eac3to - audio conversion tool

[mrr19121970]

Originally Posted by madshi

I don't know. Why do people keep asking me questions about tsMuxeR?

We're hoping that one day we can do something like this:

eac3to.exe d:\ 1) 2: movie.h264 3: english.ac3 --> movie.m2ts

:beg: :beg:

[piratburner]

Originally Posted by mrr19121970

We're hoping that one day we can do something like this:

eac3to.exe d:\ 1) 2: movie.h264 3: english.ac3 --> movie.m2ts

:beg: :beg:

That's should be something

[Thunderbolt8]

just ask the tsmuxer guys to fix their bugs and you should be fine

edit: yeah, and ask them to implement lossless -> ac3 & dts decoding :P

[alc0re]

Originally Posted by madshi

Dialnorm can be set to any value between 0 and 31. According to the AC3 specification both 0 and 31 means: No dialnorm processing. Now any dialnorm processing *lowers* the volume of the audio track. That means removing the dialnorm (which is what eac3to is doing) should result in *higher* volume. Currently eac3to sets dialnorm to 0. Unfortunetely dialnorm set to 1 means lowering volume a lot. So incorrectly working decoders might think that a dialnorm value of 0 means even lower volume than dialnorm 1. But the documentation clearly states that a dialnorm value of 0 shall be treated as "no dialnorm processing" (which means max volume). And all the PC AC3 decoders correctly see value 0 as "dialnorm processing deactivated".


IMHO the decoders in the PS3 and Panasonic are not working correctly. Or maybe my AC3 specification is outdated? Anyway, the documentation clearly says that dialnorm 0 is "reserved". So I think it's not really good that eac3to uses it. That means I'll change it to 31 in the next build. I think that should fix the problem you're seeing. However, I believe to remember that some Sony Blu-Rays had a dialnorm value of 0, too. Well, anyway...

So is there a command in eac3to to do the reverse that the dialog normalization does? As in re-apply the normalization? Say an audio track had -27 dialog normalization. If I process the .ac3 track with +27db switch (that's already had the dialog normalization removed) will that reverse the process?

Does dialog normalization only decrease the volume of the center channel? Or is it a flat volume decrease across all channels? I ask because then the +db switch might work to accomplish my goal.

Let's say you tell me the +db command will accomplish exactly what I'm trying to accomplish...do I need any of the 3rd party encoders? Because up to this point I have not needed them since I'm just extracting the audio not converting it. I'm assuming there's no conversion done with you apply a +/- audio gain, but I may be wrong.

[n0mag!c]

Originally Posted by madshi

Dialnorm can be set to any value between 0 and 31. According to the AC3 specification both 0 and 31 means: No dialnorm processing. Now any dialnorm processing *lowers* the volume of the audio track. That means removing the dialnorm (which is what eac3to is doing) should result in *higher* volume. Currently eac3to sets dialnorm to 0. Unfortunetely dialnorm set to 1 means lowering volume a lot. So incorrectly working decoders might think that a dialnorm value of 0 means even lower volume than dialnorm 1. But the documentation clearly states that a dialnorm value of 0 shall be treated as "no dialnorm processing" (which means max volume). And all the PC AC3 decoders correctly see value 0 as "dialnorm processing deactivated".


IMHO the decoders in the PS3 and Panasonic are not working correctly. Or maybe my AC3 specification is outdated? Anyway, the documentation clearly says that dialnorm 0 is "reserved". So I think it's not really good that eac3to uses it. That means I'll change it to 31 in the next build. I think that should fix the problem you're seeing. However, I believe to remember that some Sony Blu-Rays had a dialnorm value of 0, too. Well, anyway...

Please don't be hurry to change it while only one man states the issue!
I'm using PS3 and frequently removing dialog normalization with ea3to with no such issue.

alc0re
It's must certainly be your setup.
Please check up your system options in player/receiver which affecting level/compression - "dynamic range compression", "night mode".
Does your player transmit bitstream to amplifier (via HDMI? optic?) Or player decodes stream itself?
I guess the first, because when you're 1.5 FFing, player switches transmition from bitstream to LPCM.

[Atak_Snajpera]

Wake up man! Download 2.84 !!!

[jonathonsunshine]

Howdy, I just went up from 2.6 (I think) to the latest version (as of Dec 22nd, 2008) 2.84 and it doesn't seem to be correcting gaps/overlays in truehd streams anymore.

It will fix them in a standard AC3 stream but not in the TrueHD stream. It doesn't even attempt to do its "2nd pass", it just acts as if there was nothing wrong in the 1st place and that all had gone perfectly.

Does anyone know where I can get older versions of eac3to ? Is this a known problem now ? What is the latest version that doesn't have this bug ? Or am I simply stuffing something up ?

Also, Atak_Snajpera, in the post immediatly before this one, is referring to this post, cept that I deleted it and reposted because I am indeed already using 2.84 (I posted 2.83 previously)

[Easy123]

I think I got a bug with v2.84.

I have a DTS Audio track here which is about 1h 59min long and it´s 23,976fps. I want to transcode it to PAL so I use the following command: eac3to "source.dts" "taget.wavs" -speedup -2pass . 2pass because it detects clipping. At the End the Soundtrack is about 1h 50min long, but it should be about 1h 54min in Length. Can it be that during the second pass, the framerate conversion gets aplied for a second time? Meaning eac3to speeds up the 1h 54min (which is correct Runtime) for a second time?

DTS, 5.1 channels, 1:59:41, 24 bits, 768kbps, 48khz
Decoding with ArcSoft DTS Decoder...
Changing FPS from 23.976 to 25.000...
Writing WAV...
Creating file "F:\HB2-deutsch.wavs.pass1.wav"...
Clipping detected, a 2nd pass will be necessary.
Caution: The WAV file is bigger than 4GB.
Some WAV readers might not be able to handle this file correctly.
Starting 2nd pass...
Reading WAV...
Changing FPS from 23.976 to 25.000...
Reducing depth from 64 to 24 bits...
Writing WAVs...
Creating file "F:\HB2-deutsch.SL.wav"...
Creating file "F:\HB2-deutsch.L.wav"...
Creating file "F:\HB2-deutsch.C.wav"...
Creating file "F:\HB2-deutsch.R.wav"...
Creating file "F:\HB2-deutsch.LFE.wav"...
Creating file "F:\HB2-deutsch.SR.wav"...
The original audio track has a constant bit depth of 64 bits.
The processed audio track has a constant bit depth of 24 bits.
eac3to processing took 50 minutes, 22 seconds.
Done.

[73ChargerFan]

Originally Posted by alc0re

I ask because when I extract an ac3 (dolby digital) track with dialog normalization, and let eac3to remove the dialog normalization, when I play my final encoded avchd structured dvd9 in either my bluray player or my PS3 the audio is really really low and I have to crank up my receiver's volume almost all the way up to hear anything.

I have the same issue, with MPC-HC (all versions for the past year.) I set my Pioneer VSX-1015TX to a volume of about -20 for almost everything, including HD-DVDs & music from foobar & windows sounds with volume level half way up. Standard def videos and dvrms files also play fine at that volume.

For remuxed blu-ray videos , I have to boost it to at least -15, and often to -5. (dts & ac3 bitstreamed over spdif.)

My understanding is that dialnorm is signal to the receiver to lower the volume. So, removing it would make it louder. But all my bd derived videos are louder.

I don't get it. Do audio drivers / media players / audio codecs ever add a dialnorm signal where there isn't one?

[wolfbane5]

I recently started using eac3to but I have a question. Everything works great for my purposes, however eac3to thinks that surcode and mkvtoolnix aren't installed when they actually are. It somehow found that haali was installed (don't know how it found that). I'm wondering if there's a way to tell eac3to that surcode and toolnix are installed? Right now, once I have the mono .wav streams, I have to physically load them all into surcode, create the dts and then go back to eac3to to remove the padding from the dts. If eac3to knows where surcode is, then I can bypass the extra work and get an unpadded dts easily.

[alc0re]

Originally Posted by n0mag!c

Please don't be hurry to change it while only one man states the issue!
I'm using PS3 and frequently removing dialog normalization with ea3to with no such issue.

alc0re
It's must certainly be your setup.
Please check up your system options in player/receiver which affecting level/compression - "dynamic range compression", "night mode".
Does your player transmit bitstream to amplifier (via HDMI? optic?) Or player decodes stream itself?
I guess the first, because when you're 1.5 FFing, player switches transmition from bitstream to LPCM.

n0magic :

While I appreciate your feedback, I'm not the only one with the issue. 73ChargerFan posted after you that he is having the same issue. If madshi changes the code to make it 31 instead of 0 and that fixes the issue, it shouldn't effect those of you that are not seeing the issue, since as he stated both a level of 0 and 31 are "supposed" to mean no dialog normalization. So honestly I'm not sure why you are opposed to him changing that. Besides, he stated that perhaps using 31 is better since 0 is reserved. Perhaps some receivers process this correctly and others don't since 0 is supposed to be reserved.

Like 73ChargerFan, I am not seeing any issues on any other type of audio thrown at my receiver from any other source other than ac3 audio processed with eac3to with dialog normalization removed. My receiver's settings are fine, I've triple checked them. Both my PS3 and Bluray player are bitstreaming to my receiver via optical.

madshi I'd appreciate if I could still get an answer to my question about reversing the dialog normalization using -/+db...(not rushing...just making sure you see my previous reply/question to your reply)

[tebasuna51]

Originally Posted by alc0re

madshi I'd appreciate if I could still get an answer to my question about reversing the dialog normalization using -/+db...(not rushing...just making sure you see my previous reply/question to your reply)

No, you can't use this method.
You need wait until madshi change the DialNorm to 31 (-31 dB, instead 0) in next release.

[Thunderbolt8]

madshi, what exactly does "Detected PTS break, increasing PTS by 33.4ms..." mean? that the video stream gets speed up at that position by 33.4ms? Im asking because I had one example now in which I try to sync an external audio track, so this means for that track that I have to delay the track at this position by -32ms (1 ac3 frame)? if so, then it would be useful if eac3to could output the exact timestamp of that action, otherwise it can only be estimated by watching the progress bar all the time eac3to is processing that stream.

[williewonton]

madshi

The problem reported previously appears to be that eac3to borks the file. By using TsmuxeR in the first instance to remux to m2ts (from the playlist and dropping unwanted streams) a playable file is created. Using Eac3to to demux that file and then remux in TsmuxeR generates a playable file. So perhaps Eac3to is not handling the extraction of data from the original Blu-ray structure somehow (only my guess)

[yfed]

Hello,
I need help obtaining 24bit/192kHz 5.1 FLAC file from BD of Trondheim Soloists "Divertimenti"

Here's my cmd:

C:\!tools\audio\eac3to>eac3to "J:\_LINDBERG\BDMV\STREAM\00015.m2ts" 1:"G:\chapters.txt" 3:"G:\multi.flac"
M2TS, 1 video track, 2 audio tracks, 1:08:58
1: Chapters, 11 chapters
2: MPEG2, 1080p24 (16:9)
3: RAW/PCM, English, 5.1 channels, 24 bits, 192khz
4: RAW/PCM, English, 2.0 channels, 24 bits, 192khz
Creating file "G:\chapters.txt"...
[a03] Extracting audio track number 3...
[a03] Reading RAW/PCM...
[a03] Swapping endian...
[a03] Remapping channels...
[a03] Encoding FLAC with libFlac...
[a03] Creating file "G:\multi.flac"...
[a03] The original audio track has a constant bit depth of 24 bits.
Video track 2 contains 99312 frames.
eac3to processing took 45 minutes, 21 seconds.
Done.

However, the FLAC file is corrupted, when I try to seek towards the middle of it in foobar2000 - it says "unable to seek.. the file is corrupted". Maybe there are some FileSystem / FLAC limitations that eac3to ignores? Please, help

PS
flac 5.1 size: 6.46Gb
pcm 5.1 size: 13.3Gb

[madshi]

Originally Posted by n0mag!c

Please don't be hurry to change it while only one man states the issue!

Don't worry. The change I'm planning to do should have no effect at all on proper decoders. And it might help fix bad decoders.

Originally Posted by jonathonsunshine

Howdy, I just went up from 2.6 (I think) to the latest version (as of Dec 22nd, 2008) 2.84 and it doesn't seem to be correcting gaps/overlays in truehd streams anymore.

It will fix them in a standard AC3 stream but not in the TrueHD stream. It doesn't even attempt to do its "2nd pass", it just acts as if there was nothing wrong in the 1st place and that all had gone perfectly.

eac3to was never able to fix overlaps/gaps in TrueHD streams, if you just demux them to "*.thd" or "*.thd+ac3". However, eac3to can fix the overlaps/gaps if you transcode to another format (e.g. WAV or FLAC).

If you think there's a problem with eac3to v2.84, please post your full eac3to log here (in [ code] [ / code] blocks, please, without the spaces).

Originally Posted by Easy123

I think I got a bug with v2.84.

I have a DTS Audio track here which is about 1h 59min long and it´s 23,976fps. I want to transcode it to PAL so I use the following command: eac3to "source.dts" "taget.wavs" -speedup -2pass . 2pass because it detects clipping. At the End the Soundtrack is about 1h 50min long, but it should be about 1h 54min in Length. Can it be that during the second pass, the framerate conversion gets aplied for a second time? Meaning eac3to speeds up the 1h 54min (which is correct Runtime) for a second time?

Yes, your suspicion seems to be true. I think this only occurs when using the "-2pass" option, though. Just try without it and everything should probably be fine. You never have to use the "-2pass" option, eac3to will do 2 passes in any case where it's necessary, regardless of whether you used the "-2pass" option or not. The only sense of that option is to tell eac3to that probably 2 passes are necessary. In that situation eac3to does some things differently to speed up the processing a bit. It's only a performance optimization for situations where 2 passes are necessary. But it's never necessary to use that option...

Originally Posted by Thunderbolt8

madshi, what exactly does "Detected PTS break, increasing PTS by 33.4ms..." mean? that the video stream gets speed up at that position by 33.4ms? Im asking because I had one example now in which I try to sync an external audio track, so this means for that track that I have to delay the track at this position by -32ms (1 ac3 frame)? if so, then it would be useful if eac3to could output the exact timestamp of that action, otherwise it can only be estimated by watching the progress bar all the time eac3to is processing that stream.

There's nothing you have to do. This "detected PTS break" simply means that the timestamps in the source file are not continuous, but there's a jump in the timestamps somewhere. Now whenever there's such a jump in the timestamps, eac3to of course has to handle that somehow. It does so by assuming that the first video frame after the timestamp jump is supposed to be played exactly one frame after the last video frame before the timestamp jump. That's all this log message says. So again, you don't have to do anything. However, this message indicates that either the source is corrupt or that it was cut (e.g. to remove advertising). It's quite possible that you won't notice the problem and everything is just fine.

Hmmmm... You're right, though, in that I should list the runtime at which the problem occurred. At least you can then check that runtime to see whether there are any specific problems around this runtime. So I'll add that to the log output...

Originally Posted by wolfbane5

I recently started using eac3to but I have a question. Everything works great for my purposes, however eac3to thinks that surcode and mkvtoolnix aren't installed when they actually are. It somehow found that haali was installed (don't know how it found that). I'm wondering if there's a way to tell eac3to that surcode and toolnix are installed?

eac3to gets the mkvtoolnix path from one of these registry values:

HKEY_CURRENT_USER\Software\mkvmergeGUI\GUI\installation_path
HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Uninstall\MKVtoolnix\UninstallString
HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Uninstall\MKVtoolnix\DisplayIcon

If you have properly installed mkvtoolnix, all of these should be set. Surcode is located by checking these registry values:

HKEY_LOCAL_MACHINE\Software\Minnetonka Audio Software\SurCode DVD-DTS\Home
HKEY_LOCAL_MACHINE\Software\Minnetonka Audio Software\SurCode DVD DTS\Home

Again, if you have properly installed Surcode, these registry values should be set.

Originally Posted by williewonton

The problem reported previously appears to be that eac3to borks the file. By using TsmuxeR in the first instance to remux to m2ts (from the playlist and dropping unwanted streams) a playable file is created. Using Eac3to to demux that file and then remux in TsmuxeR generates a playable file. So perhaps Eac3to is not handling the extraction of data from the original Blu-ray structure somehow (only my guess)

I've checked back. Please read these posts:

http://forum.doom9.org/showpost.php?...postcount=5854
http://forum.doom9.org/showpost.php?...postcount=5954
http://forum.doom9.org/showpost.php?...postcount=5956

So, if you have the same problem, according to the posts above it's not caused by eac3to. Just try a different playlist. If you still think eac3to is responsible for the problem, then please manually play the separate m2ts parts and check whether audio is alright if you watch them separately. If that is the case, please check where the audio corruption begins. Is it at the join point of two m2ts files? If so, please send me the last 50MB of the first m2ts files and the first 50MB of the 2nd m2ts file. Thanks!

But again, if these posts above are correct, then eac3to is innocent and the problem is on the disc.

Originally Posted by yfed

Hello,
I need help obtaining 24bit/192kHz 5.1 FLAC file from BD of Trondheim Soloists "Divertimenti"

Here's my cmd:

However, the FLAC file is corrupted, when I try to seek towards the middle of it in foobar2000 - it says "unable to seek.. the file is corrupted". Maybe there are some FileSystem / FLAC limitations that eac3to ignores? Please, help

PS
flac 5.1 size: 6.46Gb
pcm 5.1 size: 13.3Gb

Does the eac3to log indicate any problems? Please try playing the FLAC file with madFlac. It's quite possible that it's a bug in foobar2000 and not in eac3to. If the problem doesn't occur with madFlac then it's probably a bug in foobar2000 and you should report it to the foobar2000 support/developers. If the problem also occurs with madFlac then I'll look into it.

[n0mag!c]

Originally Posted by alc0re

If madshi changes the code to make it 31 instead of 0 and that fixes the issue, it shouldn't effect those of you that are not seeing the issue, since as he stated both a level of 0 and 31 are "supposed" to mean no dialog normalization. Besides, he stated that perhaps using 31 is better since 0 is reserved. Perhaps some receivers process this correctly and others don't since 0 is supposed to be reserved.

Ok, let's wait for changes, wish it will help you!

Originally Posted by alc0re

Like 73ChargerFan, I am not seeing any issues on any other type of audio thrown at my receiver from any other source other than ac3 audio processed with eac3to with dialog normalization removed. My receiver's settings are fine, I've triple checked them. Both my PS3 and Bluray player are bitstreaming to my receiver via optical.

By the way, I'm bitstreaming via HDMI. And sometimes I switch to LPCM output from PS3. My receiver is Denon 2308.

[mrr19121970]

@madshi

When I demux, I'm only ever interested in the Movie, English & German audios.

nice to have #1 would be:

eac3to d:\ 1) -demux *h264* *english* *german*

so to only see the tracks that have those wildcards in.




nice to have #2 would be:

eac3to d:\ 1) 2: Movie.* 3: English.ac3 4: German.ac3

currently when you do this you get the fog-horn can't convert video. I don't want to convert it, just have it extracted as MPG2, VC1 or H264 as eac3to sees it (makes my BAT easier).



Thanks for looking & Happy Christmas.

[mrr19121970]

I just tried (Atonement HD-DVD):

eac3to q:\

and it only gave me:

1) INTRO_MAIN.EVO+MAIN_LOOP.EVO, 0:02:22



no main feature etc. it worked with earlier versions for sure (eg eac3to.v2.80.exe gives me 11 playlists).

[madshi]

Originally Posted by mrr19121970

I just tried (Atonement HD-DVD):

eac3to q:\

and it only gave me:

1) INTRO_MAIN.EVO+MAIN_LOOP.EVO, 0:02:22



no main feature etc. it worked with earlier versions for sure (eg eac3to.v2.80.exe gives me 11 playlists).

Can you please upload the XPL files? Should be just a few KBs...