Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > General > DVD2AVI / DGIndex

Closed Thread
 
Thread Tools Search this Thread Display Modes
Old 25th May 2008, 15:21   #461  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,922
Version 1.5.0 Final

* Fixed detection of field order for field structure streams.

* Fixed bug: Open file, Save Project, then File -> Open -> OK gives an error messge.

* Fixed bug: Save project with decode AC3 to WAV. Repeat that. Each time the WAV file becomes longer. The size was not reset to zero.

http://neuron2.net/dgmpgdec/dgmpgdec.html
Guest is offline  
Old 25th May 2008, 15:44   #462  |  Link
Taurus
Registered User
 
Taurus's Avatar
 
Join Date: Mar 2002
Location: Krautland
Posts: 903
It's getting better all the time!
Thanks a lot,
and looking at the change log, I can see how much work you've got.
Really appreciated
Taurus is offline  
Old 25th May 2008, 16:11   #463  |  Link
SeeMoreDigital
Life looks better in UHD
 
SeeMoreDigital's Avatar
 
Join Date: Jun 2003
Location: Notts, UK
Posts: 11,700
Hi Donald,

AAC audio (with MPEG-2) streams placed within .TS container are recognised in the "Information" bar as containing "MPA L2 2ch 44.1 free" audio.


Cheers
__________________
| I've been testing hardware media playback devices and software A/V encoders and decoders since 2001 | My Network Layout & A/V Gear |
SeeMoreDigital is offline  
Old 25th May 2008, 17:05   #464  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,922
Quote:
Originally Posted by SeeMoreDigital View Post
AAC audio (with MPEG-2) streams placed within .TS container are recognised in the "Information" bar as containing "MPA L2 2ch 44.1 free" audio.
Thanks for waiting for the release to tell me.

Do you have a stream I can use to duplicate it?
Guest is offline  
Old 25th May 2008, 17:23   #465  |  Link
SeeMoreDigital
Life looks better in UHD
 
SeeMoreDigital's Avatar
 
Join Date: Jun 2003
Location: Notts, UK
Posts: 11,700
Quote:
Originally Posted by neuron2 View Post
Thanks for waiting for the release to tell me.

Do you have a stream I can use to duplicate it?
I know, I know.... I'm useless

Here's a link to a sample: -

http://rapidshare.de/files/39518752/...h_AAC.zip.html
__________________
| I've been testing hardware media playback devices and software A/V encoders and decoders since 2001 | My Network Layout & A/V Gear |
SeeMoreDigital is offline  
Old 30th May 2008, 00:22   #466  |  Link
ChunkyNorwich
Incompetent encoder
 
Join Date: May 2008
Posts: 2
Hi neuron2,

I think I found a bug in DGIndex 1.5.0. I'm trying to use the "Analyze Sync" feature but my audio ID contains three hex digits. The output file is called "trim.timestamps.delayTb1.txt" (only two hex digits) even though the audio track is 0x4b1. The output file is virtually empty:

<file>
Delay Analysis Output (Audio ID b1)

Decode picture: temporal reference 2[I]
</file>

Here's a clip:
http://rapidshare.de/files/39561574/SpamClip.ts.html

Thanks
ChunkyNorwich is offline  
Old 30th May 2008, 01:15   #467  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,922
The audio ID is not the same as the audio PID. The audio ID is the one shown in the audio box of the info dialog. For transport streams, it is always 0. If you specify 0 as the audio ID everything will work as expected.

The users manual does state in the "Select Tracks" section that the audio ID is always 0 for transport streams, but I'll add that to the Analyze Sync section as well.
Guest is offline  
Old 30th May 2008, 11:53   #468  |  Link
ChunkyNorwich
Incompetent encoder
 
Join Date: May 2008
Posts: 2
Ah, my bad. It's working now, thanks.
ChunkyNorwich is offline  
Old 30th May 2008, 19:47   #469  |  Link
alfixdvd
Registered User
 
Join Date: Mar 2004
Posts: 243
Wit the same group of VOB's

Version 1.49 indicates colorimetry is BT.709
Version 1.50b10 indicates colorimetry is BT.709
Version 1.50 Final indicates colorimetry is BT.470

I pay attention to the final version or to the previous ones ?
alfixdvd is offline  
Old 30th May 2008, 20:01   #470  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,922
Colorimetry reporting changed. Please see the release notes.

Post a link to your stream if you want me to look at it and tell you why DGIndex reports what it does.
Guest is offline  
Old 31st May 2008, 07:58   #471  |  Link
alfixdvd
Registered User
 
Join Date: Mar 2004
Posts: 243
Quote:
21. When a stream does not declare the colorimetry, matrix_coefficients=1 is assumed for
HD video and matrix_coefficients=5 is assumed for SD video.
Quote:
DGMPGDec/DGINDEX versions starting with 1.5.0 beta 12 (released Nov 9, 2007)(including the current 1.5.0 RC2), report BT.709* for HD video and BT.470-2* [aka BT.601] for SD video when colorimetry information is not specified by 'sequence_display_extension'. The asterisk (*) means the stream did not declare the colorimetry. from http://forum.doom9.org/showpost.php?...2&postcount=93
Thanks, neuron2

Last edited by alfixdvd; 31st May 2008 at 08:01.
alfixdvd is offline  
Old 31st May 2008, 08:53   #472  |  Link
3ngel
Registered User
 
Join Date: Mar 2005
Posts: 457
@neuron2

I was reading the DgDecode manual and and i have some doubts. It says

Code:
upConv: 0 to 2 (default: 0)

Upsample from 4:2:0 to YUY2 (4:2:2) or RGB24.
- 0: Do not upsample
- 1: Upsample to YUY2 (ignored if input is already 4:2:2)
- 2: Upsample to RGB24
The points 0 and 1 are clear, but point 2 doesn't say what matrix uses in order to do YUV->RGB conversion.

Does it uses an arbitrary fixed matrix (Bt.601,Bt.709,SMPTe etc) possibly mismatching, or it does use the "declared in the stream" matrix (always correct result)?
Moreover after the conversion i get a 16-235 or 0-255 range?

Thanks
3ngel is offline  
Old 31st May 2008, 13:55   #473  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,922
The matrix is read from the GOP lines of the D2V file. The range is read from the YUVRGB_Scale line of the D2V header section. Refer to the DGIndex users manual for documentation of the D2V file format.
Guest is offline  
Old 31st May 2008, 15:09   #474  |  Link
3ngel
Registered User
 
Join Date: Mar 2005
Posts: 457
Thanks. So i see PC/TV levels refers to luma range, while the calorimetry is taken from the information provided by DgIndex.

But, at this point, looking at the information given by DgIndex when the calorimetry is not declared

Code:
Note that if the stream does not declare the colorimetry, then ITU-R BT.709* is reported for HD video,
and ITU-R BT.470-2* is reported for SD video.
The * character indicates that the stream did not declare the colorimetry.
I see from an NTSC dvd it goes on BT.470 B,G* wich means it did not declare.

But reading around many place (and i was wondering from some time about this) i read that BT.470 is used as a reference for "Analog" devices, while the BT.601 is the "Digital Counterpart". So i think this is the first "ingruence" we can say. The label would to be 601 instead of 470.
Moreover from this site wich gives a good calorimetry explanation, the ITU-R BT.470-2 System M would to be used on NTSC source while the ITU-R BT.470-2 System B, G would to be used on PAL sources.

So on my example, on NTSC dvd (altough from what i've read the 470 definition is not exactly correct) it would to use the M scale instead of the BG scale. But in the most correct way from what i've understood the BT.601 scale (or it is only a label strictly correspondy to the BG or M counterpart?) would to be used.

Moreover from this site when there are excerpt from the original "ITU Reference" wich unfortunately isn't publicy available, we have that the BG PAL assume a gamma of 2.8 while the M NTSC a gamma of 2.2.

Now because all the monitors sRGB have a gamma of 2.2 i think that it would be an even more reason to use the M matrix colorimetry instead of the BG colorimetry actually used by DGIndex on SD video.

What do you think?

Last edited by 3ngel; 31st May 2008 at 15:13.
3ngel is offline  
Old 31st May 2008, 15:57   #475  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,922
I think that if you don't like the choice that DGIndex makes, then you can upsample outside of DGDecode.

This has been discussed a lot and there is a lot of conflicting information. The choice I made is based on feedback from users.

The MPEG2 specification allows for 470-2 System B,G matrix coefficients. There is no way to specify System M for matrix coefficients. It does allow for System M for color primaries and transfer characteristics, but DGMPGDec doesn't use those in any way.

For our purposes BT.601 is synonymous with 470-2 System B,G.

Last edited by Guest; 31st May 2008 at 16:08.
Guest is offline  
Old 31st May 2008, 16:20   #476  |  Link
3ngel
Registered User
 
Join Date: Mar 2005
Posts: 457
Mine was not a critic. 'Cause i was going a more deep in the colorimetry i reported some observations of mine.

In this sense i don't know how the YUV->RGB conversion is done by DGindex but i saw in the sites above

Code:
ITU-R BT.470-2 System M (NTSC/USA)  (0.67, 0.33)  (0.21, 0.71)  (0.14, 0.08)  
ITU-R BT.470-2 System B, G (PAL)  (0.64, 0.33)  (0.29, 0.60)  (0.15, 0.06)
So i assumed that in the YUV->RGB formula those different value for the RGB cromacity could make a difference in the final result.

I obviously can choose a different colorimetry (with a conversion) if i don't like, but that was not the point

Hope you didn't misundestand me, mine was just "for the sake of the knowledge"
3ngel is offline  
Old 31st May 2008, 16:43   #477  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,922
I still don't understand your point. What is it? What are you asking me to do to DGMPGDec? This is the DGMPGDec thread, you know.

Just because you saw something in a site somewhere, it is not necessarily reliable. The only fully reliable sources are the specifications themselves.

Quote:
So i assumed that in the YUV->RGB formula those different value for the RGB cromacity could make a difference in the final result.
Did you miss the part where I told you that it is not possible to specify System M for matrix coefficients, and that it is only the matrix coefficients that are used for upsampling?

Last edited by Guest; 31st May 2008 at 16:47.
Guest is offline  
Old 31st May 2008, 17:47   #478  |  Link
3ngel
Registered User
 
Join Date: Mar 2005
Posts: 457
My point is that in the references the two BT.470-2 (M and G,B) are different and were made for two different television systems, PAL and NTSC that used two different phospores system in the consumer television, so the trasmission had 2 different color palettes.
Because you in DGIndex explicity say "System B", i made you note the fact that SystemB with a gamma of 2.3 could not be indicated on our monitors with a gamma of 2.2, while System M uses the same gamma, so using the System M pheraps would yeld more precise results.

But you say that there is no way you can use SystemM matrix (wich differentiate from the System B for the cromacity values i posted before). But at this point if you can't differentiate M from B, on wich basis you say you're using system B?
In other words if you can't differentiate from the two system, because from what you've said you don't use the cromacity values which are the difference between the twos, how can you say you use SystemB? Pheraps, you're using BT-470, but inside this there are other 2 color subsystem palettes. And if you can't differentiate between the two, pheraps saying only BT-470-2* would be more correct.

Moreover i can't find anywhere that BT.470-2 = BT.601.

So pheraps you're using BT.601 wich doesnt have the M,B differentiation being done for a digital environment.

In the end i was misleaded by the fact that you say using BT.470-2, System B when you say you are not able to distinguish from System M (part of 470-2 too), and the assertion you say BT.470-2 = BT.601 wich in my opinion (and from what i've read everyware) is not.

And because you say cromacity values dont' fit in your YUV formula i can suggest (in the end of all this) you change the label from BT.470-2 PAL G (wich refer to a color subsytem of an analogue system), to BT.601 wich doesn't have cromacity value differentiation (not referring to a analogue phosphor system) and is the right digital counterpart to BT-709.

BT.470-2 everyware i read is used in relation to analogue/signal/TV phospors systems.

BTW sorry for this long OT in the thread, if you want i'll delete all these posts.

Obviously good work and thanks.

PS: The whole point of all my dissertion was to add reproduction and formal preciseness to DGDecode.

Last edited by 3ngel; 31st May 2008 at 17:52.
3ngel is offline  
Old 31st May 2008, 18:23   #479  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,922
Why do I say System B,G?

See here:

http://neuron2.net/library/mpeg2/iso13818-2.pdf

Table 6-9 specifies B,G for value 5. So that is what I report.

Perhaps you could profit from doing a little homework on the relevance of color primaries, transfer characteristics, and matrix coefficients in the analog and digital domains. Recall that it is only the matrix coefficients that influence upsampling by DGMPGDec.

Quote:
My point is that in the references the two BT.470-2 (M and G,B) are different and were made for two different television systems
Analog television systems. DGMPGDec remains wholly in the digital domain.

Last edited by Guest; 31st May 2008 at 18:29.
Guest is offline  
Old 31st May 2008, 19:01   #480  |  Link
3ngel
Registered User
 
Join Date: Mar 2005
Posts: 457
I'm very glad you post the table because at page 68 just after table 6-9 it states:

Code:
In the case that sequence_display_extension() is not present in the bitstream or colour_description is zero
the matrix coefficients are assumed to be those corresponding to matrix_coefficients having the value 1.
That is:
if the portion of the stream containing the colorimetry is not present in the stream then switch to value 1 that is

Code:
1 Recommendation ITU-R BT.709
EóY = 0,7154 EóG + 0,0721 EóB + 0,2125 EóR
EóPB = -0,386 EóG + 0,500 EóB -0,115 EóR
EóPR = -0,454 EóG - 0,046 EóB + 0,500 EóR
This clausole of "default switching" is present in every section above the 6-9 table.

So in our case, because we were talking about "not present colorimetry" (that is no information pesent in the stream) the right setting according to the Reference would be BT.709
3ngel is offline  
Closed Thread

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 07:53.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, vBulletin Solutions Inc.