View Full Version : eac3to - audio conversion tool
madshi
28th January 2008, 10:16
It's on it's way. I was a bit fast on reporting. Should test with earlier versions of eac3to first to rule out bad lpcm source. Sorry!
www.earselect.se/bugreport.txt
www.earselect.se/test.lpcm
The crash occurs while eac3to is testing whether this file is a VC-1 stream. This explains why the problem didn't occur with v2.13 because that version wasn't able to automatically detect raw VC-1 yet. Will fix this in the next build. BTW, I'm not sure which channel mapping the Haali Media Splitter (which is used by h264tsto) uses for PCM files. So I'd suggest that you double check whether the channels are in the correct order...
Yraen
28th January 2008, 11:26
Could you please compare the length of the demuxed E-AC3 files v2.19 vs v2.20? I'm wondering whether v2.20 demuxes more or less than v2.19. Currently don't have time to look into this. Anyway, the difference should be minimal, so nothing to worry about...
Byte for byte they were the exact same size, the same goes for time. The funny thing is, I went and converted several tracks to flac (the same evo tracks converted to ac3) and different titles had the same error. I only converted to flac this round with 2.20, will try one of those two with 2.19 later today.
kmitalian
28th January 2008, 12:45
I'm gonna try it again now with that same method that has been tried with success,
i am getting a picture, the sound is perfect but it stutters, im gonna try to use the new eac3to and ill let you know, i do have POTO US Edition, and ill also try Harry Potter order of the Pheonix cause i couldnt demux that with EVODemux ok.
Jaja1
28th January 2008, 13:08
With which decoder and which eac3to version?Ah, for The Searchers I use both libav and sonic with eac3to 2.20 and earlier versions. Sonic creates
an ac3 track with clicks. However libav creates a clean ac3 track.
For Robin Hood, eac3to 2.20 and earlier, both sonic and libav gives me an ac3 track with clicks. I finally used the DVD audiotrack to mux into the mkv.
Both cases first audio track.
Regarding The Searchers, I don't know whether the file is really corrupt. As you already mentioned the eac3 track is okay and as I noticed, PDVD plays the file back fine. I got this via via from an vague acquaintance of mine, so can't rip it myself. Will ask him, but I don't know when this will be in my posession. I don't even know whether he's willing to do a rerip at all.
With eac3to v2.16 I can remux this to MKV and the resulting MKV plays perfectly fine including audio
with no artifacts and no speedup. I had to edit the EVO a bit, though, cause it was cut in such a way that the
video started at the middle of a frame.
Okay, how do I recognize frame boundaries in a hex editor. That would make life easier.
The mkv of Unforgiven seems fine now after briefly watching some parts. Great! Unfortunately extracting the audiotrack gives the not a clean audio track error. I'll try to solve that with delaycut.
I read you made your own splitter. I don't know anything of the complexity of splitters and direct show, but could this be made into a DS version? In that case VC1 evo could be played directly in ZP or MPC I guess. However, I don't know what I'm asking for in terms of time, so read the request with a grain of salt :)
And do you ever sleep? I seldom saw such a devotion to make something work. Thanks a lot for that :thanks:
rickardk
28th January 2008, 15:01
The crash occurs while eac3to is testing whether this file is a VC-1 stream. This explains why the problem didn't occur with v2.13 because that version wasn't able to automatically detect raw VC-1 yet. Will fix this in the next build. BTW, I'm not sure which channel mapping the Haali Media Splitter (which is used by h264tsto) uses for PCM files. So I'd suggest that you double check whether the channels are in the correct order...
Haali Media Splitter does not remap. So LPCM demuxed with h264tsto should be used with -blu-ray switch.
umaximus
28th January 2008, 16:23
madshi, hands down for such a great tool, would payout/donate anytime for this app..
Anyone that has done Million Dollar Baby with -skip2 option, did u maybe found out the correct delay for audio sync? how much frames does 'one skip' cut? Any good app out there, where could u import mkv video & audio in as seperates and then looking for lip sync on screen to find out correct delay needed?
rickardk
28th January 2008, 16:30
madshi, hands down for such a great tool, would payout/donate anytime for this app..
Anyone that has done Million Dollar Baby with -skip2 option, did u maybe found out the correct delay for audio sync? how much frames does 'one skip' cut? Any good app out there, where could u import mkv video & audio in as seperates and then looking for lip sync on screen to find out correct delay needed?
About -500ms. Mux audio and video with mkvmerge and then apply video delay in ffdshow. If using ffdshow for audio it important to have a preset that autoloads with diffrent codecs. AC3 will not go through ffdshow with same delay as uncompressed for example (I think).
umaximus
28th January 2008, 16:52
About -500ms. Mux audio and video with mkvmerge and then apply video delay in ffdshow. If using ffdshow for audio it important to have a preset that autoloads with diffrent codecs. AC3 will not go through ffdshow with same delay as uncompressed for example (I think).
Could i re-process the audio with eac3to and apply 500ms of delay and them mux it with mkvmerge?
rickardk
28th January 2008, 16:56
Could i re-process the audio with eac3to and apply 500ms of delay and them mux it with mkvmerge?
I think you should cut/remove 500ms from the the audio.
Don't know exactly what skip2 means in terms of delay. But there is a 133ms delay in the evo.
umaximus
28th January 2008, 16:59
yes, my bad, i meant to apply -500ms (''--500ms'') with eac3to and them mux it together in mkvmerge..
nautilus7
28th January 2008, 19:18
If it's -500ms that is needed, then you have to put "-500ms" (not --500ms) as the original delay of +133ms is only applied when the -skip option isn't used (unless i missed something).
madshi
28th January 2008, 19:32
Byte for byte they were the exact same size, the same goes for time. The funny thing is, I went and converted several tracks to flac (the same evo tracks converted to ac3) and different titles had the same error. I only converted to flac this round with 2.20, will try one of those two with 2.19 later today.
Hmmmm... I've an idea what could be going wrong. Will check that later...
umaximus
28th January 2008, 19:33
If it's -500ms that is needed, then you have to put "-500ms" (not --500ms) as the original delay of +133ms is only applied when the -skip option isn't used (unless i missed something).
ay, i thought that '-' is the 'base' symbol for any operation eac3to is working with.. in case of Million Dolla Baby, you cannot exctract audio with eac3to in the first place, since audio is also damaged at the begining..
delaycut is showing last unsynchronized frame at 624ms, could that be the 'perfect' delay time for it?
====== PROCESSING LOG ======================
Time 00:00:00.512; Frame#= 33. Crc error IGNORED: read = 21F3; calculated=4112
Time 00:00:00.528; Frame#= 34. Unsynchronized frame...REWINDED 558 bytes. Found new synch word
Time 00:00:00.544; Frame#= 35. Some basic parameters changed between Frame #41 and this frame
Time 00:00:00.544; Frame#= 35. Crc error IGNORED: read = B80B; calculated=86D4
Time 00:00:00.560; Frame#= 36. Unsynchronized frame...SKIPPED 1310 bytes. Found new synch word
Time 00:00:00.576; Frame#= 37. Some basic parameters changed between Frame #35 and this frame
Time 00:00:00.576; Frame#= 37. Crc error IGNORED: read = A5FD; calculated=76BC
Time 00:00:00.592; Frame#= 38. Unsynchronized frame...SKIPPED 16107 bytes. Found new synch word
Time 00:00:00.608; Frame#= 39. Some basic parameters changed between Frame #37 and this frame
Time 00:00:00.608; Frame#= 39. Crc error IGNORED: read = E6F1; calculated=F5ED
Time 00:00:00.624; Frame#= 40. Unsynchronized frame...SKIPPED 3819 bytes. Found new synch word
Time 00:00:00.640; Frame#= 41. Some basic parameters changed between Frame #39 and this frame
Number of written frames = 497182
Number of Errors= 12
madshi
28th January 2008, 19:33
i am getting a picture, the sound is perfect but it stutters, im gonna try to use the new eac3to and ill let you know, i do have POTO US Edition, and ill also try Harry Potter order of the Pheonix cause i couldnt demux that with EVODemux ok.
Make sure that it's not your playback system (e.g. the decoder DirectShow filter you're using, or a too slow CPU) which is responsible for the stuttering. Not every stuttering needs to be caused by a bad remux.
madshi
28th January 2008, 19:40
Ah, for The Searchers I use both libav and sonic with eac3to 2.20 and earlier versions. Sonic creates
an ac3 track with clicks. However libav creates a clean ac3 track.
For Robin Hood, eac3to 2.20 and earlier, both sonic and libav gives me an ac3 track with clicks. I finally used the DVD audiotrack to mux into the mkv.
With the Robin Hook sample you sent me decoding seems to work just fine for me with both Nero and libav. Can't hear any clicks. Don't know why it doesn't work for you.
Regarding The Searchers, I don't know whether the file is really corrupt. As you already mentioned the eac3 track is okay and as I noticed, PDVD plays the file back fine.
My current guess is that PDVD is just more error tolerant. Meaning that a bad video frame is just skipped and doesn't impact the decoder. In contrast the MS VC-1 decoder when being fed from our usual DirectShow filters seem to go totally insane when there's a damaged frame somewhere in the stream. This might be the difference. Maybe there's really one bad frame and PDVD just breezes through it, hiding the problem well.
Okay, how do I recognize frame boundaries in a hex editor. That would make life easier.
With VC-1 look for "00 00 01 0F" which is the beginning of a sequence header. Inside an EVO file there should be a "00 00 01 BA" some bytes before the sequence header, which is the EVO "pack header" structure. The "00 00 01 BA" should being on a multiply of 0x800.
I read you made your own splitter. I don't know anything of the complexity of splitters and direct show, but could this be made into a DS version? In that case VC1 evo could be played directly in ZP or MPC I guess. However, I don't know what I'm asking for in terms of time, so read the request with a grain of salt :)
Just writing a DS splitter itself is not that much of a problem. I could probably do that. But there's more than just splitting. E.g. you need to support pausing, seeking and all that kind of stuff. That's the main problem I see in writing a splitter. And that's why I'm not keen on doing that. I prefer to work no creating perfect MKV files.
And do you ever sleep? I seldom saw such a devotion to make something work. Thanks a lot for that :thanks:
Ah well, I'm *trying* to limit my eac3to programming time on Sundays cause I cannot really afford to work on it on other days because otherwise it affects my real life job. Unfortunately sometimes I can't keep myself from working on eac3to on other days, too. But I shouldn't really do that........
madshi
28th January 2008, 19:44
how much frames does 'one skip' cut?
"one skip" cuts one sequence header. There's no constant number of frames per sequence header. What I've seen it can be anything between 5-20 frames, with the average maybe being something like 8 frames. Not really sure, though. Didn't pay much attention to that. I believe there are also some movies where sequence headers are used less often. So in such cases the "one skip" might eat a lot of frames.
Any good app out there, where could u import mkv video & audio in as seperates and then looking for lip sync on screen to find out correct delay needed?
Best way would probably be to mux the audio track into the MKV and then use MPC for delay finding. With MPC you can use the [-] and [+] keys in the numpad of the keyboard to adjust audio delay in realtime during playback. Helps a lot in finding the right delay.
madshi
28th January 2008, 19:46
delaycut is showing last unsynchronized frame at 624ms, could that be the 'perfect' delay time for it?
That's seemingly the end of the corruption. It could match the end of the first two video sequence headers - or not.
Richum
28th January 2008, 21:09
D:\EAC3>eac3to F:\FEATURE_1.EVO+F:\FEATURE_2.EVO 4: D:\MKV\cream.dts -1536
EVO, 1 video track, 2 audio tracks, 2:05:42
1: Joined EVO file
2: VC-1, 1080p24 /1.001
3: RAW/PCM, 2.0 channels, 16 bits, 48khz
4: DTS Master Audio, 5.1 channels, 24 bits, 48khz, dialnorm: -4dB
Extracting audio track number 4...
Removing dialog normalization...
Decoding with DirectShow (Sonic Audio Decoder)...
DirectShow reports 5.1 channels, 24 bits, 48khz
Writing WAVs...
Creating/writing file "D:\MKV\cream.C.wav"...
Creating/writing file "D:\MKV\cream.L.wav"...
Creating/writing file "D:\MKV\cream.LFE.wav"...
Creating/writing file "D:\MKV\cream.R.wav"...
Creating/writing file "D:\MKV\cream.SR.wav"...
Creating/writing file "D:\MKV\cream.SL.wav"...
Found Surcode DTS Encoder version 1.0.21.0.
Surcode sais/asks: "At least one valid source file must be specified to encode.".
Pressing the Surcode "Encode" button didn't seem to work...
Closing Surcode...
Why does this fail? Surcode sais/asks:
All the wav files are made but are immediately deleted by the program after this error is issued. Should the sais be says?
rory83
28th January 2008, 21:31
To date, I've had no success creating a working file from my US Phantom. Eac3to, current version, produces an mkv but all the players that I have on the PC cause the usual corruption beginning after the Warner title, basically I think because of the Haali splitter destroying Madshi's good work. I don't have a method that doesn't appear to use the Haali splitter to play back mkvs. No errors are reported during eac3to's processing.
I tried taking the first 100MB of the Phantom mkv to my Popcornhour, to see how that got on with it. It's running their current beta, which does have some mkv support but that can't handle the mkv either, the screen stays blank though the network's flashing away while the play is occurring.
My other HDDVD, American Werewolf, works well using the Evodemux method, and then into VirtualDub to produce an AVI.
It would be nice if, as others have suggested, Madshi might consider a freestanding splitter. At present it would appear that Phantom is the only disk with the problem that Madshi's discovered. But I assume it may not be the last.
So, I'm just going to have to buy a super long HDMI cable so I can watch Phantom on the TV, from the HDDVD on my PC in another room!
This is a link to the first 100MB of the EVO files, which play fine in PDDVD 3319a.
http://www.sendspace.com/file/6soz6c
rickardk
28th January 2008, 22:30
Just watched Million Dollar Baby. Strange result.
From time to time it plays in slow motion. It looks like when you have the refresh rate set to 23.976 and tries to play 25 fps material.
If I press pause/play everything is fine again.
If I don't pause/play it will continue to stutter until scene change or massive motion in a scene. That solves the stuttering.
I can not reproduce this problem at the same place. It shows up from time to time. And when I rewind to watch the same scene its not there.
I will try to rewrite timestamps to see if it solves it.
Should also try to remux (with this new method in v2.20) a title that played back perfect before (Bourne Ultimatum).
nautilus7
28th January 2008, 23:09
If it's not in the same place every time, it might be because of your system/setup.
Thunderbolt8
28th January 2008, 23:15
when tried to remux the video ofarmy of shadow but got an authoring problem message:
"this evo file contains authoring faults. will try to work around that."
since this message came directly at the beginning of the remux process, does this also mean that posssible problems are at the beginning of the movie. I would like to check whether there will be any sync or other problems at that certain time where those problems take place. or would I have to watch the whole movie to be sure everything is ok?
madshi
28th January 2008, 23:17
Surcode sais/asks: "At least one valid source file must be specified to encode."
Why does this fail?
Don't know. Try "eac3to source.evo 4: D:\MKV\cream.wavs" and then start Surcode manually. Maybe your Surcode installation is bad?
madshi
28th January 2008, 23:23
To date, I've had no success creating a working file from my US Phantom.
[...]
This is a link to the first 100MB of the EVO files, which play fine in PDDVD 3319a.
Thanks for the sample!
Do you really have corruption with this sample? I'm asking because I haven't. It works perfectly fine for me. Please run "eac3to" without any parameters to make sure that you have the right version. Does it really say "eac3to v2.20"? My best guess is that your browser cache gave you an old eac3to version.
I tried taking the first 100MB of the Phantom mkv to my Popcornhour, to see how that got on with it. It's running their current beta, which does have some mkv support but that can't handle the mkv either
AFAIK the Popcornhour doesn't support MKVs with VC-1 in it. Or was that feature recently added?
It would be nice if, as others have suggested, Madshi might consider a freestanding splitter.
Nope, sorry.
madshi
28th January 2008, 23:27
Just watched Million Dollar Baby. Strange result.
From time to time it plays in slow motion. It looks like when you have the refresh rate set to 23.976 and tries to play 25 fps material.
If I press pause/play everything is fine again.
If I don't pause/play it will continue to stutter until scene change or massive motion in a scene. That solves the stuttering.
I can not reproduce this problem at the same place. It shows up from time to time. And when I rewind to watch the same scene its not there.
Hmmmm... That sounds bad... :(
Which decoders are you using? I've heard that the libav VC-1 decoder sometimes shows these effects. Try the MS VC-1 decoder (or at least the "wmv9" option in ffdshow). Also alternatively try removing ffdshow as a test.
I will try to rewrite timestamps to see if it solves it.
Should also try to remux (with this new method in v2.20) a title that played back perfect before (Bourne Ultimatum).
Yeah, those would be helpful tests. Although I don't think that timestamp rewriting will help. After all, if the problem goes away by rewinding the timestamps can't be guilty since they don't change by rewinding.
Have you tried playing the original MKV file without the audio track in it? Maybe muxing the audio track to the video introduced these problems? Hopefully not...
Richum
28th January 2008, 23:27
Don't know. Try "eac3to source.evo 4: D:\MKV\cream.wavs" and then start Surcode manually. Maybe your Surcode installation is bad?
Will do.
madshi
28th January 2008, 23:29
when tried to remux the video ofarmy of shadow but got an authoring problem message:
"this evo file contains authoring faults. will try to work around that."
since this message came directly at the beginning of the remux process, does this also mean that posssible problems are at the beginning of the movie. I would like to check whether there will be any sync or other problems at that certain time where those problems take place. or would I have to watch the whole movie to be sure everything is ok?
This message is just a hint that the EVO is not as perfect as it should be. As the message sais: eac3to tries to work around it. It's still quite possible that the remuxed MKV plays perfectly. I can't say whether titles with such authoring defects will show real problems during playback or not. I hope that they will work just fine. But we'll only find out by playing the whole movies, I guess...
Thunderbolt8
28th January 2008, 23:47
since I only seem to be able to demux all tracks of the file and I want to demux the dts-hd ma track only in order to store it for later, I'd rather do that with evodemux to save time. however, regarding that possible evo problem, would there be a difference if I make a flac file from that demuxed file, compared to a flac, which I would make with eac3to now? so could it be that there will be audio problems with that evodemux flac and none in the eac3to flac?
madshi
28th January 2008, 23:50
since I only seem to be able to demux all tracks of the file
Why is that?
I want to demux the dts-hd ma track only in order to store it for later, I'd rather do that with evodemux to save time.
Why? Is EvoDemux faster than eac3to?
however, regarding that possible evo problem, would there be a difference if I make a flac file from that demuxed file, compared to a flac, which I would make with eac3to now? so could it be that there will be audio problems with that evodemux flac and none in the eac3to flac?
Not sure. I don't know how EvoDemux handles such EVO files.
Thunderbolt8
29th January 2008, 00:00
Why is that?
Why? Is EvoDemux faster than eac3to?
Not sure. I don't know how EvoDemux handles such EVO files.
is there a possibility only to demux 1 specific track with eac3to? otherwise evodemux with only having to demux 1 track is of course faster than eac3to having to demux video + all audio tracks.
rory83
29th January 2008, 00:03
Thanks for the sample!
Do you really have corruption with this sample? I'm asking because I haven't. It works perfectly fine for me. Please run "eac3to" without any parameters to make sure that you have the right version. Does it really say "eac3to v2.20"? My best guess is that your browser cache gave you an old eac3to version.
Yes, definitely the right one, it's in the PC MKV playback things are going astray. The ripped EVOs are fine and so is a VC1 from eac3to.
Actually there has been a development tonight! I've now been pointed at the registry hack for the Sonic decoder that allows it to produce an Avisynth compatible input. So I'm experimenting with using the vc1 from eac3to to feed into Vdub, and make an AVI with divx 6.8. So far this is going with no corruption visible. I'm not actually bothered by PC playback, I want to use the PCH for everything, and the Divx AVI I did of American Werewolf is fine.
Thanks for your work!
Yraen
29th January 2008, 00:04
is there a possibility only to demux 1 specific track with eac3to? otherwise evodemux with only having to demux 1 track is of course faster than eac3to having to demux video + all audio tracks.
eac3to video.evo
That will give you all the tracks in the evo. Once you have it just use:
eac3to video.evo 2: audio.dts
That will give you the one file. Make sure you use the number (2: ) that corresponds to your track.
nautilus7
29th January 2008, 00:06
is there a possibility only to demux 1 specific track with eac3to? otherwise evodemux with only having to demux 1 track is of course faster than eac3to having to demux video + all audio tracks.
Of course you demux one track with eac3to.
eac3to 1st.evo+2nd.evo #: audio.dtshd
EDIT: Yraen was 2 minutes faster. Damn! :p
Thunderbolt8
29th January 2008, 00:18
thanks!
nautilus7
29th January 2008, 00:21
Madshi, i have a little request. When using the command
eac3to movie.evo movie.mkv
the demuxed audio tracks are named audiotrack 1, audiotrack 2, etc. I think it would be nice to change that to something responding to the name of the .mkv file (movie in this case). This way an accidental overwriting might be prevented.
umaximus
29th January 2008, 04:20
Regarding rainbow frames in some titles, like Babel for instance.
I was doing some testing of remuxed video of Babel with ffdshow & CoreAVC, and while ffdshow is showing rainbow frames on 56min22sec of movie (mexican wedding), CoreAVC is playing that with no corruption at all.
So i think its save to say that the fault is on ffdshow and not eac3to remuxing reagarding rainbow frames..
Is is possible to cut out that part, where ffdshow is picking up (or dropping) those bad frames, and check if there is some change in additional records data in header?
rickardk
29th January 2008, 04:37
I have just been watching Bourne Ultimatum and Mission Impossible 2. No problems at all. Remuxed Million Dollar Baby and gave it a 30 minutes look. No problems.
The old remux still stutters as described in earlier post. No idea why. Anyway it's safe to say that VC-1 muxing works fine now. This title must have some authoring faults.
It looks like there are "problems" with the first wave (april/may 2006) of Warner titles.
POTO (rainbow frames?)
The Last Samurai (gave micro stutters at some places before latest eac3to)
Million Dollar Baby (rainbow frames, strange stutter)
Unforgiven (rainbow frames)
**The Fugetive (Blu-ray. It will rip but m2ts will not play. Some have been able to demux it though)
Fortunately eac3to can handle them now!!
Joniii
29th January 2008, 06:32
I'm tried to encode Blu-ray 5.1 LPCM -> DTS for the first time. All I get in the DTS is this high pitching data sound. What am I doing wrong?
eac3to dracula.lpcm f:\dracula.dts
rickardk
29th January 2008, 07:08
Demuxed a TrueHD track from m2ts. When converting to flac eac3to says "Applying RAW/PCM delay..."
Does the TrueHD track contain info on alignment in m2ts or why does eac3to temper with the delay?
madshi
29th January 2008, 08:21
Yes, definitely the right one, it's in the PC MKV playback things are going astray.
Why does it work for me and for intomed? That are at least 2 people who have their POTO MKV play just fine. Maybe there's some problem with your DirectShow filters? E.g. make sure you have the latest Haali filters installed etc...
madshi
29th January 2008, 08:24
Regarding rainbow frames in some titles, like Babel for instance.
I was doing some testing of remuxed video of Babel with ffdshow & CoreAVC, and while ffdshow is showing rainbow frames on 56min22sec of movie (mexican wedding), CoreAVC is playing that with no corruption at all.
So i think its save to say that the fault is on ffdshow and not eac3to remuxing reagarding rainbow frames..
Is is possible to cut out that part, where ffdshow is picking up (or dropping) those bad frames, and check if there is some change in additional records data in header?
Samples are always welcome. I'm not working on h264 right now, though. Still if you have an interesting sample, you can always upload it for me. Please EVO/m2ts samples, though, no MKV or demuxed samples. You can cut EVO files with a hex editor (EVOs consist of blocks of 2048 bytes, so always cut on 2048 byte boundaries).
madshi
29th January 2008, 08:26
I have just been watching Bourne Ultimatum and Mission Impossible 2. No problems at all. Remuxed Million Dollar Baby and gave it a 30 minutes look. No problems.
The old remux still stutters as described in earlier post. No idea why. Anyway it's safe to say that VC-1 muxing works fine now. This title must have some authoring faults.
It looks like there are "problems" with the first wave (april/may 2006) of Warner titles.
POTO (rainbow frames?)
The Last Samurai (gave micro stutters at some places before latest eac3to)
Million Dollar Baby (rainbow frames, strange stutter)
Unforgiven (rainbow frames)
**The Fugetive (Blu-ray. It will rip but m2ts will not play. Some have been able to demux it though)
Fortunately eac3to can handle them now!!
Oh, so Million Dollar Baby is from the first wave, too? Wasn't aware of that. Yes, it does make sense that all the problems are in the first wave of titles!
So the 2nd try Million Dollar Baby remux seems to work fine? Have you already muxed the audio track into the MKV? I'm just wondering whether doing that damaged the 1st try remux? No idea...
madshi
29th January 2008, 08:27
Madshi, i have a little request. When using the command
eac3to movie.evo movie.mkv
the demuxed audio tracks are named audiotrack 1, audiotrack 2, etc. I think it would be nice to change that to something responding to the name of the .mkv file (movie in this case). This way an accidental overwriting might be prevented.
Yes, that makes sense...
madshi
29th January 2008, 08:28
I'm tried to encode Blu-ray 5.1 LPCM -> DTS for the first time. All I get in the DTS is this high pitching data sound. What am I doing wrong?
eac3to dracula.lpcm f:\dracula.dts
Please post the full eac3to processing log (see "log.txt" in "eac3to.exe" folder).
Demuxed a TrueHD track from m2ts. When converting to flac eac3to says "Applying RAW/PCM delay..."
Does the TrueHD track contain info on alignment in m2ts or why does eac3to temper with the delay?
Please post the full eac3to processing log. Thanks.
nautilus7
29th January 2008, 09:31
Why does it work for me and for intomed? That are at least 2 people who have their POTO MKV play just fine. Maybe there's some problem with your DirectShow filters? E.g. make sure you have the latest Haali filters installed etc...
The POTO sample works for me too!
madshi
29th January 2008, 09:51
The POTO sample works for me too!
Thanks for testing. So it seems that rory83 is the only one having trouble with POTO sample.
rory83
29th January 2008, 10:13
Why does it work for me and for intomed? That are at least 2 people who have their POTO MKV play just fine. Maybe there's some problem with your DirectShow filters? E.g. make sure you have the latest Haali filters installed etc...
Yes, that's what I think, but since it's playback via the PCH that I'm after, it doesn't worry me. I built the video AVI last night from the vc1 created using eac3to, using the sonic splitter and decoder, and this AM muxed in the ac3 track from eac3to. The result is excellent and split at 2GB points so I can run it from my NAS. So with the assistance of the good folks here and on the Evodemux thread, all has been accomplished :):):)
If only there was another HD/BD title I was interested in owning!
nautilus7
29th January 2008, 13:16
Madshi, i got some interesting samples!!! :)
It's a 7.1 ch Dolby Digital Plus track (http://www.sendspace.com/file/ymyvlc) and a 7.1 ch Dolby TrueHD track (http://www.sendspace.com/file/vsywrm), from The Sound of High Definition Dolby demo HD DVD.
First of all, the evos are not cut. They are complete. eac3to reports quite strange info for the audio tracks. I tried to demux the audio in order to upload just it, but eac3to and evodemux gave different files! Decoding them with nero, i got a working output but i cannot (at the moment) test whether all channels are there. Libav decoder crashed for both samples.
Hope they are really good samples and help developing. Please forward them to libav developers.
madshi
29th January 2008, 14:03
GREAT samples, thank you!! :)
I'll forward them to the libav programmers.
Kal
29th January 2008, 14:15
Hello,
Sorry to ask this question here but how can I extract ac3 audio channel from a m2ts file (blu-ray) ? Does eac3to support this task?
Kal
vBulletin® v3.8.11, Copyright ©2000-2025, vBulletin Solutions Inc.