eac3to - audio conversion tool

[itsancho]

strange, but minor bug in 2.84...

eac3to "H:\Hellboy (2004) Blu-ray AVC PCM"
1) 00030.mpls, 00032.m2ts, 2:12:29
- h264/AVC, 1080p24 /1.001 (16:9)
- AC3, English, multi-channel, 48khz
- RAW/PCM, English, multi-channel, 48khz
- AC3, French, multi-channel, 48khz
- AC3, German, multi-channel, 48khz
- RAW/PCM, German, multi-channel, 48khz
- AC3, English, stereo, 48khz

2) 00066.mpls, 2:23:08
[30+31+38+39+40+64].m2ts
- MPEG2, 480i60 /1.001 (16:9)
- AC3, English, stereo, 48khz

eac3to "H:\Hellboy (2004) Blu-ray AVC PCM" 1)
M2TS, 1 video track, 1 audio track, 5 subtitle tracks, 2:23:08
1: Chapters, 7 chapters
2: MPEG2, 480i60 /1.001 (16:9)
3: AC3 Surround, English, 2.0 channels, 192kbps, 48khz
4: Subtitle (PGS), English
5: Subtitle (PGS), French
6: Subtitle (PGS), German
7: Subtitle (PGS), Dutch
8: Subtitle (PGS), Korean

eac3to "H:\Hellboy (2004) Blu-ray AVC PCM" 2)
M2TS, 1 video track, 6 audio tracks, 21 subtitle tracks, 2:12:29
1: Chapters, 16 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: AC3, English, 5.1 channels, 448kbps, 48khz
4: RAW/PCM, English, 5.1 channels, 16 bits, 48khz
5: AC3, French, 5.1 channels, 448kbps, 48khz, dialnorm: -30dB
6: AC3, German, 5.1 channels, 448kbps, 48khz, dialnorm: -30dB
7: RAW/PCM, German, 5.1 channels, 16 bits, 48khz
8: AC3 Surround, English, 2.0 channels, 192kbps, 48khz, dialnorm: -26dB
9: Subtitle (PGS), English
...
29: Subtitle (PGS), Korean

Merry Christmas everyone! And Thank You madshi!

[tebasuna51]

Originally Posted by Thunderbolt8

got a problem with a H264 50i cap, when I try to slow it down to 23.976fps the audiotrack length increases to >5h (duration of video seems to be fine). might be a problem, because the track is rcognized by eac3to as 2.0, while its supposed to be 5.1. Already tried to fix it with delaycut, as Tebasuna once said with that flag that indicates a channel switch, but delaycut didn't report such a thing for this track here.

The ac3 extracted with eac3to from your sample with:

eac3to v2.84
command line: "D:\eac3to.exe" "D:\breakfastsplit.m2ts" zz.ac3
------------------------------------------------------------------------------
TS, 1 video track, 1 audio track, 0:00:25
1: h264/AVC, 1080i50 (16:9)
2: AC3, 2.0 channels, 384kbps, 48khz, dialnorm: -27dB, -732ms
Track 2 is used for destination file "zz.ac3".
[a02] Extracting audio track number 2...
[a02] Removing AC3 dialog normalization...
[a02] Applying (E-)AC3 delay...
[a02] Creating file "zz_e.ac3"...
[a02] The last (E-)AC3 frame is incomplete and thus gets skipped.
Video track 1 contains 1234 frames.
eac3to processing took 1 second.
Done.

have 6 frames 2.0 and after change to 5.1. Delaycut log:

[Input info]
Bitrate=384
Actual rate=384.000000
Sampling Frec=48000
TotalFrames=752
Bytesperframe=1536.0000
Filesize=1155072
FrameDuration= 32.0000
Framespersecond= 31.2500
Duration=00:00:24.064
Channels mode=2/0: L+R
LFE=LFE: Not present
[Target info]
StartFrame=0
EndFrame=751
NotFixedDelay= 0.0000
Duration=00:00:24.064
====== PROCESSING LOG ======================
Time 00:00:00.192; Frame#= 7. Some basic parameters changed between Frame #1 and this frame
Number of written frames = 752
Number of Errors= 1

[Snowknight26]

I'm still a bit confused about changing channel mapping. With the -0,1,2,3,4,5 option, how do I know which number corresponds to which channel? Or better yet, how would I switch the LFE and Back Left channels using that switch (assuming I'm dealing with 5.1 content)?

[tebasuna51]

With -0,1,2,4,3,5 the change is LFE <-> BackLeft

[dorati]

@madshi:
I have a question to the Subtitle (PGS):
Is it possible to implement a switch, to extract only forced suptitles in the SUP-File?

Sometimes the forced-suptitles are one extra stream - this is no problem!
But Sometimes the forced-subtitles in the normal suptitle stream -(.
SupRip has a switch for forced subtitle and can extract only the forced to srt.
But I will have the graphic-forced-subtiteles to remux in the m2ts-container.

Merry X-Mas

[Thunderbolt8]

Originally Posted by tebasuna51

The ac3 extracted with eac3to from your sample with:

eac3to v2.84
command line: "D:\eac3to.exe" "D:\breakfastsplit.m2ts" zz.ac3
------------------------------------------------------------------------------
TS, 1 video track, 1 audio track, 0:00:25
1: h264/AVC, 1080i50 (16:9)
2: AC3, 2.0 channels, 384kbps, 48khz, dialnorm: -27dB, -732ms
Track 2 is used for destination file "zz.ac3".
[a02] Extracting audio track number 2...
[a02] Removing AC3 dialog normalization...
[a02] Applying (E-)AC3 delay...
[a02] Creating file "zz_e.ac3"...
[a02] The last (E-)AC3 frame is incomplete and thus gets skipped.
Video track 1 contains 1234 frames.
eac3to processing took 1 second.
Done.

have 6 frames 2.0 and after change to 5.1. Delaycut log:

as I said the line 'Time 00:00:00.192; Frame#= 7. Some basic parameters changed between Frame #1 and this frame' didnt appear in my case, there was no problem delaycut reported at all for me with this track :S
will try again

EDIT: SORRY, I NAMED THE FILE EXTENSION OF THE SAMPLE WRONG, ITS .TS NOT .M2TS HERE

edit²: got it working now, tried to process the file which was already slowed and in that case he didnt find that line with delaycut. when I tried to process the demuxed, non-slowed ac3 track then it worked.

[ggking7]

Should I use -keepPulldown with my 120Hz TV so the de-judder mechanism can do its thing?

[odin24]

Could someone please help me out, I'm a little uncertain about the description of the -logdts output. In the case below, does the "Core+Xch" mean the HD data is applied to the entire 6.1 channels, or just the extra Cs channel?

Thanks,

+ DTS-Core
  - frameSize            2012
  - DTS-ES               +
  - channelNo            6
  - lfe                  1
  - channelDescr         6.1
  - samplingRate         48000
  - bitDepth             24
  - bitrate              1509000
  - dialNorm             4
  - extAudio             XCh
  - samplesPerFrame      512
  - copyHistory          1
+ DTS-HD
  - fullSize             2040
  - headerSize           28
  - refClockCode         1/48000
  - frameDurationCode    1
  - activeMasks          [1], [[1]]
  + Asset [0]
    - fullSize             2012
    - headerSize           10
    - corePackets          Core+XCh
    - extSubStrPackets     XBR
    - bitResolution        24
    - maxSampleRate        48000
    - totalNumChannels     7
    - activeSpeakers       C L R Ls Rs LFE Cs ($1f)

[Atak_Snajpera]

Should I use -keepPulldown with my 120Hz TV so the de-judder mechanism can do its thing?

120hz / 23.976 = ~ 5
so the answer is NO

[Snowknight26]

Originally Posted by tebasuna51

With -0,1,2,4,3,5 the change is LFE <-> BackLeft

So going by what you said, the following is correct?:
0 = Left
1 = Right
2 = Center
3 = LFE
4 = Back Left
5 = Back Right

Also, can you change the channel order for 7.1 channel audio with eac3to?

Edit: It seems that when I change the channel order (-0,1,2,4,3,5), the resulting file still has the same channel order when I check with ffdshow.

[vucloutr]

hi there, i found a minor bug with .aac encoding.
eac3to handles "-quality=0.01" like "-quality=0.10", "-quality=0.02" like "-quality=0.20", .. , "-quality=0.09" like "-quality=0.90".

i noticed it when i saw these assumingly wrong filesizes:

[tebasuna51]

Originally Posted by Snowknight26

So going by what you said, the following is correct?:
0 = Left
1 = Right
2 = Center
3 = LFE
4 = Back Left
5 = Back Right

Of course.

Also, can you change the channel order for 7.1 channel audio with eac3to?

Yes. Now:
6 = Side Left
7 = Side Right

Edit: It seems that when I change the channel order (-0,1,2,4,3,5), the resulting file still has the same channel order when I check with ffdshow.

Nope, the eac3to remapping works fine.

[Snowknight26]

G:\Encoding Tools\eac3to>eac3to.exe ..\bhd.broken.cut.flac
FLAC, 5.1 channels, 16 bits, 48khz

G:\Encoding Tools\eac3to>eac3to.exe ..\bhd.broken.cut.flac bhd.broken.cut.flac
FLAC, 5.1 channels, 16 bits, 48khz
Decoding FLAC...
Encoding FLAC with libFlac...
Creating file "bhd.broken.cut.flac"...
The original audio track has a constant bit depth of 16 bits.
eac3to processing took 5 seconds.
Done.

G:\Encoding Tools\eac3to>eac3to.exe bhd.broken.cut.flac
FLAC, 5.1 channels, 0:01:05, 16 bits, 1233kbps, 48khz

Any reason why the 1st track didn't show the length?..

G:\Encoding Tools\eac3to>eac3to.exe bhd.cut.flac
FLAC, 5.1 channels, 2:24:20, 16 bits, 9kbps, 48khz

G:\Encoding Tools\eac3to>eac3to.exe ..\bhd.cut.flac bhd.cut.flac
FLAC, 5.1 channels, 2:24:20, 16 bits, 9kbps, 48khz
Decoding FLAC...
Encoding FLAC with libFlac...
Creating file "bhd.cut.flac"...
The original audio track has a constant bit depth of 16 bits.
eac3to processing took 5 seconds.
Done.

G:\Encoding Tools\eac3to>eac3to.exe bhd.cut.flac
FLAC, 5.1 channels, 0:01:05, 16 bits, 1235kbps, 48khz

..and this one shows different ones?

http://www.stfcc.org/misc/bhd.cut.flac (before reencoding to FLAC)
http://www.stfcc.org/misc/bhd.broken.cut.flac (before reencoding to FLAC)

(Don't mind the broken in the name, I was messing with channel orders - broken refers to the one with the wrong channel order.)

[madshi]

Originally Posted by wolfbane5

In that case, I'm assuming there should be a folder labeled SurCode DVD-DTS? and that SurCode DVD-DTS and SurCode DVD DTS are both missing a Home folder?

If you own a proper Surcode DTS encoder license and if you have installed it properly, these folders are supposed to be there. So my suggestion would be to uninstall, get the latest version from Surcode and install it. Please note that AFAIK there are two different products: One for CD (stereo) encoding and one for movie encoding. eac3to only supports the latter one...

Originally Posted by Thunderbolt8

eac3to lists the 7.1 DTS-HD MA track of the nordic black hawk down edition with "(strange setup)" and "CAUTION: Decoding this track with ArcSoft results in low volume" [...].

so I guess arcsoft does not process this track very well, is the channel order wrong?

No. The log already tells you what is wrong: The volume is lower than it should be. There's no problem with the channel order...

Originally Posted by rickardk

Alos tried this one and got som kind of static noise in the by eac3to created FLAC. I guess the ArcSoft decoder does something wrong ...

Hmmmm... Could you guys please report this problem to ArcSoft? I mean it should also occur if you play the full physical Blu-Ray disc with the ArcSoft player (as long as your speakers are setup to be 7.1), right? So this is something ArcSoft should really be willing to fix...

I've already seen this problem once. I'm not sure if the DTS 5.1 -> 7.1 channel extension is broken in this specific movie or whether there's a bug in the ArcSoft decoder. Probably the latter. You can work around it by either using Sonic for decoding or by using the "-6" parameter to force ArcSoft to output 6 channels, only. I'm not fully sure whether the "-6" removes the problem completely. Maybe it does, maybe not. You'll have to check that yourself. Using Sonic for this track might be safer. Of you can decode only the core with ArcSoft by using the "-core" switch. That's safe, too, but of course loses the Master Audio information...

Originally Posted by itsancho

strange, but minor bug in 2.84...

Thanks for the report, will be fixed in the next build (I hope).

Originally Posted by dorati

I have a question to the Subtitle (PGS):
Is it possible to implement a switch, to extract only forced suptitles in the SUP-File?

Sometimes the forced-suptitles are one extra stream - this is no problem!
But Sometimes the forced-subtitles in the normal suptitle stream -(.
SupRip has a switch for forced subtitle and can extract only the forced to srt.
But I will have the graphic-forced-subtiteles to remux in the m2ts-container.

I'm currently only demuxing the subtitle stream. I'm not even trying to "understand" it. So I currently don't have the capability to extract only a part of the subtitle stream. Of course it would be possible to add. But I'm not sure how difficult it is, and to be honest, I still have lots of more important things on my to do list, which are probably easier to implement...

Originally Posted by ggking7

Should I use -keepPulldown with my 120Hz TV so the de-judder mechanism can do its thing?

I'm not a fan of storing video/audio in a suboptimal format just to work around some special requirements imposed on you by your hardware. What happens if in 2 years from now you have a different display which works better with pulldown stripped? IMHO you should store video/audio in the best possible format and that is with pulldown removed. But that's just my opinion.

Now specifically about 120Hz: With pulldown you have 2*60Hz = 120Hz. Without pulldown you have 5*24Hz = 120Hz. So in theory a good 120Hz display should be able to handle the video track fine, regardless of whether you remove the pulldown or not. With pulldown removed it should even be better. But in pratical life things are often different than theory suggests...

Originally Posted by odin24

I'm a little uncertain about the description of the -logdts output. In the case below, does the "Core+Xch" mean the HD data is applied to the entire 6.1 channels, or just the extra Cs channel?

    - corePackets          Core+XCh
    - extSubStrPackets     XBR

It means that the core already contains the back channel ("XCh") extension. The DTS-HD data only contains an XBR extension, which is additional bitrate. I don't know if the additional bitrate is only for 5.1 or whether it's for 6.1, but that doesn't really matter. The DTS-HD data does improve audio quality of the 5.1 channels in your specific case.

Originally Posted by vucloutr

hi there, i found a minor bug with .aac encoding.
eac3to handles "-quality=0.01" like "-quality=0.10", "-quality=0.02" like "-quality=0.20", .. , "-quality=0.09" like "-quality=0.90".

i noticed it when i saw these assumingly wrong filesizes:

Thanks for the report, will be fixed in the next build.

Originally Posted by Snowknight26

Edit: It seems that when I change the channel order (-0,1,2,4,3,5), the resulting file still has the same channel order when I check with ffdshow.

Should work just fine. If you can reproduce that it doesn't work, then please post your eac3to log. You should check channel order in a WAV editor like e.g. Audacity.

Originally Posted by Snowknight26

Any reason why the 1st track didn't show the length?

Yes, the length information is not stored in that file. It's as easy as that.

[madshi]

Originally Posted by zeropc

unfortunately this doesn't help me cause i wanna decode aac and not encode.

The problem is that the standalone Nero AAC decoder only supports *.mp4 files as input, while eac3to only supports demuxed raw AAC streams. So those two tools don't match well together, unfortunately.

[n0mag!c]

Originally Posted by madshi

Now specifically about 120Hz: With pulldown you have 2*60Hz = 120Hz. Without pulldown you have 5*24Hz = 120Hz.

Unless 24p mode is activated both on player and TV, player transmits content at 30/25fps, and TV is refreshing with multiplier of 4 at 120/100 Hz. And in 24p mode there is no matter with what multiplier TV is refreshing - 4 or 5 (I ain't trust TV in this mode, that its refresh rate stays 120Hz), because surely no additional frames are interpolated, the same frame is shown continually.
But 24p mode can only be activated when pulldown is removed, so this is another reason why pulldown can be removed.

Originally Posted by madshi

So in theory a good 120Hz display should be able to handle the video track fine, regardless of whether you remove the pulldown or not. With pulldown removed it should even be better. But in pratical life things are often different than theory suggests...

[wolfbane5]

Originally Posted by madshi

If you own a proper Surcode DTS encoder license and if you have installed it properly, these folders are supposed to be there. So my suggestion would be to uninstall, get the latest version from Surcode and install it. Please note that AFAIK there are two different products: One for CD (stereo) encoding and one for movie encoding. eac3to only supports the latter one...

It's possible that I don't have those folders since I'm using v1.0.23 which I believe is not the latest version. However, when I fire up the program it does say 'Surcode DVD-Pro DTS Encoder' so I'm assuming I'm using the movie encoding one. Truth be told, I borrowed SurCode and ArcSoft from a buddy of mine who's a sound engineer since he hardly uses them. I wanted to see if they'd actually work for my purposes, but apparently they're throwing some curve balls. By the way, what about ArcSoft? Where do I find that in the registry? I'd like to be able to use '-core' in eac3to.

[madshi]

Originally Posted by wolfbane5

It's possible that I don't have those folders since I'm using v1.0.23 which I believe is not the latest version.

v1.0.23 is supposed to work. Try reinstalling with admin rights...

Originally Posted by wolfbane5

By the way, what about ArcSoft? Where do I find that in the registry? I'd like to be able to use '-core' in eac3to.

The "-core" switch doesn't need ArcSoft. You only need ArcSoft for best quality decoding...

[madshi]

eac3to v2.85 released

http://madshi.net/eac3to.zip

* using "eac3to source video.h264" doesn't demux audio/subtitle tracks, anymore
* using "eac3to source movie.*" demuxes video, audio and subtitle tracks
* using "eac3to source 1: video.* 2: audio.*" demuxes the specified tracks
* AC3 and E-AC3 dialnorm removal now uses "-31db" instead of "-0db"
* workaround for DTS files where last byte is missing in each audio frame
* fixed: v2.84 sometimes crashed when parsing HD DVD XML files
* fixed: v2.84 sometimes chose incorrect XML file
* fixed: v2.84 sometimes chose wrong m2ts playlist file
* fixed: some actions were eventually applied twice when "-2pass" was used
* fixed: AAC encoding quality "quality=0.0x" was passed to Nero as "0.x"

On a side note, I looked into making Nero 9 DirectShow decoders work with eac3to, but I gave up on it because of these reasons:

(1) AC3 decoding works, but gives slightly different results compared to Nero 7 and Sonic.
(2) E-AC3 decoding is not supported at all, as far as I can say (which kinda makes sense, since Nero 9 doesn't support HD DVD, anymore).
(3) The decoder didn't accept my TrueHD data. Not sure, maybe they changed something and maybe it would be possible to make it work again with more effort.

All in all I just don't see the value of making Nero 9 work, since E-AC3 support seems to be missing, anyway. Instead I'll try to make the ArcSoft (E-)AC3 decoder work, sooner or later...

[mrr19121970]

Great, Atonment HD-DVD works again now.

Thanks.