All,
I've created a new version of IFOUpdate (attached). The main change in this version is the inclusion of a new mode call quot;Adjusted Cell.quot; Other changes are listed in the README file.
You may have noticed that a lot of DVDs have been coming out lately with cells that are a part of more than one PGC (especially in the extras -- a good example is quot;Unfaithfulquot;). When you break out the PGCs you find that you might have 3 hours of video for what was only 2 hours on the original -- because the duplicate cells are treated as independent when you pull out the PGCs. Using the Maestro mode of IFOUpdate, these blocks can even be out of sequence...
This new mode analyzes the IFO at the cell level, it organizes them based upon sector location so they will always be in order according to the way in which DVD2AVI reads them. It also recognizes duplicates sectors (between or within PGCs) and flags them internally so they won't be repeated. That way you can simply let DVD2AVI create a D2V for then entire VTS and let IFOUpdate worry about the structure... That also means that you can author a multiple PGC title with a package that only supports single PGCs (like SpruceUp).
IFOUpdate does the analysis when it loads the IFO. The secret is to set IFOUpdate to quot;Adjusted Cellquot; mode and then write the chapter files/CCE files from the quot;Save Chapter Filesquot; dialog. Then author the new VTS using the chapters provided --- NO OTHERS -- the chapters have to match exactly when you run IFOUpdate to rebuild the original IFO, or it will bail out...
Please note that this is a new feature -- I've tested it, but not thoroughly. Let me know if you run into problems so I can fix them.
An example Sequence (with little detail) for using this method:
1. Rip the DVD to a directory using SmartRipper or DVDDecrypter.
2. Look at the directory and pick out the VTS (the big one) holding the movie
3. Open the IFO for that VTS with IFOUpdate.
4. Set to quot;Adjusted Cellquot; mode, and write the chapter files.
5. Run DVD2AVI against the entire VTS set
6. Create an AVISYNTH .AVS file to feed to CCE.
7. Open up CCE, load the .AVS, set parameters, and save the .ECL file.
8. With a text editor, open the .ECL file -- and add the CCE chapter points - If FILM source use the FILM .TXT file else use the other - If you're new to this -- search the forum
9. Open CCE again and reload the ECL -- encode the movie
10. Run PULLDOWN if necessary.
11. Load the file into an authoring package with audio
12. Set the chapters based on the output of step 4. - The file will import directly into Maestro
13. After you've completed, save he orginal .BUP and IFO files - This is not required but is a good way to be safe
14. Run IFOUpdate and update the original file (in quot;Adjusted Cellquot; mode).
15. All done. Do any other VTS sets that may need to shrink.
I'm still struggling with the problem encountered when one of the PGCs has no sound and that causes all the following audio to be out of sync. Look for a fix in the next IFOUpdate version... For now you may want to choose the DoItFast4U route on those...EDITED [18 Jan]: Please download version 0.60 below -- it fixes two important bugs.
Thank you jdobbs for your continuing effort in enhancing ifoupdate.
Bear
Amazing conceptual idea! I'm trying it out on a DVD I've waited a long time to backup. I will let you know my results soon!
Great job visualizing a simplistic approach and then implementing it. You are amazing!
That said, let me ask you about something else. Some DVDs I've seen have both pgc1 and pgc2 with a full movie (think regular and director's cut). Then pgcs 3+ (in one DVD there were 30 more PGCs!) are parts of vobids found from both PGC1 and PGC2. Any idea of a way to make IFOUpdate work with this? An analysis of the entire IFO perhaps, so that we can author PGC1 and PGC2 in Scenarist and then IFOUpdate can find the vobids from both and use these to do the rest? I'm sure there could be some DVDs that have 3 PGCs with different VOB IDs too. Is it possible for a revision to check the entire .ifo? We can, of course make sure the vobids match in the PGCs we author, to make it easier if need be.
Eyes`Only
Jdobbs,
How in the heck do you use this 'adjusted cell method'?
Case scenario:
Original .ifo has 18 PGCs, the first PGC has 50 cells, 29 chapters.
I've tried creating both a single PGC .ifo with 50 chapters in PGC1. IFOUpdate gave me the error 'Original/New Cell Counts do not match, aborting'. Checked the .ifos, both had exactly 50 cells for PGC1. Note: Original .ifo has 18 PGCs, new .ifo has 1 PGC.
So OK I said to myself.. maybe i have to have the same number of chapters(?). So I recreated the exact structure of PGC1 (with Scenarist this is easy.. I don't use Maestro, I rather refuse to touch it because to me it's a toy). Soooo... now my reauthored .ifo has 50 cells and 29 chapters in PGC1. Looks exactly like the original except that I didn't put it together by vobid, I just threw in one .m2v and added 50 chapters. Anyway, it looks just like the original except for that. Here's an example:
---------------------------
Original .ifo:
[Ch 01] [Pg 01] [Cell 01] [V/C Id: 1/ 1] : time: 00:01:06.00 / 30 fps [Pos: 00:01:06.00] [Frames: 1980] [Cell 02] [V/C Id: 2/ 1] : time: 00:03:25.11 / 30 fps [Pos: 00:04:31.11] [Frames: 8141] [Cell 03] [V/C Id: 3/ 1] : time: 00:00:15.00 / 30 fps [Pos: 00:04:46.11] [Frames: 8591] [Ch 04] [Pg 02] [Cell 04] [V/C Id: 3/ 2] : time: 00:01:39.16 / 30 fps [Pos: 00:06:25.27] [Frames: 11577] [Cell 05] [V/C Id: 3/ 3] : time: 00:00:37.23 / 30 fps [Pos: 00:07:03.20] [Frames: 12710] [Ch 05] [Pg 03] [Cell 06] [V/C Id: 4/ 1] : time: 00:01:41.00 / 30 fps [Pos: 00:08:44.20] [Frames: 15740]
----------------------------
Reauthored .ifo: [Ch 01] [Pg 01] [Cell 01] [V/C Id: 1/ 1] : time: 00:01:05.25 / 30 fps [Pos: 00:01:05.25] [Frames: 1975] [Cell 02] [V/C Id: 1/ 2] : time: 00:03:25.14 / 30 fps [Pos: 00:04:31.09] [Frames: 8139] [Cell 03] [V/C Id: 1/ 3] : time: 00:00:15.00 / 30 fps [Pos: 00:04:46.09] [Frames: 8589] [Ch 02] [Pg 02] [Cell 04] [V/C Id: 1/ 4] : time: 00:01:39.15 / 30 fps [Pos: 00:06:25.24] [Frames: 11574] [Cell 05] [V/C Id: 1/ 5] : time: 00:00:37.25 / 30 fps [Pos: 00:07:03.19] [Frames: 12709] [Ch 03] [Pg 03] [Cell 06] [V/C Id: 1/ 6] : time: 00:01:40.28 / 30 fps [Pos: 00:08:44.17] [Frames: 15737]
----------------------------
So now I have the same exact # of cells AND chapters in my reauthored ifo for PGC1 (still no pgc2+ in the reauthored). I try again and get the same error ('Original/New Cell Counts do not match, aborting').
What am I doing wrong? On a whim, i used your program to create a .chp file and then took a look at that.. seems like you're supposed to create one pgc only, according to that ADJUSTED-MAESTRO.CHP file that was output.
So, please. Tell me what I'm doing wrong. Give us an idea of what the heck is supposed to be the structure in the reauthored .ifo, as well as how your program performs this miraculous feature. I'd LOVE to use it, I just can't figure out HOW. And if it can be done in Maestro, it can definitely be done in Scenarist (heck Maestro can't even get multipgc right without mucking about and adding a pgc that you didnt ask for), and you know if it can be done in Scenarist, most likely I already know how to do it! So HELP me please!
(I have attached both the original and reauthored files to this posting)
Eyes' Only,
Regardless of the PGC count you should create a single PGC in your newly authored file -- you can create multiples and it would still work, but it's a waste of time since this mode only looks at the cell level. But the number of cells must match. So for most simple authoring packages that would mean you create a chapter for each cell (since they create them at a 1:1 ratio). I realize you have more control than that in Scenarist. In fact using scenarist you wouldn't even have to create any chapters -- as long as it has the correct number of cells and they begin at the correct point (based upon where they truly sit in the VOB files as defined by the sector pointers).
I'll look at your files and see what I come up with -- gotta be a bug in there somewhere...
@Eyes Only
Thanks for finding this... I had two really off-the-wall bugs (I was working late when I finished that version ) that was causing your problem. Sorry, I'm usually a little better at cleaning up the code before I post. But they didn't occur in the version I was using (one of them would only show its ugly head when the chapter-to-cell ratio wasn't 1:1 -- doesn't that figure)
Attached is an updated version. You will see that in your example --although there are only 50 cells -- you would actually be updating 172 cells across the PGCs after all the duplicate pointers are adjusted...
Thanks again.
@Everyone -- please download this version -- the last one sucks.
@Eyes Only,
That said, let me ask you about something else. Some DVDs I've seen have both pgc1 and pgc2 with a full movie (think regular and director's cut). Then pgcs 3+ (in one DVD there were 30 more PGCs!) are parts of vobids found from both PGC1 and PGC2. Any idea of a way to make IFOUpdate work with this? An analysis of the entire IFO perhaps, so that we can author PGC1 and PGC2 in Scenarist and then IFOUpdate can find the vobids from both and use these to do the rest? I'm sure there could be some DVDs that have 3 PGCs with different VOB IDs too. Is it possible for a revision to check the entire .ifo? We can, of course make sure the vobids match in the PGCs we author, to make it easier if need be.
That is exactly what quot;Adjusted Modequot; is meant to do. It works across the entire IFO. It will get the cell information from all the PGCs in a VTS (outlined in the IFO) and then you author it all as one. The VOB and CELL IDs are corrected to match the ones found in the new VOBs that were created at authoring.
If you run into any more problems let me know. I have a three day weekend to fix all that needs fixing.
Also: Would you find it useful if I made this mode work on a single PGC as well, perhaps in support of use with DoItFast4U?Ok.. to report back.. It worked! However, one major snag...
The DVD i am trying to back up has angles originally. I tried to remove them via IFOEdit, but we all know what a crappy job IFOEdit does keeping chapter points for something like this.. and this time was no exception. The damn cells are all off after IFOEdit messed with em while it was removing angle2.
So.. your tool works great. However I still don't have a good DVDr to burn, because ifoupdate worked correctly, however the chapters in the de-angled ifo are off. Any chance on a 'angle 1 only' type of feature so that I could use my reauthored ifo with 50 cells with an original ifo that has 60 cells but only 50 if you don't count angle 2?
(did that make sense?)
Thanks,
Eyes
That's the great area of potential about this method. Since this works at the cell level, you shouldn't really have to remove the angles -- they should roll into the reassembly. The cell entry points output have to be frame accurate. I'm positive the CCE chapter files are, because I've tested them and verified.
I'm not as confident with the ones I output as Maestro chapter points. I've never been able to get those exact. I'm guessing it has to do with drop frame but I still haven't broken the formula. It's close enough for almost everything, but angle breaks have to be perfect.
That's the goal I'd like to meet. In the interim, you could break out the VTS by cell, reencode all the cells (angles and all), reauthor in Scenarist, and it should fit back together perfectly. That would have been a real pain in the past -- but DoItFast4U kills a lot of the backbreaking legwork now! Can it do cells?
What I can do is modify the IFO so the angles disappear -- then DVD Decrypter (in IFO mode) or SmartRipper could rip just that angle.
If your program had an option to remove cells before it started the cell analysis, it seems it would work fine. I've never thought of making DoItFast4U demux a video with all angles in it... always thought that was a giant waste of time, but I can see how now it could actually help.
Yes, DoItFast4U already has the option to demux by cell id, even removing duplicate cells for you automatically just like it does for vobid mode. I have some ppl currently using this demux mode to analyze the ability to reauthor an Infinifilm title.
Does the Adjusted Cell mode work using the CCE 2.66 ??
I ask this because in can save the ecl file and i caní insert the chapters directly in CCE.
Does the Adjusted Cell mode work using the CCE 2.66 ??
I ask this because in can´t save the ecl file and i can´t insert the chapters directly in CCE.
It should still work. You can either manually set the chapters in CCE, or you can just be satisfied when the newly authored breakpoints are up to a half-second off (depending upon where the nearest IFRAME is).
I tried with CCE 2.66 and it WORKS !!!
Outstanding !!! It saves a lot of trouble in multi PGC.
Only one thing:
The Copy Audio and Subpictures tables are always on im teh Adjusted mode cell, isn´t it ???
I tried a VTS with one subpicture (in the original) and in the authored one i didn´t put it, and it shows in the created one a not specific subpicture.
Can you fix it or is it right ???
Does that mean that if I have a VTS title set for the main movie with 5 PGCs in it, I just need to rip the title set, use ifoupdate in adjusted cell s mode, write down the chapters and reauthor the title set by creating only one PGC ?
At the moment with multiple PGCs, I ripped each PGC individually and then recreate all the PGCs in scenarist ?
So your tool, if I am right, will save me time by only ripping the entire VTS and not each PGC adn reauthor it very simply
using IfOupdate should fix multiple PGC problem
Is it correct ?
emistral, this new feature is only for seamless branching titles. If you're looking for the 'quick fix' for multipgc, you've not found it.
another question I forgot to ask
For the chapters, I currently use chapterxtractor with a filter set up for CCE then scenarist
Before encoding in CCE I use the CCE filter in chapterxtractor and then modify manually the ECL file
Then when it comes to author, I create a new project, drag the audio+video and eventually the subtitles, export the scenario.
Once it is exported, I used then Scenchap and I load the exported scenario and previously I was using chapterxtractor with scenarist setting to import the chapter points from the VTS title set.
So I guess with your new version of ifoupdate I'll have to use the chapter points given by adjusted cells mode instead of chapterxtractor
I might probably have to manually format them to be inserted in the scenarist scenario file but do you think it will work ?
Yes you will have to use the chapters provided by IFOUpdate unless you can create cell points with some other method (Scenarist). ChapterXtractor will create the quot;chaptersquot; of an original single PGC, not the common points for cells.
@emistral
I disagree with Eyes' Only that this won't work on Multiple PGCs. That is the main reason it was created. So you could do multiple PGCs -- and not have to worry whether their individual chapters/cells point to the same sectors. I have used the tool on several multiple PGC DVDs and it works every time except for those annoying instances where there is a PGC in the middle that has no sound. I'm working now on a method to fix that... It may take a little while.
To do a multiple PGC with this you set the mode to quot;Adjustedquot; -- and save the quot;Adjustedquot; files. It is then important that you use the .TXT files and insert them into CCE's ECL file. They are dead-on accurate, so you can be assured a GOP starts at the Cell point. Then use the chapters provided in the .CHP file when authoring. You may have to adjust these one GOP one way or the other to be exact. I'm still trying to figure out why it doesn't hit exactly when Maestro imports.
Anyway -- the result (from Maestro, ReelDVD, SpruceUp, Scenarist) is a single PGC that has cells all properly located at the correct GOP. Then when IFOUpdate moves them back to the original, it simply uses their sector, VOB, and Cell IDs and inserts them at the correct location in the receiving PGCs.
I'm trying to get DVD backing down to its simplest level which is cells and sectors. It leaves out a lot of the guesswork.
@FilipeAmadeuO
The Audio and Subpictures, as well as the color tables are not moved at all in this method (that's why they're grayed out). I could copy them, but since I'm going from one PGC to multiples and these tables are in each PGC, I'd just have to duplicate it across the destination PGCs. My intention was to leave them alone and use the originals -- but if you have a need I can do this.
@jdobbs:
ok I understand your method. However I have another question. The movie I am trying to backup has 5 PGCs. PGC1 is the main movie with 33 chapters and 34 cells (there is an extra one in chapter 13). Then there are 4 more cells for PGC 2,3,4 and 5.
With the method I am using now, I reauthor each PGC. For PGC 1, after importing the scenes chapters form Scenchap, I have 34 entries. Then I drag and drop each scene into the program folder under PGC one to create a chapter (except for chapter 13 for which I drag and drop 2 entry points)
For the other PGCs, I don't have any entry.
so now if I use your method and the new IFOUpdate in Adjusted Cell modes, this should give me a total of 34 + 4 cells (PGC1 = 34 + 4 PGC 2 to 5)
So how do I rearrange these cells ?
If I had to recreate the chapters by dragging and droping the cells in the PGCs, what do I have to do with the 4 extra cells for PGC 2 to 5 ?
Where so they fit in ?
Unless with your method I don't need to do that ?
Thanks |
Post at 2011-4-15 15:03
| This Author's Posts Only
