DGMPGDec 1.4.4 Final!

[neuron2]

http://neuron2.net/dgmpgdec/dgmpgdec144.zip

DGMPGDec 1.4.4 Final has these changes versus 1.4.3:

1. If a CLI D2V creation is invoked on a transport stream with the wrong pids
specified, when you enter the right PIDs, a bug caused an invalid filename
to be generated for the D2V file (".d2v"). This has been repaired.

2. Audio 48KHz -> 44.1KHz and Normalization have been fixed (they were
badly broken but Pre-Scale Decision was OK).

3. The __src__ AVS template macro was changed to __vid__ and an
__aud__ macro was added. Refer to the DGIndex manual for details.

4. The Audio menu options are enabled/grayed based on the selected audio
output method.

5. AVS script file generation was added to the DGIndex manual.

6. A complete and definitive account of the undocumented audio options was
added to the DGIndex user manual. I hope it may dispel some confusion and
misconceptions.

7. Fixed a bug that stopped AVS script creation when -EXIT was used in the CLI.

8. Added -MINIMIZE and -HIDE options to the CLI.

9. The DGIndex INI file is now always loaded from, and saved to, the same
directory that contains DGIndex.exe.

10. The error message for bad D2V opening has been broken into four to
make it easier to troubleshoot: D2V file is empty, DGIndex/DGDecode
mismatch, cannot open one of the source files listed in the D2V file, and
source stream has no sequence header.

11. Fixed an indexing bug in DGIndex that caused incorrect random frame
access in DGDecode in some rare cases.

12. Fixed a problem that caused stepping backward by GOP in DGIndex to fail
in some cases.

13. Added "m1v" to the open dialog "Files of type" string.

14. Increased from 100 to 500 the maximum number of PIDs that can be
displayed when performing Detect PIDs: Raw.

15. "Decode AC3 to WAV" was totally broken for tracks 2-8 (1 was OK). This
has been repaired.

16. A deficiency in the implementation of pulldown in DGDecode was repaired.
The result is that decoding is now resilient to changes of field order. This
makes Fix D2V superfluous, so it has been removed.

17. The right arrow key can now be used to single step in play/preview mode
if the playback speed is set to Single Step (previously just the right arrow button worked).

18. Colorimetry was reported incorrectly for MPEG2 in some cases. This has been fixed.

19. HOME key is now a synonym for the [ button, and END key is a synonym for the ] button.

[Cyberia]

Originally Posted by neuron2

6. A complete and definitive account of the undocumented audio options was added to the DGIndex user manual. I hope it may dispel some confusion and misconceptions.

Givin away all our secrets, man. shhhhh.

[mgh]

Thanks for adding the __aud__ micro.

[Zep]

Originally Posted by neuron2

DGMPGDec 1.4.4 beta 1 has these changes versus 1.4.3:

2. Audio 48KHz -> 44.1KHz and Normalization have been fixed (they were
badly broken but Pre-Scale Decision was OK).

wow just how bad were they broken? though I rarely if ever use
48KHz -> 44.1KHz, i did and do use Normalization a lot and i'm very
curious as to what it messed up in those encodes.

if i use projectx then I use besweet to do the audio but many
times i go straight to DGindex when i know the .ts had no drops.

Drops are why i'm stuck using projectx. to fix and sync but it is
slow and one more app/step i would rather avoid using if possible
and why i hope some day DGindex will fix streams then index the
fixed stream etc... then convert to wav with Normalization for editing.

I have to admit it would be nice to be able to edit (even at GOP level)
in DGindex. Then DGindex would only have to fix/index 40 minutes of
the show instead of 60.

yes i realize I may be dreaming the impossible dream (zep breaks out
into song even though he is clearly not julie andrews)

[neuron2]

@Zep

Let's not dwell on it. Let's just say that those two audio features were accidentally disabled in my source code [see ThreadKill()]. I assume the executable matches the source code.

Cutting and error preprocessing are coming; it's not a dream.

[GZZ]

When will a CLI option to totally hide the application doing D2V creation using commando line be added ?

a commando like "-Hide".

GZZ

[neuron2]

@GZZ

That is on the development list but is not a high priority. I did some experiments with it and it is not as easy as it sounds. I can't give you any schedule for its implementation.

[neuron2]

@GZZ

Lucky you. I dusted off my previous attempt and seem to have got it working OK, but that's what betas are for -- you tell me.

So, here is version 1.4.4 beta 2. It adds the CLI options -minimize and -hide. It also fixes a bug that stopped creation of the Avisynth script when -exit is used.

http://neuron2.net/dgmpgdec/dgmpgdec144b2.zip

[lineman]

Originally Posted by neuron2

@Zep


Cutting and error preprocessing are coming; it's not a dream.

This is awesome news, thanks for all your great work!

[iradic]

why is the "Output_Method=3" for "Demux all tracks" in ini file (i'm still using 1.4.3 final) - intention or mistake?
and why does dgindex leaves ini file behind (creates one in current dir) when started from other than install dir?

sorry if this have been answered before...

[neuron2]

Originally Posted by iradic

why is the "Output_Method=3" for "Demux all tracks" in ini file (i'm still using 1.4.3 final) - intention or mistake?

#define AUDIO_NONE 0
#define AUDIO_DEMUX 1
#define AUDIO_DECODE 2
#define AUDIO_DEMUXALL 3

Do you have a problem with that?

and why does dgindex leaves ini file behind (creates one in current dir) when started from other than install dir?

Because it's always been coded that way. Do you have a problem with that?

[iradic]

Originally Posted by neuron2

#define AUDIO_NONE 0
#define AUDIO_DEMUX 1
#define AUDIO_DECODE 2
#define AUDIO_DEMUXALL 3

Do you have a problem with that?

no, no problems

but cli is different, can be confusing ...
Output Method
-OM=method
0=None, 1=Demux Tracks, 2=Demux All Tracks, 3=Decode AC3 to WAV

forget i asked about ini file

bye

[Matthew]

Originally Posted by neuron2

Lucky you. I dusted off my previous attempt and seem to have got it working OK, but that's what betas are for -- you tell me.

Lovely. Works just fine and dandy for me. Calling it in a dos box, the only quibbles are:
a) It doesn't seem to wait for the project file to be saved. This actually suits what I'm using it for but as a general rule doesn't make sense.
b) The cmd window loses focus.

[neuron2]

Originally Posted by iradic

but cli is different, can be confusing ...
Output Method
-OM=method
0=None, 1=Demux Tracks, 2=Demux All Tracks, 3=Decode AC3 to WAV

Oh, I see. I've fixed it in the beta below.

forget i asked about ini file

Actually, I think you are right. I have fixed it.

Here is version 1.4.4 beta 3. It has these changes:

* The DGIndex INI file is now always loaded from, and saved to, the same directory that contains DGIndex.exe.

* The error message for bad D2V opening has been broken into four to make it easier to troubleshoot: D2V file is empty, DGIndex/DGDecode mismatch, cannot open one of the source files listed in the D2V file, and source stream has no sequence header.

http://neuron2.net/dgmpgdec/dgmpgdec144b3.zip

[Zyphon]

Wow! Yet another update of your excellent tool. I wonder Donald, do you ever sleep. It's great to see this excellent tool going strength to strength with regular updates.

Great work Donald.

[Stereodude]

I have a transport file recorded by my DVico Fusion 5 Gold HDTV tuner card from a Comcast's QAM channel that BBDMUX says has a PAT table in it (@ PID 0x0000), that DGIndex 1.43 says doesn't have a PAT/PMT table in it. 1.44b3 also says the same thing. I'm not sure if you're interested in taking a look at a small 10 second recorded file (~35MB) of this. If you are. Let me know.

BTW, thanks for all the hard work on this very useful tool.

[neuron2]

Yes, I am interested. Please upload it to my server ASAP and inform me of the filename.

[Stereodude]

Originally Posted by neuron2

Yes, I am interested. Please upload it to my server ASAP and inform me of the filename.

The file has been uploaded. The name is 2005-09-06(Tue)PM0834-CHD935.tp

[neuron2]

That stream has private tables for PMT, for which I don't know the semantics. Does BBDMUX actually decode the PAT/PMT and give you the program number-pid pairs, or does it just tell you a PAT is present as a stream?

My parser reports only the first 100 pids it finds. When I increased it to 500, it found the PAT but then failed to find usable PMTs because they carry the private table id 0x40.

If BBDMUX decodes them I can see how from the source, but first tell me what it can tell you. Thank you.

EDIT: Never mind. I tried it and it just prints the same stuff I do after I increased the maximum number of PIDs. All it tells you is that PAT sections are present. But as I said, they point to PMT sections that are coded with private data tables. Neither BBDMUX nor DGIndex can decode those private tables. Sorry.

[Zep]

Originally Posted by neuron2

@Zep

Let's not dwell on it. Let's just say that those two audio features were accidentally disabled in my source code [see ThreadKill()]. I assume the executable matches the source code.

oh

good thing i kept using 1.2.1 for so long for encodes huh LOL

Originally Posted by neuron2

@Zep
Cutting and error preprocessing are coming; it's not a dream.

Well i knew it was on your plate but your plate is already so full
and it appears to be a major and complex feature addition(s).

have to admit with those features added DGindex would
be the #1 app I would use by far