DVDRebuilder 1.10.8 Error Report

[godhead]

Ok -- fixed it. It wasn't really related to custom matrices, but they made it more obvious. The problem was stray characters from previous encodes being left at the end of ECL files. It was caused by the way the buffering was done for read/writes of the ECL files.

I'm looking at one more quick fix and hopefully will get v1.10.9 out today.

I was just about to ask if you needed any information from my latest encode of "Three...Extremes" that presented the same Corrupt Stream message. I will wait for v1.10.9 and try again to see if this corrected the error.

[stereo]

OK, good to know. I've aborted my project, and am looking forward to the next update.

[jdobbs]

Just as a note, If you don't get the "Corrupt Stream" error -- your encode should be fine. The error is also very unlikely to occur unless you have matrices set for some segments and not others. That's why I had so much trouble repeating it. I usually use default matrices.

[Trahald]

i use ultra low on everything but the feature.

and just to be on the same page.. i did the entire project with that setting (one click) and it was fine.. then i blanked some segments.. turned off one click and immediately hit rebuild. then i got the errors. then i unchecked them and immediately hit rebuild.. and no errors.

[Sharc]

Got a working backup now, even with all the WARNINGS:Corrupt stream detected.

I sent you the files by mail.

[jdobbs]

I'd be very careful... If you got the warnings that means there is a lot of superfluous video data included in the backup. It isn't likely to have been truly successful. You may want to wait for the new version.

[jdobbs]

i use ultra low on everything but the feature.

and just to be on the same page.. i did the entire project with that setting (one click) and it was fine.. then i blanked some segments.. turned off one click and immediately hit rebuild. then i got the errors. then i unchecked them and immediately hit rebuild.. and no errors.You didn't reencode on the second REBUILD attempt?

In your case the error message is correct then, and the source is corrupted. Any time you save changes from the Viewer/Editor, you have to ENCODE again. What DVD-RB saw in your case was segments that had large numbers of frames in them -- but had been set to be blanked by the editor.

I could prevent the editor from running after an ENCODE is complete -- but that would then prevent any changes/reencoding after the first time through. The only other choice would be to delete all the M2V files and set the progress flag to "PREPARE complete" when you save Viewer/Editor changes. I worry, though, that people won't like it very much when they can accidentally delete three hours worth of encoding.

So what you are seeing is truly a "Corrupt Stream", based upon what the INF file says the files should look like compared to what they do.

[Trahald]

ahh.. hah. well.. i guess if it accidently hinted toward the right direction then it was a good thing. but yeah.. a warning or something would be good. i assumed the encode would happen automatically if it needed to (dropping down to 2-click mode)

[jdobbs]

@SpazzHH

I left you a PM...

[jdobbs]

ahh.. hah. well.. i guess if it accidently hinted toward the right direction then it was a good thing. but yeah.. a warning or something would be good. i assumed the encode would happen automatically if it needed to (dropping down to 2-click mode)I may have to look at it and maybe reset it to reencode with an appropriate warning.

[Sharc]

If it's in the VIDEO_TS.VOB or the VTS is smaller than the minimum size setting (default of 50MB) it won't show up in the viewer/editor and is copied intact.

It's in the VIDEO_TS.VOB.
Does it mean that the content of VIDEO_TS.VOB is never visible in the Viewer/Editor, is always copied intact and hence cannot be blanked or bitrate modified? Any reason for this limitation?

[jdobbs]

Yes. The VIDEO_TS is one of the things on my "to do" list.

[Sharc]

Great. Looking forward.

[SpazzHH]

@jdobbs

Thought you might want to know that I ran into this "corrupt stream detected" again with Keeping Up With the Steins-R1(VTS 01). I'm sure it's probably in the rip(DVD Fab 3.0.3.5 and AnyDVD), but also knew you've been trying to prevent this error in RB, so thought I'd report it.

[jdobbs]

Unfortunately in most cases the "corrupt stream detected" is just what it says. The source is corrupt -- and there's not a lot DVD-RB can do about it. I have added a few routines over time to catch certain types of corrupt streams and get around them, but DVD-RB isn't the source of the problem -- it's just the messenger.

[SpazzHH]

I'll trust you totally on that. I just thought it might be something different, since just like last time, the encode goes without a hitch and the offending segment seems to play fine with MPC.

[jdobbs]

Does your log show missing frames?

[SpazzHH]

No. This is as far as it gets:

[05:41:47] Phase III, REBUILD started.
- Copying IFO, BUP, and unaltered files...
- Processing VTS_01
- Reading/processing TMAP table...
- Rebuilding seg 0 VOBID 1 CELLID 1
- Updating NAVPACKS for VOBID_01
- Rebuilding seg 1 VOBID 2 CELLID 1
- Rebuilding seg 2 VOBID 2 CELLID 2
- Rebuilding seg 3 VOBID 2 CELLID 3
- Rebuilding seg 4 VOBID 2 CELLID 4
- WARNING: Corrupt stream detected - [1047/1037/10]
- WARNING: Corrupt stream detected - [1047/1049/2]
- WARNING: Corrupt stream detected - [1047/1049/0]
- WARNING: Corrupt stream detected - [1047/1049/1]
- WARNING: Corrupt stream detected - [1047/1049/5]
- WARNING: Corrupt stream detected - [1047/1049/3]
- WARNING: Corrupt stream detected - [1047/1049/4]
- WARNING: Corrupt stream detected - [1047/1049/8]
- WARNING: Corrupt stream detected - [1047/1049/6]
- WARNING: Corrupt stream detected - [1047/1049/7]
- WARNING: Corrupt stream detected - [1047/1049/9]
[05:42:54] 00051 00005 -- error attempting to write NAVPACK.
Aborted.

[jdobbs]

I'll see if I can find that disc and give it a test run.

Looking at your error message -- it appears your input M2V file for a segment has more frames than the oriignal -- something has definitely gone wrong. The one time I've seen that happen before the original stream was corrupt (with a group of illegal picture start codes).

[jdobbs]

I'm really confused by your report. Have you done any preprocessing? You're using DVD-RB Pro, right?

I just looked at that disc, and VTS_01/VOBID 2 is unreferenced and is not processed by DVD-RB Pro (it was probably garbage as a part of some protection mechanism on the original). The IFO doesn't have any reference to a VOBID 02 or any of its CELLIDs.

You mentioned that you ran it with AnyDVD? Because that's what I used to rip it also. I'll try DVDFab now.

[Rippraff]

You're using DVD-RB Pro, right?
Yes, he is (http://forum.doom9.org/showthread.php?p=884310#post884310). :)

You mentioned that you ran it with AnyDVD? Because that's what I used to rip it also. I'll try DVDFab now.
Maybe try RipIt4Me too.

Cu Rippraff

[jdobbs]

I know he's a VIP user, but I was still wondering how he got that result with Pro. That's what's confusing. How can you get an error on a cell that is ignored? I just ran that disc and the cell is (as it should be) ignored.

The only thing I can think is that possibly DVDFab renumbered the VOBIDs (which is sometimes does)...

[jdobbs]

Yep. DVDFab Decrypter has renumbered the VOBIDs... I'll see what happens when I try to run DVD-RB against the DVDFab rip.

[jdobbs]

Well, I just ran that disc ("Keeping Up With The Steins", R1) and it finished REBUILD without error. So I have to come to the conclusion that it was a bad rip.

My guess is that DVDFab Decrypter makes the assumption that any/all errors received when reading from a disc is copy protection. The problem with that is that occasionally you get a read error because of a small smudge on the disc -- so in most instances you just wipe the disc off, reinsert it, and it works fine the next time. But if DVDFab saw it as copy protection -- it will put a null or padding sector in its place, even though it may fall right in the middle of a referenced stream.

I'm just guessing, of course, but the fact is that the disc worked fine in my tests using both DVDFab Decrypter v3.0.3.5 and AnyDVD v6.0.8.2.

Are you using any special settings? Movie only?

[SpazzHH]

Sorry to have even bothered you on this one, you were right about a bad rip. I just got home from work, and because of what you said this morning, I ran the ripped source through V-Dub MPEG2, and it reported 7 bad frames at the beginning of that segment. I then re-ripped without DVD Fab, and the new source passed with flying colors. It's encoding now, and I'm sure it will come out fine.

Sorry again to have taken up your time, but Thanks so much for being willing to check things out so quick.

Weird thing about the bad rip, the frames just looked like both DVD Fab and AnyDVD just gave up the ghost for awhile. You know that green blocky look of still encrypted video?

[jdobbs]

That would match the "error" scenario. You probably had a couple of sectors read as bad -- and you ended up with some orphaned frames without a reference.