PgcEdit 0.5.1 beta (with the Menu Buttons Editor) [Archive]

jeanl

15th March 2005, 22:32

r0lZ,

First of all, THANK YOU! and congrats on the menu editor. It's absolutely fantastic (like schwartzenegger would say). I should have expected no less from you, but I"m really amazed: it looks like you left nothing behind. I'm not a pro with menus, but it looks to me like you can edit every possible aspect of menus, including the color schemes! The possibilities are endless!!! I'm amazed!

I also need to report what I think is a small bug.
- open the menu editor, and set the force action button after 1 second,
- click OK then click Edit again. All good.
- click OK then click Edit again. The force action after is now -1 seconds.
Of course, this had to be the first thing I checked ;)

The same thing happens on the 2 DVDs I tried....
Jeanl

r0lZ

15th March 2005, 22:41

Thanks, jeanl.

Yes, the 1 is changed to -1 the second time you enter the editor. Strange! Thanks for the bug repport.

[EDIT:] Stupid bug fixed!

ShawnFumo

15th March 2005, 23:34

Just a quick question, since I'm not totally sure how the DVD specs work for this. If you had two subpictures muxed into the video, would the latest beta of PgcEdit actually be able to create a button from that? Or does that have to be encoded as a button from the start? I just ask because I've been wanting to do Buttons Over Video, but I don't think any of the free authoring tools support it yet, but of course you can mux in regular subpictures. I'm guessing it probably isn't possible, but just wanted to make sure. :)

Shawn

r0lZ

15th March 2005, 23:46

Good question. Honnestly, I don't know the answer!
PgcEdit do not scan the VOBs for subpics. You may probably mux a subpic stream, and then use the menu editor to create a new button around the visible part of the subpic, but you will have to guess where the subpic is positionned on screen. Note that I have never tested this method.

It is also possible to create transparent or translucent menu buttons without any subpic at all. Note that if there are no subpic streams defined in the IFOs, PgcEdit will create one stream definition automatically in the current PGC and in the domain streams attributes section, but will not create a new stream in the VOB.

jsoto

16th March 2005, 00:41

@r0lZ,
A big thank you.

jsoto

blutach

16th March 2005, 01:53

r0lZ - How can we say thank you enough?

Hope your paypal is still working!

Regards

r0lZ

16th March 2005, 02:04

Thanks for the thanks, guys!

blutach

16th March 2005, 02:47

@r0lZ

If I have previously hidden a button and wish to reinstate it, how can I change the width back to its regular size?

How does one change the x value?

These also apply with the New Button command.

Regards

jeanl

16th March 2005, 03:03

you have to select one of the groups, not "all". Then the x slider is enabled.
jeanl

blutach

16th March 2005, 03:07

Of course! What an idiot I am

Regards

blutach

16th March 2005, 03:11

@r0lZ - the export / import feature makes copying buttons between menu cells an absolute breeze! However, when exporting and importing buttons, at first it appears the buttons's location and size info (x, y, height, width) is kept. However, on reopening the menu or just looking in the menu viewer, these settings don't seem to have "taken". Is this intended?

One suggestion - can we also have a box next to the x, y, height, width sliders, so a value can be manually input? This would be particularly useful for menu design, when you have done the first button's position and want to position the others relative to it. Maybe also a "nudge" slider with the keyboard direction keys for small adjustments (to the pixel).

Also, when we double click on a menu button in the menu viewer, it would be nice if it brought up the editor dialogue box (without having to push the Edit button). What do you think?

Regards

DMagic1

16th March 2005, 06:57

Very nice.
I just had a project that could have used this.
Will it be able to add buttons soon?

blutach

16th March 2005, 06:59

It already does that. Ctrl-M to bring up the relevant menu. Press Edit --> Button --> New Button

Regards

jeanl

16th March 2005, 07:24

It's scary, it does everything! :)
jeanl

blutach

16th March 2005, 08:32

It sure does - it just made me a nice plate of scrambled eggs! :D

Regards

M7S

16th March 2005, 09:49

A long wait is over, but it was for sure worth it!

The menueditor is just wonderful.

Thank you r0lZ!!

r0lZ

16th March 2005, 10:19

Originally posted by blutach
@r0lZ - the export / import feature makes copying buttons between menu cells an absolute breeze! However, when exporting and importing buttons, at first it appears the buttons's location and size info (x, y, height, width) is kept. However, on reopening the menu or just looking in the menu viewer, these settings don't seem to have "taken". Is this intended? No! It should be a bug. Will see...

[EDIT:] OK. Another stupid bug fixed.
Note that if you import the menu button parameters in All Groups mode, the values of the disabled sliders are not imported! (But the current values are always exported, including the disabled sliders in All Groups mode.)

One suggestion - can we also have a box next to the x, y, height, width sliders, so a value can be manually input? This would be particularly useful for menu design, when you have done the first button's position and want to position the others relative to it. Maybe also a "nudge" slider with the keyboard direction keys for small adjustments (to the pixel). Difficult to add an input box. I need to verify the values so that the upper left corner is really to the left and up of the oposite corner, and that the position + width is not too much. Easy to do with sliders, but not with input boxes. However, I'll see what I can do. Maybe I can add some copy/paste buttons?

You can already make small adjustments by clicking in the 'through', white part of the sliders. One click increase/decrease the value by 1, and a long click moves the slider slowly.

The cursor keys are already used to select the adjacent buttons, relative to the current button. Handy to test if the adjacent links are setupped correctly. (Same in the viewer.)

Also, when we double click on a menu button in the menu viewer, it would be nice if it brought up the editor dialogue box (without having to push the Edit button). What do you think? Right. It was my intention to do that, but I forgot. Will be in the next release.

[EDIT:] Done. You'll have to use the right button. Double-clicking with the left button doesn't work for technical reasons.

Also, if you doesn't have already noted: the command button in the viewer is a shortcut to edit the command without having to go in the editor.

If I have previously hidden a button and wish to reinstate it, how can I change the width back to its regular size?
How does one change the x value?
These also apply with the New Button command. As jeanl explained, you have to select another group. It is only possible to edit the X or Y direction in Group All mode if the same value is shared by all groups. But I may add a computation of the correct values for all groups in the future.
Note that the X, Y, width and height values are not changed in All Groups mode if you doesn't move the sliders. This is to allow different values in several groups. So, use the position sliders in All Groups mode with care.

blutach

16th March 2005, 13:43

Thanks r0lZ - I'll try that and await the next version with the "fixes".

Regards

iliketowatch

16th March 2005, 17:59

In Linux, this happens with both PgcEdit 0.5.0 and 0.5.1 beta 1; 0.4.9 works fine.

When I run pgcedit, there is no window but it takes about 95% CPU and memory (Athlon XP 1800+, 1G RAM) and it really loads the system:
load average: 11.66, 9.42, 6.96

If I let it run to completion I get this in /var/log/messages:
kernel: __alloc_pages: 0-order allocation failed (gfp=0x1d2/0)

Any ideas?

r0lZ

16th March 2005, 19:24

Hum! I have a new PC, and my Linux partition is not yet installed. I'll see that soon. Thanks for the bug report.

r0lZ

17th March 2005, 14:00

@iliketowatch and all Linux users

Problem fixed. See here (http://forum.doom9.org/showthread.php?s=&postid=626061#post626061) for the explanation.

r0lZ

17th March 2005, 14:04

- Fixes the two bugs in the menu editor reported by jeanl and blutach.
- Allow to right click on a button in the viewer to go in the editor.
- Fixes the bug starting PgcEdit 0.5 standalone on Linux platforms.

Tobii

17th March 2005, 19:36

I have a small problem with a menu.

The structure of the menu:

Program 1, Cell 1: VOB ID 6 ,Cell ID 1 (0 buttons)
Program 2, Cell 2: VOB ID 6 ,Cell ID 2 (5 buttons)
Program 3, Cell 3: VOB ID 6 ,Cell ID 3 (5 buttons)
Program 4, Cell 4: VOB ID 6 ,Cell ID 4 (6 buttons)
Program 5, Cell 5: VOB ID 6 ,Cell ID 5 (0 buttons)
Program 6, Cell 6: VOB ID 6 ,Cell ID 6 (0 buttons)
Program 7, Cell 7: VOB ID 6 ,Cell ID 7 (0 buttons)

If I delete the second button in the Program 2 and indicate the third button as forced selection, I cannot delete the same button into Program 3 and 4.
PgcEdit then freezes. Any ideas?

DMagic1

17th March 2005, 19:56

Maybe someone can help with this. I've wondered this even from using MenuEdit in the past. Which is better, to delete a button or to hide it?
Does it matter? I've looked at the changes made when doing each. It seems like both ways would work well.

jeanl

17th March 2005, 20:00

Originally posted by DMagic1
Maybe someone can help with this. I've wondered this even from using MenuEdit in the past. Which is better, to delete a button or to hide it?
Does it matter? I've looked at the changes made when doing each. It seems like both ways would work well.
Hiding provides an easy "undo" mechanism... But other than that i don't see any difference... Except some programs will behave differently depending on whether there is a button (like MenuShrink and vobblanker I think) or not (they're not smart enough to see the button is "hidden").
jeanl

M7S

17th March 2005, 22:35

Isn't there a posibility that the wrong button gets highlighted if you delete buttons?

jeanl

17th March 2005, 22:45

I see what you mean. Yes, if the button you're deleting isn't the last one, and you're not updating the highlighting commands, then you're right, there's a possiblity the wrong button will be highlighted. I guess it all depends on what "deleting" means! I haven't checked what PgcEdit does, but I'm guessing trying to readjust all the highlight commands would be very hard... This would indicate that it's better to hide, right? (although then, there's a possibility that no highlight will show if the hidden button is supposed to be highlighted)...
jeanl

r0lZ

17th March 2005, 23:56

Deleting a button is dangerous, even if it's the last one.
If a command highlights the deleted button before entering the menu, the player may hang.
And if the deleted button is not the last one and there are post commands doind specific actions based of the SPRM 8 value, the jump may occur at the wrong place. This is certainly the case when the same command is assigned to all buttons (most often the command is LinkTailPGC.)

It is normally safe to delete a button if you force the highlighted button to a non-deleted one, and if the commands of the remaining buttons are direct jumps to different PGCs.

Of course, you may safely delete a button created with the New Button function.

In the other hand, hiding a button is always safe, but if a SetHL_BTN highlights an hidden button, it will be selected by default, and nothing will be visible. This is why there are 3 choices to replace the command of the hidden button.

- Keep original command: if you press enter on the deleted button, the original command will be executed. This will probably lead to blanked material, or to a killed PGC.

- Jump to non deleted button: restarts the cell, but this time, a non-hidden button will be selected by default.

- NOP: Nothing happens. The menu continues normally. You have to use the cursors to highlight a non-deleted button.

It is possible to force the highlighted button to a non-hidden one with the Force Highlight parameter. But the same button will always be selected by default, regardless of the value of SPRM 8.

r0lZ

18th March 2005, 00:00

@Tobi
Please send me your IFOs and the menu VOB with at least the three cells with buttons.
I need also a description of what you did, step by step.
Thanks.

M7S

18th March 2005, 00:51

When I try to hide a button in Group 2 with PgcEdit that was already hidden with MenuEdit (Demo) in Group 1, PgcEdit freezes.

If I choose a command different commands for group 1 and 2 and then click on all, the command of the prevously chosen group is set for both groups. Is this behaivour intended?

I tried to make a BOV in the mainmovie with the subpictures that already existed, just to check if it was posible. I wasn't able to see the highlight. I checked the navpack with vobedit and saw that the "highlight endtime" was -1, could this be why it doesn't work, or does highlight endtime -1 mean eternal highlight?

Regards,
M7S

blutach

18th March 2005, 04:57

Re: Deleting/Hiding

In some players, I have found that a NOP hangs the player! I think the best is hide and (auto-activated) jump to non-deleted button. This way if that button is highlighted via a HL button pre-command, it simply jumps nicely to the next.

Alternatively, not hiding the button at all and replacing the command with LinkTopCell, button (next unhidden one) - which is the command for jump to non-deleted button - works real well and saves you the pain of reinstating the button later, if you need to.

BTW: r0lZ - is there an easy "unhide" function in PgcEdit's button editor?

Regards

blutach

18th March 2005, 08:18

@r0lZ - When blanking a whole titleset, the audio and subs are kept in the domain stream attributes. This seems weird since the titles in that VTST have only a NAVPack and 4 frames of video.

As well, PgcEdit offers to "correct" the number of streams back to what they were before blanking.

http://img225.exs.cx/img225/6823/untitled8ei.png (http://www.imageshack.us)

I think it needs to be the other way round - to delete any audio and subtitles in the stream attibutes.

Your thoughts?

Regards

r0lZ

18th March 2005, 10:30

Originally posted by blutach
Re: Deleting/Hiding

In some players, I have found that a NOP hangs the player! I think the best is hide and (auto-activated) jump to non-deleted button. This way if that button is highlighted via a HL button pre-command, it simply jumps nicely to the next.

Alternatively, not hiding the button at all and replacing the command with LinkTopCell, button (next unhidden one) - which is the command for jump to non-deleted button - works real well and saves you the pain of reinstating the button later, if you need to.

BTW: r0lZ - is there an easy "unhide" function in PgcEdit's button editor?

Regards NOP hangs on some players? Strange. I've seen many commercial DVDs with BOVs with NOP commands, for 'White Rabbit' like navigation.

Unfortunately, Auto Activate doesn't work for the button that is highlighted automatically when entering the menu. You have to manually select a button to trigger the auto activation.

No, there are no unhide facility. I doesn't have a place to store a backup of the deleted informations. The simplest way to revert to the old menu is to open the PgcEdit_backup in another instance of PgcEdit, export the button definition to disk, and import it back.
You may also systematically export the whole menu before doing any edit, as a backup.
Note that the exported file is an ASCII file, and therefore may be edited with a text editor. You may, for example, remove the definitions of the buttons you don't want to import, and leave only the definitions of the buttons you want to unhide.

blutach

18th March 2005, 10:36

Thanks for that info r0lZ.

Yes, my old SONY hung on a NOP (can you believe it?).

So, if auto-activate doesn't work, and we want the button to highlight, then can we force activation somehow?

Regards

r0lZ

18th March 2005, 10:59

Originally posted by blutach
Thanks for that info r0lZ.

Yes, my old SONY hung on a NOP (can you believe it?).

So, if auto-activate doesn't work, and we want the button to highlight, then can we force activation somehow?

Regards No. You may force auto activation by changing the Force Action button number and Force Action Button after N seconds, but these settings are global, and not programmable. You will therefore always automatically select the same (hidden) button when the delay is over.
But you may safely change the Force Select Button number. With that setup, you can force a non-hidden button to be highlighted when entering the menu. Drawback: the same button will always be selected when entering the menu, regardless of the value of SPRM 8. In many DVDs, when you return to a menu, the last selected button (or the next one) is automatically selected. If you force the Select button, you will loose this 'intelligence'.

Unfortunately, there is no good way to be sure that an hidden button is never selected. :(

r0lZ

18th March 2005, 11:14

Originally posted by blutach
@r0lZ - When blanking a whole titleset, the audio and subs are kept in the domain stream attributes. This seems weird since the titles in that VTST have only a NAVPack and 4 frames of video.

As well, PgcEdit offers to "correct" the number of streams back to what they were before blanking.

I think it needs to be the other way round - to delete any audio and subtitles in the stream attibutes.

Your thoughts?

Regards
There is indeed a problem discovered by jsoto:
Suppose that a DVD is authored with the same number of audio and subpic streams, with the same language codes, in each VTS. The SetSTN may be used only once, because when you go to another VTS, the same setup is still valid.
But if you have blanked a VTS, and removed the streams, the default audio and subpic streams will be automatically resetted to the default values by the player. When you return to a non-blanked PGC, the streams assignments will therefore no longer be the right ones.

So, although a little bit inconsistant, it is safer to leave the streams definitions as they were originally. This is why I don't change the subpics attributes in the blanked PGCs.
I agree that removing the streams definitions in the VTSI_MAT and VMG_VTS_ATRT tables while leaving the streams assignments in the PGCs is obviously inconsistant. I think I will remove the code that resets the streams numbers to 0 in the global tables after a blank out.

blutach

18th March 2005, 13:14

re streams

Yes, I saw that thread started by jsoto about a month or so back. It just seems to me that there won't be much damage because when you return from a totally blanked titleset, and want to play a title in a non-blanked one, then you will naturally select a stream present in the non-blanked one.

I understand if, for example, you selected DTS, and then the DVD played the (blanked) DTS intro (in a different VTST) and returned that you might lose the selected stream, but in general, I have not seen this happen.

I actually like to have them removed in VTSI_MAT and VMG_VTS_ATRT. Maybe, for people like me, you could set an option to have them automatically removed?

Regards

r0lZ

18th March 2005, 13:53

maybe... ;)

blutach

18th March 2005, 22:46

r0lZ - never mind! I'd rather be safe than sorry and have decided that (at least for me) it doesn't matter that there are a few extra stripped streams present and looks "untidy".

Regards

nwg

19th March 2005, 13:05

r0lZ

I think your software is fantastic and especially like your new menu editor.

I do have a suggestion if you don't mind.

I often use PgcEdit to create an ISO and Decrypter to burn it. I noticed there isn't an option to use Decrypters test mode. I would like to tick Burn ISO image with Decrypter and then tick use Test mode and then burn automatically after a successful test write.

Is that possible?

Thanks.

r0lZ

19th March 2005, 13:32

Thanks, nwg :)

Unfortunately, it is not possible to set the test mode in PgcEdit, because there is no TEST option in the command line arguments of DVDDecrypter.
If you really want to do that, uncheck the Start Burn Immediately option in PgcEdit, and select the test mode manually in DVDDecrypter.

r0lZ

19th March 2005, 13:35

Originally posted by Tobii
If I delete the second button in the Program 2 and indicate the third button as forced selection, I cannot delete the same button into Program 3 and 4.
PgcEdit then freezes. Any ideas? OK. I've found the problem.
It is an infinite loop in the routine that fixes the adjacent buttons links after a delete or hide button.
It's fixed in beta 3.
Thanks for your help, Tobi.

r0lZ

19th March 2005, 13:42

Fixes the bug discovered by Tobi (see above), and a small problem with the Group radiobutton being highlighted in red after a Delete Button, even when the All Groups mode was current.

Tobii

19th March 2005, 14:01

I try it out just.
BTW, It has noticed which disc this was? The part is heavy to reauthoring.

How always...Thanks.

Edit: Am now ready, work without errors now. :)

ukendt

19th March 2005, 17:08

PgcEdit 0.5.1 beta 3 available
where?:rolleyes:
Never mind, found it as announced in post 1(I'm too old and getting blind :( )
(BTW great work, this proggy is going to kill many other commercial progs and has alredy valuable features to subs colours and so)

blutach

21st March 2005, 13:32

@r0lZ

Is it possible to skip a menu (eg Language Selection/Aspect Ratio) using the force select and force action functions?

I tried this - the forced select worked just fine, but the whole cell played first before the forced action. How would this go with a still menu of infinite time? Or would the idea be to reduce still time to 0?

Regards

nwg

21st March 2005, 15:43

Originally posted by r0lZ
Thanks, nwg :)

Unfortunately, it is not possible to set the test mode in PgcEdit, because there is no TEST option in the command line arguments of DVDDecrypter.
If you really want to do that, uncheck the Start Burn Immediately option in PgcEdit, and select the test mode manually in DVDDecrypter.

Thanks for the reply. I already do it the way you have said.

r0lZ

21st March 2005, 16:50

Originally posted by blutach
@r0lZ

Is it possible to skip a menu (eg Language Selection/Aspect Ratio) using the force select and force action functions?

I tried this - the forced select worked just fine, but the whole cell played first before the forced action. How would this go with a still menu of infinite time? Or would the idea be to reduce still time to 0?

Regards Yes, normally it's possible. You have to use the Force Action button, and the Force Action after... delay. The Force Select doesn't matter (although if the delay is not zero, it is better to see what button will be activated by default). The still time do not need to be changed.
The force action delay must be < total playback time (ie motion or audio length + still time)
If you want to skip the menu, the action delay should be set to 0.

Note that I have heard that in some cases, this doesn't work as expected. But all test I've made were conclusives, including with my Sony.

jeanl

21st March 2005, 19:03

I confirm what r0lZ says. It's possible and it seems to work OK (in the cases I tried). See this (http://forum.doom9.org/showthread.php?s=&threadid=91495) thread and the links therein, it's an "old" idea...
Jeanl

blutach

21st March 2005, 22:51

Thanks guys. In PowerDVD, it waits till the end of the cell and then for the force action delay time (I had set it to 1).

I will try again with 0.

Regards

MarioGarcia

25th March 2005, 01:08

hello
I went to the webpage and all I could download is the version 0.5.0, how can I get this new version so I can test?
thank you

r0lZ

25th March 2005, 08:17

You must go to the VideoHelp mirror (http://www.videohelp.com/~r0lZ/pgcedit/). Download the beta in the folder pointed to by the "All Versions" link.

Anyway, the final version will be released in a couple of days...

CoNS

25th March 2005, 09:56

A request:

As you know, I use Paddington nice tool SubtitleCreator alot for adding custom subtitles to my DVDs. I use the default subtitle colours in SC, which is located in the file VTS_01_0.IFO in the examples folder.

After muxing the disc (with MuxMan) I always load the disc in PgcEdit to add a SetSTN precommand to turn on the new subtitles as default. I then have to exit PgcEdit and load the disc in IfoEdit to be able to copy and paste the subtitle colours from SC's .ifo file, as PgcEdit only allows copying of subtitle colours between two full DVDs - it cannot load the separate .ifo file from the SC folder.

In connection with the subtitle colours paste function, could you add an option to import subtitle colours to the clipboard from a single .ifo file, i.e. not load the whole disc. As an alternative to starting PgcEdit two times and copying between full discs?

r0lZ

25th March 2005, 11:15

It's more difficult than that. In most IFOs, there are several PGCs. What PGC should I use to retrieve the color table?
And if there are already some original subtitles in the PGC, you will paste wrong colors.

My advice: create a small DVD (with only a black frame as title 1), then import manually the color table of subtitle creator in this new DVD, or create your own table, and save it. You may copy the whole DVD in the examples folder of subtitle creator, and change his setup. (Edit the XML file, which holds the defaults values, including the IFO filename.)
Now, you will be able to load the new DVD in PgcEdit to copy/paste the color table easily.

CoNS

25th March 2005, 15:24

I see your point about cases with more PGCs in a VTS. I admit that I didn't think of that. However, you could bring up a small simple window and let the user select the PGC he wants from a dropdown menu, like in PgcDemux... (select by looking at the playback time for the PGCs)

It's just that the solution with loading two instances of PgcEdit at the same time to copy the colours in a way seems a bit off the very high quality level of the application in every other respect...

My request is inspired of ReJig and SubtitleCreator, which both have a function where the user can load an .ifo file into the program and retrieve the subtitle colours (from the largest PGC, if more than one are present?) so that they're used when muxing and creating subtitles respectively with the programs. It works like a charm in these two proggys, and I was just thinking it would be very cool to have in PgcEdit, too. :)

doomer2

25th March 2005, 20:11

I'd like to be able to move/add/delete program points in an existing TTN, but can't figure out how to do it. In the streams screen you can change the VOB/Cell id's and the program mapping and in the Chapter window you can add/delete chapters, but is there a way to move the chapter/program points around?

r0lZ

25th March 2005, 20:14

No. You have to reauthor the DVD to insert program (or chapter) points where you want. A program must begin at the beginning of a new VOB/Cell ID, therefore it cannot be changed by modifying the IFOs only.

doomer2

25th March 2005, 20:40

Ok, then how about a way to add an empty (blank) program point at the end of a title for the purposes of being able to skip all the way to the end using the next chapter button? There's already a method to replace a vob/cell link with a new blank vob cell, how hard would it be to add a new blank vob/cell and be able to put it at the end (or beginning) of the cell chain?

I have a dvd that has two titles that I'd like to be able to change to play mostly seamlessly (and nav between with chapter keys). Being able to add a new program point would allow me to attach a cell command to chain to the next title but make the chapters look seamles (other than a momentary black screen between).

Possible?

r0lZ

25th March 2005, 21:02

You may use AddChapter (by Jsoto) that does just what you want.

It's not possible with PgcEdit. I prefer the method of changing the NextPGC link to be able to skip the last chapter with the remote.

doomer2

25th March 2005, 21:33

Thanks, I'll look into that.

Unfortunately, changing the NextPGC link only works for chapter forward, not for previous chapter, which I think would work with an empty cell with a cell command to link to the previous title.

doomer2

25th March 2005, 23:41

Looks like I was wrong.

I figured out how to do it with the NextPGC and PrevPGC links, and an added dummy pgc for each title that linked to a dummy VTSM that jumped to the right title. Add a GP register to flag if I needed to link to the last chapter when going from second title to the first, and it works great.

Thanks for your help!

doomer2

26th March 2005, 01:14

I think I found a bug in 0.5.1beta3.

If you add a dummy PGC to a title, and then delete that same pgc, save and reload the dvd, then something gets screwed up in the IFO's and the last title in that VTS disappears. The PGC that was the title is still there, it's just no longer recognized as a Title, though it does still show up in the Title Play Map Table display.

I also noticed something else, which may or may not be a bug. If you add a dummy pgc to a title that is not the last title in the VTS, it will do it. But after that, you cannot delete that added pgc, as PgcEdit complains "Can only remove the last TTN in the titleset". Is that restriction really necessary?

doomer2

26th March 2005, 01:23

Ok, I have a question that I'm not sure where the problem is.

I've created a new dummy pgc on a title and made the PrevPGCN for the title point to it. Then in the new pgc, I issue a CallSS to a VMGM pgc that links to another title in a different titleset. In some software players this works, but in WinDVD it causes a crash. It also doesn't seem to work in my standalone Hitachi player.

Is there a problem linking to a higher numbered PGCN with a PrevPGCN link? Or maybe something else I've missed?

I've also tried setting the PrevPGCN link to the current PGC, but unfortunately, this doesn't seem to perform the same shortcut of jumping to the post command chain that doing it in the NextPGCN link does. At least, not in the players I've tested with. Or is there a way to make that work?

Thanks in advance.

r0lZ

26th March 2005, 11:09

Originally posted by doomer2
I think I found a bug in 0.5.1beta3.

If you add a dummy PGC to a title, and then delete that same pgc, save and reload the dvd, then something gets screwed up in the IFO's and the last title in that VTS disappears. The PGC that was the title is still there, it's just no longer recognized as a Title, though it does still show up in the Title Play Map Table display. I will look into it.

[EDIT:] OK. Bug fixed. The number of TTNs was decremented also if the TTN number of the deleted PGC was shared by several PGCs. Now, the number of TTNs is computed correctly.

I also noticed something else, which may or may not be a bug. If you add a dummy pgc to a title that is not the last title in the VTS, it will do it. But after that, you cannot delete that added pgc, as PgcEdit complains "Can only remove the last TTN in the titleset". Is that restriction really necessary? This restriction is necessary to keep the TTN numbers unchanged (except the last one, of course.) If a TTN is deleted before the last one, all jumps to the following TTNs will be wrong.

Apparently, there is a problem when trying to delete a PGC which share the same TTN number with other PGCs. In that precise case, PgcEdit should allow to remove the PGC, since the TTN will not be deleted completely.

[EDIT:] Bug fixed too. Now, PgcEdit looks if there are several PGCs with the same TTN number, and allow to delete the last PGC regardless of his TTN number if that TTN is still used by another PGC.
But you will still not be able to delete the last PGC if his TTN number is not shared with another PGC, and if that TTN number is not the last one in the domain.

r0lZ

26th March 2005, 11:21

Originally posted by doomer2
Ok, I have a question that I'm not sure where the problem is.

I've created a new dummy pgc on a title and made the PrevPGCN for the title point to it. Then in the new pgc, I issue a CallSS to a VMGM pgc that links to another title in a different titleset. In some software players this works, but in WinDVD it causes a crash. It also doesn't seem to work in my standalone Hitachi player.

Is there a problem linking to a higher numbered PGCN with a PrevPGCN link? Or maybe something else I've missed?

I've also tried setting the PrevPGCN link to the current PGC, but unfortunately, this doesn't seem to perform the same shortcut of jumping to the post command chain that doing it in the NextPGCN link does. At least, not in the players I've tested with. Or is there a way to make that work?

Thanks in advance. With the Next/Prev/GoUp links, in the title domain, you may only jump to another PGC of the same TTN. You must select the main PGC before using the 'New Dummy PGC' function, so that the dummy will share the same TTN number. Is is what you did? If so, it should work. :confused:

If you set the PrevPGC link to the number of the current PGC, then, when you press the Prev button of the remote when in program 1, you will jump back to the beginning of the program. The pre commands are not executed, nor, of course, the post commands.

r0lZ

26th March 2005, 11:23

Originally posted by doomer2
Looks like I was wrong.

I figured out how to do it with the NextPGC and PrevPGC links, and an added dummy pgc for each title that linked to a dummy VTSM that jumped to the right title. Add a GP register to flag if I needed to link to the last chapter when going from second title to the first, and it works great.

Thanks for your help! 2COOL has analysed this method. There is a thread somewhere here at Doom9 with his conclusions. I don't remember where...

doomer2

26th March 2005, 20:23

Originally posted by r0lZ
With the Next/Prev/GoUp links, in the title domain, you may only jump to another PGC of the same TTN. You must select the main PGC before using the 'New Dummy PGC' function, so that the dummy will share the same TTN number. Is is what you did? If so, it should work. :confused:

Yes, that is what I did. It looks kind of strange because there is another title after the one I need to bridge that is in the same VTS, and the new added chain has the TTN of the first, which makes them out of order.

I believe it works in all cases using the NextPGC link. The problem is when trying to do it with the PrevPGC and linking back to a prior title. That's why I was wondering if there is something requiring the PrevPGC link to go to a lesser numbered PGC than the current one. But I don't know of a way to insert a dummy pgc so it is in order.

The Prev link does seem to work in IfoEdit's DVD preview, and it works in PowerDVD. It doesn't work in WinDVD (causes a crash), and in my standalone, it stops playing the dvd when trying to go from the second part to the first with the prevchap button.

r0lZ

26th March 2005, 20:48

Hum. This problem is verry strange. :confused:

IMHO, it is perfectly legal to jump to any PGC number, as far as it share the same TTN.
I will try to do some tests...

sweetness

27th March 2005, 00:38

what does it mean when the group ALL is highlighted red? only one program does that the others don't.

and can i request something too?
will it be possible to export the button layout as a bmp too. if i move the buttons around i can put that layout in photoshop to fix were the subs go. just makes it a little easier. no big deal.

r0lZ

27th March 2005, 02:33

The All radiobutton is highlighted in red when at least one parameter of the button is different in each group. Also, in this case, the All group is not selected by default when you enter the editor.
If you go to the All group when it is red, you will loose this difference, because the parameters are saved for all groups.
Of course, the button position parameters are normally different in each groups, and are therefore ignored.
Most of the time, all groups are identical (execpt, again, for the buttons positions) but sometime there may be some important differences. For example, the button command to launch a movie may jump to different PGCs to show the movie in 16:9 or in pan&scan, depending on the display mode of the menu.
So, if you see the All button is red, you should normally never enable it, or you should carefully examine all groups to see if it is safe to use the All mode.
Note that there is a tooltip on the All button when it is red.

It's not easy to export the button layout, because the button highlights are not parts of the image. They are overlays lines, not included in the background image. Also, the image is always scaled to be shown in the correct display aspect ratio on your PC monitor. Saving the bitmap as it is is therefore useless, as it will not have the correct size, and no button highlights. Of course, you may use the print screen windows function to export the bitmap as it is on screen to your drawing program. Also, the positions of each button are saved in the exported menu or button files, in text format. You may retrieve the coordinated from these files to draw the rectangles with precision in your paint program.

sweetness

27th March 2005, 17:34

Originally posted by r0lZ
Note that there is a tooltip on the All button when it is red.
I didn't see that. thanks
like i said no big deal about the bmp i do a lot of print screen stuff.
thank you sir.

blutach

28th March 2005, 06:06

Hi r0lZ

Suppose I have a DVD that has many subtitle streams, most of which have been stripped. Is there a way I can use the Menu Editor to first select all the buttons which no longer represent valid streams (i.e. a number of buttons at all once) and then only do a single Hide Button command on all of these at once?

Regards

r0lZ

28th March 2005, 10:01

No, there is no way to multi-select the menu buttons.
But, if you want so, I may change the behavior of the Hide Button routine so that the next button is automatically selected when the current button is hidden. It's the way Delete Button works. This way, it will be faster to hide a sequence of buttons.
Also, don't forget that you may use the tearoff function on the "Hide Button" menu (the dotted line of the menu), so that all you will have to do is to click several times on the menu item to hide a consecutive sequence of buttons.

doomer2

28th March 2005, 11:10

Originally posted by r0lZ
Hum. This problem is verry strange. :confused:

IMHO, it is perfectly legal to jump to any PGC number, as far as it share the same TTN.
I will try to do some tests...

After a lot of time testing with various players, I think I've figured out what the problem is. Some significant number of players consider it illegal to have a PrevPGCN link to a dummy PGC. A chain called in a PrevPGC link MUST have at least one cell defined to work properly in several of the players I tested.

Strangely enough, this restriction doesn't seem to be enforced for NextPGCN in any players I've tested, which seems inconsistant.

This did bring up an idea for a feature request though. In getting around this restriction, I used the Add Play All function, to create a chain with cells to comply with this requirement. However, that left me with a ton of cells defined in a chain that should never be used.

I would have much preferred being able to add a chain with a single blank cell, or with a selectable chain of the cells from the current VTS. Do you think it would be much trouble to add something similar to the Add Play All function, but that lets you pick which cells to include and in what order? Or is there already some way to do this?

Thanks for your help with this.

blutach

28th March 2005, 11:25

@r0lZ

Sounds like a good method to me. Will look for this implementation in next version. Thanks.

Regards

blutach

28th March 2005, 13:34

Layer Break

@r0lZ - My understanding is that the layer break is more than just a flag (a non-seamless joint) - it is a few sectors of 00s that the player never really reads. Please correct me if I am wrong.

Now, if I remove the LB flag with PgcEdit, what happens to those sectors? Do they just stay there? Are they played? Is this a safe way of actually removing the layer break?

Assume that the VOBID is unchanged at the layer break (because if it does change, there can't be seamless playback anyway, right?).

Regards

r0lZ

28th March 2005, 14:48

Yes, the layer break flag in PgcEdit is not the real LB. In fact, it's the inverse of the Seamless flag. I have added it in the cell list for the user convenience.

The 'real' LB is, according to mpucoder, a sequence of several blank sectors. Unfortunately, there is no way to read them from the original DVD. Therefore, the VOB files are saved without those empty sectors. But the LB's cell must have the Seamless flag off. This is why it is possible to guess where the LB is.

Anyway, the real LB is lost when you rip the DVD. So, it is always safe to remove the LB flag in PgcEdit (assuming that the cells can be played seamlessly.)

When burning a DVD, I don't add the LB myself, because it is too difficult to safely figure out its place. Furthermore, I don't know how to inform a burner of the location of the LB. There is no flag in the command line of mkisofs to inform it where to put the LB. I suppose that it is the responsability of the burner program to find a good place, or maybe to search for a non-seamless cell to use as the LB. I don't know what's the behavior of DVDDecrypter on this matter.

blutach

28th March 2005, 14:59

@r0lZ - Thanks for that - I was not sure if the rip included the blank sectors but since it doesn't then it's easy to uncheck the flag.

AFAIK, DVD Decrypter in ISO Read mode preserves everything about the DVD, including the layer break info, and can, in ISO write mode, reinsert it (I also understand that it has the ability to change it, subject to the rules of layers 0 and 1). This is DL burners, obviously.

In any event, for us DVD-5 types, we can uncheck the LB flag. Thank you.

Regards

mpucoder

28th March 2005, 15:06

Originally posted by doomer2
After a lot of time testing with various players, I think I've figured out what the problem is. Some significant number of players consider it illegal to have a PrevPGCN link to a dummy PGC. A chain called in a PrevPGC link MUST have at least one cell defined to work properly in several of the players I tested.
My first reaction was "da" (not Russian)
Chained PGCs must have video, it's what the links are for. If you want to execute commands in a dummy PGC, you must jump to it.
A side effect of chaining is that the title can no longer be considered "one_sequential", which means no time map.