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 > (HD) DVD, Blu-ray & (S)VCD > (HD) DVD & Blu-ray authoring

Reply
 
Thread Tools Search this Thread Display Modes
Old 2nd March 2007, 17:53   #781  |  Link
JnZ
Registered User
 
JnZ's Avatar
 
Join Date: Jan 2004
Location: Czech
Posts: 181
Quote:
Originally Posted by Deckard2019 View Post
I've done a 2h20 movie with this method. No trouble. To be continued
Few minutes ago I tested AC3 produced from FFdShow and BeSweet and must say, that from FFDShow is totally crappy. LFE, SL and SR are totally different than original source. From BeSweet all OK.

So I strongly do not recomended use FFDshow output AC3!!!

Tested on UNILOGO.EVO
__________________
(Sorry for my bad english, I'm czech, not englishman... :))
JnZ is offline   Reply With Quote
Old 2nd March 2007, 18:19   #782  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 8,864
I've used EncWAVtoAC3 (which internally uses Aften) und the results look quite fine to me. In Audacity the graphs (pre and post AC3 compression) look nearly identical to me.
madshi is online now   Reply With Quote
Old 2nd March 2007, 23:01   #783  |  Link
Deckard2019
Registered User
 
Join Date: Jan 2005
Posts: 110
Quote:
So I strongly do not recomended use FFDshow output AC3!!!
So you mean, ffdshow can't encode AC3 properly ? So liba52 ?

Well, I've made several tries with different methods.
As we don't have any good DTS encoder, my goal is to produce an AC3@640kbps file directly from AVS script, without create a big WAV file. I can now do it with BeHappy, using Aften or ffmpeg.

How can I hear if the created AC3 file match perfectly the source ?
madshi, what are your Aften settings ant what is your Audacity method to compare the files ?
Deckard2019 is offline   Reply With Quote
Old 3rd March 2007, 01:48   #784  |  Link
orbitlee
Registered User
 
Join Date: Apr 2006
Posts: 78
Quote:
Originally Posted by Pelican9 View Post
I think the XPL's info is irrelevant.
The duration info comes from the DSI packet(s) of the EVOB (difference of the PTMs). I think this is the real value. The length of the demuxed audio is the same.

I think the time difference is caused by drop-frame timecode vs non-drop timecode. So nothing wrong at here.

You may find this article helpful for understanding.
http://www.csif.org/html/dropframe.html
orbitlee is offline   Reply With Quote
Old 3rd March 2007, 04:19   #785  |  Link
juhu
Registered User
 
Join Date: Aug 2003
Posts: 37
Quote:
Originally Posted by orbitlee View Post
I think the time difference is caused by drop-frame timecode vs non-drop timecode. So nothing wrong at here.

You may find this article helpful for understanding.
http://www.csif.org/html/dropframe.html
thanks... but there's another "curiosity" :
I finally decoded DD+ (as .wav mono ) with sonic decoder 4.2->avs->behappy... And it appears sound ran slightly faster than images (@ 23.976). Choosing 24 fps for video didn't help either, so I toyed with timestrech dsp available via behappy, and finally found the perfect stretching value : 24.000 -> 23.990 to keep the A/v synched all along the film
Don't ask me why...

Now, I'm working on "the thing" ... and then again, audio decoded (with Aften this time) runs faster.
For this one, the correct audio stretching is 24.000->23.988 to have a constant sync with video (still encoded @ 23.976)
so weird...

Of course, I'm talking about progressive decay in all cases, nothing to do with the 100-150 ms a/v starting delay

Last edited by juhu; 3rd March 2007 at 04:35.
juhu is offline   Reply With Quote
Old 3rd March 2007, 05:41   #786  |  Link
Icemaan
Registered User
 
Join Date: Mar 2006
Posts: 39
Morning

Can everybody say me where i can download behappy . the webside i cant download the file is there a alternatvie download link

Thanks

Icemaan
Icemaan is offline   Reply With Quote
Old 3rd March 2007, 06:24   #787  |  Link
2439rf8hasjdert4
Registered User
 
2439rf8hasjdert4's Avatar
 
Join Date: Jan 2007
Posts: 14
Quote:
Originally Posted by Icemaan View Post
Morning

Can everybody say me where i can download behappy . the webside i cant download the file is there a alternatvie download link

Thanks

Icemaan
Well, I don't think everybody can say you where download behappy, whatever that is, but I'm sure someone might tell you.
2439rf8hasjdert4 is offline   Reply With Quote
Old 3rd March 2007, 07:05   #788  |  Link
Chumbo
Registered User
 
Chumbo's Avatar
 
Join Date: Feb 2005
Posts: 585
I believe there's a link a few posts up, otherwise google should prove useful.
__________________
Chumbo
Chumbo is offline   Reply With Quote
Old 3rd March 2007, 09:39   #789  |  Link
LRN
TeMa=) team member
 
LRN's Avatar
 
Join Date: Apr 2005
Location: Universe, Galaxy, Solar System, Earth, Russia, Moscow
Posts: 212
Google for two keywords: "behappy" and "avisynth". It's 99.9% that you'll find BeHappy - Avisynth-based audio encoder rather than something else

Awesome work around here! Following guides from this thread i was able to decode and reencode Riddick HD-DVD into x264+AAC (only stereo ATM) (essentially, reencoded 16-GB VC-1 stream into 14-GB x264 stream (0.99 SSIM), so with AAC audio tracks it will fit on 15-GB HD-DVD disk). While i have some problems with fps, resync, decoding (from my usual playing toolkit [MPC/ffdshow + VLC + MPUI] no player is able to play this movie properly - either it hungs on some moment, or plays video faster than 23.976 fps, or can't open mp4 container) - it's all minor technical issues and could be solved. Currently i'm trying to smooth the whole process and to make it more open/free (i.e. without expensive software filters/decoders).
Nevertheless it works. We can decode and reencode HD-DVDs! CooL!
__________________
Without effort, you can't even pull a fish out of the pond.
Without a pond you can't pull it out even with effort.

Last edited by LRN; 3rd March 2007 at 09:49. Reason: addition
LRN is offline   Reply With Quote
Old 3rd March 2007, 11:23   #790  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 8,864
I have the same problem with using BeHappy that I had when I originally tried to create a wav directly with the Sonic Audio Decoder, only. Namely I got 2 good channels and 4 bad channels. The bad channels looked like loud noise to me with a tiny bit of audio information in the first 1-3 seconds of the track. Adding ffdshow to the graph behind the Sonic HD Audio Decoder solved the problem for me when outputting a wav file through "WAV Dest -> FileWriter". Unfortunately BeHappy doesn't work with a graph "DTS/AC3 Source -> Sonic HD Audio Decoder -> ffdShow". It complains about not being able to find a class or something like that. If I remove ffdShow from the graph, BeHappy works, but then only 2 channels are good.

Just as a warning to you guys. If it works for you that way, just BeHappy about that. But better make sure that you check whether all 6 channels are good. You can check that by converting the final AC3 file to 6 mono WAV channels by using Tranzcode and then by looking at the WAV files in Audacity.
madshi is online now   Reply With Quote
Old 3rd March 2007, 11:50   #791  |  Link
Deckard2019
Registered User
 
Join Date: Jan 2005
Posts: 110
I tried SoundOut with this AVS :
Code:
DirectShowSource("aud.grf", video=false)
AudioDub(BlankClip(), last)
SoundOut()
Open it in VirtualDub and SoundOut pops out. Choose AC3 (Aften) and start.
But SoundOut crashes ...
Deckard2019 is offline   Reply With Quote
Old 3rd March 2007, 11:54   #792  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 8,864
@Pelican9,

here's a list of little bugs I found in your great EvoDemux tool:

(1) If I open the "file open" dialog to load a new EVO file - but abort the "file open" dialog, the EVO is read again. This should happen, aborting the "file open" dialog should behave as if I didn't even open the dialog in the first place.

(2) If I delete the EVO file which is currently selected by EvoDemux and then open the "file open" dialog and then abort the "file open" dialog, there's a crash and afterwards all buttons are disabled.

(3) If I rebuild the EVO with only 1 audio track, the other audio tracks are still listed in the EVO. It seems that their content is removed, but they're still listed in the header or something like that.

(4) If there's some kind of action running and I try to close the whole program, the closing stalls until I press the "Cancel" button. Trying to close the whole program should either automatically cancel the running action. Or alternatively you could open up a MessageBox asking whether the running action should be cancelled.

Keep up the great work! Thanks also to Ron/drmpeg for the original demuxing code.

P.S: Does EvoDemux support being controlled by command line parameters yet? That would help a lot in automating things.
madshi is online now   Reply With Quote
Old 3rd March 2007, 11:56   #793  |  Link
Pelican9
Coder
 
Pelican9's Avatar
 
Join Date: Jan 2007
Location: Around the World
Posts: 697
Quote:
Originally Posted by orbitlee View Post
I think the time difference is caused by drop-frame timecode vs non-drop timecode. So nothing wrong at here.

You may find this article helpful for understanding.
http://www.csif.org/html/dropframe.html
Thanks,
it was really helpful for me.
Pelican9 is offline   Reply With Quote
Old 3rd March 2007, 12:01   #794  |  Link
Pelican9
Coder
 
Pelican9's Avatar
 
Join Date: Jan 2007
Location: Around the World
Posts: 697
Quote:
Originally Posted by madshi View Post
(1) If I open the "file open" dialog to load a new EVO file - but abort the "file open" dialog, the EVO is read again. This should happen, aborting the "file open" dialog should behave as if I didn't even open the dialog in the first place.

(2) If I delete the EVO file which is currently selected by EvoDemux and then open the "file open" dialog and then abort the "file open" dialog, there's a crash and afterwards all buttons are disabled.

(3) If I rebuild the EVO with only 1 audio track, the other audio tracks are still listed in the EVO. It seems that their content is removed, but they're still listed in the header or something like that.

(4) If there's some kind of action running and I try to close the whole program, the closing stalls until I press the "Cancel" button. Trying to close the whole program should either automatically cancel the running action. Or alternatively you could open up a MessageBox asking whether the running action should be cancelled.
Thanks, I'm working on it...

Edit:
1, 2, 4: Fixed.
3: Could you send me the status window of EVOdemux?
For me it's working perfectly

Last edited by Pelican9; 3rd March 2007 at 15:31.
Pelican9 is offline   Reply With Quote
Old 3rd March 2007, 17:23   #795  |  Link
zgx
Registered User
 
Join Date: Feb 2007
Posts: 94
Quote:
Originally Posted by madshi View Post
(3) If I rebuild the EVO with only 1 audio track, the other audio tracks are still listed in the EVO. It seems that their content is removed, but they're still listed in the header or something like that.
I am getting other bug when it comes to audio streams. I have played around alot with my test movie Batman Begins. EVOdemux detects:

"VOB number 4 contains 2 video , 5 audio and 4 subpicture streams."

It contains three DD+ tracks and one Dolby TrueHD track. I select the main video stream, one of the DD+ tracks, the Dolby TrueHD track and a subtitle and press rebuild.

The new rebuilt file contains all the selected streams (as it should) but if I run EVOdemux on the new file it cannot find the Dolby TrueHD track but it's there as the file size is much larger then if I do not select it.
zgx is offline   Reply With Quote
Old 3rd March 2007, 18:35   #796  |  Link
puppydg68
Registered User
 
Join Date: Dec 2006
Posts: 43
Quote:
Originally Posted by kornesque View Post
thanks to the efforts, input, arguments and collaborations of everyone on this thread, i've managed to successfully create an unadultered .wmv (playable on a 360) from a vc-1 .evo. it's only video, but baby steps always get you to where you're headed. all that remains is getting the damn audio downmixed and muxed. i'll be following closely until we get it all hammered out. thanks especially to Penguin9, drmpeg and Isochroma.
Could you please share this guide on getting the VC-1 video muxed unadultered into a Xbox 360 .wmv playable format?
It would be great if you could post it here http://forum.doom9.org/showthread.php?t=121168 or as an addition to this thread. Thanks in advance.
puppydg68 is offline   Reply With Quote
Old 3rd March 2007, 19:24   #797  |  Link
juhu
Registered User
 
Join Date: Aug 2003
Posts: 37
as noticed 2 posts above by ZGX, evodemux gives curious reports on rebuilt .evos

Once again : Casablanca.
Builiding one single evo with the 2 parts gives me this :

--------------------
Opening file Feature_1_Feature_2.rebuilt.EVO
File size: 13851 Mbytes.
VOB number 4 contains 1 video , 5 audio and 4 subpicture streams.
PTM of first video frame = 00000D91
PTM of last video frame = 105CFFB1
Duration = 0:50:50.297
VC-1 video stream 0 found!
First PTS = 00000D91
Substream id = 55
Sequence Header found
0 frames before first I-frame
Advanced Profile
Level = 3
Chroma Format = 4:2:0
Size = 1920x1080
Display Horizontal size = 1920
Display Vertical size = 1080
Aspect ratio = 1:1 (square samples)
Frame Rate = 23,976 (24000/1001)
(audio/sub parts removed)

147590 counted frames (1:42:35.733) in video stream 0.
73134 calculated frames.

in other words, it gives me stats from the first part, even after appending the 2nd (apart from the counted frames, but we know it's much less useful than calculated frames...)
juhu is offline   Reply With Quote
Old 3rd March 2007, 19:38   #798  |  Link
_xxl
ffdshow user
 
_xxl's Avatar
 
Join Date: Oct 2005
Location: Romania
Posts: 818
Quote:
I test AC3 via FFDshow, and it works...but not for long DD+ (only for unilogo.ddp).
It works or not?
Quote:
that from FFDShow is totally crappy
?
Quote:
LFE, SL and SR are totally different than original source
I would like to know how much different?
1%,30%,80%,100%?
Any samples?

Last edited by _xxl; 3rd March 2007 at 19:45.
_xxl is offline   Reply With Quote
Old 3rd March 2007, 19:47   #799  |  Link
Deckard2019
Registered User
 
Join Date: Jan 2005
Posts: 110
Quote:
Originally Posted by JnZ View Post
Hi folks,
after experimenting few hours, I found this working procedure to convert DD+ to anything else:

So here is updated Guide (old guide can't go over 4G barrier):
1) Create Graph filter: DTS/AC3/DD+ Source -> Sonic Cinemaster Audio Decoder 4.2 -> Dump (for DTS/AC3/DD+ open demuxed DD+ track, for Dump open output RAW (test.raw)). This grab RAW PCM data from Sonic decoder and put into file.
2) Pres play to render RAW PCM data. This rawdata file is depend on Sonic setting, so make sure you set 5.1 output. You've get 5.1ch 16bit (depends maybe on soundcard type), 48khz 6ch RAW data file without header. (it is critical to use Dump filter,coz it can take You over 4G barrier...which file writer evidently not). Wait until filesize of rawfile stop raising, than you can stop graphedit.
3) Use SoX to convert PCM to WAVe: sox -r48000 -t .raw -c 6 -w -s test.raw x:\test.wav (try set output to another physical disk, it speedups the process)
4) If you want 6 separate mono WAVes (for SurCode or other like-this stuff), use Wavewizard. In preferences sets Ignore size in header -> always, Stream manipulation -> Mono streams, Output format -> Wave PCM and finaly output dir. Open test.wav and press "Convert". It creates six WAVes 0-1-2-3-4-5 which match these channel order: FL-FR-C-LFE-SL-SR.
5)Encode final WAVes to anything you want.... for Besweet You need create .mux file for six mono WAVes (for AC3 encodes, etc), it contain lines like this (for AC3 encode):
Code:
x:\test_ch0.wav
x:\test_ch1.wav
x:\test_ch2.wav
x:\test_ch3.wav
x:\test_ch4.wav
x:\test_ch5.wav
Note 1: In my test, SurCode is crappy encoder, it can't handle 48khz WAVes correctly...I recomennded use other DTS encoder.

Note 2: Things You need:
- Graphfilter: http://www.doom9.org/Soft21/Filters/graphedit.rar
- SoX: http://sox.sourceforge.net/
- Dump ds filter (in graphedit package), use "register.bat".
- Wavewizard: http://www.rarewares.org/wavewiz/wavewizardv0.54b.zip

Note 3: First of all, prepare a lot of disk space: 2 hours of 6ch PCM or WAVe takes aprox 4G disk space. So You need double (for PCM and for WAVe).

EDIT: Channel order for BeSweet corrected!
I've just tested this on the same 2h20 movie. test.wav size is 7GB, just like test.raw.
But absolutely no sound in test.wav And duration is 1h23

Quote:
It works or not?
Well it works for me. AC3 file sounds ok to me.
Quote:
I would like to know how much different?
1%,30%,80%,100%?
Any samples?
+1

Last edited by Deckard2019; 3rd March 2007 at 19:54.
Deckard2019 is offline   Reply With Quote
Old 3rd March 2007, 20:39   #800  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 8,864
Quote:
Originally Posted by Pelican9 View Post
3: Could you send me the status window of EVOdemux?
Before or after the rebuild? Or maybe after reading the combined EVO?
madshi is online now   Reply With Quote
Reply

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 17:05.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2017, vBulletin Solutions Inc.