DGMPEGDec 1.0.13 beta 11

[neuron2]

I'm releasing a beta for feedback. Here is the link:

http://neuron2.net/fixd2v/dgmpgdec1013b11.zip

EDIT: Link updated to latest beta version.

Please refer to the Changes text file for details.

[Boulder]

Originally posted by neuron2
4. When quantizer matrix changes occur in a stream, random access would fail
in some cases, producing a corrupted frame. This has been repaired with the caveat
that any file that changes quantizer matrices on the fly must send a sequence
header with every GOP. I have never seen any file that does not do this. Please
notify me if you find one (i.e., one that does not have a sequence header for
every GOP *and* changes quantizer matrices).

If you're interested in checking this behaviour just for the hell of it, the latest CCE SP trial has the option for adaptive quantization, which I assume means changing the matrix in the stream. I don't know if using it will also force inserting the sequence header to every GOP.

Good to see you again:)

[Cyberia]

Originally posted by neuron2
1. A new CLI option for loading input files from a file list contained in a text file
was added as follows:

-Batch-Files -BF ([filelist]) Load batch of files.
The filelist file contains the list of files to load,
one file per line with no extra white space.
There is a very easy way to automatically generate an error-free file list:dir *.vob /b > filelist.txt
Especially great for long lists.

[neuron2]

It just occurred to me that the GUI should be able to load from the file list too. Then we would finesse the autoload issues. Users could use your trick to make a list, edit it as they like, and then load it through the GUI.

[Cyberia]

The GUI could even make the file list...

[neuron2]

To be consistent, then I'd have to give an editing capability. :)

[neuron2]

Here is DGMPGDec 1.0.13b5:

http://neuron2.net/fixd2v/dgmpgdec1013b5.zip

This version adds tritical's changes as follows (I really like the info overlay!):

A) added info option to mpeg2source() to display info on frames
- example usage: mpeg2source(info=true)

B) iPP is a bool again. If it is not explicitly set to true or false then it
defaults to auto which switches between field/frame based pp based on the
progressive_frame flag

C) 4:2:2 input support
- can now correctly decode 4:2:2 input sources and correctly output it
to AviSynth as YUY2

*NOTE: mpeg2source() now throws an error on 4:4:4 input, it could handle it
but AviSynth 2.5 does not have internal YUV 4:4:4 support

3a.) 4:2:2 PLANAR post-processing support

D) Upsampling to 4:2:2 from 4:2:0 based on progressive_frame flag. Uses new
upConv parameter of mpeg2source().

- example mpeg2source(upConv=true)

*NOTES: it will only work if input is 4:2:0, if input is anything else
then the upConv parameter is ignored

E) BlindPP now supports YUY2 colorspace

F) other stuff:

- fixed blindPP syntax bug (x and X)
- fixed a small bug with showQ option
- fixed small memory leak with FrameList/GOPList not being free'd
- fixed a bug in the vertical chroma deblocking postprocessing
QP pointer being passed was incorrect for 4:2:0
- faster mmx 4:2:2 to packed YUY2 and YUY2 to planar 4:2:2 conversions
- info output would not work correctly if temporal_reference was not zero
based at the beginning of gops, it does now

[zettai]

These changes are excellent! Everything is working fine for me so far.

I was wondering if it is possible to include the color matrix used by the vob in the info list?

[neuron2]

Originally posted by zettai
I was wondering if it is possible to include the color matrix used by the vob in the info list? DGDecode is of course not limited to VOBs. But what is the "color matrix used by the vob"?

Thank you for your feedback.

[zettai]

Ah sorry, yes I mean the mpeg2 coefficients - i.e. Rec.709 vs Rec.601. Gspot can identify these in mpeg2 headers. The reason I mention it is because it's useful to know when converting to RGB in avisynth as by default avisynth uses rec.601.

All mpeg2 files should be decoded rec.709 unless specified in the header and it's that header info that I'd find handy in dgindex or dgdecode.

There is a little discussion of this in the colormatrix filter thread:

http://forum.doom9.org/showthread.php?s=&threadid=82217


(Note this is really low priority :P)

[neuron2]

Can you make it easy for me by specifying the specific syntax element in MPEG2, so that I may look it up in the ISO 13818 spec? Thank you.

[Inc]

BlindPP now supports YUY2 colorspace
NICE!

Thanx a lot! :)

[Wilbert]

Neuron, I'm looking at iso138182 (http://le-hacker.org/hacks/mpeg-drafts/is138182.pdf).

It's the matrix_coefficients in the sequence_display_extension() in table 6-9 (of section 6.3.6), page 56.

[neuron2]

Originally posted by incredible
NICE!

Thanx a lot! :) Be sure to thank tritical for that!

[neuron2]

Originally posted by Wilbert
Neuron, I'm looking at iso138182 (http://le-hacker.org/hacks/mpeg-drafts/is138182.pdf).

It's the matrix_coefficients in the sequence_display_extension() in table 6-9 (of section 6.3.6), page 56. Perfect. Thank you, Wilbert.

[neuron2]

Originally posted by Wilbert
Neuron, I'm looking at iso138182 (http://le-hacker.org/hacks/mpeg-drafts/is138182.pdf).

It's the matrix_coefficients in the sequence_display_extension() in table 6-9 (of section 6.3.6), page 56. Wilbert, I have never seen a stream with a sequence display extension in it. Are you saying such things are out there? How can I report this field if it is never present? Is there some other way?

EDIT: Can you send me a stream that you think uses something other than 709 so I can check if it has the extension?

[Wilbert]

Many mpeg2s from dvd don't have it, although some do. I will look them up.

Many transport streams have it though (can also be seen with GSpot). I can send you a few samples, but I assume you have enough of those on your hdd :)

[Cyberia]

Originally posted by neuron2
To be consistent, then I'd have to give an editing capability. :)
I wasn't trying to lump work upon work. It just seemed like we could use the auto-loading feature of the GUI to get the filelist and save it for the CLI version.

However, typing: "dir *.vob /b > list.txt" is pretty darn fast, certainly faster that a GUI would be. Plus, if you are already using the CLI then you're not shy at the DOS prompt.

Typical of me... suggest a feature and then play devil's advocate against it!

[neuron2]

Originally posted by Wilbert
Many mpeg2s from dvd don't have it, although some do. I will look them up.

Many transport streams have it though (can also be seen with GSpot). I can send you a few samples, but I assume you have enough of those on your hdd :) I'll have a look. But, sure, I can display the matrix coefficients type, and if there is no extension, I have to report 709, per ISO spec.

[Boulder]

Originally posted by neuron2

EDIT: Can you send me a stream that you think uses something other than 709 so I can check if it has the extension?
This clip shows I470 (MPEG-1 coeffs) when opened in GSpot:

http://www.saunalahti.fi/sam08/adapt.mpv

[neuron2]

Originally posted by Cyberia
Typical of me... suggest a feature and then play devil's advocate against it! So typical. :)

I think I'll solve this loading issue once and for all.

[TEB]

thx for the new version. I tested it on my SWEp6 grabbed from DVB-sattelite and it crashes on detecting pids for video and audio. Just hangs. PM and ill give u a place to dl.

edit2: Well i tried it in Nic's version also and that one just bombs after opening the file.. The file works fine in videolan and all my other TS players.. So i have no idea whats wrong..

[Wilbert]

@neuron2,

I couldn't find any dvd samples. But if you create m2v with TMPGEnc or QuEnc it will have it:

http://forum.doom9.org/showthread.php?s=&postid=539167#post539167

The first post of http://forum.doom9.org/showthread.php?s=&threadid=81191 contains some QuEnc encodes.

[neuron2]

Originally posted by TEB
PM and ill give u a place to dl. Done.

[neuron2]

Originally posted by Wilbert
@neuron2,

I couldn't find any dvd samples. But if you create m2v with TMPGEnc or QuEnc it will have it:

http://forum.doom9.org/showthread.php?s=&postid=539167#post539167

The first post of http://forum.doom9.org/showthread.php?s=&threadid=81191 contains some QuEnc encodes. OK. Thank you. I'll just go ahead and add the feature.

[tritical]

That was fast :). Just wanted to say keep up the good work on dgindex/dgdecode it is much appreciated.

[neuron2]

Originally posted by tritical
That was fast :). Just wanted to say keep up the good work on dgindex/dgdecode it is much appreciated. In case you hadn't noticed, you are now in the credits list at the end of the DGDecode help file. If you prefer to have your real name, let me know. And if you have any more great ideas, be sure to speak up. :)

[neuron2]

Here's beta 6 with the colorimetry info added to the info display. Seems to work fine.

http://neuron2.net/fixd2v/dgmpgdec1013b6.zip

[neuron2]

Originally posted by Boulder
This clip shows I470 (MPEG-1 coeffs) when opened in GSpot:

http://www.saunalahti.fi/sam08/adapt.mpv It shows up that way in DGDecode too. :)

[Zep]

Originally posted by neuron2
In case you hadn't noticed, you are now in the credits list at the end of the DGDecode help file. If you prefer to have your real name, let me know. And if you have any more great ideas, be sure to speak up. :)


how about adding a decimate frames to use output file like
trictical does in his tdecimate

1st pass
decimate(5,framesout="theframesToUse.txt")

2nd Pass
decimate(5,framesin="theframesToUse.txt")

it speeds up pass 2 a TON not having to re calc all the dupes.

[neuron2]

Originally posted by Zep
how about adding a decimate frames to use output file like
trictical does in his tdecimate Not germane for DGMPGDec as far as I can see. Am I missing something?

[neuron2]

Originally posted by TEB
thx for the new version. I tested it on my SWEp6 grabbed from DVB-sattelite and it crashes on detecting pids for video and audio. Just hangs. PM and ill give u a place to dl.

edit2: Well i tried it in Nic's version also and that one just bombs after opening the file.. The file works fine in videolan and all my other TS players.. So i have no idea whats wrong.. Let me take a wild guess. :D

You have to re-name the files to ***.ts from ***.mpg.

It's time to implement autodetection of transport streams as Nic proposed.

[Zep]

Originally posted by neuron2
Not germane for DGMPGDec as far as I can see. Am I missing something?


heh yeah i guess this is not the thread to ask that but i tend
to group DGMPGDec and Decomb into the same DG package mindset.

because For me

mpeg2source() # is always followed by
Decimate() # when doing HDTV caps.

I must have DG on the brain huh lol


Anyway, what about adding that feature to decimate?
yes i will ask in a better thread if you want :D


Thanks and welcome back

[Leak]

Originally posted by Zep
Anyway, what about adding that feature to decimate?
yes i will ask in a better thread if you want :D

Ummm - if you want to have that feature now, why not just use TIVTC? *wonders*

np: Iwasaki Taku - Abandon The Books, Go Out On The Town (R.O.D The TV Original Soundtrack 2)

[Zep]

Originally posted by neuron2
In case you hadn't noticed, you are now in the credits list at the end of the DGDecode help file. If you prefer to have your real name, let me know. And if you have any more great ideas, be sure to speak up. :)

darn if i knew you were back i would have saved a .ts stream
I rawed last week that DGindex froze on but worked fine in
in everything else i ran on it.

I ended up having to use HDTVtompeg to strip the ts
and covert to mpeg so DGindex would index it which worked fine.

i will try and get another sample. Note i ran DGfix on it also
but it found no errors but still crashed/froze DGindex.

times like these i wish i had more drive space :)

[Zep]

Originally posted by Leak
Ummm - if you want to have that feature now, why not just use TIVTC? *wonders*


oh that is easy to answer. TIVTC is slower with same settings,
(well as close as i can get them to DG's decimate) DG's is 35% faster.
So using TIVTC with that feature is a wash and only gets
me to the same speed of DG's decimate which does not have that
feature.

So i figured if DG could add it we would be golden :D

if i recall correctly, tritical's version is c++ only? perhaps
that is the reason for the slower run times? i do use TIVTC when
i need the better options it has though which does a better job at
wacky patterns or other options that helps when dupes are edited out
causing 5 orignal frames in a row and i want NONE of them decimated
else you get jerky motion. tdecimate raised the cycle limit to 100
frames too (though i need 374 for UPN pattern cycle count with some
of their shows)

Anyway, i use both.

depends on what i'm doing and how fast i want need to do it :)

[tritical]

In case you hadn't noticed, you are now in the credits list at the end of the DGDecode help file. If you prefer to have your real name, let me know. And if you have any more great ideas, be sure to speak up.

Nick is fine, and I will... though none of them were really my ideas, just looked at what was on the development list and chose a few that I could manage :).

[neuron2]

@tritical

OK, then. If you want to claim any more items from the work list, please feel free. Just inform me so we can avoid duplicated work. I'll send you the current source if you want to do anything (usually I release the source only for release versions).

@all

Here is beta 7. It adds a fix for autoloading through the CLI, and does transport file auto detection, i.e., any file extension can now be used for transport files.

http://neuron2.net/fixd2v/dgmpgdec1013b7.zip

I will now address the PP deblocking issue and the autoloading general solution.

[Cyberia]

At some point it would be nice if Telecide could explicitly pick up the field order from DGDecode, so we'd never (rarely) have to set that option again.

Don't stop what you're doing to do this though :)

[neuron2]

Originally posted by Cyberia
At some point it would be nice if Telecide could explicitly pick up the field order from DGDecode, so we'd never (rarely) have to set that option again. Hints to the wise. ;)

[neuron2]

1.0.13b8 is now available. It fixes the postprocessing problem found by Janzki, and is the first step in the auto load solution.

http://neuron2.net/fixd2v/dgmpgdec1013b8.zip

Regarding the latter, the way it works now is that the Windows file open dialog allows multiple selection. And of course drag and drop can add multiple files. That gives the way to get the desired files into the file list. For both of these I automatically sort the files intelligently using strverscmp(). If you want to avoid the sorting, use repeated single file adds.

The CLI is unchanged.

This is the first step and I imagine there may be some whining. What I want to hear is *practical* cases where this new method causes problems. Not some obscure what-ifs. :)

I'll refine it as needed.

[Cyberia]

Drag and drop files need to go into the list. They are skipping that step and auto-loading.

There is a strange sorting related issue. Here is my file list:
VTS_01_VOBID_003_CELLID_013_1.VOB
VTS_01_VOBID_003_CELLID_014_1.VOB
VTS_01_VOBID_003_CELLID_015_1.VOB
VTS_01_VOBID_003_CELLID_015_2.VOB
VTS_01_VOBID_003_CELLID_016_1.VOB
VTS_01_VOBID_003_CELLID_017_1.VOB
VTS_01_VOBID_003_CELLID_018_1.VOB
VTS_01_VOBID_003_CELLID_019_1.VOB
VTS_01_VOBID_003_CELLID_020_1.VOB
VTS_01_VOBID_004_CELLID_001_1.VOB
VTS_01_VOBID_004_CELLID_002_1.VOB
VTS_01_VOBID_004_CELLID_003_1.VOB
VTS_01_VOBID_004_CELLID_004_1.VOB
VTS_01_VOBID_004_CELLID_005_1.VOB
VTS_01_VOBID_004_CELLID_006_1.VOB
VTS_01_VOBID_004_CELLID_007_1.VOB
VTS_01_VOBID_004_CELLID_008_1.VOB
VTS_01_VOBID_004_CELLID_009_1.VOB
VTS_01_VOBID_004_CELLID_010_1.VOB
VTS_01_VOBID_004_CELLID_011_1.VOB
VTS_01_VOBID_004_CELLID_012_1.VOB
VTS_01_VOBID_004_CELLID_013_1.VOB
VTS_01_VOBID_004_CELLID_014_1.VOB
VTS_01_VOBID_004_CELLID_015_1.VOB
VTS_01_VOBID_004_CELLID_016_1.VOB
Notice the fourth vob name.

I drag-n-dropped these files, so they autoload. The filenames are too long to see them in the list, so I am using the slider bar to see if they load correctly. The vob names are listed on the title bar.

I am grabbing the slider bar and holding it.

When I reach the spot where vob VTS_01_VOBID_003_CELLID_015_1.VOB should begin, the title bar says: VTS_01_VOBID_003_CELLID_015_2.VOB

It only displays this vob name for a few frames, and then changes to VTS_01_VOBID_003_CELLID_015_1.VOB. After this vob, then vob VTS_01_VOBID_003_CELLID_015_2.VOB is correctly listed.

The video in the display window also incorrectly using the wrong vob.

But, if I release the slider bar when it indicates the wrong vob, it changes back to the correct vob and the display changes to be correct. If I step through the video, the vobs ARE in the right order.

In fact, once I release the slider bar the first time, everything seems OK.

Finally, I cannot reproduce this using ONLY vobs 015_1 and 015_2!!!!! I need at least the surrounding four vobs to make it happen. Weird.


Still needed:
Horizontal scroll bar
Move Up & Move Down
Save Filelist (maybe)

[neuron2]

Here's beta 9 with the scroll bar, to help you test.

http://neuron2.net/fixd2v/dgmpgdec1013b9.zip

I'll look at the issue you reported.

[neuron2]

>Drag and drop files need to go into the list.
>They are skipping that step and auto-loading.

Please explain. I don't understand. It works for me.

[Cyberia]

Originally posted by neuron2
>Drag and drop files need to go into the list.
>They are skipping that step and auto-loading.

Please explain. I don't understand. It works for me.

I tested Beta 9 and it has the same problems. But the scrollbar works! Thanks :)

When you click File/Open, the File List dialog opens with the list, add, delete etc... Right? This dialog should open and the files should populate the list, when you drag files into the GUI.

They aren't, they are being auto-sorted and directly loading.

Multi-select using the ADD button, works correctly though. The file list is populated correctly.

The slider bar issue still occurs the sme way. (FYI, cause I know you haven't looked at it yet)

[Cyberia]

Two other minor issues:

The horizontal scrollbar is enabled even if it isnt necessary. The existing vertical one disables if not needed.

Ideally, neither should even appear if not needed. But the hor bar should at least disable if not needed.

Also, could you add some file filters to the Windows Open File dialog? Currently, you list either:

All MPEG-2 files
All files

But *.vob, *.ts, *.m1v, *m2v, and any others would sometimes be convenient.

[neuron2]

Drag and drop is like multiselect plus OK. You can open the file list if you like to add more. Is this a problem or just a consistency issue?

I don't know how to make the H scrollbar like the vertical. There is no Win32 control property like there is for the V one. Anyone know?

[Cyberia]

Well, I really just think dragondrop should work just like multiselect. You would want it for the same reason as you would for multiselect.

Think of drag-n-drop as an automatic multiselect.

[neuron2]

Beta 10:

http://neuron2.net/fixd2v/dgmpgdec1013b10.zip

This has:

Drag and drop now opens the file list box.
More file type choices.
Move up and move down buttons.

The trackbar slider has been screwed up from day 1 of DVD2AVI. I'm sure that is the cause of the funniness. Maybe I'll give it some attention; it's about time it got cleaned up.

Can we say autoloading is solved now?

[Cyberia]

Lookin good!

Two comments:

I would keep the ADD, DEL and DEL ALL buttons grouped together at the top. Then put the M-up and M-down.

Also, the new file filters are excellent! But... keep the default filter of "All Openable Files" (or whatever its called). Users will initially see all openable files, but can narrow (or expand) that filter if desired.


Now, just talking out loud here.... how about adding a .d2v filter to File Open and not have a Load Project menu option? Thoughts? Been mulling that over some today.

FYI: There are slider/button items on the dev list too. (Frame stepping, multiple selection ranges, seek to end buttons). May want keep them in mind if you do slider bar work. Even if you don't add these yet, maybe you can build it to support them easily. (or at least don't make it harder :) )

[len0x]

I've noticed that long time ago, but only recently managed to understood why it is happening: if you start DGIndex from CLI and for some reasons it fails to write output D2V file(most common reason - output directory doesn't exist or it doesn't have rights to write in the output dir) then it hangs forever (and save menu doesn't work) even though "-Exit" option was specified in CLI. I think it would be nice if DGIndex could respect "-Exit" option in case something goes wrong during save operation.

[neuron2]

@len0x

Beta 11 will work this way: If -Exit is specified, DGIndex will exit on failure to write the -OF specified D2V file. If it is not specified, then DGIndex remains open, an error box pops up, and DGIndex reverts to normal non-CLI operation to allow the user to save the D2V elsewhere. Is that satisfactory for your purposes?

@Cyberia

I like the current button arrangement. :p

I will add your default filter.

I'll keep the Load Project as is in order not to confuse the two functions: loading input files and loading a project.

[Boulder]

Don,

did you do some changes which may have affected the audio delay detected during project creation? The version you released before your sabbatical always gave me a -44ms delay on my PVR-250 captured MPEG-2's, and now the newer ones state -84ms. That extra 40ms would be exactly one frame in PAL video, maybe that's the point where it comes from?

I'm unable to say whether it's causing any a/v sync loss or not, of course. My brain has usually very bad a/v sync:p

EDIT: Oh, and decoding the resulting MPA track with MADPlay always reports "Lost synchronization" at the very last audio frame. The earlier versions didn't cause this.

[len0x]

Originally posted by neuron2
Beta 11 will work this way: If -Exit is specified, DGIndex will exit on failure to write the -OF specified D2V file. If it is not specified, then DGIndex remains open, an error box pops up, and DGIndex reverts to normal non-CLI operation to allow the user to save the D2V elsewhere. Is that satisfactory for your purposes?


Yes, perfect. Thanks.

[neuron2]

Originally posted by Boulder
did you do some changes which may have affected the audio delay detected during project creation? The version you released before your sabbatical always gave me a -44ms delay on my PVR-250 captured MPEG-2's, and now the newer ones state -84ms. That extra 40ms would be exactly one frame in PAL video, maybe that's the point where it comes from? I'm not aware of any changes that would be *expected* to cause reported delay differences, but side effects are always possible, especially in the sphagetti bowl of code that DGIndex started from.

Can you please provide a source file that shows differences under 1.0.12 and 1.0.13b10? Thank you.

Regarding the lost sync at the end, was the source file cut at the end? The changes I made do affect EOF handling, so if you do not have a clean cut of the input stream at the end, there may be issues.

@len0x

You're welcome.

[Boulder]

Originally posted by neuron2
Can you please provide a source file that shows differences under 1.0.12 and 1.0.13b10? Thank you.

Here you are : http://www.saunalahti.fi/sam08/debug_clip.mpg (~1.4MB). It shows -80ms delay with 1.0.13b10 and -40ms with 1.0.12. The gibberish you may hear when you play the audio track is Finnish:D

Regarding the lost sync at the end, was the source file cut at the end? The changes I made do affect EOF handling, so if you do not have a clean cut of the input stream at the end, there may be issues.

Nope, I always feed the whole several GB clip to DGIndex. It's possible that the capture program itself has produced some garbage at the last frame. Anyway, it doesn't affect anything since there's nothing after that particular audio frame.

Thanks!

[gizmau]

@neuron2

when i stop play or preview by pressing escape the "play/preview completed"-requester pops up every time, which i have to confirm. could you change this behaviour back to the old style of 1.0.12 without the requester, please?

[TEB]

Neuron: Back in town and tested the autodetect = works perfect.
But ive still got problems that it crashes when i click on the videopid and select videopid = bombs..
Ive uploaded the sw4.ts file for ure pleasure.

[neuron2]

@Boulder

I did change something that caused it. I have changed it back, but I'm not sure which is actually correct. At least it will be consistent with earlier versions. I am going to revisit all this AV sync stuff very soon. Will be fixed in beta 11.

@gizmau

Great idea. I had added that because I had a little bug that I could never find and that disguised it. :) But I just looked again with a fresh mind and found the problem straight away. So it'll now not pop up a box and the trackbar will not return to the beginning. You'll have to dismiss the info box yourself if you want. Will be fixed in beta 11.

@TEB

I wondered why you hadn't picked up your PM. I am downloading now and will hold up beta 11 until I have a look.

[neuron2]

@TEB

It almost feels like you are intentionally trying to torment me. :D

This latest stream of yours has some kind of private data stream on pid 0x11. Of course DGIndex defaults to that PID for video. So the video parser goes looking for video start codes and there aren't any in that stream. Because of the way I loop at EOF (don't ask!), it tries forever.

You can make your stream work by setting the video PID to something else, like 0x02 (never appears in a real stream), and then loading up your file.

As a temporary expedient, I made DGIndex default to reserved PIDs, with the idea that you should do a detect PIDs first. But refer to the Transport Stream Notes file for caveats. I'm thinking about the best long-term solution.

@all

So, here is beta 11:

http://neuron2.net/fixd2v/dgmpgdec1013b11.zip

It has len0x's CLI fix, a good "all decodable files" filter for the list box, Boulder's audio delay fix, gizmau's behavior at end of Save D2V/Preview/Play, and this change for default PIDs.

And now I am off for my din-dins.

[Cyberia]

Beta 11 permanently freezes if you try to do something stupid like load a jpeg rather than a MPEG-2 file and then do a Save Project.

1.0.012 does not hang, although it saves an empty d2v (Project Successful!)

And yes, I m aware of what the obvious solution is, but it shouldn't hang.

[neuron2]

Didn't you hear me say I'm off for my din-dins? :)

It's worth fixing. Next beta.

[MacAddict]

Beta 11 no longer presents itself in the foreground with a completed message it seems. Auto-incremented open files is missing now? Thanks for the continued updates!

[Cyberia]

Originally posted by MacAddict
Beta 11 no longer presents itself in the foreground with a completed message it seems.
Apparently true.

Originally posted by MacAddict
Auto-incremented open files is missing now?
You can now multi-select or drag-and drag files instead.

[MacAddict]

Originally posted by Cyberia
You can now multi-select or drag-and drag files instead. [/B] IIRC multi-select using the shift key always worked or maybe I'm wrong and thinking of something else. Guess I got use to it auto-selecting all of my VOBs in order :) Thanks.

[B.F.]

to neuron2:

I want to ask you to make a DGMPEGDec and Decomb hybrid.
It would be usefull for hybrid type (Film + NTSC)content, so I can use Decomb only for nonfilm video but Film pieces process normal (Force Film) way.

[neuron2]

Originally posted by MacAddict
IIRC multi-select using the shift key always worked or maybe I'm wrong and thinking of something else. Guess I got use to it auto-selecting all of my VOBs in order :) Thanks. No, drag and drop multiple files worked, but they didn't get sorted. You couldn't multiselect in the file open dialog. We changed the philosophy a bit so that now you explicitly multiselect all the files and then DGIndex sorts them. This was done because the autoload method did not always succeed, and was limited to a fixed number of files.

[TEB]

@Neuron2 : Yupp hehe.. And ill keep on doing it :) I have access at work on DVB-feeds from satellite for testing and Ill grab some more clips for testing. They contain lotsa goodie substreams that are bound to crash dgindex.

[suspiciousBob]

footling gui suggestion:

when Audio > Output Method = Disabled or Demux All Tracks, then grey out menu Audio > Track Number

[Cyberia]

The whole audio subsection (code and GUI) need work. We know. Not gonna be in 1.0.13 I bet. :)