Strange Scenarist error with normal solution not working.

[Chu]

Hello all. When I try to build the current DVD I am working on, I get the following errors :

[QUOTE]
Info Multiplexing
Info Multiplexing VOB Main Stream
Error The number(0.0) of frame in a VOBU is less than 11.99 (around 00:00:33:01).
Warning Exiting is canceled because "ExitVobuSizeError" is 0 in the file Scenarist.ini.
Error The number(0.0) of frame in a VOBU is less than 11.99 (around 00:00:33:01).
Warning Exiting is canceled because "ExitVobuSizeError" is 0 in the file Scenarist.ini.
Error Can not read a GOP at 00:00:33:01 (try to read = 12624 B; read done = 11092 B).
Error Please check the video stream (H:\ARJUNA_V4\VTS06\VTS__06_P01.I-TFF.16~9_1.NOPULL.MPV) or the .vif file or re-generate them.
Error dvd_mux : DoMux Multiplexing Error
Error Terminated Multiplexing (Arjuna4.scn-VTS_06_PGC_01_t.vob).
Error Can't read the input stream.
Error Multiplex is failed.
Info Multiplexing failed, Track<VTS_06_PGC_01>

Error MuxFromDB:MuxVTS failed
Error Multiplex failed
Error DVD files could not be created
[\QUOTE]

The normal problem is for some reason the length reported by Scenarist doesn't match the actuial length, and the normal solution being to edit the according values in the SCP files and re-import. The problem is that the values are correct.

There is one oddity though that might help. The length on the original video file is NOT the value reported in ReAuthorist (it's 00:00:00:04 longer), but does match the value of the encoded video. I don't know if this is normal or not, but it was suspicious.

Also, the last time I had an error similar to this (when the normal method worked), someone suggested making sure that no chapter points were beyond the end of the video. None are.

The last thing is that when importating a file that will generate this error on build, the import reports that the times do not match. EVERY importated VTS throws this error. There are 22 total, with me reauthoring 16 of them. One is the standard "Dolby Digital Comet" video you've probably seen on tons of DVD's, which I think I can safely send someone if they want to take a look at the vob/ifo/bup since your almost gaurenteed to have a DVD with this track.

-Chu

[influenza]

I don't mind taking a look at the ifo files.

[Chu]

The ifo is here (http://web.ics.purdue.edu/~milutino/arjuna_v4/VIDEO_TS/VTS_06_0.IFO). If you (or anyone else) is interested, the "root" of my working directory only including the files that have something to do with this problem (the 6th VTS) is located here (http://web.ics.purdue.edu/~milutino/arjuna_v4/). Thanks for taking a look at this, and please tell me if I am just doing something really stupid.

-Chu

[influenza]

I'v taken a look at the files and could not detect anything strange. The files compile fine on their own. (I've imported audio and video into scenarist and compiled it). What if you try this? So import audio/vid and chapters and compile it. If this works out fine then I guess it's safe to say somethings wrong with your reauthorist file.

[Chu]

Originally posted by influenza
I'v taken a look at the files and could not detect anything strange. The files compile fine on their own. (I've imported audio and video into scenarist and compiled it). What if you try this? So import audio/vid and chapters and compile it. If this works out fine then I guess it's safe to say somethings wrong with your reauthorist file.

Your right, they do compile fine on their own. After poking around a TON, I relize that the problem really is Chapter points past the end of the video. EVERY SINGLE TRACK has three chapters, each 00:00:00.12 in length, and the very end of the video file. I really wish I knew why they did this :(

What scenarist did though (2.7) is just scoot them back to the end of the VTS, and "Stack" them. I have no clue how to tell what the end of a scene is in Scenarist is, since it just gives me a Scene time and not a Start time, but I guess it is safe to say they are extending past the end of the VTS.

Anyways, my gut instict tells me I just can't nuke them in Scenarist/IfoEDIT, or bad things will happen. The problem is I really don't know enough about DVD's to feel my way around this. Any chance you can suggest a solution to this? Is it safe to just nuke the chapter at the end of each VTS, is it safe to somehow move them a bit back in scenarist and hope for the best, or is something more drastic required?

-Chu

[influenza]

I would just extract the chapter points with ifoupdate and the create the tracks myself in scenarist. Export the script in scenarist and use scenchap to add the saved chapterpoints from ifoupdate. Then compile it.

[Chu]

Originally posted by influenza
I would just extract the chapter points with ifoupdate and the create the tracks myself in scenarist. Export the script in scenarist and use scenchap to add the saved chapterpoints from ifoupdate. Then compile it.

I've found the root cause of the problem (finally). Scenarist won't let you make a chapter shorter then 0.4 seconds. It takes the correct times from the .SCP files, and them alters them to try to adhere to the 0.4 second rule, but apparently fails miserably. Scenchap doesn't seem like it will get around this problem :(

I really don't know much about how the DVD strucuture works, but form poking around some more in IFOEdit it appears that I could probably get around this by setting the chapter points wherever I wanted, and then edit the VTS_PGCITI after I compiled the vob's. The problem is this will be incredibly tedious, and looks incredibly prone to OBOE errors considering the way the data is laid out :(

I was wondering if you knew a better way then editing the VTS_PGCITI table by hand?

[Chu]

You know, I just realized an extremly trivial solution to this -- if possible -- would just to somehow add those extra 3 frame or so that dissapeared when reauthoring back to the video, and then edit the .SCP file to account for the new time. Is there any utility out there that will let me add ANYTHING (preferrably something sensable ...) to the end of my m2v to get those last extra frames back?

If not, I suppose I could add them in the avs files, however that would be even more time consuming to do in a sensable way, even though it would be much less error prone the whipping out IFOEdit.

-Chu

P.S. :

I am curious, somehow ReAuthorist knew these frames would be dropped since the time in ReAuthorist reflects the authored time, not the original time. How does it do this? Might be useful to write a small script to check for this behaviour and then just pad the .avs files with the missing frames.