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. |
4th May 2005, 01:28 | #1 | Link |
Just a Member
Join Date: Nov 2002
Location: Spain
Posts: 2,067
|
VTS_TMAP discontinuity flag
Hi all,
It's not clear to me when this flag (bit 31 of each VTS_TMAP entry) should be set. On mpucoder's site it can be read "bit 31 set if VOBU time codes are discontinuous with previous" which is not clear to me. MuxMan does not mark any discontinuity in TMAP table (testing the case of one PGC with several chapters) but, looking what IFOEdit does, seems IFOEdit marks as "discontinuous" the last VOBU in the table for each cell, including the last one Looking into a couple of originals, I've seen the same that IFOEdit does: The bit is set in the last VOBU in the table of each cell. Could anyone clarify this point? jsoto EDIT: Found other original marking as "discontinuous" the first VOBU of each cell. But, even in this case, the last entry (which is the last VOBU of the last cell) is also marked as discountinuos.
__________________
Web sites with my tools http://download.videohelp.com/jsoto http://jsoto.posunplugged.com/ Last edited by jsoto; 4th May 2005 at 11:26. |
4th May 2005, 02:03 | #2 | Link |
Moderator
Join Date: Oct 2001
Posts: 3,530
|
Funny you should mention this as I was looking at it myself today. Up until now, if the bit means a clock discontinuity, there was no reason to set it. But 0.15 works with multiple VOBs, so needs to set the bit.
After looking at some DVDs I agree, it means something other than a clock discontinuity. What exactly I'm not sure yet, as there does not seem to be a clear pattern. |
4th May 2005, 11:37 | #3 | Link |
Just a Member
Join Date: Nov 2002
Location: Spain
Posts: 2,067
|
Could you check what Scenarist does in a simple case? (one PGC with multiple chapters).
jsoto
__________________
Web sites with my tools http://download.videohelp.com/jsoto http://jsoto.posunplugged.com/ |
4th May 2005, 15:53 | #4 | Link |
Moderator
Join Date: Oct 2001
Posts: 3,530
|
I have many Scenarist examples made by myself, and I still see no pattern. On one project I had 2 tracks, and titles - one for each track and a "play all" joined seamlessly. The flag was set on the last time entry for all three PGCs. For the play-all it was not set on the end of the first track. I also have some non-seamless examples that do that. So it is not reflecting any property of the multiplex (such as seamless or discontinous).
Another example with multiple uses of the same track has the flag set on a vobu that is 1.2 seconds from the end. On the play-all that sector is not flagged, but the next entry (which was not present on the single play map, but is part of the first track) is flagged. The only consistency I'm seeing is that the interval following a flagged entry is not the time interval, but shorter or longer, and that there is a minimum duration required in order to make an entry. The verifier is no help for this, it only wants entries to be exact using 90000 ticks as one second. btw, it totally screws up if the first PGC does not use the first multiplexed cell (which is legal, that's why there is a map for each PGC. But the verifier ignores the PGC and treats the table like it was in VobID/CellID order) Last edited by mpucoder; 4th May 2005 at 15:56. |
4th May 2005, 17:33 | #5 | Link |
Moderator
Join Date: Oct 2001
Posts: 3,530
|
OK, I just created a couple Scenarist projects to test a theory, but discovered something different altogether (like most scientific experiments!). I looked at all my Scenarist projects to confirm this - Scenarist sets the flag on the last entry for each PGC, and nowhere else. This includes multiple VOB titles, multiple cells, and even angles in the middle of the movie.
So what we are looking at is how different authoring programs interpret when to set this flag. On most DVDs I have the flag is set if the time span includes the end of one or more cells. With this flag being subject to so much interpretation, it is likely that DVD players ignore it altogether for being inconsistent and unreliable. |
4th May 2005, 21:58 | #6 | Link | |
Just a Member
Join Date: Nov 2002
Location: Spain
Posts: 2,067
|
Quote:
In any case, I agree, seems it has no effect in any settop. jsoto
__________________
Web sites with my tools http://download.videohelp.com/jsoto http://jsoto.posunplugged.com/ |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|