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.

 

Go Back   Doom9's Forum > (HD) DVD, Blu-ray & (S)VCD > IFO/VOB Editors

Reply
 
Thread Tools Search this Thread Display Modes
Old 13th April 2007, 08:24   #1  |  Link
brownstem
Registered User
 
Join Date: May 2003
Posts: 58
Help splitting cell to set layer break and get PGCEdit to burn a DL disc

Trying to burn a DVD Remake Pro merged DL disc using PGCEdit (I've done dozens of "regular" DL discs successfully with PGCEdit / ImgBurn). The size of the merged disc isn't the problem, but the lack of a suitable layer break point is. Here is the message PGCEdit gives me when I try to create and burn the image:

Can't find a suitable cell for the layer break!

Try to change the order of the titlesets with DVD -> Remap Titlesets, or remove some unnecessary material.

Note for expert users:
Given the minimal ISO file size (4101367 sectors) and the capacity of your DL DVD+R, the LB must be between absolute sectors 2014464 and 2086912 in the ISO. The absolute starting sector of the VOB files of the Title domain of VTS 2 is 1819628. The LB cell must begin between relative sectors 194836 and 267284 in that domain.

Therefore, try to split cell 3 of VTST 2, 1 TTN 1 (19:34) Title 6.


Can anyone give me instructions or direct me to a guide? I routinely use PGCEdit, but only for basic tasks (making a dvd start with the menu as opposed to previews or bypassing one of those annoying widescreen / fullscreen initial menus...Paramount or somebody). I also have VobBlanker but haven't split any cells using that, either.

Thanks to anyone who can help.
brownstem is offline   Reply With Quote
Old 13th April 2007, 08:40   #2  |  Link
blutach
Country Member
 
blutach's Avatar
 
Join Date: Sep 2004
Location: is everything!
Posts: 6,499
Split cell 3 of the 2nd titleset (title 6) in DVDRMP. The help file shows you exactly how to do it. But it is as easy as moving the slider in the cell preview to a split point then right clicking and selecting Split here.

You also know exactly where to split it, thanks to PgcEdit.

Alternatively, you can use VobBlanker to split. You need to work you way down thru the cells dialogue box but it is as easy as pie after that.

Regards
__________________
Les

Only use genuine Verbatim or Taiyo Yuden media.

Last edited by blutach; 13th April 2007 at 08:49.
blutach is offline   Reply With Quote
Old 13th April 2007, 09:17   #3  |  Link
brownstem
Registered User
 
Join Date: May 2003
Posts: 58
Thanks, will give it a shot tomorrow.
brownstem is offline   Reply With Quote
Old 13th April 2007, 10:52   #4  |  Link
r0lZ
PgcEdit daemon
 
r0lZ's Avatar
 
Join Date: Jul 2003
Posts: 7,469
Not sure about DVDRMP, but VobBlanker displays now the relative sector of the current position in its GUI. You have to cut cell 3 of title 6 somewhere between sectors 194836 and 267284.

You might also try to swap titlesets 1 and 2. With some luck, it might be possible to find a LB point without having to cut a cell.
Use DVD -> Remap Titlesets.
__________________
r0lZ
PgcEdit homepage (hosted by VideoHelp)
BD3D2MK3D A tool to convert 3D blu-rays to SBS, T&B or FS MKV
r0lZ is offline   Reply With Quote
Old 13th April 2007, 21:33   #5  |  Link
Surf
Senior Kindergarten
 
Join Date: Sep 2004
Location: here & there
Posts: 417
I too am trying to comprehend sectors & dual layers:

Without getting technical, lets say there are 2000 sectors in a dl disc, 1000 for each layer.

Brownstem got DVD remake pro successfully merged the disk, but the "1st half" of the movie went beyond the 1000 sectors mark, am I assuming correctly? Is that why he is stuck at "Can't find a suitable cell for the layer break!" ??

Last edited by Surf; 13th April 2007 at 21:36.
Surf is offline   Reply With Quote
Old 13th April 2007, 23:16   #6  |  Link
r0lZ
PgcEdit daemon
 
r0lZ's Avatar
 
Join Date: Jul 2003
Posts: 7,469
Well, it's more complex.

Usually, there is still enough free space on the DL to be able to shift all files (except VIDEO_TS.IFO) sufficiently to align the beginning of a cell with the layer break position. In other words, the layer break doesn't have to be at the end of the first layer.

For example, if your DVD is 1800 sectors long (still assuming that a DL is 2000 sectors, and ignoring the space needed for the file system at the beginning of the DVD) the LB can be between sectors 800 and 1000.

For 800, a big gap of 200 sectors is created between VIDEO_TS.IFO and the next file, VIDEO_TS.VOB or .BUP (but for clarity, let's assume that the gap is created at the beginning of L0). The layer break is therefore at physical sector 1000, and the entire layer 2 is filled with the remaining 1000 sectors:
Code:
L0: [gap of 200 sectors][800 sectors of data]
L1: [1000 sectors of data                   ]
LB:                                         ^
On the other hand, if the LB is at sector 1000, the files are contiguous, and the second layer contains only 800 sectors (but is padded with null bytes till the end.)
Code:
L0: [1000 sectors of data                    ]
L1: [200 padding sectors][800 sectors of data]
LB:                                          ^
Of course, if the LB is at sector 900, there is no need to create a gap, as it is possible to place the LB at sector 900, with 900 sectors per layer. That's the best option, as it is not necessary to add a gap or padding bytes. A part of the disc, near the edges (sensible to scratches, defects and fingerprints), is simply not written.
Code:
L0: [900 sectors of data ][100 unused sectors]
L1: [900 sectors of data ][100 unused sectors]
LB:                      ^
Of course, another position is also possible, like in this example:
Code:
L0: [gap:100][850 sectors of data ][50 unused sectors]
L1: [    950 sectors of data      ][50 unused sectors]
LB:                               ^
It is easy to understand that if there is less free space in the compilation, it is harder to find a suitable cell. For example, when the compilation is 1990 sectors long, the layer break must be between sectors 990 and 1000. Of course, if the chapters (more exactly, the cells) are long, it is sometimes impossible to find a suitable cell for the LB.

Let's suppose now that the joined DVDs are, say, 800 and 1050 sectors long. The first disc fits entierly on L0, but disc 2 must be cut near the beginning, between sectors 50 and 200:
Code:
L0: [150 sectors of gap][800 sectors of disc 1][50 sectors of Disc 2]
L1: [1000 sectors of disc 2                                         ]
or
Code:
L0: [  800 sectors of disc 1   ][200 sectors of Disc 2]
L1: [150 sectors of padding][  850 sectors of disc 2  ]
If there are no cells beginning between sectors 50 and 200 of disc 2, PgcEdit can't set the layer break.
(Of course, you can replace "disc 1" and "disc 2" by part 1 and 2 of a long movie. The logic is identical.)
In this case, PgcEdit computes the relative start and end sectors in the current domain, corresponding to sectors 50 and 200, and displays them. You have to cut the cell between those positions to be able to burn the DVD.

Notes:

This method concerns mainly the DL DVD+R(W). On -R, it is necessary to put the layer break at the end of L0, exactly at the edge of the DVD, unless a special technique, not easy to master, is used. That's stupid! It's why, among other things, I don't recommend using -R DLs (although ImgBurn has implemented this technique.)

In many technical explanation about the LB position, you will see that it is always necessary to have MORE data on L0 than on L1. It is therefore only necessary to pad L1 to have a working DVD. In the first example of a compilation of 1800 sectors, that would mean that the LB has to be between sectors 900 and 1000. That's not true. By creating a large gap on L0, it is possible to have less data on L0, and more on L1. This is why PgcEdit offers twice as much possibilities than many authoring/burning programs for the LB. (BTW, ImgBurn has implemented exactly the same method. ImgBurn is excellent! )

Note also that it is almost always necessary to create a little gap, or to enlarge the existing gap a little bit, because the LB must be placed at the beginning of an ECC bloc (for error correction.) Since an ECC bloc is 16 sectors long, the LB position must always be divisible by 16. (I have not taken this requirement into account in the explanation above, for clarity.)
The 32K gap technique (used to ensure that an IFO and its BUP backup are not sharing the same ECC bloc) is also responsible of some additional gaps between files.
This is why it is very complex to compute the right position of the cut manually.

For a complete technical discussion on how PgcEdit has implemented the layer break, and its history, see this thread.
__________________
r0lZ
PgcEdit homepage (hosted by VideoHelp)
BD3D2MK3D A tool to convert 3D blu-rays to SBS, T&B or FS MKV

Last edited by r0lZ; 29th September 2009 at 00:02. Reason: typo
r0lZ is offline   Reply With Quote
Old 15th April 2007, 13:42   #7  |  Link
Dr.Khron
Registered User
 
Dr.Khron's Avatar
 
Join Date: Oct 2006
Location: Gotham City, USA
Posts: 389
YAY! An insanely detailed response from r0lZ!

This is what I love about this board. Thanks for the explantion, that cleared up a lot of things.
Dr.Khron is offline   Reply With Quote
Old 15th April 2007, 14:13   #8  |  Link
r0lZ
PgcEdit daemon
 
r0lZ's Avatar
 
Join Date: Jul 2003
Posts: 7,469
You're welcome!
__________________
r0lZ
PgcEdit homepage (hosted by VideoHelp)
BD3D2MK3D A tool to convert 3D blu-rays to SBS, T&B or FS MKV
r0lZ is offline   Reply With Quote
Old 30th September 2007, 12:22   #9  |  Link
Cela
Registered User
 
Join Date: May 2004
Posts: 185
Processing a long opera recording I got the
"Can't find a suitable cell for layer break!"
error message. Along with an indication of a suitable LB slot expressed in relative sector numbers:
"... The LB cell must begin between relative sectors 2071988 and 2086036 in that domain.
Therefore, try to split cell 25 of VTST 1,1 TTN 1 (2:49:24) Title 1. ..."

That is already a great help.
Unfortunately, cutting and editing programs don't accept relative sector numbers as input, but accept time code information. My favorite program, Mpeg2Schnitt, likes frame numbers or time values expressen in hh:mm:ss.milliseconds.

Would it be possible to provide in addition to the sector numbers the equivalent time slot information and add a line to the error message.
For example:
"The timeslot for LB is between 01:45:27.840 and 01:46:10.520 [hh:mm:ss.mss]"

Thank you very much in advance!

Regards
Cela
Attached Images
 

Last edited by Cela; 30th September 2007 at 12:32.
Cela is offline   Reply With Quote
Old 30th September 2007, 12:27   #10  |  Link
blutach
Country Member
 
blutach's Avatar
 
Join Date: Sep 2004
Location: is everything!
Posts: 6,499
Why not open the project in VobBlanker and split the cell there. Very simple.

Regards
__________________
Les

Only use genuine Verbatim or Taiyo Yuden media.
blutach is offline   Reply With Quote
Old 30th September 2007, 12:35   #11  |  Link
Cela
Registered User
 
Join Date: May 2004
Posts: 185
Quote:
Originally Posted by blutach View Post
Why not open the project in VobBlanker and split the cell there. Very simple.

Regards
Because I cannot split it anywhere in the LB slot but have to find a place where the video and especially the audio content is not destorted at replay by the layer break gap. So I have to listen for an I-frame where the music is very low and the singer is not in the middle of a sentence or a word.

P.S. Just downloaded newest VobBlanker. Would be fine if I could switch on audio replay. Would be really very handy to do it with VobBlanker. Please tell me how to hear the audio.

Regards
Cela

Last edited by Cela; 30th September 2007 at 12:51.
Cela is offline   Reply With Quote
Old 30th September 2007, 15:34   #12  |  Link
blutach
Country Member
 
blutach's Avatar
 
Join Date: Sep 2004
Location: is everything!
Posts: 6,499
VobBlanker doesn't preview audio. But I have a solution for you. Make the layerbreak seamless. You won't notice it. Just tick the seamless box in PgcEdit afterwards.

If you have DVD Remake Pro, it has audio preview and a split function, too.

Regards
__________________
Les

Only use genuine Verbatim or Taiyo Yuden media.
blutach is offline   Reply With Quote
Old 30th September 2007, 19:13   #13  |  Link
Cela
Registered User
 
Join Date: May 2004
Posts: 185
@blutach
Thank you very much for your help. I appreciate it a lot!

@r0lZ
Please consider to add the time code information for the LB time slot.

Regards,
Cela
Cela is offline   Reply With Quote
Old 1st October 2007, 00:51   #14  |  Link
blutach
Country Member
 
blutach's Avatar
 
Join Date: Sep 2004
Location: is everything!
Posts: 6,499
You're welcome.

In respect of time, remember it doesn't correlate directly with sectors. Yes, you can extrapolate within a cell and this should normally be reasonably OK, but the method you are using (ie find the sectors from the start of the cell) is the most accurate. Then perhaps just find a place where lips are not moving.

Regards
__________________
Les

Only use genuine Verbatim or Taiyo Yuden media.
blutach is offline   Reply With Quote
Old 1st October 2007, 08:17   #15  |  Link
r0lZ
PgcEdit daemon
 
r0lZ's Avatar
 
Join Date: Jul 2003
Posts: 7,469
I agree that that would be a neat addition, but as blutach said, it's not as easy as it sounds. To find the exact time codes, I would have to analyze the VOB file itself, which is not really simple. Maybe I'll add that later, but honestly, as PgcEdit is not really necessary any more to burn the DL DVDs (since ImgBurn v2 does the same job very well) I do not want to spend much time on the burn function of PgcEdit any more.

Anyway, you can also note the relative sectors and use VobBlanker to convert them to time codes. Use the slider to go to the absolute sector you want to convert, and VB will display the relative time code in the current title.

You can also use the PgcEdit preview to convert the sector number to the time code. (The absolute sector number is in the titlebar. You have to tick the "info" checkbox to see the time code.)
__________________
r0lZ
PgcEdit homepage (hosted by VideoHelp)
BD3D2MK3D A tool to convert 3D blu-rays to SBS, T&B or FS MKV

Last edited by r0lZ; 1st October 2007 at 08:22.
r0lZ is offline   Reply With Quote
Old 1st October 2007, 20:33   #16  |  Link
Cela
Registered User
 
Join Date: May 2004
Posts: 185
Quote:
Originally Posted by r0lZ View Post
I agree that that would be a neat addition, but as blutach said, it's not as easy as it sounds...
Thank you very much for your valuable suggestions.

In the meantime I helped myself for my 25 fps PAL material interpolating timecodes based on average values of frames per sector with respect to the suggested cell using data found in pgcedit (error message and VTST cell table) as documented in the screenshot in my earlier posting #9.

Step 1: I take the Playback Time from the suggested cell and convert it into 25 fps PAL frames.
Step 2: Then I calculate the cell length from the provided relative sector numbers and divde it by the cell length in frames from ytep 1 yielding the average frames per sector value of that particular cell.
Step 3: with this information I calculate the relatice playback times for the start and endpoint of the LB slot with respect to the beginning of the cell in hh:mm:ss.mss (mss ... milliseconds)
Step 4: Adding the End Time of the predecessor cell to the relative times yields the timecodes with respect to the beginning of the Title.
So I got
ta = 1:45:27.840 (decimal point)
te = 1:46:10.737
lenght = 0:00:42.897 [hh:mm:ss.mss]
These ta and te values are directly accepted by Mpeg2Schnitt as marker values in the marker list. Double clicking on these values defines a cut-segment for easy inspection. Preview and pre-listening of this segment showed a good cut point near the I-frame at time position 01:45:37.160 which is frame number 158429. The resulting DL DVD playes faultlessly in my home DVD player.


I guess, my simple procedure is not an exact solution but an appropriately close interpolation within that cell, and should not take too much time and effort to implement. It only employs very basic arithmetics plus conversion of [hh:mm:ss.fr] to [hh:mm:ss.mss] to [seconds] and vice versa.

I would be very grateful if something like this could be integrated in the error message yielding the [hh:mm:ss.mss] time values.

Regards
Cela

Last edited by Cela; 1st October 2007 at 20:35.
Cela is offline   Reply With Quote
Old 1st October 2007, 20:45   #17  |  Link
r0lZ
PgcEdit daemon
 
r0lZ's Avatar
 
Join Date: Jul 2003
Posts: 7,469
I understand the method but the user will probably want precise timings and not an approximation. I could also use the time map of the IFO to evaluate the timing of a specific sector, but again it's only an approximation. I will see what I can do...
__________________
r0lZ
PgcEdit homepage (hosted by VideoHelp)
BD3D2MK3D A tool to convert 3D blu-rays to SBS, T&B or FS MKV
r0lZ is offline   Reply With Quote
Old 3rd October 2007, 12:47   #18  |  Link
mpucoder
Moderator
 
Join Date: Oct 2001
Posts: 3,530
Wouldn't finding the nearest or next NAV pack and reporting the cell elapsed time (c_eltm, offset 0x045-0x048) be sufficient?
mpucoder is offline   Reply With Quote
Old 3rd October 2007, 13:35   #19  |  Link
r0lZ
PgcEdit daemon
 
r0lZ's Avatar
 
Join Date: Jul 2003
Posts: 7,469
Yes, I suppose so. It's easier to do for me. Thanks for the suggestion!
__________________
r0lZ
PgcEdit homepage (hosted by VideoHelp)
BD3D2MK3D A tool to convert 3D blu-rays to SBS, T&B or FS MKV
r0lZ is offline   Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 11:37.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.