailman/listinfo/xvid-devel
>
>