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 > (HD) DVD, Blu-ray & (S)VCD > One click suites for DVD backup and DVD creation

Reply
 
Thread Tools Search this Thread Display Modes
Old 24th May 2007, 13:51   #81  |  Link
Sir Didymus
Registered User
 
Join Date: Mar 2004
Location: Italy
Posts: 948
Some encoders produce clips without the sequence end code at the end...

Release 1.0 of the BatchMux.exe program (it is the authoring wrapper of the plugin) have a patch for this very specific condition...

This release is not included in the plugin download (I was planning to update it in the next - needed - release of the plugin itself...)...

Do you mind checking if this release of BatchMux.exe:

http://forum.doom9.org/showthread.ph...light=BatchMux

is fixing the problem ?

Cheers,
SD
Sir Didymus is offline   Reply With Quote
Old 24th May 2007, 14:16   #82  |  Link
Fishman0919
FishmanMod Android Dev
 
Fishman0919's Avatar
 
Join Date: Apr 2004
Location: Somewhere else, maybe Arizona Bay
Posts: 1,143
Quote:
Originally Posted by manolito View Post
It does not look like this is caused by BatchMux. Muxman detected a couple of buffer underflows, and if something like this happens, Muxman does not finish the authoring correctly. It does not behave like DVDAuthor which will finish and leave it up to you if you want to keep the result.

In your case Muxman seems to be in an endless loop. This has not happend to me so far (maybe it was introduced with the latest version 0.15R), but if underflows were present, the VIDEO_TS folder would never contain all the files necessary for burning.

Bottom line: TMPGEnc did a bad job encoding your video. Switch to CCE, HC or QuEnc...


Cheers
manolito
True about TMPGEnc... just want to try something new/diff for testing. With TMPGEnc, DVD2DVD doesn't do the pulldown it's done with TMPGEnc...just trying diff stuff to see if something would fail
__________________
"Cinderella story, out of nowhere,
former greenskeeper, now about
to become the Masters champion.
It looks like a mirac- it's in the hole!"

Last edited by Fishman0919; 24th May 2007 at 14:20.
Fishman0919 is offline   Reply With Quote
Old 24th May 2007, 16:51   #83  |  Link
Sir Didymus
Registered User
 
Join Date: Mar 2004
Location: Italy
Posts: 948
Quote:
In your case Muxman seems to be in an endless loop. This has not happend to me so far (maybe it was introduced with the latest version 0.15R)...
A similar situation was present in FAVC (and kindly reported by Mr_Odwin some days ago).

Apparently the trouble holds just when using the free version of MuxMan+BatchMux v0.9. The glitch seems to be not present in the professional version of MuxMan (18.8).

By loading (and running) the BatchMux.Mxp project file produced by DVD2SVCD with MuxMan 18.8, I am quite sure that the issue will disappear.

I implemented a patch in BatchMux 1.0 (see my post just above) that should be able to recover from this condition, but it's not a definitive solution. In order to solve the issue in a radical way MuxMan 15R should be updated; however I am not sure if this is in the priorities of MpuCoder...
Sir Didymus is offline   Reply With Quote
Old 24th May 2007, 19:09   #84  |  Link
Fishman0919
FishmanMod Android Dev
 
Fishman0919's Avatar
 
Join Date: Apr 2004
Location: Somewhere else, maybe Arizona Bay
Posts: 1,143
re-ran the authoring part with new patch in BatchMux 1.0, worked perfect.

VIDEO_TS = 4.33 GB (4,650,047,488 bytes)
__________________
"Cinderella story, out of nowhere,
former greenskeeper, now about
to become the Masters champion.
It looks like a mirac- it's in the hole!"
Fishman0919 is offline   Reply With Quote
Old 24th May 2007, 20:55   #85  |  Link
Sir Didymus
Registered User
 
Join Date: Mar 2004
Location: Italy
Posts: 948
Hei Fishman0919, in the last days you gave to me some very nice news (and you performed some excellent tests with the BatchMux plugin...)...



I think it's time to upgrade the plugin to release 1.5 (it should include the new BatchMux.exe 1.0 and the new Mplex_hack 1.0)... Next week, most probably...
Sir Didymus is offline   Reply With Quote
Old 24th May 2007, 21:19   #86  |  Link
Fishman0919
FishmanMod Android Dev
 
Fishman0919's Avatar
 
Join Date: Apr 2004
Location: Somewhere else, maybe Arizona Bay
Posts: 1,143
Quote:
Originally Posted by Sir Didymus View Post
Hei Fishman0919, in the last days you gave to me some very nice news (and you performed some excellent tests with the BatchMux plugin...)...



I think it's time to upgrade the plugin to release 1.5 (it should include the new BatchMux.exe 1.0 and the new Mplex_hack 1.0)... Next week, most probably...
NP, thank you for making an excellent plug-in... glad I could help
__________________
"Cinderella story, out of nowhere,
former greenskeeper, now about
to become the Masters champion.
It looks like a mirac- it's in the hole!"
Fishman0919 is offline   Reply With Quote
Old 24th May 2007, 21:50   #87  |  Link
manolito
Registered User
 
manolito's Avatar
 
Join Date: Sep 2003
Location: Berlin, Germany
Posts: 3,093
Here are some test results using the new Mplex_hack Beta:

Setup:
BatchMux 1.4
New Mplex_hack.exe Beta1
BatchMux.exe 1.0
Encoder: QuEnc

AVI2DVD mode, NTSC (Pulldown ON): OK
(thanks to Nick for providing a NTSC source AVI)
AVI2DVD mode PAL: OK

DVD2DVD mode PAL, DVD Chapters: OK
DVD2DVD mode PAL, Fixed Chapters: OK

DVD2DVD mode NTSC Film (Pulldown ON), DVD chapters: OK
DVD2DVD mode NTSC Film (Pulldown ON), Fixed chapters: OK


What is missing so far is NTSC interlaced Video (Pulldown OFF), I will test this tomorrow.

So far the new Mplex_hack.exe looks like winner.


@Sir Didymus:
I have no idea how you implemented the fixed chapter fix when using CCE 2.70. Do you think it is necessary to test the new Mplex_hack.exe with CCE 2.70 and maybe earlier CCE versions also?


Cheers
manolito
manolito is offline   Reply With Quote
Old 24th May 2007, 23:16   #88  |  Link
manolito
Registered User
 
manolito's Avatar
 
Join Date: Sep 2003
Location: Berlin, Germany
Posts: 3,093
Test Update:

For NTSC Interlaced @29.97 the new Mplex_hack.exe also delivers the correct chapter points. No matter if Fixed Chapters or DVD Chapters are selected.


Cheers
manolito
manolito is offline   Reply With Quote
Old 25th May 2007, 08:20   #89  |  Link
Sir Didymus
Registered User
 
Join Date: Mar 2004
Location: Italy
Posts: 948
Here is the way the Pulldown fix is implemented:

For DVD chapters, there is no much to say: the lines of the chapter file are represented as Encoded Frames and MuxMan need a representation in terms of Presentation Frames. When Pulldown flag is on, all entries of this file should be scaled by 1,25.

For fixed chapters, in the previous version, in the step 6 of the code, the following parameters were extracted from the DVD2SVCD files (this is just the initialisation of the involved variables...):

Code:
	PalFlg = 0;
	MovieFrames = 0;
	hh = mm = ss = ff = 0;
now also pulldown is extracted:

Code:
	PalFlg = 0;
	PulldownFlg = 0;
	MovieFrames = 0;
	hh = mm = ss = ff = 0;
The two timing representations of the movie length (in frames and in ND time code - hh, mm, ss, ff) are extracted and - in the new beta of Mplex_hack, when the Pulldown flag is set, the MovieFrames variable is updated in order to represent Presentation Frames (not Encoded Frames):

Code:
	if (PulldownFlg)
	{
		MovieFrames = MovieFrames + MovieFrames/4;
	}
These two distinct representations are both needed, since, depending on the different DVD2SVCD modes, just one of the two is present in the configuration files of DVS2SVCD. For the production of the proper Celltimes.txt just MovieFrames is used, but this variable should represent Presentation Frames, not Encoded Frames...

In case MovieFrames is derived from the ND time code (hh:mm:ss:ff) then this variable is already representing Presentation Frames, so there is no need to scale it by 1,25.

In case MovieFrames is not derived from the ND time code - e.g. when it is fetched from the configuration files of DVD2SVCD, then it represents Encoded Frames, so it is necessary to scale it by 1,25.

Before this step, it take place the segment of code for supporting CCE 2.70:

Code:
	// In case MovieFrames == 0 try to get this parameter from the ECL file generated by CCE
	if (MovieFrames==0)
	{
		MovieFrames=GetMovieFramesFromEcl();
	}
Where GetMovieFramesFromEcl() basically performs the following:

Code:
	MovieFrames=encode_last-encode_first+1;
But also in this case, the MovieFrames extracted from the ECL expresses Encoded Frames.

So, to answer your question, an impact of the presence of the Pulldown handling on the CCE 2.70 exist (I would say more than just an impact it is a straight relationship). However, due to way the code is implemented I am very confident it should work. I have to admit I did not test this condition...

Hope this quick description is sufficient to clarify...

Thanks a lot for your further testing and for the very nice news and confirmation of the good behaviour of Mplex_hack_beta1.0 !
It is really appreciated !

Cheers,
SD

Last edited by Sir Didymus; 25th May 2007 at 08:25.
Sir Didymus is offline   Reply With Quote
Old 25th May 2007, 08:25   #90  |  Link
ChickenMan
Registered User
 
ChickenMan's Avatar
 
Join Date: Oct 2001
Location: Australia
Posts: 350
Been away for a couple of weeks on holidays and I see all this new action Keep it up guys. Will do some testing also as soon as I get myself re-organised.
__________________
Cheers,
CM
ChickenMan is offline   Reply With Quote
Old 25th May 2007, 08:38   #91  |  Link
Sir Didymus
Registered User
 
Join Date: Mar 2004
Location: Italy
Posts: 948
Hi ChickenMan!

Take your time... I will wait for some feedback also from your side before updating the whole plugin.

For making it short, and for the benefit of possible other casual readers, the two component (Mplex_hack.exe and BatchMux.exe) need to be updated. You may get the most recent releases here:


Edit 20-06-2007 Link removed. See first page for the last available release...

Cheers,
SD

Last edited by Sir Didymus; 20th June 2007 at 11:10.
Sir Didymus is offline   Reply With Quote
Old 25th May 2007, 21:27   #92  |  Link
manolito
Registered User
 
manolito's Avatar
 
Join Date: Sep 2003
Location: Berlin, Germany
Posts: 3,093
@Sir Didymus

Thanks for the explanation about how CCE 2.70 chapters are implemented.

I still had the urge to test it, and -Bingo- I did find bugs!

Mode DVD2DVD, source was NTSC FILM, Pulldown ON.
For fixed chapters the chapter points were at the correct position, but the last chapters of the movie were missing.

For DVD chapters all chapter points were there, but at the wrong positions. They should have been expanded by 1.25, but obviously they were not.

Just to make sure my source was OK I crosschecked using QuEnc, and here everything was perfect.


And I also found another issue which is not related to chapter creation at all and which has been there all the time:

Mplex_hack.exe seems to assume that the Mplex command line always starts with the parameter "-f". This is not always true. In my tests I often use "Keep original audio", and many times DVD2SVCD determines that there is an audio delay which must be corrected. In my latest test the Mplex command line issued by DVD2SVCD was like this:

"E:\Programme\DVD2SVCD\DVDAuthor\Mplex_hack.exe" -1 17ms -f 8 -o "I:\Movies\MPlex_Muxed_File00.mpg" "I:\Movies\Pulldown_Encoded_Video_NTSC.mpv" "I:\Movies\Extracted_audio_1.ac3".

Please notice that there are two additional parameters in front of the "-f" parameter. In this case Mplex_hack.exe fails to determine the needed folder information.

I found that the solution is to simply discard the delay parameters in front of the "-f" parameter. I never got a sync problem when I just ignored these parameters.


And last something OT:
I always have the urge to try newer versions of the SVD2SVCD helper applications. AviSynth 2.57 works well with D2S, but for DGIndex / DGDecode it is different. Version 1.45 which comes with D2S is the only version which works. Newer versions (from 1.46 to 1.49) all have issues:
For PAL sources they all crash as soon as the encoder starts. For NTSC sources they seem to work at first glance. But they all detect NTSC interlaced video falsely as Film which leads to very ugly results. So do yourself a favor and stick with 1.45....


Cheers
manolito
manolito is offline   Reply With Quote
Old 25th May 2007, 22:56   #93  |  Link
Sir Didymus
Registered User
 
Join Date: Mar 2004
Location: Italy
Posts: 948
Quote:
Originally Posted by manolito
...
Mode DVD2DVD, source was NTSC FILM, Pulldown ON.
For fixed chapters the chapter points were at the correct position, but the last chapters of the movie were missing.

For DVD chapters all chapter points were there, but at the wrong positions. They should have been expanded by 1.25, but obviously they were not.

Just to make sure my source was OK I crosschecked using QuEnc, and here everything was perfect.
This is possible only if the Pulldown flag, in the "dvd2svcd project file.d2s" file is (wrongly ?) set to "No"...
Could you post the Mplex_hack.log file ?

Quote:
And I also found another issue which is not related to chapter creation at all and which has been there all the time:

Mplex_hack.exe seems to assume that the Mplex command line always starts with the parameter "-f".
That's right...

Quote:
This is not always true. In my tests I often use "Keep original audio", and many times DVD2SVCD determines that there is an audio delay which must be corrected. In my latest test the Mplex command line issued by DVD2SVCD was like this:

"E:\Programme\DVD2SVCD\DVDAuthor\Mplex_hack.exe" -1 17ms -f 8 -o "I:\Movies\MPlex_Muxed_File00.mpg" "I:\Movies\Pulldown_Encoded_Video_NTSC.mpv" "I:\Movies\Extracted_audio_1.ac3".

Please notice that there are two additional parameters in front of the "-f" parameter. In this case Mplex_hack.exe fails to determine the needed folder information.

I found that the solution is to simply discard the delay parameters in front of the "-f" parameter. I never got a sync problem when I just ignored these parameters.
Good to know...
I simple was not aware of the different variants of the command lines passed to Mplex...
Do you know if there are other arguments like the one you have discovered, that should be taken into account ?
It will be quite easy to implement the delay parameter (BatchMux.exe always is perfectly adequate to support it...), but it would be important to know what is the overall set of arguments generated by DVD2SVCD.

Quote:
And last something OT:
I always have the urge to try newer versions of the SVD2SVCD helper applications. AviSynth 2.57 works well with D2S, but for DGIndex / DGDecode it is different. Version 1.45 which comes with D2S is the only version which works. Newer versions (from 1.46 to 1.49) all have issues:
For PAL sources they all crash as soon as the encoder starts. For NTSC sources they seem to work at first glance. But they all detect NTSC interlaced video falsely as Film which leads to very ugly results. So do yourself a favor and stick with 1.45....

Cheers
manolito
Nice to know... Never tryed anything else than the DGDecode version bundled with DVD2SVCD... Since it seems this version works quite well, have you specific needs/good reasons for upgrading ?

Anyway, well, it seems there is still some work to perform before releasing a new version of the plugin...



Cheers,
SD

Last edited by Sir Didymus; 25th May 2007 at 23:03.
Sir Didymus is offline   Reply With Quote
Old 26th May 2007, 02:07   #94  |  Link
manolito
Registered User
 
manolito's Avatar
 
Join Date: Sep 2003
Location: Berlin, Germany
Posts: 3,093
Quote:
Originally Posted by Sir Didymus View Post
This is possible only if the Pulldown flag, in the "dvd2svcd project file.d2s" file is (wrongly ?) set to "No"...
Could you post the Mplex_hack.log file ?
Yes , you are right (as usual). When using CCE 2.70 with DVD2SVCD the pulldown flag in the D2S project file is not reliable. In all my tests with CCE 2.70 using NTSC film sources DGIndex determined Film and activated "Force Film". D2S did apply pulldown, but in the D2S project file the pulldown flag was set to "No". No wonder that Mplex_hack.exe got it all wrong.

But there is a very easy fix. Whenever DVD2SVCD applies pulldown it renames the video stream. If the name of the original video stream was "XYZ.mpv" it will be renamed to "Pulldown_XYZ.mpv". I believe that this would be a foolproof way for Mplex_hack.exe to determine if pulldown was used or not. Just parse the video stream file name for "Pulldown" and set the pulldown flag accordingly.


For the Mplex parameters all I could find so far is that D2S can issue delay parameters for the two supported audio streams which appear in front of the "-f" parameter. The format is "-1 xxms -2 yyms". As I already said I never had any sync problems just ignoring these parameters.


Cheers
manolito
manolito is offline   Reply With Quote
Old 26th May 2007, 11:11   #95  |  Link
ChickenMan
Registered User
 
ChickenMan's Avatar
 
Join Date: Oct 2001
Location: Australia
Posts: 350
I just tried my first AVI2DVD encode using BatchMux.exe ver 1.0, Mplex_hack.exe ver1.0, MuxMan 0.15R, Nicks latest Ac3Enc and CCE 2.70. Chapter points set to 5min (300sec).

AVI was 1hr 41min 15sec long NTSC @ 23.976fps with MP3 audio. All went fine. Played final VIDEO_TS folder and all is well and in sync, however, chapter points were at exactly 5min intervals but stopped at 1hr 20min.

A quick calculation shows 101min * 0.8 (thats 1 / 1.25 ) = 80.8min and thats where the last chapter point was put in, at 80min.

EDIT: Checked the Project file and Pulldown was set to NO, even though a Pulldown_Encoded_Video_NTSC.mpv was correctly produced.
__________________
Cheers,
CM

Last edited by ChickenMan; 26th May 2007 at 11:29.
ChickenMan is offline   Reply With Quote
Old 26th May 2007, 19:20   #96  |  Link
Sir Didymus
Registered User
 
Join Date: Mar 2004
Location: Italy
Posts: 948
Quote:
Originally Posted by manolito
...
But there is a very easy fix. Whenever DVD2SVCD applies pulldown it renames the video stream. If the name of the original video stream was "XYZ.mpv" it will be renamed to "Pulldown_XYZ.mpv". I believe that this would be a foolproof way for Mplex_hack.exe to determine if pulldown was used or not. Just parse the video stream file name for "Pulldown" and set the pulldown flag accordingly.
I see...
However, if we change the way Mplex_hack.exe is acquiring the PulldownFlg (using the avi filename instead of some configuration parameters), this have the consequence that avi files whose original filename is "Pulldown_xxx.avi" can not be allowed anymore within DVD2SVCD... Right ?
It would be nice to understand what is the rule adopted by DVD2SVCD when it applies pulldown and implement something similar instead...

@ChickenMan
OK. I see. Thanks for confirming the issue discovered by manolito...
Sir Didymus is offline   Reply With Quote
Old 26th May 2007, 19:40   #97  |  Link
Nick
One click is all it takes
 
Nick's Avatar
 
Join Date: Mar 2002
Location: Heart of the Broad Acres
Posts: 1,404
No - this is not the case.
It is not the original AVI that is renamed.
What he means by the "original video stream" is the MPV file - the encoded MPEG2 video stream created by D2S.

The video encoder is called to create a stream called Encoded_Video_CCE(or whichever encoder)_PAL(or NTSC).mpv

If pulldown is to be applied - which as I understand it is only in the case of a 23.976fps source or use of the IVTC function - the pulldown command line is set to input Encoded_Video_*.mpv and output Pulldown_Encoded_Video_*.mpv . It is this latter file that is used for muxing.

You can therefore tell whether Pulldown has been applied from the parameters in the Mplex_hack commandline - if the filename of the video stream to be multiplexed has "Pulldown" as the first word, pulldown has been applied. If it doesn't, it hasn't.

Hope this clarifies!
Cheers
Nick

Last edited by Nick; 26th May 2007 at 19:52.
Nick is offline   Reply With Quote
Old 26th May 2007, 19:49   #98  |  Link
manolito
Registered User
 
manolito's Avatar
 
Join Date: Sep 2003
Location: Berlin, Germany
Posts: 3,093
Exactly!
Thanks for the clarification...

Cheers
manolito
manolito is offline   Reply With Quote
Old 26th May 2007, 20:14   #99  |  Link
Sir Didymus
Registered User
 
Join Date: Mar 2004
Location: Italy
Posts: 948
OK! I see! Thanks for the clarification Nick...

So, there is no drawback in fetching the PulldownFlg variable from the Mplex command line, as suggested by manolito... Let's code it...



Cheers,
SD

Edit. OK. here it is the link for a quick fix of the two problems reported above. Pulldown is now fetched from the command line (video file argument) and the presence of optional audio delay arguments should not break the plugin anymore.
For the moment, these arguments, if present, are just discarded - I will include them in the process in a future time...

Edit 20-06-2007 Link removed. See first page for the last available release...

Last edited by Sir Didymus; 20th June 2007 at 11:09.
Sir Didymus is offline   Reply With Quote
Old 27th May 2007, 13:06   #100  |  Link
ChickenMan
Registered User
 
ChickenMan's Avatar
 
Join Date: Oct 2001
Location: Australia
Posts: 350
Just downed Beta 3, trying it now

EDIT: Just tried on an NTSC 29.97 avi (no pulldown ) and all chapter points present and correct. Bed time, more testing tomorrow.
__________________
Cheers,
CM

Last edited by ChickenMan; 27th May 2007 at 15:07.
ChickenMan 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 Off
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 09:53.


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