PGCEDIT dual layer project to DDP Images in GEAR PRO Mastering?

[joybreaker]

Hi.

My situation is this. Having a DVD-9 DL project in PGCEDit, where I set the layer break successfully. I'm also able to burn the DVD through PGCEDit (with IMGBurn). The ending result seems to be flawless, the layer break is in a non-seamless cell spotted by PGCEdit, and you can't see a pause in the layer break when watching the DVD. Perfect.

But, I need to deliver this DVD to replication factory as DDP images through Gear mastering software. How to accomplish this?

here are the details of the .iso file from PGCEDIT ISO creation log:

Running under windows (OS type: Windows NT)

mkisofs 2.01-bootcd.ru (i686-pc-mingw32)
Layer break is at absolute sector 1860080. (Offset in L0 is 215213.)
Layer break cell:
1644867: VTST 1 , 1 TTN 1 (1:50:21) Title 1 Cell 11 (00:04:42.18), V/CID: 1/11 high mux rate

The pad was 215224 for file VIDEO_TS.VOB
Total translation table size: 0
Total rockridge attributes bytes: 0
Total directory bytes: 4248
Path table size(bytes): 42
3720147 extents written (7265 MB)

ISO created OK.

...Ok, when I import the same VIDEO_TS folder to gear I'm able to set the start sector offset (PGCEDIT uses the term padding for this, right?), so I set the offset to this 215213. And then in the GEAR project settings I define the layer break sector to this 1860080.

All seems fine, screenshot from GEAR here: http://www.box.net/shared/hyeabascgk

BUT, it says that the VIDEO_TS.VOB is at sector 215221 (instead of 215224 as PGCEdit sets this), why is that? and what is the reason for it and can it be fixed / does it matter?

Anyhow, when I burn this project with GEAR, I get a slight pause during the layer break (ok it's quite minimal around 0.5secs, could likely because of the written disc?) but a player called UNITED freezes in this LB. But not in the project that I create with PGCEDIT and burn through PGCEDit in IMGBURN, what might be the reason for this? SO, is there a way how to get the PGCEDit to GEAR so it would work similarly perfect, and then I would be able to create the DDP images from GEAR of this project, and all would be fine?

The reason why I'm not setting the LB in GEAR is that using the GEAR LB spreadsheet, it doesn't offer me to use the cell11 as a layer break (which is the only cell with the non-seamless flag there), I guess because it doesn't provide a similar kind of big offset / padding way to get this cell 11 to the layer break window as does PGCEDit and IMGBurn. Screenshot of layer break window of GEAR, IMGBURN and PGCEDIT: http://www.box.net/shared/krjo4hh8g8 <!--end msg--> (the only reason why cell 19 is selected in the screenshot is because it's the only suitable cell in the layer break window which GEAR is providing me in terms of less action in the movie, but it is flagged as seamless so not a cell that PGCEDit would recommend for the LB)

[blutach]

If the LB is is a non-seamless cell, you will get a pause. This is normal and is the usual way a commercial DVD is authored.

The 2 or 3 sector differences may be due to filesystem size.

Regards

[joybreaker]

Well, it's colored as blue in PGCEdit LB panel. And when I use it as a the cell for the LB, I don't get a pause when it's played in DVD player and it doesn't freeze on any player.

The 2 or 3 sector differences may be due to filesystem size.

So can I set the layer break to the cell sector that PGCEDit is telling me then in GEAR, or will it be eventually a miss when using this sector (that PGCEdit tells me) in GEAR if their file systems are somehow different?

[r0lZ]

The method used by PgcEdit is totally dependent of mkisofs, and cannot be used to create the iso/burn with another program, unless that program respects the information in the IFOs. It's not the case of Gear!

Since the other DVD files are referenced by VIDEO_TS.IFO, relatively to its position, it is essential to know its exact position in the ISO. PgcEdit doesn't compute this position. It launches mkisofs and grabs its output to analyze it. As soon as the beginning of VIDEO_TS.IFO is detected, mkisofs is killed, and PgcEdit can compute the offset to use to place the other files relatively to VIDEO_TS.IFO. The gap is created after VIDEO_TS.IFO. It can be very large. Other (small) gaps are inserted when necessary after each IFO file. For a DL, PgcEdit can then analyze the VOB cells and propose the suitable cells to the user, and adapt the size of the gaps according to his choice. On the other hand, Gear places the gap before VIDEO_TS.IFO, and it can only create a relatively small gap. As far as I know, it cannot add gaps after each IFO file.

As you understand now, using PgcEdit to generate the offsets for another program cannot work. You have to use Gear's method to use Gear.

This thread might be useful for you (although it is very old!)

[joybreaker]

In Gear you can set the offset to as high value as in my sample, another thing is does it have any effect in a way that the cell 11 would become to the layer break window (at least if you follow the steps in Gear How to set the layer break, you set the offset after you have chosen the layer break cell). Too bad if this cannot be achieved somehow, as now if I don't get cell 11 as the layer break cell, I have to satisfy to this secondary option, which I could get in PGCEdit and which proves to be working perfectly. If you have any tips how to achieve getting the layer break in Gear to the cell 11 I would appreciate your help more than you can think of.

After all I have a perfectly working DVD with a legal layer break set in PGCEDit, just would need to get it to DDP image and eventually to the replication factory. Would you think it would be possible to achieve this using the burned DVD through this: http://www.gearsoftware.com/howtoguides/ddpimage.cfm

Or is this a similar case, that a DVD with layer break set in and created in PGCEdit this way cannot be used in Gear in any way?

[r0lZ]

When the DVD contains several different titlesets, it is sometimes possible to swap them to place more or less data before the main movie VTS, and make some other cells available for the LB. Unfortunately, your DVD has only one titleset. Perhaps you could try to create a new dummy titleset, with some blank video, and move it before VTS 1 to create an additional gap. It is not easy to compute the ideal length of the VOB file, though, and I don't like this kind of trick. But I don't think there is another solution.

BTW, which are the cell numbers you can use for the LB with Gear?

[joybreaker]

rolz, see figure http://www.box.net/shared/krjo4hh8g8
The first image is the LB excel sheet of this GEAR project.

After inputting the chosen video object VTS_xx_1.VOB startsector (which is 138,615) gear tells me to use IFOEdit to view the VTS_xx_0.IFO to find an entry point in the following range: "Chosen cell must have a starting sector address between 1614046 and 1935881 sectors"

when VTS_01_0.IFO is viewed then in IFOEDIT the start sector of cell 11 is 1506602, and start sector of cell 19 is 1704345

BTW. the cell 19 has seamless flag in its cell, according to DVD spec the layer break has to be defined to a non-seamless cell (cell 11 is non-seamless naturally), so is it ok to clear the seamless flag of cell 19 in PGCEDIT and then use that as the LB? Would this be my best option if I don't want to risk player freezes after the disc has been replicated and if I can't eventually use cell 11 in Gear as the LB?

[r0lZ]

Yes. I don't know why cell 11 is already flagged as non-seamless, but it is not required to use that cell for the LB. You can normally tick the seamless flag of cell 11, as having a pause here is useless, and theoretically, you must untick the flag of the layer break cell (although most players accept the seamless break method.)

Seems they have made no progress at all at Gear! They are still requesting to use IfoEdit and complex computations with a spreadsheet, and they are still unable to propose the full range of suitable cells! For a professional and expensive software, it's a shame!

[joybreaker]

Yes, changed the cell 19 to non-seamless and set the layer break with Gear and burned it with Gear.

Now the results are these:

With rather cheap Samsung player (A bit newer player) it hesitates very long after it starts to play the DVD. The layer break is 1 to 2 secs which I guess is ok as this still is a written DVD+R DL at the moment.

With the a bit older cheap United player the disc starts to play instantly, but it freezes at the layer break, but you are able to pass the freeze with NEXT CHAPTER button so it then moves along (This happens with this player btw. with all the tests I've done with GEAR so far).

Then if I set the layer break to cell 19 in PGCEDit and burn it with IMGBURN, there is no layer break pause at all with United player and it the disc plays instantly. But then this disc won't start play at all with the Samsung player.

EDIT: The Samsung player managed to start playing the PGCEDit & Imgburn LB @ cell 19 disc after testing a couple of times Opening and closing the tray, and it does play the cell 19 break without a pause. So, looks like at least that that it didn't start playing the disc right away was because of the written disc..right? SO, in both players the cell 19 break created with PGCEDit and burned with IMGBURN did work perfectly (as did with the cell 11 LB disc ).

Just that, based on this experience can I trust that this is just due to the written disc, that gear burns it badly or soething? So that if I create the DDP images in Gear and set the layer break to cell 19 in Gear, this would eventually work in all players after it's replicated? And the replicator doesn't provide a check disc...

[joybreaker]

New results added to the previous post above.

[joybreaker]

r0lz,

About changing the seamless joint of a cell in PGCEDit. Isn't the procedure exactly like this: you double click the VTST and see all the cells. Pick one cell (that exists in the layer break window you can use) and de-tick the "seamless joint" checkbox of the cell you want to use as the LB cell, then choose OK to exit the VTST panel. Save your DVD (remember to have at the Options -> Output -> leave 32k of space, selected when saving). Now you have a dvd standard non-seamless cell for the layer break.

A couple of questions rise:

Is this 100% legal and risk free to change the cell type like this or is there something more to it? Do you also have to uncheck from the cell's "Flag type panel" in PGCEdit the "seamless angle change" and the rest of the flags in order for this to work the way fully intended?

[blutach]

Yes, ticking the seamless joint box will make the cell seamless, so long as your burning software doesn't make it non-seamless again. Don't worry about seamless angle (does your project have angles?).

Is it legal and 100% risk free? Well, it seems to be as most superbit DVDs are authored this way. I haven't come across a player that can't play it either but that doesn't mean it is 100% risk free. Having said all that, most DVDs are authored with the LB as non-seamless. Just find a good place where the motion fades to black and the audio is gone, if you can.

Regards

[joybreaker]

Ok, as GEAR didn't allow me to use cell 11, in the layer break window, I used this cell 19 and de-ticked its seamless joint checkbox with PGCEDit. Now the cell is non-seamless. Fine, so I set the layer break with Gear to that and build DDP images, which I send to replication factory. They run the DDPs through Eclipse Image Analysis and come out with a one warning from the whole project, that being:

"Seamless playback flag not set correctly at layer break."

This error indicates that the layer break is not set properly on a DVD-VIDEO image.

DVD spec has no definition for layer break. However, the do specify the following two rules:

1. A cell shall not span across layers.

2. Two cells which are presented seamlessly shall be allocated on the same layer.

Eclipse's interpretation of these two rules is as follows:

1. The layer break shall no occur within a cell.
When this rule is triggered, it indicates that a cell spans between both layers. This is know to cause playability problems. When this condition occurs, the Additional Info column in the analysis will display the following message:

"Layer break address is not a cells starting cell"

Layer 1 must start with a new cell.
This condition should be considered serious since it is known to cause playability problems.

2. The seamless playback flag of the first playback cell on the second layer should be set to non-seamless.

Since there is a physical layer break, the last cell in Layer 0 and the first cell in layer 1 should be played back non-seamlessly. The effect on this condition may be a pause in the video during the layer change.

--------------

SO, what is wrong with the cell 19? I've used its starting sector (which I read with IFOEDit the way GEAR's layer break setting guide advices) as the layer break sector (so the LB is in the beginning of the cell, not within the cell but it is indeed the cell's starting address). And the resulting sector is dividable by 16.

Here is a screenshot to the image that Imgburn tells about the cells in it's Build mode (just so that you can see infos about the cells). http://www.box.net/shared/md8kuzio8g

Here is a screencapture from IFOEdit about cell 19 and its preceding cells:

http://www.box.net/shared/8pql2428so

Here is a screencapture of the GEAR LB spredsheet with all required values of cell 19 inputted:

http://www.box.net/shared/2xkkvmfwgk

Can you tell me based on this, that what is wrong with the cell? What should I do else, or which other cell should I pick instead?

Ps. And yes, the IMAGE.DAT of the Layer 0 inside the DDP did match the layer break sector so it is 1:1 with that (after and before burning the each DDP to DVD-5 UDF DVD-ROM), so I think the problem cannot be because of the DDP writing failure.

[r0lZ]

Honestly, I really don't know. I have never used Gear or Eclipse Image Analysis. If you are sure that your computations are right, that should work. Anyway, it is sufficient to turn off the seamless joint flag to be standard compliant. I can't help much more.