eac3to - audio conversion tool [Archive] - Page 135

rebkell

26th October 2008, 16:06

Correct, you only need to rerun those tracks which you want to have gap/overlap-fixed. It's just easier to rerun the same command line. Also it would be difficult to explain what you need to do exactly to save space. That's why eac3to simply asks to rerun the same command line.

You really can't save space can you? You can run the same command line without the video, but on the first pass you still have to have the video included to fix the gaps, so you have to have enough space on the first pass to hold the video.

There isn't anyway to fix the gaps without demuxing the video, at least on the first pass, is there? If there is I'd be very interested in knowing how to do it.

rica

26th October 2008, 16:44

There isn't anyway to fix the gaps without demuxing the video, at least on the first pass, is there?

"0n the first pass" correct,
"without demuxing the video" incorrect.

Boulder

26th October 2008, 17:35

I've searched for comparisons but found none, except the infinitewave site you linked to. Where did you read that people prefer SSRC over r8brain?There are no exact comparisons, I'm afraid. Reading through posts at hydrogenaudio.org shows that SSRC is recommended most of the time. This topic particularly points out one problem with r8brain: http://www.hydrogenaudio.org/forums/index.php?showtopic=64989&hl=r8brain (see post 7).

By the way, KikeG recommends these command lines for SSRC: http://www.hydrogenaudio.org/forums/index.php?showtopic=7133 (see post 8)

madshi

26th October 2008, 17:50

You really can't save space can you? You can run the same command line without the video, but on the first pass you still have to have the video included to fix the gaps, so you have to have enough space on the first pass to hold the video.

There isn't anyway to fix the gaps without demuxing the video, at least on the first pass, is there? If there is I'd be very interested in knowing how to do it.
I don't understand what you're aiming it. Either you want the video or you don't want it. If you want it, you need the space on the harddisk. If you don't want it, then simply don't ask eac3to to demux it!

There are no exact comparisons, I'm afraid. Reading through posts at hydrogenaudio.org shows that SSRC is recommended most of the time. This topic particularly points out one problem with r8brain: http://www.hydrogenaudio.org/forums/index.php?showtopic=64989&hl=r8brain (see post 7).
Thanks. Well, the guys talking about r8brain there are just looking at the graphs and comment on that. That still doesn't fully satisfy me. But anyway, SSRC is default in eac3to now, while r8brain can still optionally be used. So everybody should be happy.

By the way, KikeG recommends these command lines for SSRC: http://www.hydrogenaudio.org/forums/index.php?showtopic=7133 (see post 8)
The settings listed there are mainly for dithering. That doesn't apply here because eac3to fetches the full floating point processing result from SSRC. So SSRC's dithering functionality isn't used at all.

Boulder

26th October 2008, 18:13

The settings listed there are mainly for dithering. That doesn't apply here because eac3to fetches the full floating point processing result from SSRC. So SSRC's dithering functionality isn't used at all.What about eac3to's own dithering, could the method KikeG describes for dither type 3 be used when resampling:

ssrc --rate 44100 --bits 16 --dither 3 --pdf 1 0.7 infile.wav outfile.wav

rebkell

26th October 2008, 18:20

I don't understand what you're aiming it. Either you want the video or you don't want it. If you want it, you need the space on the harddisk. If you don't want it, then simply don't ask eac3to to demux it!

What I'm aiming at, which may not be feasible anyway, is that I do a lot of capturing with the hauppauge hdpvr unit and it tends to have drop outs, which can cause sync problems(mostly gaps). I use neuron2's DGAVCDec to index the video and then re-encode with x264, so in essence(I think it will work out) I will be creating a video file straight from the .ts file, but a straight demux of the audio doesn't always match up, because of the gaps and such when it has the momentary dropouts.

The way it is now, I have to create the mkv file, then demux that to .h264 before DGAVCDec can handle the video, but it can handle the video directly from the .ts file.

73ChargerFan

26th October 2008, 21:45

I don't think MKV supports such a kind of thing in any reasonable way.
Actually, it does and it works perfectly. The pieces are stored as separate mkv files, then are linked. Haali's muxer & splitter can handle them.
VERY SLICK.

Check out these pages:
101 things you never knew you could do with Matroska (http://www.mod16.org/hurfdurf/?p=8)
MKV Ordered Chapters: User's Manual (http://forums.animesuki.com/showthread.php?t=66444)

lchiu7

26th October 2008, 23:17

Since your captured streams are h.264, you'd have to use tools like TSPE, found at http://www.bitstreamtools.com, to cut the file, i.e., remove the commercials and leave only the 2.0 or 5.1 programming which will help keep things in sync.

I use TSPE to edit out the ads. It works fine. The resultant file plays fine and in sync.

But I try to recompress the video (extracted from the cut file) and mux it back with original audio, it goes out of sync. And Txremuxer (and MKVMerge) report bad AC-3 frames in the stream.

lchiu7

26th October 2008, 23:32

these streams are invalid because they switch from 2.0 to 5.1 and back and forth?

What program are you using to extract the audio with? I've had plenty of sync problems and those are a pain to deal with, but try dgavcdec to extract the ac3 stream with, if it can be extracted I wouldn't expect any problems muxing it back in.

Sorry madshi, I know this is off topic.

No - there are two streams in the audio AC-3 (DD 5.1) and AAC-HE LATM. Txmuxer just ignores the AAC audio (says invalid) and dgavcindex offers the choice to demux either (I always choose AC-3).

Think as you say we are getting a bit OT. The on topic part is"

1. demux the audio out using tsremux or dgavcinxex
2. mux it back with the original video

Get AC-3 CRC frame errors and out of sync

Try to fix with eacc3to and it says bad data or whatever, use delaycut to fix. Try do do that and no matter which option I choose with delaycut, (skip, fix, silence), while no more errors, muxed audio/video is out of sync

madshi

26th October 2008, 23:34

What I'm aiming at, which may not be feasible anyway, is that I do a lot of capturing with the hauppauge hdpvr unit and it tends to have drop outs, which can cause sync problems(mostly gaps). I use neuron2's DGAVCDec to index the video and then re-encode with x264, so in essence(I think it will work out) I will be creating a video file straight from the .ts file, but a straight demux of the audio doesn't always match up, because of the gaps and such when it has the momentary dropouts.

The way it is now, I have to create the mkv file, then demux that to .h264 before DGAVCDec can handle the video, but it can handle the video directly from the .ts file.
eac3to "fixes" gaps/overlaps in video tracks by setting the timestamps in the MKV container accordingly. If you demux the video track from the MKV container the gap/overlap fixes get lost.

Actually, it does and it works perfectly. The pieces are stored as separate mkv files, then are linked. Haali's muxer & splitter can handle them.
VERY SLICK.

Check out these pages:
101 things you never knew you could do with Matroska (http://www.mod16.org/hurfdurf/?p=8)
MKV Ordered Chapters: User's Manual (http://forums.animesuki.com/showthread.php?t=66444)
Cool. Didn't know that!

However, I'm using the Haali Matroska Muxer and while the Haali Splitter might support all these fancy features, I don't think the Haali Muxer does. At least I wouldn't know how to activate and control these features. Also this sounds like a really tough job for a muxer and so I don't plan to seriously look into this.

nwg

26th October 2008, 23:50

rebkell

26th October 2008, 23:53

eac3to "fixes" gaps/overlaps in video tracks by setting the timestamps in the MKV container accordingly. If you demux the video track from the MKV container the gap/overlap fixes get lost.

Ahh, Ok, I didn't realize that, I always thought you actually padded the audio stream or removed some it to fit.

madshi

27th October 2008, 00:01

eac3to v2.70 released

http://madshi.net/eac3to.zip

* added floating point support to the complete audio processing chain
* added gain functionality, e.g. "-3db" or "+1db"
* bitdepth analyzation is now done separately for each channel
* fixed: when decoding lossy audio with libav, peaks were clipped incorrectly
* fixed: libav MP1/2/3 decoder output was cut down to 24bit
* fixed: with some EVO sources the AC3 track was not listed
* fixed: if no key frame was found, h264 track in m2(ts) was not listed
* fixed: video/audio data before first PAT/PMT was discarded
* Blu-Ray chapters now don't contain link points, anymore, unless necessary
* added 10db boost to LFE channel, when "-down2" and "-mixlfe" are used
* ArcSoft output can now be overwritten to "-2", "-6", "-7" or "-8" channels
Some more detailed comments:

* added floating point support to the complete audio processing chain
Some more related changes coming in a future build. E.g. currently I'm still passing 24bit integer to audio encoders. In a future build I'll pass floating point, whereever possible.

The changes I've done so far seem to work just fine, but there might be some new bugs.

* bitdepth analyzation is now done separately for each channel
I've checked about 20 tracks but found none where any of the channels differed from the others. If all channels share the same max bitdepth, eac3to will report the statistics as it always did. However, if some of the channels have different max bitdepth values, eac3to will now report the bitdepth statistics seperately for every channel group which has identical attributes.

* ArcSoft output can now be overwritten to "-2", "-6", "-7" or "-8" channels
This one is for xkodi and especially for tebasuna51. You can ask the ArcSoft decoder to output 2, 6, 7 or 8 channels, regardless of which channel configuration the source has. Interesting tests would be:

2.0 -> 6
2.0 -> 7
2.0 -> 8
5.1 -> 2
5.1 -> 7
5.1 -> 8
6.1 -> 2
6.1 -> 6
6.1 -> 8
7.1 -> 2
7.1 -> 6
7.1 -> 7

So there's a lot to test. Maybe we can figure out how down- and upmixing should ideally be done. But the big question is: Does the ArcSoft decoder do all that in an ideal way? Maybe the first check should be to see what 5.1 -> 2 does? If that uses proper Dolby Pro Logic II then maybe that's a good sign that the general up-/downmixing quality should be ok?

madshi

27th October 2008, 00:03

What does sync byte missing mean? I have been getting it on a few discs when trying to convert TruHD to Ac3 and also DTS to Ac3.

I have been using several GUI's and it looks like it is eac3to itself.
It means that the source file is corrupt. Try reripping the disc. If that doesn't help the disc might have an authoring fault.

Ahh, Ok, I didn't realize that, I always thought you actually padded the audio stream or removed some it to fit.
It does actually change the audio stream. I was talking about the VIDEO stream.

rebkell

27th October 2008, 00:12

It does actually change the audio stream. I was talking about the VIDEO stream.

Good, I was wondering how I had done any encodes with audio gaps in them. I guess that all the gap/overlap problems I've had must have been corrected with the audio portion of the program.

nwg

27th October 2008, 01:13

It means that the source file is corrupt. Try reripping the disc. If that doesn't help the disc might have an authoring fault.

I have had it on the last three blu rays and tried re-ripping and still does it. It plays fine in the BD player.

BlackJack1

27th October 2008, 02:21

Tried to remux to mkv "The Fugitive" HD-DVD:
eac3to v2.70
command line: eac3to\eac3to K:\HD\BD_THE_FUGITIVE_HDDVD\HVDVD_TS 1) 3: o:\ fugitive.mkv 4: o:\fugitive.ac3 2: o:\fugitive_chpts.txt
------------------------------------------------------------------------------
EVO, 1 video track, 4 audio tracks, 5 subtitle tracks, 2:10:14
"Feature Presentation"
1: Joined EVO file
2: Chapters, 42 chapters without names
3: VC-1, 1080p24 /1.001 (16:9) with pulldown flags
4: E-AC3, English, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB
5: E-AC3, French, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB
6: E-AC3, German, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB
7: E-AC3, English, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB
"Commentary"
8: Subtitle, English
9: Subtitle, French
10: Subtitle, German
11: Subtitle, English, "SDH"
12: Subtitle, German, "SDH"
This video conversion is not supported.
What I'm doing wrong...?

rebkell

27th October 2008, 02:25

Tried to remux to mkv "The Fugitive" HD-DVD:

What I'm doing wrong...?

It looks like you have a space after the first o:\

command line: eac3to\eac3to K:\HD\BD_THE_FUGITIVE_HDDVD\HVDVD_TS 1) 3: o:\ fugitive.mkv 4: o:\fugitive.ac3 2: o:\fugitive_chpts.txt

BlackJack1

27th October 2008, 02:29

Oooo shit...
Thx. My fault :(

rica

27th October 2008, 02:37

It looks like you have a space after the first o:\

command line: eac3to\eac3to K:\HD\BD_THE_FUGITIVE_HDDVD\HVDVD_TS 1) 3: o:\ fugitive.mkv 4: o:\fugitive.ac3 2: o:\fugitive_chpts.txt

Yes;
"o:\ fugitive.mkv " must have been:
"o:\fugitive.mkv "

This is one of the most common errors which might be done.

Simple but easy to be made.

(I've made lots of similars.)

itsancho

27th October 2008, 03:04

oops, just to report some small bug(s) in 2.70
eac3to 2.69
"K:\Mannen som elsket Yngve (2008) (The Man Who Loved Ynge) (NOR)Blu-ray AVC dts-HD Hi-Res\MANNEN SOM ELSKET YNGVE" 1)
M2TS, 1 video track, 4 audio tracks, 4 subtitle tracks, 1:34:23
1: Chapters, 13 chapters
2: h264/AVC, 1080i50 (16:9)
3: AC3, Norwegian, 5.1 channels, 448kbps, 48khz, dialnorm: -27dB
4: DTS Hi-Res, Norwegian, 5.1 channels, 16 bits, 1774kbps, 48khz, dialnorm: -4dB
5: AC3, Norwegian, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB
6: AC3, Norwegian, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB
7: Subtitle (PGS), Danish
8: Subtitle (PGS), Norwegian
9: Subtitle (PGS), Swedish
10: Subtitle (PGS), Finnish

eac3to 2.70
"K:\Mannen som elsket Yngve (2008) (The Man Who Loved Ynge) (NOR)Blu-ray AVC dts-HD Hi-Res\MANNEN SOM ELSKET YNGVE" 1)
TS, 1 video track, 4 audio tracks, 4 subtitle tracks, 1:34:23
1: h264/AVC, 1080i50 (16:9)
2: AC3, 5.1 channels, 448kbps, 48khz, dialnorm: -27dB
3: DTS Hi-Res, 5.1 channels, 16 bits, 1774kbps, 48khz, dialnorm: -4dB
4: AC3, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB
5: AC3, 2.0 channels, 192kbps, 48khz, dialnorm: -27dB
6: Subtitle (PGS)
7: Subtitle (PGS)
8: Subtitle (PGS)
9: Subtitle (PGS) and 10x once again 4 the great tool, madshi!

jeffnoone

27th October 2008, 05:30

Extracting from Dave Matthews BR disc, which has 5.1 AC3, PCM 2-ch, and TrueHD 5.1

Used eac3to 2.69 to extract the VC1 and all three audio tracks in one run; used DOS commandline, and -i 24 -c 6 -s 48000

For the TrueHD, I just extracted using the stdout technique, and saved as 6-channel PCM

Anyway tsMuxeR will import and correctly recognize the video and all three audio tracks (after I ran the 2-channel PCM thru PCM2TSMU)

But there is some problem with the BR files I produce
When running tsMuxer, it says at one point about processing a certain number of video frames (as if it has finished video early)

When you let thing complete, there are no declared errors, but PowerDVD reports as 5.38.04 i length (way too long), and video seems to "run out" about half way thru.
The AC3 track plays fine and is synced
There is sound on the 6-ch PCM but it appears very slow

I have successfully extracted 6 and 8 channel PCM before from other formats including TrueHD, but something seems to have gone wrong here - any ideas

Thanks

Unhandled

27th October 2008, 06:02

First of all, thank you madshi. eac3to is a great piece of software. :)

I believe I found a regression in downmixing. Seems to have been introduced in v2.64. When I try to downmix a TrueHD 5.1 track to 2.0 (-down2), eac3to reports it can't downmix this channel configuration. It worked in v2.63:

C:\Video Stuff\eac3to_263>eac3to.exe f:\content\hddvd\HALO_22\HVDVD_TS 1) 4: ..\test.wav -down2
EVO, 1 video track, 3 audio tracks, 1 subtitle track, 1:32:28
"WinterTour"
1: Chapters, 21 chapters with names
2: VC-1, 1080p30 /1.001 (16:9)
3: AC3, English, 5.1 channels, 448kbps, 48khz
4: TrueHD, English, 5.1 channels, 48khz
5: AC3, English, 2.0 channels, 256kbps, 48khz
6: Subtitle, English
[a04] Extracting audio track number 4...
[a04] Decoding with libav/ffmpeg...
[a04] Downmixing multi channel audio to stereo...
Loading white noise (needed for dithering)...
[a04] Writing WAV...
[a04] This audio track contains more than 16 bits of information.
[a04] Creating file "..\test.24bit.wav"...
^C

C:\Video Stuff\eac3to_264>eac3to.exe f:\content\hddvd\HALO_22\HVDVD_TS 1) 4: ..\test.wav -down2
EVO, 1 video track, 3 audio tracks, 1 subtitle track, 1:32:28
"WinterTour"
1: Chapters, 21 chapters with names
2: VC-1, 1080p30 /1.001 (16:9)
3: AC3, English, 5.1 channels, 448kbps, 48khz
4: TrueHD, English, 5.1 channels, 48khz
5: AC3, English, 2.0 channels, 256kbps, 48khz
6: Subtitle, English
[a04] Can't downmix this channel configuration.
[a04] Internal error - unknown audio output format!

Same problem occurs when trying to downmix 5.1 WAV or 5.1 FLAC file created by eac3to v2.64 from the TrueHD track.
My source is an HD-DVD folder. The title is "Nine Inch Nails - Beside You In Time". Can anyone else confirm this behavior on similar tracks?
Thank you.

svgame

27th October 2008, 06:13

madshi i want you help me
In order to import TrueHD stream from BD to Scenarist HDMV you have to do the following:
- run TrueHD track via eac3to twice:
1. eac3to track.thd audio1.ac3
2. eac3to track.thd audio1.thd
Then rename audio1.thd to .mlp and feed MUI Generator with audio1.ac3 specifying it as Dolby Lossless.
In next window be sure, but at last the error message"External component has thrown an exception"
please help me``madshi

madshi

27th October 2008, 08:46

oops, just to report some small bug(s) in 2.70
and 10x once again 4 the great tool, madshi!
Ouch. Thanks, will look into this.

But there is some problem with the BR files I produce
When running tsMuxer
This is not the tsMuxeR support thread.

I believe I found a regression in downmixing. Seems to have been introduced in v2.64. When I try to downmix a TrueHD 5.1 track to 2.0 (-down2), eac3to reports it can't downmix this channel configuration.
Thanks for the report. Problem will be fixed in the next build.

In order to import TrueHD stream from BD to Scenarist HDMV you have to do the following
No. Just do "eac3to track.thd audio1.thd+ac3".

madshi

27th October 2008, 09:01

eac3to v2.71 released

http://madshi.net/eac3to.zip

* fixed: v2.70 detected Blu-Rays as "TS" without chapters and track languages
* fixed: TrueHD downmixing to 2.0 didn't work

evdberg

27th October 2008, 10:44

eac3to v2.70 released

* bitdepth analyzation is now done separately for each channel
Thanks!

I've checked about 20 tracks but found none where any of the channels differed from the others. If all channels share the same max bitdepth, eac3to will report the statistics as it always did. However, if some of the channels have different max bitdepth values, eac3to will now report the bitdepth statistics seperately for every channel group which has identical attributes.

I retested the same Universal movie and again got the same result:
"Original audio track: max 24 bits, average 19 bits, most common 16 bits."
So this means that the result is valid for all channels, and not some channels 16 bits and other 24 bits.

shambles

27th October 2008, 10:49

* bitdepth analyzation is now done separately for each channel
I've checked about 20 tracks but found none where any of the channels differed from the others. If all channels share the same max bitdepth, eac3to will report the statistics as it always did. However, if some of the channels have different max bitdepth values, eac3to will now report the bitdepth statistics seperately for every channel group which has identical attributes.

doesn't seem to work.. i checked out a few tracks where the channel bit depths differ and eac3to still reports the same it did before

C:\Program Files\eac3to>eac3to e:\inside.flac -analyzebitdepth
FLAC, 5.1 channels, 2:08:44, 24 bits, 2097kbps, 48khz
Decoding FLAC...
Original audio track: max 24 bits, average 19 bits, most common 16 bits.
eac3to processing took 2 minutes, 18 seconds.
Done.

C:\Program Files\eac3to>eac3to e:\inside.flac f:\inside.wavs
FLAC, 5.1 channels, 2:08:44, 24 bits, 2097kbps, 48khz
Decoding FLAC...
Writing WAVs...
Creating file "f:\inside.LFE.wav"...
Creating file "f:\inside.R.wav"...
Creating file "f:\inside.C.wav"...
Creating file "f:\inside.L.wav"...
Creating file "f:\inside.SL.wav"...
Creating file "f:\inside.SR.wav"...
Original audio track: max 24 bits, average 19 bits, most common 16 bits.
eac3to processing took 3 minutes, 15 seconds.
Done.

C:\Program Files\eac3to>eac3to f:\inside.c.wav -analyzebitdepth
WAV, 1.0 channels, 2:08:44, 24 bits, 1152kbps, 48khz
Reading WAV...
The original audio track has a constant bit depth of 16 bits.
eac3to processing took 22 seconds.
Done.

C:\Program Files\eac3to>eac3to f:\inside.l.wav -analyzebitdepth
WAV, 1.0 channels, 2:08:44, 24 bits, 1152kbps, 48khz
Reading WAV...
The original audio track has a constant bit depth of 16 bits.
eac3to processing took 22 seconds.
Done.

C:\Program Files\eac3to>eac3to f:\inside.r.wav -analyzebitdepth
WAV, 1.0 channels, 2:08:44, 24 bits, 1152kbps, 48khz
Reading WAV...
The original audio track has a constant bit depth of 16 bits.
eac3to processing took 22 seconds.
Done.

C:\Program Files\eac3to>eac3to f:\inside.lfe.wav -analyzebitdepth
WAV, 1.0 channels, 2:08:44, 24 bits, 1152kbps, 48khz
Reading WAV...
The original audio track has a constant bit depth of 16 bits.
eac3to processing took 28 seconds.
Done.

C:\Program Files\eac3to>eac3to f:\inside.sl.wav -analyzebitdepth
WAV, 1.0 channels, 2:08:44, 24 bits, 1152kbps, 48khz
Reading WAV...
The original audio track has a constant bit depth of 24 bits.
eac3to processing took 25 seconds.
Done.

C:\Program Files\eac3to>eac3to f:\inside.sr.wav -analyzebitdepth
WAV, 1.0 channels, 2:08:44, 24 bits, 1152kbps, 48khz
Reading WAV...
The original audio track has a constant bit depth of 24 bits.
eac3to processing took 30 seconds.
Done.

sample, just in case you need one: http://www.sendspace.com/file/fvrzoi

madshi

27th October 2008, 11:18

doesn't seem to work.. i checked out a few tracks where the channel bit depths differ and eac3to still reports the same it did before
Thanks for the sample. There was a bug in the bitdepth analyzer's channel splitter. Will be fixed in the next build. And this is what I'm now getting for your sample and also for Notting Hill:

eac3to "Notting Hill (1999)\Englisch.flac" -analyzebitdepth -50mb
FLAC, 5.1 channels, 2:04:03, 24 bits, 1842kbps, 48khz
Decoding FLAC...
Original audio track, L+R+C+LFE: constant bit depth of 16 bits.
Original audio track, SL+SR: constant bit depth of 24 bits.
eac3to processing took 5 seconds.
Done.

shambles

27th October 2008, 11:21

awesome, thanks! :D

madshi

27th October 2008, 11:22

BlackJack1

27th October 2008, 12:12

ACrowley

27th October 2008, 12:20

eac3to v2.72 released

http://madshi.net/eac3to.zip

* fixed: per channel bitdepth analyzation didn't work correctly
Enough updates now! Next build will have to wait until next week, unless some serious new bugs show up...

WOW ! Nice now i can get 7Ch Output from, DTS ES 6.1 Matrix Tracks.
I made a few Tests with DTS ES 6.1 Matrix and it looks right to me ?! The BC output from arcsoft -7 looks/sound Ok

Thunderbolt8

27th October 2008, 12:21

* fixed: when decoding lossy audio with libav, peaks were clipped incorrectly
* fixed: video/audio data before first PAT/PMT was discarded
2 questions regarding these: did those incorrectly clipped peaks result in loss of quality, would we have to redo these tracks?
the same for the PAT/PMT thing, did this result in some minor loss of quality or possible syncing mistakes?
thanks!

tebasuna51

27th October 2008, 13:23

...
Used eac3to 2.69 to extract the VC1 and all three audio tracks in one run; used DOS commandline, and -i 24 -c 6 -s 48000
...
There is sound on the 6-ch PCM but it appears very slow

Please put your eac3to log files and exact command lines used at this thread (http://forum.doom9.org/showthread.php?t=138417)

deank

27th October 2008, 13:46

Is there a way to make eac3to.exe to treat parameters enclosed in quotes?

So a command line: eac3to.exe "test.dts" "test.ac3" "-448" to be processed correctly?

i.e. if it is possible to make it recognise "-parameters" with quotes...

Thanks for this great tool!

madshi

27th October 2008, 13:53

I've got Sonic audio decoder 4.3.0.169 and Sonic DTS encoder 1.0.23.0 and would like to remux video and DTS-ES 6.1 24bit 48kHz 1536kbps track. Do I need Arcsoft DTS decoder or my Sonic will do the job?
You don't need any decoders for demuxing. You only need decoders for decoding.

WOW ! Nice now i can get 7Ch Output from, DTS ES 6.1 Matrix Tracks.
I made a few Tests with DTS ES 6.1 Matrix and it looks right to me ?! The BC output from arcsoft -7 looks/sound Ok
I'll wait for tebasuna51's check... :)

2 questions regarding these: did those incorrectly clipped peaks result in loss of quality, would we have to redo these tracks?
If you had a lossy (!) track which you decoded with libav and which had very high peaks then yes, the new eac3to version outputs better quality. However, this doesn't apply to TrueHD. Also it doesn't apply to Nero or Sonic decoders.

the same for the PAT/PMT thing, did this result in some minor loss of quality or possible syncing mistakes?
No sync mistakes. And no loss of quality. Just a loss of a few video frames at the beginning of the file. And this should only occur with broadcasts, if at all. It's not a very important fix.

Is there a way to make eac3to.exe to treat parameters enclosed in quotes?
What would you need that for?

PHD_1976

27th October 2008, 14:01

did those incorrectly clipped peaks result in loss of quality, would we have to redo these tracks?would we have to redo these tracks?

Madshi's already answered, just want to add a few words:

These peak overflow gives a crackle noise and its clearly audible. I found this with libav decoder and DD 5.1 EX track. So if you can hear crackles while watching a movie I guess you should redo your tracks. :-)

This evening I gonna check new build and report about clipping.

To Madshi:
1. Thanx a lot for a new build
2. Is there a way with eac3to to demux in 1 step TrueHD track directly from BD/HD into ac3 core and lossless part (for Scenarist)?
I tried it with HD DVD yesterday and i made 2 steps: 1) thd+ac3 track creation 2) demuxing to ac3 and mlp. It's not a problem for me, just curious.

BlackJack1

27th October 2008, 14:53

You don't need any decoders for demuxing. You only need decoders for decoding.
I need to remux no. 4: DTS-ES 6.1 track 1536k and 3: DD 5.1EX 640k track.
Here is the log:
eac3to v2.72
command line: eac3to "O:\# HD New\The_Untouchables_Blu-ray_1987-PerfectionHD\BDMV" 1) 1: o:\untouchables.txt 2: o:\untouchables.mkv 4: o:\untouchables.dts 3: o:\untouchables.ac3
------------------------------------------------------------------------------
M2TS, 1 video track, 4 audio tracks, 4 subtitle tracks, 1:59:27
1: Chapters, 24 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: AC3, English, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB
4: DTS-ES, English, 6.1 channels, 24 bits, 1536kbps, 48khz, dialnorm: -4dB
5: AC3, French, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB
6: AC3, Spanish, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB
7: Subtitle (PGS), English
8: Subtitle (PGS), English
9: Subtitle (PGS), French
10: Subtitle (PGS), Spanish
Creating file "o:\untouchables.txt"...
[v02] Extracting video track number 2...
[a04] Extracting audio track number 4...
[a03] Extracting audio track number 3...
[a04] Removing DTS dialog normalization...
[a03] Removing AC3 dialog normalization...
[v02] Muxing video to Matroska...
[a04] Creating file "o:\untouchables.dts"...
[a03] Creating file "o:\untouchables.ac3"...
Added fps value to MKV header.
Video track 2 contains 171840 frames.
eac3to processing took 37 minutes, 8 seconds.
Done.

but... PDVD reports that I've got DTS-ES 5.1 1536k track and not 6.1...
Previously during playing BD movie using the same PDVD - reported: DTS-ES 6.1 1536k.
Please explain what I did wrong.

ACrowley

27th October 2008, 15:05

I'll wait for tebasuna51's check..

@madshi

Arcsoft -7 on DTS ES 6.1 "Matrix" looks very good !

I made a Test :

DTS ES 6.1 "Discrete" to .wavs with eac3to/Arcsoft. Then i made a reencode from these 7 mono waves to DTS ES 6.1 "Matrix" with DTS Pro Series.
After that i made a decoding from the new DTS ES 6.1 "Matrix" with eac3to/Arcsoft Decoder and -7

Then i compared both Waves from the Back Center Channels (the BC from original discrete and BC from the Matrix reencode) in Audacity. They look fully identical :)

http://www.abload.de/thumb/sshot-1ytcb.png (http://www.abload.de/image.php?img=sshot-1ytcb.png)
http://www.abload.de/thumb/sshot-25i2g.png (http://www.abload.de/image.php?img=sshot-25i2g.png)

itsancho

27th October 2008, 15:43

btw, madshi, is there any chance eac3to to report if AC3 track is DD-EX?
10x again and again...
;-)

svgame

27th October 2008, 16:06

thank you Madshi,just there has a problem.i make 3 step
1) use eac3to demux TrueHD track directly from BD/HD
2) type command "eac3to 00001.thd audio.thd" "eac3to 00001.thd audio.ac3",demuxing to ac3 core and lossless part
3) rename audio.thd to audio.mlp,and import to Scenarist.at 100%,its hung!!!about 2 minute，get a error message "External component has thrown an exception"
where is problem?Madshi

tebasuna51

27th October 2008, 17:45

Originally Posted by ACrowley
WOW ! Nice now i can get 7Ch Output from, DTS ES 6.1 Matrix Tracks.
I made a few Tests with DTS ES 6.1 Matrix and it looks right to me ?! The BC output from arcsoft -7 looks/sound Ok

I'll wait for tebasuna51's check... :)

Not all Ok here.
If we name source channels like Rear: RL, RR, RC and

RL' = +90º phase shift RL
RR' = -90º phase shift RR

seems the phase shift is make at encoder phase and is the same for 'matrix' and 'discrete' DTS-ES in all the tests.

The output channels after encode to 3/3.1 DTS-ES discrete and decoded with ArcSoft:

1) 3/3.1 -> 5.1 (like previous test)

SL = RL' + 0.71 x RC
SR = RR' + 0.71 x RC
Ok

2) 3/3.1 -> 6.1

BC = RC
SL = RL'
SR = RR'
Ok

3) 3/3.1 -> 7.1

SL = 0.40 x RL' + 0.60 x RC
SR = 0.40 x RR' + 0.60 x RC
BL = 0.75 x RL'
BR = 0.75 x RR'
Seems wrong: RC only in Side and most Rear part in Back, I think must be the other way, Side <-> Back

----------------------------
The output channels after encode to 3/3.1 DTS-ES matrix and decoded with ArcSoft:

1) 3/3.1 -> 5.1

SL = RL' + 0.71 x RC
SR = RR' + 0.71 x RC
Ok

2) 3/3.1 -> 7.1

SL = 0.71 x RL' + 0.50 x RC
SR = 0.71 x RR' + 0.50 x RC
BL = 0.71 x RL' + 0.50 x RC
BR = 0.71 x RR' + 0.50 x RC
For me seems ArcSoft can't recover the RC mixed in RL-RR because RC must be present only in Back channels.

3) 3/3.1 -> 6.1, eac3to crash with this bugreport.txt (http://pastebin.com/m276c45a0)

The report was made with 2.70 but is the same with 2.72, I send the 2.70 bug because was made after a reset with the minimum of soft running.