eac3to - audio conversion tool

[DrNein]

Okay, so I may as well enter the nearest myself. However, is it invalid to use other delays and does the delaycut program allow them or also automatically adjust to the nearest?

[kartman_canada]

I'm trying to convert my HD-DVDs to something that is good for the PS3. I've been using EAC3TO to demux the VC-1 and demux the audio and convert to AC3 5.1 @ 640kbps. The final step is to convert the VC-1 to AVC and mux these into an M2TS.

I thought all was good until I tried watching a converted copy of Hot Fuzz. The video looks good and the audio is all there and in sync but I'm hearing some weirdness in the balance between the channels... dialog too quiet and fronts/sides too strong. I keep trying to turn up the volume to hear the dialog and then panic looking for the remote when an explosion happens.

Below is a listing from the filter test:

eac3to v2.84
command line: "E:\Video_Encoding\eac3to\eac3to.exe" -test
------------------------------------------------------------------------------
Nero Audio Decoder (Nero 7) works fine
ArcSoft DTS Decoder (1.1.0.0) works fine
Sonic Audio Decoder (2.87.0.0) doesn't seem to be installed
Haali Matroska Muxer (2008-03-29) is up to date
Nero AAC Encoder (1.1.34.2) is installed
There's a new version (1.3.3.0) available
http://www.nero.com/eng/downloads-nerodigital-nero-aac-codec.php
Surcode DTS Encoder (1.0.23.0) is installed
MkvToolnix (2.4.1.0, beta 2008-12-07) is installed

This particular movie is 5.1 E-AC3 EX. Are my filter versions the issue? Are there any EAC3TO audio switches that I should be using?

Here is an example of what I've been doing... in this case I'm only processing the audio and I experimented with not removing the dialog normalization. It didn't seem to help. I'd really appreciate any input the more experienced users out there might be able to offer!

Many thanx to Madshi too...

eac3to v2.84
command line: "E:\eac3to.exe" "E:\FEATURE_1.EVO"+"E:\FEATURE_2.EVO" 4: "E:\movie.ac3" -640 -keepDialnorm
------------------------------------------------------------------------------
EVO, 1 video track, 6 audio tracks, 7 subtitle tracks, 2:00:52
"Main Movie"
1: Joined EVO file
2: Chapters, 28 chapters with names
3: VC-1, 1080p24 /1.001 (16:9) with pulldown flags
4: E-AC3 EX, English, 5.1 channels, 1536kbps, 48khz, dialnorm: -27dB, 133ms
5: E-AC3 EX, French, 5.1 channels, 768kbps, 48khz, dialnorm: -27dB, 133ms
6: E-AC3 Surround, English, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB, 133ms
"Feature Commentary w/ Simon Pegg and Edgar Wright"
7: E-AC3 Surround, English, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB, 133ms
"Feature Commentary w/ Sandford Police Service"
8: E-AC3 Surround, English, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB, 133ms
"Feature Commentary w/ Sandford Village People"
9: E-AC3 Surround, English, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB, 133ms
"Feature Commentary w/ The Real Fuzz"
10: Subtitle, English, "SDH"
11: Subtitle, French
12: Subtitle
13: Subtitle
14: Subtitle
15: Subtitle
16: Subtitle, English, "Fuzz-O-Meter: Trivia Track"
[a04] Extracting audio track number 4...
[a04] Decoding with DirectShow (Nero Audio Decoder 2)...
[a04] DirectShow reports 5.1 channels, 24 bits, 48khz
[a04] Applying RAW/PCM delay...
[a04] Encoding AC3 <640kbps> with libAften...
[a04] Creating file "E:\movie.ac3"...
Video track 3 contains 173880 frames.
eac3to processing took 15 minutes, 54 seconds.
Done.

[Nullity]

I'm trying to convert my HD-DVDs to something that is good for the PS3. I've been using EAC3TO to demux the VC-1 and demux the audio and convert to AC3 5.1 @ 640kbps. The final step is to convert the VC-1 to AVC and mux these into an M2TS.

I thought all was good until I tried watching a converted copy of Hot Fuzz. The video looks good and the audio is all there and in sync but I'm hearing some weirdness in the balance between the channels... dialog too quiet and fronts/sides too strong. I keep trying to turn up the volume to hear the dialog and then panic looking for the remote when an explosion happens.
Don't use the "-keepDialnorm" switch, it lowers the volume of the center channel. Or to be more correct, not using that switch allows eac3to to remove the dialog normalization which increases the volume of the center channel.

[tebasuna51]

... dialog too quiet and fronts/sides too strong. I keep trying to turn up the volume to hear the dialog and then panic looking for the remote when an explosion happens...

This is called 'Full Dynamic Range Preserved'.
Your command line are ok and -keepDialnorm don't help. This is the original sound.

By default many players apply a 'Dynamic Range Compression' than reduce the differences between low and high volume values (the quality is also reduced).

If you want this effect the receivers have a 'Night Mode' to use.

EDIT: @Nullity, the DialNorm value lower the volume of all channels not only center channel.

[Nullity]

@Nullity, the DialNorm value lower the volume of all channels not only center channel.

Ah, thanks for the clarification.

[kartman_canada]

This is called 'Full Dynamic Range Preserved'.
Your command line are ok and -keepDialnorm don't help. This is the original sound.


OK... The initial converted audio that I listened to (and found that I didn't like the level of the center channel) was processed using the command line below.
command line: "E:\eac3to.exe" "E:\FEATURE_1.EVO"+"E:\FEATURE_2.EVO" 4: "E:\movie.ac3" -640

I'm glad to hear that this is "correct" but I still think the center channel is too weak. Seeing as it's coming from an HD-DVD it's very hard for me to arrange a proper A/B comparison between the converted AC3 audio and the original E-AC3.

Is there anything else that might cause the level of the center to be reduced (or the other channels to be boosted)? It seems wrong but I guess that I could look to tweak the AC3 before muxing...

Any other comments or suggestions?

[ultratoto14]

I think that that was already asked but could it be possible to convert h264-dts-movie.mkv to h264-ac3-movie.mkv or just video-with-dts-movie.mkv to audio.ac3. I know that reading mkv is not the same as writing but it could be very interesting.

[MichaelAnders]

Mashi, found a bug in v2.87 with a BD, wrong play duration is displayed.

O:\bd\2>eac3to h:
1) 00021.mpls, 00005.m2ts+00006.m2ts, 1:53:44
- VC-1, 1080p24 /1.001 (16:9)
- DTS, German, multi-channel, 48khz
- DTS, English, multi-channel, 48khz
- RAW/PCM, German, multi-channel, 48khz
- RAW/PCM, English, multi-channel, 48khz

2) 00006.mpls, 00006.m2ts, 1:53:39
- VC-1, 1080p24 /1.001 (16:9)
- DTS Master Audio, German, multi-channel, 48khz
- DTS Master Audio, English, multi-channel, 48khz
- RAW/PCM, German, multi-channel, 48khz
- RAW/PCM, English, multi-channel, 48khz

3) 00014.mpls, 00014.m2ts, 0:21:15
- MPEG2, 480i60 /1.001 (16:9)
- DTS, English, stereo, 48khz

4) 00011.mpls, 00011.m2ts, 0:20:40
- MPEG2, 480i60 /1.001 (16:9)
- DTS, English, stereo, 48khz

O:\bd\2>eac3to h: 1)
M2TS, 1 video track, 4 audio tracks, 2 subtitle tracks, 0:00:05
1: Chapters, 22 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: DTS, German, 5.1 channels, 24 bits, 768kbps, 48khz
4: DTS, English, 5.1 channels, 24 bits, 768kbps, 48khz
5: RAW/PCM, German, 5.1 channels, 16 bits, 48khz
6: RAW/PCM, English, 5.1 channels, 16 bits, 48khz
7: Subtitle (PGS), German
8: Subtitle (PGS), German

Here the duration is now just 5 sec - seems this is the time for 00005.m2ts file which just displays "Senator"?

O:\bd\2>eac3to h: 2)
M2TS, 1 video track, 4 audio tracks, 2 subtitle tracks, 1:53:39
1: Chapters, 22 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: DTS Master Audio, German, 5.1 channels, 24 bits, 48khz
(core: DTS, 5.1 channels, 24 bits, 768kbps, 48khz)
4: DTS Master Audio, English, 5.1 channels, 24 bits, 48khz
(core: DTS, 5.1 channels, 24 bits, 768kbps, 48khz)
5: RAW/PCM, German, 5.1 channels, 16 bits, 48khz
6: RAW/PCM, English, 5.1 channels, 16 bits, 48khz
7: Subtitle (PGS), German
8: Subtitle (PGS), German

Here the time is correct...

[madshi]

However, is it invalid to use other delays and does the delaycut program allow them or also automatically adjust to the nearest?
But eac3to and delaycut accept any number, but round it to the nearest possible value.

I'm glad to hear that this is "correct" but I still think the center channel is too weak.
Maybe the studio mix was bad? Or maybe there's something else going on with your playback chain? It's really hard to diagnose for us without having access to your PC and your playback hardware. What I can say is that you've used a Dolby reference decoder for decoding the E-AC3 track and you've used a well tested and usually very reliable AC3 encoder for encoding. So the tracks you created should be fine.

Is there anything else that might cause the level of the center to be reduced (or the other channels to be boosted)? It seems wrong but I guess that I could look to tweak the AC3 before muxing...
You could decode to WAV and then tweak the track in e.g. Audacity.

I think that that was already asked but could it be possible to convert h264-dts-movie.mkv to h264-ac3-movie.mkv or just video-with-dts-movie.mkv to audio.ac3. I know that reading mkv is not the same as writing but it could be very interesting.
It is not possible with the current eac3to version. You can use mkvextract (part of mkvtoolnix) to extract tracks from MKV container. Then you can use eac3to to convert the extracted audio tracks to whatever format you prefer. And then you can use mkvmerge (part of mkvtoolnix again) to mux the final track back into the MKV and to remove the tracks you don't like.

Mashi, found a bug in v2.87 with a BD, wrong play duration is displayed.
Could you please zip the "CLIPINF" and "PLAYLIST" folders and upload them? Should be just a few KBs. Thanks!

[~bT~]

@ madshi

i'm sorry if this has been asked before but i was wondering if its poss to output celltimes.txt or not? cheers!

ps. if not, will it be poss to implement it?

[ultratoto14]

@madshi

Thanks for the answer.
I already do that.
I use a DLNA media server for the PS3 that can in real time send mkv remuxed to ts or m2ts. The server can transcode DTS to ac3 in real time during muxing as the PS3 does not handle DTS via streaming.
This is the only server i know that can do this and do not loose original mkv quality. But i prefer the sound quality of eac3to compared to the one used in this server (ffmpeg).

The writer told me that he could use eac3to if eac3to could be able to handle mkv or pipes as input/output.
Do you do some check on existing files before running ?
Can eac3to work with named pipes ?

[shambles]

eac3to seems not to strip zero bytes from truehd->flac anymore:

eac3to v2.87
command line: eac3to test 1) 3: test.flac
------------------------------------------------------------------------------
M2TS, 1 video track, 8 audio tracks, 18 subtitle tracks, 0:09:34
1: VC-1, 1080p24 /1.001 (16:9)
2: AC3, English, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB
3: TrueHD/AC3, English, 5.1 channels, 48khz, dialnorm: -27dB
(embedded: AC3, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB)
-snip-
[a03] Extracting audio track number 3...
[a03] Extracting TrueHD stream...
[a03] Removing TrueHD dialog normalization...
[a03] Decoding with libav/ffmpeg...
[a03] Encoding FLAC with libFlac...
[a03] Creating file "test.flac"...
[a03] [libav] End of stream indicated
[a03] [libav] Lossless check failed - expected 0, calculated fe
[a03] The original audio track has a constant bit depth of 16 bits.
Video track 1 contains 13762 frames.
eac3to processing took 1 minute, 55 seconds.
Done.

command line: eac3to test.flac test.wav
------------------------------------------------------------------------------
FLAC, 5.1 channels, 0:09:34, 16/24 bits, 2034kbps, 48khz
Decoding FLAC...
Stripping zero bytes...
Writing WAV...
Creating file "test.wav"...
The original audio track has a constant bit depth of 16 bits.
eac3to processing took 13 seconds.
Done.

command line: eac3to test.wav test.flac
------------------------------------------------------------------------------
WAV, 5.1 channels, 0:09:34, 16 bits, 4608kbps, 48khz
Reading WAV...
Encoding FLAC with libFlac...
Creating file "test.flac"...
The original audio track has a constant bit depth of 16 bits.
eac3to processing took 29 seconds.
Done.

eac3to test.flac
FLAC, 5.1 channels, 0:09:34, 16 bits, 2033kbps, 48khz

[MichaelAnders]

Could you please zip the "CLIPINF" and "PLAYLIST" folders and upload them? Should be just a few KBs. Thanks!

http://rapidshare.com/files/180787649/BDMV.zip.html

[n0mag!c]

eac3to v2.87
command line: D:\PROGRAMS\AUDIO\EAC3TO\eac3to.EXE 5e.ac3 j:5e.wav
------------------------------------------------------------------------------
AC3, 2.0 channels, 0:28:46, 192kbps, 48khz, dialnorm: -27dB
The Nero decoder doesn't seem to work, will use libav instead.
Removing AC3 dialog normalization...
Decoding with libav/ffmpeg...
Reducing depth from 64 to 24 bits...
Writing WAV...
Creating file "j5e.wav"...
eac3to processing took 20 seconds.
Done.
Now "eac3to" creates file "j5e.wav" in current folder of current disk, but v.2.80 (I've upgraded from this version to 2.87) has created file "5e.wav" in current folder of disk "j", as it's supposed to be. Can you please fix it back?!

And second problem - "sound forge 8" can't open stereo wav-files, generated by "eac3to".
Headers for 16-bit wav-files:
sforge:
0000000000: 52 49 46 46 24 C8 C0 13 │ 57 41 56 45 66 6D 74 20 RIFF$ИА‼WAVEfmt
0000000010: 10 00 00 00 01 00 02 00 │ 80 BB 00 00 00 EE 02 00 ► ☺ ☻ _> о☻
0000000020: 04 00 10 00 64 61 74 61 │ 00 C8 C0 13 01 00 00 00 ♦ ► data ИА‼☺eac3to:
0000000000: 52 49 46 46 3C C8 C0 13 │ 57 41 56 45 66 6D 74 20 RIFF<ИА‼WAVEfmt
0000000010: 28 00 00 00 FE FF 02 00 │ 80 BB 00 00 00 EE 02 00 ( юя☻ _> о☻
0000000020: 04 00 10 00 16 00 10 00 │ 03 00 00 00 01 00 00 00 ♦ ► ■ ► ♥ ☺
0000000030: 00 00 10 00 80 00 00 AA │ 00 38 9B 71 64 61 74 61 ► _ Є 8>qdataeac3to mono:
0000000000: 52 49 46 46 24 64 E0 09 │ 57 41 56 45 66 6D 74 20 RIFF$dа○WAVEfmt
0000000010: 10 00 00 00 01 00 01 00 │ 80 BB 00 00 00 77 01 00 ► ☺ ☺ _> w☺
0000000020: 02 00 10 00 64 61 74 61 │ 00 64 E0 09 01 00 00 00 ☻ ► data dа○☺
Is there special reason that "eac3to" creates different header for stereo file?

[tebasuna51]

@n0mag!c
About the wav headers:

v2.57
...
* parameter "-extensible" is no longer supported (it's default now)
* new parameter "-simple" can be used to disable the "-extensible" wav header

I obtain also mono with WAVE_FORMAT_EXTENSIBLE headers.

[Snowknight26]

Now "eac3to" creates file "j5e.wav" in current folder of current disk, but v.2.80 (I've upgraded from this version to 2.87) has created file "5e.wav" in current folder of disk "j", as it's supposed to be. Can you please fix it back?!

You should be using J:\5e.wav.

[n0mag!c]

"j:\5e.wav" means create file in ROOT folder of disk "j", but I want to create file in CURRENT folder of disk "j".
Obviously Madshi added/change parser, which removes ":" in file names now.

[n0mag!c]

* new parameter "-simple" can be used to disable the "-extensible" wav header
Thanks for the tip!
Can this switch be mentioned on the first page, as like as all other yet unmentioned switches? I forgot in time the switch, with which I can get float-wav. Can somebody please remind me? If I ain't wrong about its existing.

I obtain also mono with WAVE_FORMAT_EXTENSIBLE headers.Why?


Another problem - "eac3to" crashes with this "m2ts"-file (http://rapidshare.de/files/41302230/00001.m2ts.html).
VC-1 stream was incorporated from WMV-file.

[nautilus7]

"j:\5e.wav" means create file in ROOT folder of disk "j", but I want to create file in CURRENT folder of disk "j".
Obviously Madshi added/change parser, which removes ":" in file names now.
The output file is written in the folder where eac3to.exe is called from or in whatever folder you set.

[Chumbo]

...Can this switch be mentioned on the first page, as like as all other yet unmentioned switches?...
Read your changelog.txt that's included with every release.

And for all those that don't use this file as a reference, it's the least you can do to show madshi you appreciate all his efforts.

[n0mag!c]

The output file is written in the folder where eac3to.exe is called from or in whatever folder you set.
I fully understand the logic of creation files, don't worry about that. I state, that now "eac3to" removes ":" from file name like this "j:5e.wav" and creates file "j5e.wav" instead.
Read your changelog.txt that's included with every release.
This is evil logic to keep description of application in "changelog.txt". Especially, while help screen exists.

And another little bug, as far as it goes. :)
When capturing output from "eac3to" like this: "eac3to>help.txt", file is filled with 08h characters, but they must have not be there.

And for all those that don't use this file as a reference, it's the least you can do to show madshi you appreciate all his efforts.
P.S. Madshi, I appreciate your work very-very much, there's no doubt! Making world a little more perfect is great.

[rebkell]

I would guess the bug or change happened because of this problem:

http://forum.doom9.org/showpost.php?p=1229768&postcount=7607

[alc0re]

madshi,

Could you add a command that forces dialog normalization...something like -forceDialnorm

Or even a command that would me me specify what level of dialog normalization an ac3 file has? Either way would work. Yet another option would be if eac3to detected a -0 dialog normalization level as having dialog normalization so that when eac3to extracted the file it would change it to -31. Reason I ask is so that I can "fix" my ac3 tracks with -0 dialog normalization without having to re-encode them. I believe that this wouldn't be too big of a feature to add since removing dialog normalization doesn't seem to require re-encoding (seems like its just changing the value from whatever the file was to -0 pre v2.85 or -31 post v2.85 .)

One other question : When I do an eac3to command on a bluray structure, it tells me my main movie track's video is something like 1080p24. lets say that was feature 1. So I do a eac3to 1) command and again is says the video track is 1080p24. When I check the m2ts file with mediainfo it says the video is 23.976 FPS. MeGUI also detects the video as 23.976 FPS. I transcode my video down to 720p and create a bluray structure with tsMuxer. When I play my avchd discs on my bluray player, the video stutters when there's movement...like if the camera is following the actor and the background is moving, the background seems to stutter. Now I can't figure out for the life of me why this is happening. Only difference I have noticed is the following. eac3to reports 1080p24 on both the regular eac3to command against the original bluray and when I look at the main movie feature (ie eac3to 1) ) When I do a eac3to command on my bluray structured transcoded bluray, it reports 720p24. But when I look at the main movie feature (eac3to 1)) it reports 720p23. I looked at my bluray structure with bdedit.exe on both the original bluray and my transcoded bluray and I don't see any difference. Where does eac3to get the information from that would be having eac3to say its 720p23 instead of 720p24? I think this might have something to do with why my video seems to stutter. PS. doesnt happen on my pc when I play the m2ts file directly. And yes when I transcode my video it is transcoded at 23.976 fps.

[n0mag!c]

When I play my avchd discs on my bluray player, the video stutters when there's movement...like if the camera is following the actor and the background is moving, the background seems to stutter. Now I can't figure out for the life of me why this is happening.
I guess it's just your TV issue. 24 fps is too slow for human eye. Size of your PC monitor is smaller and refresh rate is greater, that's why this stutter is less noticable on PC.
Modern TVs use different techniques to eliminate this effect, personally, I prefer interpolation of additional frames, similar to Philips' DNM (digital natural motion). But many others don't like it for the "theatre" look.

[nautilus7]

I fully understand the logic of creation files, don't worry about that. I state, that now "eac3to" removes ":" from file name like this "j:5e.wav" and creates file "j5e.wav" instead.
Have you ever tried using : for a filename in windows? Guess what happens. ;)

[Chumbo]

...This is evil logic to keep description of application in "changelog.txt". Especially, while help screen exists....
The help screen is for existing switches and, at the developer's discretion, may not show hidden switches. It's not for "updates," i.e., switches may be added, removed, renamed, etc. That's why you need to check the changes log. ;)

...And another little bug, as far as it goes. :)
When capturing output from "eac3to" like this: "eac3to>help.txt", file is filled with 08h characters, but they must have not be there.
...
Use the -log switch. There's no need to pipe the output since the -log now does everything and madshi was kind enough to allow multiple running instances to have their own logs.

It's not a bug, it's just what gets piped from the screen/console to the file.

All you need to do is run this: eac3to -log=help.txt

[n0mag!c]

Have you ever tried using : for a filename in windows? Guess what happens. ;)
Of course, constantly! I'm using "Far manager" as "GUI" :) for "DOS" environment. And using "j:5e.wav" as file name helps me to avoid using full path. So current paths for all disks live long lifes in this console application. ;)

It's not a bug, it's just what gets piped from the screen/console to the file.
But this happens only with "eac3to"...

[asarian]

@Madshi,

When I convert a DTS Hi-Res stream (from Predator 2) to LPCM, eac3to doesn't seem to list the final bit-depth:


eac3to 00013.m2ts 3: c:\video\temp.pcm
M2TS, 1 video track, 9 audio tracks, 24 subtitle tracks, 1:48:00
1: Chapters, 33 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: DTS Hi-Res, English, 5.1 channels, 24 bits, 3018kbps, 48khz
(core: DTS, 5.1 channels, 24 bits, 1509kbps, 48khz)
...
[a03] Extracting audio track number 3...
[a03] Decoding with ArcSoft DTS Decoder...
[a03] Swapping endian...
[a03] Remapping channels...
[a03] Creating file "c:\video\temp.pcm"...
[a03] The last DTS frame is incomplete and thus gets skipped.
Video track 2 contains 155364 frames.
eac3to processing took 17 minutes, 40 seconds.
Done.


Normally, it gives me a message saying the final output file had 16/24 bit depth, etc. This time not. Is that a small bug?

Thanks

[alc0re]

I appreciate your reply n0magic but that's not taking into account that I said it doesn't happen when i'm watching non-transcoded retail bluray or hd-dvds on my tv...

[n0mag!c]

I appreciate your reply n0magic but that's not taking into account that I said it doesn't happen when i'm watching non-transcoded retail bluray or hd-dvds on my tv...
I'm sorry, all ordinary TVs seem like stuttering to me. :) Ok, then here's my second guess - bitrate overflow. Do you have ability to burn your transcodings on BD-RE which support higher transfer rate to test this?
Sometimes 23, 24, 23.976 are used as synonyms.

[alc0re]

no i dont have that ability. To me I see stuttering a lot too but not like this. I can take the same movie and watch it from the retail bluray or hd-dvd and then transcode it and see a big difference.

BD-RE is not an option for me.

I thought maybe the bitrate was too high also for a dvd, but I've been using Ryu77's profiles lately for avchd content. he's got a lower value in his profile. Ryu's profile : VBVBuffersize = 15000 and VBVMaxbirate = 17500. Those values are a lot lower than max 40000 for 4.1 (which is max supported.) I'm still curious why it says 1080p24 on the original bluray and not on the transcoded.

If I can't figure this out I'm going to try playing the avchd disc on another brand bluray player. Problem with that is that it won't be on my tv.

EDIT/UPDATE : This morning I transcoded The Matrix Revolutions using RipBot264 instead of MeGUI. My normal process is using MeGUI with Ryu77's AVCHD profile, with a target size of 8152 to fit on a dual layer dvd. I have been burning my movies to memorex dual layer dvds. I wanted to try something different so I transcoded the matrix 3 this morning using ripbot instead of megui. I also changed the target size to fit on a single layer dvd. I used ripbot's avchd/bluray profiles. I also burnt the movie to a Verbatim single layer dvd that I just bought cause I researched and found that Verbatim make better blank dvds. I'm watching that ripbot transcoded movie right now and I don't see the same stuttering issue. Unfortunately there were a few different things that are different now. RipBot did the transcoding, albeit with the same program that MeGUI does so that shouldn't be the difference. The bitrate is a lot lower than all my other transcodes since I made this one fit on a single layer dvd (dual layers are freakin expensive.) The x264 encoding profile is different. The level is different (4.0 as opposed to 4.1) I just now have to try different things to figure out what made the difference. Also, RipBot does the muxing through tsMuxer as opposed to me doing it manually. I need to figure out if RipBot changes any of the default muxing options in tsMuxer, like the "continually insert SPS/PPS or Add Picture Timing Info or or Use ASync I/O.

EDIT/UPDATE 2 : Demuxed a movie that had the stuttering issue. Remuxed the movie with tsRemux and on the video stream options, I changed the level from High@4.1 to High@4.0. Tested. Works perfectly. So for anyone interested, the panasonic bd35 does not play nice with video at level 4.1.

[idbirch2]

When I convert a DTS Hi-Res stream (from Predator 2) to LPCM, eac3to doesn't seem to list the final bit-depth:

Normally, it gives me a message saying the final output file had 16/24 bit depth, etc. This time not. Is that a small bug?Just run eac3to "c:\video\temp.pcm"

[shanghai2004]

Madshi,

Had finally time to make a sample to reproduce the issue from post #7225.
(issue: cannot extract EAC3 5.1 audio track from EVO, got endless 13ms audio overlap warnings, preventing eac3to from completing)

However, while testing the sample, eac3to doesn't like incomplete EAC3 packets anymore seems. Got crash.

Sample (part1.evo, 33MB):

http://www.mediafire.com/?c1sn23b2kxr

See also crash report send by email

[n0mag!c]

I've never got "Clipping detected" message on AC3-file before, is it normal?
eac3to v2.87
command line: D:\PROGRAMS\AUDIO\EAC3TO\eac3to.EXE en.ac3 J:\temp\trans\en.wavs
------------------------------------------------------------------------------
AC3, 5.1 channels, 2:23:27, 448kbps, 48khz, dialnorm: -27dB
The Nero decoder doesn't seem to work, will use libav instead.
Removing AC3 dialog normalization...
Decoding with libav/ffmpeg...
Remapping channels...
Reducing depth from 64 to 24 bits...
Writing WAVs...
Creating file "J:\temp\trans\en.C.wav"...
Creating file "J:\temp\trans\en.LFE.wav"...
Creating file "J:\temp\trans\en.L.wav"...
Creating file "J:\temp\trans\en.R.wav"...
Creating file "J:\temp\trans\en.SL.wav"...
Creating file "J:\temp\trans\en.SR.wav"...
Clipping detected, a 2nd pass will be necessary.
Starting 2nd pass...
Removing AC3 dialog normalization...
Decoding with libav/ffmpeg...
Remapping channels...
Reducing depth from 64 to 24 bits...
Writing WAVs...
Creating file "J:\temp\trans\en.L.wav"...
Creating file "J:\temp\trans\en.C.wav"...
Creating file "J:\temp\trans\en.R.wav"...
Creating file "J:\temp\trans\en.LFE.wav"...
Creating file "J:\temp\trans\en.SL.wav"...
Creating file "J:\temp\trans\en.SR.wav"...
eac3to processing took 9 minutes, 13 seconds.
Done.

"DelayCut" didn't find any problem:====== INPUT FILE INFO ========================
File is ac3
Bitrate (kbit/s) 448
Act rate (kbit/s) 448.000
File size (bytes) 481990656
Channels mode 3/2: L+C+R+SL+SR
Sampling Frec 48000
Low Frec Effects LFE: Present
Duration 02:23:26.976
Frame length (ms) 32.000000
Frames/second 31.250000
Num of frames 268968
Bytes per Frame 1792.0000
Size % Framesize 0
CRC present: YES
=============================================
====== TARGET FILE INFO ======================
Start Frame 0
End Frame 268967
Num of Frames 268968
Duration 02:23:26.976
NotFixedDelay 0.0000
=============================================
====== PROCESSING LOG ======================
Though "eac3to" created files with exactly the same length - 02:23:26.976

P.S. Gosh, I'd rather be using 2.80... :confused:

P.P.S. Here is that sample. (http://www.megaupload.com/?d=IOXZ7I1W)

[Thunderbolt8]

is there actually any way to utilize DTS Express tracks somehow atm? using ac3filter, I only get garbage sound and muxing them with mkvmerge doesnt work as well yet. so is converting them to normal DTS or ac3 is the only choice I have atm?

[nautilus7]

I've never got "Clipping detected" message on AC3-file before, is it normal?
Yes, it is. eac3to can now detect audio peaks above "maximum level" and if necessary reduce the volume of the track so the clipping is gone.

(hope i explained this correctly, madshi)

is there actually any way to utilize DTS Express tracks somehow atm? using ac3filter, I only get garbage sound and muxing them with mkvmerge doesnt work as well yet. so is converting them to normal DTS or ac3 the only choice I have atm?Can't you use nero decoder with your media player (mpc)? I think is the only dshow decoder available for dts express. Otherwise conversion is the only way.

[asarian]

Just run eac3to "c:\video\temp.pcm"

Never knew you could do that. :) (since pcm/raw has no header and all). If at all possible, though, I'd like eac3to to include that information for DTS-HD tracks too, like it does for DTS-MA/TrueHD, etc.

[n0mag!c]

eac3to can now detect audio peaks above "maximum level" and if necessary reduce the volume of the track so the clipping is gone.Thanks! I incorrectly believed that this is equal to gap/overlap detection. :o (on coincidence, my two audio streams go out of sync at this point). I really don't like that the second pass is needed to eliminate clipping.

[Thunderbolt8]

Yes, it is. eac3to can now detect audio peaks above "maximum level" and if necessary reduce the volume of the track so the clipping is gone.

(hope i explained this correctly, madshi)

Can't you use nero decoder with your media player (mpc)? I think is the only dshow decoder available for dts express. Otherwise conversion is the only way.
i deactivated all other audio filters, but still only got garbage noise with it. same for arcsoft and arcsoft HD decoder

[73ChargerFan]

When capturing output from "eac3to" like this: "eac3to>help.txt", file is filled with 08h characters, but they must have not be there.
Yup, funny.
All you need to do is run this: eac3to -log=help.txt
By convention,
command > log.txt
should work with any console command line application. For some reason, eac3to prints 80 backspace characters at the beginning of each line. Wierd.

[rebkell]

Yup, funny.

By convention,
command > log.txt
should work with any console command line application. For some reason, eac3to prints 80 backspace characters at the beginning of each line. Wierd.

That's how he keeps the status line on the console in the same position, he writes 80 characters, then backspaces them out before he writes the next update line.

[Jaja1]

I really don't like that the second pass is needed to eliminate clipping.Try to run eac3to while keeping dialnorm. Perhaps that prevents the clipping.

[asarian]

Try to run eac3to while keeping dialnorm. Perhaps that prevents the clipping.

Hmm. So, are you saying that the normal procedure of removing dialog normalization causes/can cause the track to become damaged? (as in: parts of audio being cut off, above a certain threshold). That would not be good.

Here's to hoping I misunderstood. :)

[n0mag!c]

That's how he keeps the status line on the console in the same position, he writes 80 characters, then backspaces them out before he writes the next update line.
Common practice is to output "CR" (carriage return) character (0Dh) for that purpose. But in this case only 79 characters in line can be used to return back. But there is no need for returning back at help screen.

[n0mag!c]

Try to run eac3to while keeping dialnorm. Perhaps that prevents the clipping.
It makes no difference. I provided a sample, you can try for yourself.

[tebasuna51]

Try to run eac3to while keeping dialnorm. Perhaps that prevents the clipping.

Not at all. DialNorm is a field in header not used in this process.

[weedenbc]

Quick question - I'm trying to convert an m2ts files with a TrueHD audio track using eac3to. It keeps failing with error message "This audio conversion is not supported."

I thought the libav that shipped with eac3to was supposed to handle TrueHD? I also have Arcsoft TMT installed but that doesn't seem to be helping.

----
Brian

[Snowknight26]

Helps to post the command you're trying.

[asarian]

Quick question - I'm trying to convert an m2ts files with a TrueHD audio track using eac3to. It keeps failing with error message "This audio conversion is not supported."

I thought the libav that shipped with eac3to was supposed to handle TrueHD? I also have Arcsoft TMT installed but that doesn't seem to be helping.

Two likely causes:

1) Wrong track number (you selected a video track, instead of audio, for instance).

2) Unrecognized extension for the audio output file.

So, show us what ya tried to do. :)

[weedenbc]

Yeah, you see it helps if I put a .mkv on the end of the second parameter telling eac3to where to write the video stream to.

Sorry :(