DGMPGDec 1.5.3 Final

[damjang]

Oh, I watch the video more acurately and found some glitches. And here is the sample (http://www.zshare.net/download/65495003728202/) (cut with projectX)

[LordIntruder]

Hi,


Some months ago I ripped a movie, PAL and progressive, around 1h50 long. From 1h40 to 1h45 suddenly the sequence was interlaced and then the last 5 minutes progressive like the rest of the movie.

This week I encountered the exact same problem with an old serie. In the middle of the episode, around 10 min, suddenly everything is interlaced.

Is there a way with DGIndex that could tell us automatically that the video has both progressive and interlaced content? Because as usual, I always see that AFTER the encoding ;) Grrr... ;)

I thought that maybe when we save the project (F5), at the end in the log file, we could have a report saying:

- Both progressive and interlaced

So that you know you have to use a deinterlacer. We already have such a report in the log, but I tried with my episode and it only report progressive and no interlaced because the final frames are not.

Is it possible to implement that?

At this stage the only way I know is to watch closely all the video and see if a sequence is not interlaced. As I'm rather lazy, I would enjoy an automatic way. :)

Unless there is already a tool that allow to do that?


Thanks :)

[SCIF]

neuron2, check your PM, please. I have a question about translate dgindexmanual.html

[signal]

@LordIntruder

I know it's PAL but when you previewed or played the material wouldn't the "Video Type" in the information window still switch to % Film if interlacing showed up?

[Boulder]

AFAIK, DGIndex uses the flags of the stream to determine progressive/interlaced, it doesn't do any in-depth analysis. Most PAL DVDs are simply flagged as interlaced and the flag doesn't change during the video.

[LordIntruder]

I know it's PAL but when you previewed or played the material wouldn't the "Video Type" in the information window still switch to % Film if interlacing showed up?

Hi,

Yes if you preview at the correct sequence, you have to press "[" and start preview, here the information change from progressive to interlaced, you are right. But if you know that, this is because you already made a search and this is what I want to avoid.

In my case, I open the VOB, I press the Preview button, DGindex does not report me that in the middle of the movie/episode a small sequence is interlaced, I presume it can't know it. It has to go through each frame to determine that.

My guess is when we save the project, DGindex looks at each frame (I suppose it works like that) and knows that some frames are not progressive like the others but interlaced, so that it could reports it in the final log. But my guess may be completely wrong.

If that is impossible while saving the project, maybe a new option similar to preview could be implemented that would do an in-depth analysis and making reports. Because doing a preview at maximum speed, on my machine (4200+) an episode of 50min needs 15min to be fully previewed. Almost half an hour for a 100min movie. I can't stay in front of my machine so long, looking at the information tab without moving my eyes for an hypothetical change here. We could have an option doing preview at maximum speed and get an automatic report log that would tell us:

"oh oh, some progressive and interlaced frames detected". And we could also have the start of those frames.

Say a movie has 150000 frames. Only the changing frame state could be reported

- frame 1: progressive
- frame 58000: interlaced
- frame 65001: progressive
- frame 150000: progressive

So you conclude that between 58000 and 65000 it is interlaced.

You can then examine with your rip program (Gknot, etc...) or directly VirtualDub via an avs those frames.

A friend of mine reported me he also encoutered the same problem 2-3 times on NTSC stuff where 2 min where interlaced. So you can control closely and still miss that unique and very short interlaced sequence.

If that can be done a day, that would be great. If not, I'll continue using my magnifying glass. ;)


Thanks.

[hkazemi]

Here's a keyboard usability note for DGINDEX 1.5.0 rc2:

1.) Using the keyboard navigate the following sequence: Alt-F | File | Open | select a file | hit Enter

2.) you should now see a window called 'File List' come up with ADD/UP/DOWN/DEL/DEL ALL on the right side

3.) observe that you can't use the keyboard to move between the controls. The 'tab' key should work to do this, but it doesn't. If you use the mouse to manually click on 'ADD' and then 'Cancel', you can see the button focus is now on 'ADD', and you can use the arrow keys to sort of select between the options, although they're not selected in order, and the focus gets stuck in the file list again.

[Brewskie]

Hi,

The last version I had where the splash.bmp was working was 1.5.0 beta 10. It does not seem to work in 1.5.0 RC 2. If there was an intentional change that I am unaware of, I am sorry I didn't notice it in the thread. Also, it's not like this hurts DGIndex functionality so it's not too big of a deal! ;)

Thanks!

[Guest]

I'm 15000 miles away from my source code right now, but IIRC, rename the splash.bmp to dgindex.bmp. I did it to allow coexisting with DGAVCIndex. The DGAVCIndex bmp should be called dgavcindex.bmp. I'll fix the manual when I return. Thank you for pointing it out.

[hajj_3]

sse4 optimisation would be nice, divx 6.7 and other programs that use it get UPTO 80% increased performance which is insane. sse4 will be on all 45nm quad core/dual core cpu's that are out in 2 weeks time.

Keep up the good work neuron.

[squid_80]

They use it for motion estimation. As far as I'm aware decoding doesn't need that.

[Atak_Snajpera]

I have one question.

Why MPEG Layer-2 audio is demuxed to *.mpa extension instead of *.mp2 ?

[buzzqw]

@Atak_Snajpera

from vob and ts (mpeg2 video) the audio extension is MPA

BHH

[Atak_Snajpera]

from vob and ts (mpeg2 video) the audio extension is MPA

aaaaaand.... What is your point?

If audio is detected as MPEG Layer-2 it should have *.mp2 extension.

[buzzqw]

look here

http://img122.imageshack.us/img122/7835/screenshot3101200813590ki2.th.jpg (http://img122.imageshack.us/my.php?image=screenshot3101200813590ki2.jpg)

audio reported as mpa, extracted as mpa

(if you want i can post the sample vob , 7mb)

BHH

[Atak_Snajpera]

MPA L2 = MPEG Layer-2 = *.mp2
Do you understand me know?

[buzzqw]

.. i loose a "Why" .. fast read as "How"

BHH

[Guest]

Is it a problem for you?

MPA covers all layers.

[Atak_Snajpera]

I think *.mp2 extension is more logical instead of mpa. If I have Dolby digital stream then I have ac3 extension. If I have PCM then I have wav and so on.

[Guest]

I didn't ask whether you thought it was logical.

If I have MPEG audio then I have *.mpa and so on.

[Atak_Snajpera]

If I have MPEG audio then I have *.mpa and so on.

If this "too" difficult to change then forget it. It looks like I won't convince you to use correct extension.

[Guest]

I asked you a simple question and you refuse to answer. Who is the intransigent one?

http://www.fileinfo.net/extension/mpa

[Atak_Snajpera]

I asked you a simple question and you refuse to answer

Some apps (h264tsto or xport) cannot detect audio type so demuxed stream has always .mpa extension. I use NicMPASource only if I have .mp2 extension so I'm sure that stream will be recognized otherwise DirectShowSource is used (if .mpa). You may ask then Why not use DirectShowSource for .mpa and .mp2? Because I've noticed that Nic's plugin is more stable. Besides It would be easier for user if he had .mp2 extension so knew immediately that he is dealing with MPEG Layer-2 compression.
All I'm asking to be more precise.

http://www.fileinfo.net/extension/mpa
http://www.fileinfo.net/extension/mp2

[buzzqw]

@Atak_Snajpera

you can use a Try / Catch avisynth script, if nicaudio fails go to dss
or try first with Nic, check for audio channels, and if not 2/error -> go to dss

BHH

[Atak_Snajpera]

you can use a Try / Catch avisynth script, if nicaudio fails go to dss
or try first with Nic, check for audio channels, and if not 2/error -> go to dss

It's not very professional solution in my opinion. I will use MediaInfo to rename demuxed mpa to mp2.

[Inventive Software]

WTF? How hard is it to rename it to MP2?

[Guest]

Besides It would be easier for user if he had .mp2 extension so knew immediately that he is dealing with MPEG Layer-2 compression. ...
All I'm asking to be more precise. The layer is already specified in the file name. But I suppose it is not difficult to print an extension of mp1, mp2, or mp3, so I will implement that for you. Thank you for the suggestion.

[The Scientist]

The layer is already specified in the file name. But I suppose it is not difficult to print an extension of mp1, mp2, or mp3, so I will implement that for you.oh no, can we have a vote, lol

[The Scientist]

I suppose it is not difficult to print an extension of mp1, mp2, or mp3, so I will implement that for you.ermmmm, can we vote on that? lol

[Sharktooth]

@neuron2: you may be interested in this: https://forum.doom9.org/showthread.php?t=135055

[Guest]

Here's a keyboard usability note for DGINDEX 1.5.0 rc2 All dialogs fixed per your suggestion. Will be in the next release (probably RC3). Thank you.

[Guest]

This week I encountered the exact same problem with an old serie. In the middle of the episode, around 10 min, suddenly everything is interlaced.

Is there a way with DGIndex that could tell us automatically that the video has both progressive and interlaced content? Because as usual, I always see that AFTER the encoding ;) Grrr... ;)

I thought that maybe when we save the project (F5), at the end in the log file, we could have a report saying:

- Both progressive and interlaced

So that you know you have to use a deinterlacer. We already have such a report in the log, but I tried with my episode and it only report progressive and no interlaced because the final frames are not.

Is it possible to implement that? The problem is that the report of interlaced/progressive is not really telling you if the content of the video is interlaced, strangely enough. It tells you only how the stream was *encoded*. So it's not really suitable for basing a decision on whether a deinterlacer is needed or not. Given, that, it seems to me that your suggestion is moot.

There are tools for determining whether streams are actually interlaced, such as Megui.

[Guest]

But it would be great if the user could specify a permanent folder to store BMPs, or at a minimum, have the program remember a selected folder during an editing session and not go back to the source folder on each save. I have implemented an option to specify an initial directory that should be used for the BMP save dialog. It is stored in the INI file. This will be in the next release. Thank you for the suggestion.

[The Scientist]

The layer is already specified in the file name. But I suppose it is not difficult to print an extension of mp1, mp2, or mp3, so I will implement that for you. Thank you for the suggestion.Can someone please tell me why on TWO seperate occasions my reply to this message has been removed?
My reply was something like "can we have a vote on that... lol" hardly something worth deleting......

[Atak_Snajpera]

I see no reason. What is your problem? Finally DGIndex will use correct extensions instead of mpa which doesn't mean anything. You don't know if you have Layer 2 or Layer 3.
BTW. DGAVCDec has been already adjusted and nobody was complaining.

[hkazemi]

I see no reason. What is your problem? Finally DGIndex will use correct extensions instead of mpa which doesn't mean anything. You don't know if you have Layer 2 or Layer 3.
BTW. DGAVCDec has been already adjusted and nobody was complaining.

I don't know why the posts were removed, but I also don't see why the change would need a vote either.

If you have a script that depends on the .mpa extension, you can workaround this change by adding a script step to rename the .mp2/.mp3/.ac3 to .mpa before going to the next step.

[Guest]

@The Scientist

I deleted them because they appeared to be useless chat to me (rules 3 and 11). Apparently, I made a mistake thinking you were just making jokes in the development thread. So, are you really arguing for the extension to be left as MPA? I know some apps allow the user to define the extensions. Do you think we need something like that?

I restored your messages so that you may decide whether they live or die. I should have PM'ed you about it. My bad.

Oh, I also deleted one of Atak's, so it wasn't personal.

Note to self: Don't put your hand in the suit spinner until it stops spinning. Blood is hard to wash off clothes.

[jeffy]

@neuron2: Unless this is a big issue for you, I agree (this is not a joke) with The Scientist. I have a lot of batch scripts depending on the mpa extension and I don't think I am alone. If it is not much of a hassle for you, would it be possible to have another checkbox or DGIndex.ini option, please? Thank you.

[remiche]

I hate this new version!

[25.2.2008 ã. 00:58:18] AutoGK 2.45
[25.2.2008 ã. 00:58:18] OS: WinXP (5.1.2600).2
[25.2.2008 ã. 00:58:18] Job started.
[25.2.2008 ã. 00:58:18] Input file: D:\Movies\VIDEO_TS\VTS_01_0.IFO
[25.2.2008 ã. 00:58:18] Output file: D:\Movies\VIDEO_TS\Movies.avi
[25.2.2008 ã. 00:58:18] Output codec: XviD
[25.2.2008 ã. 00:58:18] Audio 1: English AC3 6ch
[25.2.2008 ã. 00:58:18] Subtitles: none
[25.2.2008 ã. 00:58:18] Format: AVI
[25.2.2008 ã. 00:58:18] Target size: 1493Mb
[25.2.2008 ã. 00:58:18] Custom resolution settings: fixed width of 720 pixels
[25.2.2008 ã. 00:58:18] Audio 1 settings: Original
[25.2.2008 ã. 00:58:18] Started encoding.
[25.2.2008 ã. 00:58:18] Demuxing and indexing.
[25.2.2008 ã. 00:59:34] Processing file: D:\Movies\VIDEO_TS\VTS_01_1.VOB
[25.2.2008 ã. 00:59:34] Processing file: D:\Movies\VIDEO_TS\VTS_01_2.VOB
[25.2.2008 ã. 00:59:34] Processing file: D:\Movies\VIDEO_TS\VTS_01_3.VOB
[25.2.2008 ã. 00:59:34] Processing file: D:\Movies\VIDEO_TS\VTS_01_4.VOB
[25.2.2008 ã. 00:59:34] Source resolution: 720x480
[25.2.2008 ã. 00:59:34] Found NTSC source.
[25.2.2008 ã. 00:59:34] Source aspect ratio: 16:9
[25.2.2008 ã. 00:59:34] Analyzing source.
[25.2.2008 ã. 01:05:56] Source has percentage of interlacing in motion areas: 45,51
[25.2.2008 ã. 01:05:56] Source has percentage of telecined patterns: 96,97
[25.2.2008 ã. 01:05:56] Source has percentage of progressive patterns: 3,03
[25.2.2008 ã. 01:05:56] Source has percentage of interlaced patterns: 0,00
[25.2.2008 ã. 01:05:56] Source is considered to be FILM.
[25.2.2008 ã. 01:05:56] Output will contain 193339 frames
*************************************
EXCEPTION: Audio is not found.
*************************************
[25.2.2008 ã. 01:05:56] Job finished. Total time: 7 minutes 38 seconds

[The Scientist]

I see no reason. What is your problem?My problem is you moaned and argued like a child to get the option included.

I deleted them because they appeared to be useless chat to me (rules 3 and 11).I see... fine line between "moderater" and "censor"? I was making a 'light-hearted' comment to a slightly more serious issue, if that requires deletion, then expect to lose a chunk of the forum. Anyway this is starting to go off-topic, the point I am trying to make is: It doesn't matter what your program calls the audio files because the programmers that use DGIndex as a third party application can amend their programme code to suit, but you appeared to be 'bullied' into doing so from one person.

[hkazemi]

I hate this new version!

[25.2.2008 ã. 00:58:18] AutoGK 2.45
[25.2.2008 ã. 00:58:18] OS: WinXP (5.1.2600).2
...

*************************************
EXCEPTION: Audio is not found.
*************************************
[25.2.2008 ã. 01:05:56] Job finished. Total time: 7 minutes 38 seconds

The audio filename change wasn't suggested until after RC2, and RC2 is the version currently posted. Where'd you find RC3? If you're on RC2, you may have a different problem.

[Guest]

[25.2.2008 ã. 00:58:18] AutoGK 2.45
*************************************
EXCEPTION: Audio is not found.
*************************************
[25.2.2008 ã. 01:05:56] Job finished. Total time: 7 minutes 38 seconds I am not the author of AutoGK, nor do I support it. If you load your VOB(s) directly in DGIndex, is the audio found? If not please provide a VOB fragment that I may use to duplicate and fix your issue.

As AutoGK is no longer developed, compatibility with evolving versions of DGIndex can no longer be guaranteed. It might be best to stick with the version that comes with AutoGK.

[Guest]

Would you MPA loving guys be happy if I just added a line to the INI file to allow you to revert the extension to the old way, and not bothered making it controllable through the GUI?

[Atak_Snajpera]

Would you MPA loving guys be happy if I just added a line to the INI file to allow you to revert the extension to the old way, and not bothered making it controllable through the GUI?

I think this will be the best solution for those guys

[Taurus]

Would you MPA loving guys be happy if I just added a line to the INI file to allow you to revert the extension to the old way, and not bothered making it controllable through the GUI?
This would be the second best choice.
But this will not prevent some lamers and non-readers to lament and clutter the DGMPGDec threads.
Just my two cents.
@ neuron2: And a big "Thank You" for your ongoing work.

[jeffy]

Would you MPA loving guys be happy if I just added a line to the INI file to allow you to revert the extension to the old way, and not bothered making it controllable through the GUI?
For me yes, and a big "Thank you!"

[Guest]

OK, it's done. INI control only.

[For the record, I prefer not to view any DGMPGDec users as "lamers". Any and all feedback is gratefully received and I will do my best to respond. Sometimes, though, the tone one takes in making a request can be mirrored in the response. :)]

[Guest]

I've redemuxed the audio, but in mode "select track"
instead of "All tracks" (even if there was only one),
and the wave was correct. I was not able to reproduce any problems with WAV headers in 1.5.0 RC2, irrespective of demuxing method. Do you still have an issue with this? If so, may I have a test stream? I want to get RC3 out soon, so your response will be appreciated. Thank you.

[Taurus]

Sometimes, though, the tone one takes in making a request can be mirrored in the response. :)

:goodpost::) :p

[remiche]

http://img168.imageshack.us/img168/5821/99562952jg3.png