Maestro and multiPGCs... again.

[Fmazzanti]

Hi everybody,
I've trying to backup a multiPGC movie with teh IfoUpdate+CCE+Maestro method explained in earlier posts. Actually my VTS_01 containing the main movies has 7 PGCs. Actually IfoUpdate analyzes this VTS and creates the following file
Chapter assignment information for DVDMaestro Playlists:

- Playlist #1 (PGC #1); [VOB Sector: 0]
- 00:00:00:00
- 00:11:19:05
- 00:19:27:18
- 00:32:24:12
- 00:41:34:11
- 00:49:39:21
- 00:57:55:11
- 01:07:52:02
- 01:16:21:09
- 01:23:50:13
- 01:31:52:23
- 01:44:10:20
- Playlist #2 (PGC #2); [VOB Sector: 2916480]
- 01:47:00:00
- Playlist #3 (PGC #3); [VOB Sector: 2922748]
- 01:47:19:00
- Playlist #4 (PGC #4); [VOB Sector: 2922771]
- 01:47:24:00
- Playlist #5 (PGC #5); [VOB Sector: 2938461]
- 01:48:12:00
- Playlist #6 (PGC #6); [VOB Sector: 2938466]
- 01:48:12:12
- Playlist #7 (PGC #7); [VOB Sector: 2944961]
- 01:48:31:12
Any unlisted PGCs have been determined to be nulls (dummys)
and will remain intact in the original file.

Now I create the DVD2AVI project, feed that to CCE to compress the file and start the Maestro party. I load the .mpv created by CCE and... it says video length its length is 01:48:25:23 !!
That's not possible since then PGC7 (=playlist 7) would be OUT of the stream. Actually I've tried to put a final chapter n Maestro by hand and compile the project, and you see in playback that actually PGC6(playlist 6) really gets truncated at sec 12 when it was supposed to last for 19 secs or so...

So does anybody know what's going on here?
Thanks...

[jdobbs]

Use "Adjusted Cell Mode" instead. You probably have PGCs that point to the same cells.

[Fmazzanti]

Can you give details on what to do what you say? As far as what I read on the guides, I *must* use 'Maestro Mode' in IfoUpdate to create the information about playlists. If I set IfoUpdate to 'Adjust cells mode' then I loose that info, and the 'Adjusted-maestro.chp' file now reads

$Spruce_IFrame_List

00:00:00:00
00:11:19:05
00:19:27:18
00:32:24:12
00:41:34:11
00:49:39:21
00:57:55:11
01:07:52:02
01:16:21:09
01:23:50:13
01:31:52:23
01:44:10:20
01:47:00:00
01:47:19:00
01:47:24:00
01:48:12:00
01:48:12:12
01:48:31:12

that is, a usual maestro chapters file. I can't check the 'maestro files for multiPGC' checkbox in 'adjust cells mode'. Besides, this chapters file also shows the last chapter at 01:48:31:12, a time that doesn't seem to exist in the .mpv file imported into Maestro.

Furthermore, if what you want to say is that I must run IfoUpdate in 'adjust cell mode' once the Maestro project is done when finishing the project, it doesn't work neither. That's because the 'original IFO file' has 7 PGCs, while the one created by maestro has 8 (this is the well known story about Maestro putting an extra PGC at the beginning as you know), so IfoUpdate refuses to update the IFO...

Any help? Thanks!

[jdobbs]

Playlists, Shmaylists. That's too much trouble. Adjusted Cell Mode is an improved way to do Maestro multiple PGCs (although it isn't specific to just Maestro). I will assume you know a lot of this so I'm condensing the steps:

1. Rip to your hard drive as usual.

2. Put IFOUpdate in "Adjusted Cell Mode"

3. Got to "File" and then "Save Chapters"

4. Click on the "Adjusted Mode Chapter Files" checkbox and "Save"

5. The output files include chapter files for CCE and for Maestro.

6. Use DVD2AVI and create a .D2V, AVISYNTH to create a .AVS

7. Create an appropriate .ECL in CCE, then edit it, and add the CCE chapters from the .TXT file.

8. Run CCE against it (RoBa or Robshot, whatever)

9. Do Pulldown.exe if necessary on the resulting video file.

10. Import the Video and .AC3 into Maestro.

11. Import the .CHP file into Maestro (check the last chapter -- as always it can sometimes not come in correctly if it is too close to the end)

12. Compile

13. Now, replace the original .VOB set with the new one (save the old one somewhere in case it gets screwed up). Also save the .IFO and .BUP files... to be safe.

14. Open IFOUpdate (still in Adjusted Cell Mode) and point to the original and newly compiled IFO files -- make sure the only options checked are "Autocorrect VTS Sectors" and "Autoanalyze Original IFO" .

15. Push the "Update IFO" button. All done.

The same exceptions exist as in Maestro mode. Sometimes a PGC with a missing soundtrack can cause audio sync loss.

[Fmazzanti]

Thanks! I'll try that... BTW, regarding step 7, where do I have to put the chapter information in the ecl file? Or it doesn't matter the exact place where you put that in the file?
Thanks again...

[jdobbs]

It doesn't matter. I usually put it in the same directory as the AVS and the D2V.

[Fmazzanti]

Uhhh... I'm asking about where INSIDE the .ecl file should I place the chapter information, not where should I place the file with the chapter info generated by IfoUpdate :D

[jdobbs]

Just a note for understanding... After the compile you will have a single PGC... don't worry. Adjusted Cell Mode works at the cell level and points all the cells to the correct PGC when you do the update.

Areas to watch: Multiple Angles won't work. (They wouldn't under Maestro mode either). You can, however, remove the angles using DVDDecrypter and then select "Ignore Angles (Adjusted Cell Mode)" as an option. It will make all the selected angles play the same one.

Also don't do "The Matrix" with it... I have yet to come up with a way to duplicate the interleaving using the CCE method. A transcoder like DVD2one will do it though (with the expected quality drop).

[jdobbs]

Put it right after the [File] area -- it will look something like this:

[file]
name=D:\Image3\vts1.avs
type=0
frame_first=0
frame_last=135942
encode_first=0
encode_last=135942

[jdobbs]

To clarify a little more -- after you add it it might look like this:

[file]
name=D:\Image3\vts1.avs
type=0
frame_first=0
frame_last=135942
encode_first=0
encode_last=135942
chapter=0:
chapter=985:
chapter=1865:
chapter=5897:
chapter=8301:
chapter=10189:
chapter=17441:
chapter=20069:
chapter=25541:
chapter=30865:
chapter=36605:
chapter=37885:
chapter=43569:
chapter=54181:
chapter=54925:
chapter=71185:
chapter=75693:
chapter=78897:
chapter=96385:
chapter=99905:
chapter=101189:
chapter=102705:
chapter=126704:
chapter=135933:

[Fmazzanti]

Ok thanks! You're helpful and damn fast replying. I'll try that and tell you what ;)

[Fmazzanti]

OK, here I report again. I tried what you said and it is certainly easier than the Playlist method... but unfortunately yields (in this case) exactly the same result. PGC1 (movie) and PGC2 (logo) get at the right places, PGC3 doesn't show up at all, and its place is taken by the first 4 secs of PGC4. Then PGC4 starts 4 secs late and its last 4 secs are the beginning of PGC5 and so on and so forth.
Everything gets screwed up due to this missing PGC3... which I don't know why gets missed. It is made of an still image telling the name of the movie...

[oddyseus]

I use Adjusted Cell mode a lot and I have to say it always worked as it should provided that u follow the guidelines of CCE as well as Maestro chapter placement.

From what u r describing there must be something wrong with the CCE chapter list.

Can u post the ifo and CCE chapter list for us to study?

[Fmazzanti]

I don't know. I included the chapters in *exactly* the way jdobbs told me. And please confirm I can post the whole Ifo file here (aren't there restrictions on length, etc?).
One thing I noticed is the following: PGC3 is supposed to display a still image with the name of the movie and last 5 secs. However by looking at the .mpv file generated by CCE, I see that image for 1/2sec or so... and that's what makes all PGCs be screwed up from then on.
Appart from that, I can't tell.
Thanks again.

[oddyseus]

It really doesn't matter how long CCE has encoded the pgc3 as long as in the timeline is contained between 2 chapter points that define the start and end of it.

U can paste the header of the ifo, the screen that contains the details about each pgc. Or u can attach it.

[Fmazzanti]

Originally posted by oddyseus
It really doesn't matter how long CCE has encoded the pgc3 as long as in the timeline is contained between 2 chapter points that define the start and end of it.

you must be right, I'm not going to discuss that. However I'm convinced that the problem is there (or either that I must do something else): PGC3 is supposed to last 5 secs while the image encoded by CCE takes 0.5secs, so the whole PGC3 takes 4.5 secs from PGC4. According to what you say, I could add by hand a last chapter deffining the end of PGC3, but then I guess IfoUpdate would complain about chapters in the original and authored IFO files not being the same...


U can paste the header of the ifo, the screen that contains the details about each pgc. Or u can attach it.
Attachments should be approved by mods, isn't it? I'll see what I can do tonight (now I'm at work).

Thanks for your help.

[oddyseus]

It isn't .5sec. It is 40ms that is 0.04secs. Its a single frame that is displayed for 5 secs. It is a limitation of DVD2AVI. even if you can get past it, I bet u ll be having issues with audio sync after this pgc.

I ll be waiting 4 your data.

[Fmazzanti]

Touche :sly: Audio synch problems are there too.
Right now this is not a big problem because I can strip out the other PGCs (which are not that important after all), but still I'd like to know what's going on there and how to fix these problems...
Tonight, tonight...

[Fmazzanti]

Ok, stupid as it may seem, I don't know how to put the info here.
How do I post images? And here at the bottom says I can not post attachments, so what...

[Xayd]

Host the image on your own somewhere and link us to it.

BTW I've seen the same problem you have here. oddyseus' explanation makes sense.

Since it's a DVD2AVI limitation I assume demuxing the video via some other application and then re-encoding from the demuxed source would overcome it?

[Fmazzanti]

Ok, here I come again. I've hosted a couple of IFO files in the links listed below. Both files correspond to VTS_01_0.IFO, but the former is the original one and the later what I get after IfoUpdating in 'Adjust Cell mode' as siuggested
Original: http://www.salleurl.edu/~mazzanti/VTS_orig.IFO
IfoUpdated: http://www.salleurl.edu/~mazzanti/VTS_author.IFO
Please let me know if you encounter any problem downloading it. Also let me know if you find a nice solution :D

[Fmazzanti]

BTW, you'll notice the number of precommands in the authored file does not match the same number in the original file. That's ok, I did that to finally bypass this PGC, but restoring that to the original values does not solve the problem (actually it is then when troubles show up...)

[oddyseus]

Well yesterday we were celebrating the Greek Revolution of 1821 against the Turks and since I don't currently have a working Internet connection at Home I couldn't reply back to your posts.

However today is another day at work and here I am.

Afaik the ifoupdate WASN'T done in adjusted cell mode, is was in another one, Maestro Mode most probably since it has distinguised the playlists as pgcs and updated the relevent times.

From what I can tell the update went OK. The VTS_author.ifo should work without any navigational problem except the audio sync that might exist after pgc3. Did u 'Get VTS Sectors' it?

[Fmazzanti]

Originally posted by oddyseus
Well yesterday we were celebrating the Greek Revolution of 1821 against the Turks and since I don't currently have a working Internet connection at Home I could reply back to your posts.

Hope you had a nice day then :D


Afaik the ifoupdate WASN'T done in adjusted cell mode, is was in another one, Maestro Mode most probably since it has distinguised the playlists as pgcs and updated the relevent times.

Uhmm... I can't tell anymore, since I've done many tests and the first guide I found searching the forum says you should put it in Maestro mode. However I then tried what was suggested in this thread and I sure used the 'Adjust cell mode' afterwards. I would say what I put in the link was the one obtained in 'adjust cell mode', but I'm not going to fight for that.


From what I can tell the update went OK. The VTS_author.ifo should work without any navigational problem except the audio sync that might exist after pgc3. Did u 'Get VTS Sectors' it?

Well, there are no navigation problems other than the ones I said before: PGC3 plays for 0.5 secs while in the original it plays for 5 secs. The final product is not that nice then, and I can't find the reason/cure for that. And the audio synch problems are still there...
My test in 'adjust cell mode' (which I did for sure) rendered exactly the same results as long as DVD playback is concerned, as I said before...

But let me finish saying that I do use the search button before posting as a rule. Many times I get answers, many other times I don't. And of course I do not ask if I find the solution already posted...
Thanks anyway.

[oddyseus]

I am studying the posts and the ifos and I must say I am a bit confused.

From the original ifo we can see that there isn't any cell that is used twice, all the vob-cell ids r unique.

Assuming that the d2v project is correct u must have other pgcs, not only no3, that r stills displayed for a given time.

Given that dvd2avi cann't cope with this situation yet u will face more audio sync problem later on in the proccess.

Your situation and everybody's else that have a stil image as a pgc in the middle of a vts can be solved only by vob-id ripping. Right now u can try to by-pass the offending parts in the timeline by dragging the start-end of the stream to the desired timecode, remove the part that is now pgc3 and replace it with a still image with duration of 5secs. U should do also the same with any other pgc that falls in the still image category.

[Fmazzanti]

@odysseus,
thanks for the reply. I'm glad you said this is somewhat more starnge than other backups, since this is the first time I face a multiPGC backup, following the guides, and failing. I'm a little at lost with all this stuff and first time I try something like this... bang! hit the wrong target. Ahh... anyway, you learn from everything.
In fact I must admit that my *real* problem has ben solved, since all the PGCs that were affected by short-time displaying or audio synch problems were not that important and I could get along by stripping or bypassing them. The final product still has the original menues and the main movie without problems, and that's what I was looking for after all. However, I was interested (and still I am) in getting to knowhow to do the backup without these problems...

[Ak47]

I've looked at the orig IFO...to re-auth this title you'll have to

1. Rip it to separate PGC's

2. encode the larger PGC's.. leave the v small ones alone tor time being.

3. Demux the content from all very small PGC's

4. Line PGC's up in correct order on maestro timeline

5. MAKE SURE PGC CONTENT LENGTH MATCHES IFO FRAME TIMINGS. substitue black stills if need be, as length can be easily adjusted.. and maestro will accept on timeline mixed with video NP.

6. Compile and IFOupdate as normal...

generally the prob with stills/audio sync is easy to fix once you understand what's going on.... this title falls into the structure type that this fix won't resolve.. see last 2 paragraphs below.

Stills are the cause of sync loss because IFOupdate (brilliant as it is) is reading the CONTROL information from the IFO and not the actual content from the VOB... DVD2AVI on the other hand ignores control information and bases the d2v on actual content of the VOB(s)

Stills usually occupy a single VOB sector...2 video fields, one of which is often black... it's easy to see this if you pop open the status panel in DVD2AVI (press F6) and scroll through the VOB. 2 fields make up a single frame.

Usually the IFO will tell the player to display the still for a number of seconds/frames, which means that when you IFOupdate you actually force the audio to lose sync... the hold in the IFO would normally just keep the silence of the single vob sector for as long as the IFO stipulates, in titles of this type VOB ID is used to differentiate stills content... even if theres a 5 second hold there are still only 2 fields of content...

an IFOupdate of 5 secs actually inserts a chapter at 5 secs so to make the audio line up you would require 5 secs of actual audio content... or a delay of 5 secs on the audio file....

the workaround is relatively easy... zero any delay on the audio (ac3 delay corrector) then insert delay to match the hold time in the IFO, less the actual stills content time...

i.e for a still that holds for 00:00:00:10 there will be 1 frame of content (single vob sector). An additional offset of 00:00:00:09 frames is required to make the audio line up.

In the case of PAL a single frame is about 40ms, so 9 x 40ms = 360ms of delay to add to the audio file.. and then it will line up NP

I use IFOupdate adjusted Cell mode chaps with IFOEdit multiplex of the m2v/ac3 and subs... the method described resolves any stills delay probs I've ever encountered.... even when multiple stills are present. the only thing that will beat this method is structure like

still/video/still/movie

don't yet have a way of inserting delay for the 2nd still... without making use of an authoring app....suggestions welcome as IFOEdit mutliplex beats the tar out of any authoring based method for speed and simplicity.. just load m2v/ac3/sup and chaps then mux

[oddyseus]

Quite long and detailed analysis.

Congrats