View Single Post
Old 11th February 2009, 16:48   #8199  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 9,140
Quote:
Originally Posted by Jeff Flowerday View Post
Actually an option to output a cue sheet from eac3to would take all the pain away.

http://wiki.hydrogenaudio.org/index.php?title=Cuesheet

Performer, title and track names aren't really important since we need to tag the resultant flac anyway. "FEATURE" , "FEATURE ARTIST" and chapter numbers for track names would suffice.

If you don't see value in it, I'll right a program to convert your chapter files to cue sheets.
I was already asked to offer an option to produce tsMuxeR compatible chapter files. But to be honest, I don't really like the idea to add one option for every tool out there which may use its own private chapter format...

Quote:
Originally Posted by laserfan View Post
But then even if I force eac3to to look at 00001 it still gives me 51 chapters?
Oh, that seems to be a bug, will have to check that...

Quote:
Originally Posted by magic144 View Post
Are you saying that doing something like this:-

Code:
eac3to L: 1) 2: video.mkv
rather than a blanket disc title demux would keep a VC1 stream in a container (.mkv, albeit a different one from its original .m2ts housing) and allow eac3 to fix video gaps/overlaps, and is that an existing feature or a future planned capability?
Yes. It's an existing feature.

Quote:
Originally Posted by magic144 View Post
If I remember rightly, you said the use of video gaps/overlaps is incredibly rare in source material - but you have seen it?
It's rare, but not unheard of. Especially broken broadcasts can have such problems.

Quote:
Originally Posted by 0xdeadbeef View Post
Well, indeed it doesn't seem to make much sense to have a seperate speedup option for each stream, either. I guess it's highly unlikely that someone wants to speed up just some streams.
Actually it does make sense to only speed up some streams. E.g. you may want to store the original audio track as it is (for future use) and speed it up at the same time. If I made "-speedup" a global option, doing such a thing would not be possible, anymore...

Quote:
Originally Posted by Snowknight26 View Post
The video was encoded using 23.976fps, just like all my encodes I have done. Should I be specifying 24000/1001 instead? Or does it not matter because its only a cosmetic issue?
It's mostly a cosmetic issue. However, all Blu-Rays I've seen are using 24000/1001 (or 24000/1000, but never 23976/1000). And all broadcasts I've seen use 60000/1001. So I do consider 23976/1000 "non-standard". But again: It's mostly a cosmetic issue...

Quote:
Originally Posted by honai View Post
would it require a lot of effort to include a simple video cutting mechanism, e.g. remove the first 2000ms of the video stream when demuxing and muxing from/to MKV?
"A lot of effort" is a relative term. It would be possible and doable, but it would cost some time, and I still have so many other things on my to do list. However, I could use some very limited cutting functionality myself for cleaning up the beginning of self-recorded broadcasts. So it might come sooner or later, but not too soon...

Quote:
Originally Posted by hubblec4 View Post
I found some old videos with sound.ogg
Is it possible to support them. It's not important but will be nice.
You mean Vorbis, right? Ogg is just a container. I will probably not add full support for Vorbis audio tracks. Eventually I might add support for decoding it through libav (if libav actually contains a Vorbis decoder, don't know). But I'm not sure yet...

Quote:
Originally Posted by tebasuna51 View Post
There are a problem with 'stdout.wav' and automatic second pass for overflow.
Didn't think about that. Will check that out later...

Quote:
Originally Posted by tebasuna51 View Post
There are a method to cancel the automatic second pass for overflow?
Many times are little peaks (<0.1dB) due to imperfections in lossy codecs than can be assumed.
You can use "-3db" or even "-1db" to avoid 2nd pass in most cases. Since clipping can only occur with floating point input, anyway, there should be no measurable loss in audio quality caused by the volume change.

Quote:
Originally Posted by KevinMcPool View Post
I have a 5.1 set-up with a modern hdmi receiver that can decode Dtshd and truehd, a PCH A110 and i have a number of movies with 7.1 dtshd sound..would it be better to demux with eac3to and the sonic decoder to 5.1 dtshd or let my receiver matrix the 7.1 channels to my 5.1 setup???Thanks.
Let your receiver do the work.

Quote:
Originally Posted by bobberty View Post
I'd like to take a BD or HDDVD and demux the main movie video stream and audio streams. I'd like to keep them in their original format, without processing or doing anything to the streams, ideally just a bit for bit extraction.
Why extracting HD DVD video tracks bit for bit? I think it would make more sense to let eac3to remove the pulldown. If you let eac3to do that, you'll usually end up with fully Blu-Ray compatible streams. In the early days of the HD DVD / Blu-Ray war Microsoft had supplied the Dual-Format-Studios with a little tool which would convert their HD DVD style VC-1 tracks to Blu-Ray style VC-1. eac3to does basically the same thing. I see no sense in keeping the pulldown flags in the video stream...

Quote:
Originally Posted by bobberty View Post
Is just passing -demux to eac3to the best way to do this? Would I need to pass other options, like -keepDialnorm and -keepPulldown?
You can use these options, but I don't really recommend that. The manipulations eac3to does by default usually have their purpose. But if you insist, of course you can stop eac3to from "improving" the video/audio data.

Quote:
Originally Posted by bobberty View Post
Also, for demuxing, does it make a difference if I use eac3to by itself without installing the Nero, Arcsoft, and Sonic filters? Or does having these filters installed give better demuxing results?
No need for any of those filters if you just demux.

Quote:
Originally Posted by Snowknight26 View Post
Just a couple of issues:
Code:
C:\unzipped\eac3to>eac3to.exe "C:\temp\i'm on a boat.mkv" 1: C:\temp\boat.h264 -
changeto29.970
MKV, 1 video track, 1 audio track, 0:03:09, 29.989p
1: h264/AVC, 720p29.989
2: AAC, 2.0 channels, 44.1khz
v01 The video bitstream is encoded in a non-standard framerate.
Was asked to modify track 1: to 29.970, but the original FPS is not supported.
Strange source. Will check that out later...

Edit: I can download that Youtube video as FLV, but not as MP4. How can I get that in MP4 format?

Quote:
Originally Posted by VonZippa View Post
While decoding a TrueHD audio track using eac3to, I received this error:

A search through the forums revealed that this is likely a bug in the libav TrueHD decoder. I am providing a 10 MB sample which reproduces this bug.
Thanks for the sample. But your search should also have revealed that it's no problem if you just receive one of those libav warnings. That means that max 0.8ms of your audio track is not lossless (if at all). All the other millions of milliseconds are still perfect.

Last edited by madshi; 11th February 2009 at 16:53.
madshi is offline