eac3to - audio conversion tool [Archive] - Page 107

EPiPH0NE

28th June 2008, 03:20

madshi, i've solved the problem:

the only different file in C:\Program Files\Common Files\ArcSoft\MPEG Engine

between the "Intro version" and the "Trial version" is: ArcDemux.ax

eac3to doesn't work with from the "Trial version".

P.S. both ArcDemux.ax are version 1.0.2.68, the working one has MD5 sum: abf8200f9d35b7be9dee79ff23a4ba52 and the broken one is with MD5 sum ef37767a2bd8faa20b13722a05ec2ae5. maybe the difference is because of the registration capabilities of the "Trial version".

I am also having this problem:

P:\>P:\eac3to_2.49\eac3to.exe -test
Nero Audio Decoder (Nero 7 or older) works fine
Sonic Audio Decoder (4.3.0.169) works fine
ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) is not working correctly
Haali Media Splitter (2007-11-18) is installed
Surcode DTS Encoder (1.0.23.0) is installed
MkvToolnix (v2.2.0) is installed

P:\>P:\eac3to_2.49\eac3to.exe -test
Sonic Audio Decoder (4.3.0.169) works fine
ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) is not working correctly
Nero Audio Decoder is not working correctly
Haali Media Splitter (2007-11-18) is installed
Surcode DTS Encoder (1.0.23.0) is installed
MkvToolnix (v2.2.0) is installed

P:\>P:\eac3to_2.49\eac3to.exe -test
Nero Audio Decoder (Nero 7 or older) works fine
Sonic Audio Decoder (4.3.0.169) works fine
ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) is not working correctly
Haali Media Splitter (2007-11-18) is installed
Surcode DTS Encoder (1.0.23.0) is installed
MkvToolnix (v2.2.0) is installed

How exactly did you solve it? I have 1.0.6.28 + update -> 1.0.6.29. Can you upload the proper ArcDemux.ax and PM me the link? Would I just have to replace this file? I don't care about the actual app I'm only installing it for the filters so if it screws up that app I don't care.

xkodi

28th June 2008, 06:30

I am also having this problem:

How exactly did you solve it? I have 1.0.6.28 + update -> 1.0.6.29. Can you upload the proper ArcDemux.ax and PM me the link? Would I just have to replace this file? I don't care about the actual app I'm only installing it for the filters so if it screws up that app I don't care.

it turns out that the root cause of the problem is not the ArcDemux.ax, i am still searching for the cause and what really fix the problem on my computer, but i am not sure. however, i think it is something to do with the following DLLs:

MagCore.dll
MagUIInter.dll
MagUIEngine.dll
MagPCMac.dll

so, please, copy them from "C:\Program Files\ArcSoft\TotalMedia Extreme" to the directory where your eac3to.exe is, try again with "eac3to.exe -test" and report back the results.

@madshi

so, until now 3 problems with 2.49:

1) http://forum.doom9.org/showthread.php?p=1153102#post1153102
2) "ArcSoft Audio Decoder is not working correctly" - can't find what exactly fix initially the problem, but still it is here and has random effect in one minute it is there in the next it is gone
3) "ArcSoft Audio Decoder doesn't seem to be installed" - on my other computer, no matter what i do eac3to reports this error

madshi, please, tell more about how eac3to test for 2) and 3), so we can figure out what is wrong.

itsancho

28th June 2008, 07:44

i think SP3 is doing some shit...
i did exactly the same installation of ArcSoft TotalMedia Extreme, first 1.0.6.28 then 1.0.6.29 update on 2 PCs, one with SP2 and the other with SP3, guess what...

with SP2 :eac3to -test
Nero Audio Decoder (Nero 7 or older) works fine
Sonic Audio Decoder (4.3.0.169) works fine
ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) works fine
Haali Media Splitter (2008-3-29) is installed
Surcode DTS Encoder (1.0.29.0) is installed
MkvToolnix (v2.2.0) is installed with SP3 eac3to -test
ArcSoft Audio Decoder (2.49.0.0, 2.49.0.0) doesn't seem to be installed
Nero Audio Decoder (Nero 7 or older) works fine
Sonic Audio Decoder (4.3.0.169) works fine
Haali Media Splitter (2008-3-29) is installed
Surcode DTS Encoder (1.0.29.0) is installed
MkvToolnix (v2.2.0) is installed

xkodi

28th June 2008, 07:48

i think SP3 is doing some shit...
i did exactly the same installation of ArcSoft TotalMedia Extreme, first 1.0.6.28 then 1.0.6.29 update on 2 PCs, one with SP2 and the other with SP3, guess what...

with SP2 : with SP3

if you run "eac3to.exe -test" many times is always reported:

"ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) works fine"

because on my computer sometimes it is:

"ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) works fine"

and sometimes it is:

"ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) is not working correctly"

itsancho

28th June 2008, 07:59

if you run "eac3to.exe -test" many times is always reported:
"ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) works fine"
because on my computer sometimes it is:
"ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) works fine"
and sometimes it is:
"ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) is not working correctly" yeap! i've run it 10 times one after another and the message always was that is working fine...

ACrowley

28th June 2008, 09:29

Will ArcSoft TotalMedia™ Extreme work with Vista x64 ? So far i can see not..too bad
Maybe the single .ax Decoders are working when i register them manually ?

Maybe this helps ?
http://www.arcsoft.com/support/downloads/download_patches/vista.asp

What is the newest Version ? 1.0.6.29/ build 113 ?
On the Website/Update are newer Versions
http://www.arcsoft.com/support/downloads/download_patches_menu.asp

EDIT:
All Filters are working on Vista X64 :)

Can i use The AudioHH Decoder for THD,EAC3 in mpc ?

Encoder888

28th June 2008, 14:20

im getting the lossless check failed error again when i tried to convert TrueHD to flac for some reason.
Where is the progress bar when this complaint appears?

it happens immediately after the conversion is complete, not until the progress bar has gone all the way across. today i got the same error with I Am Legend (with 2.49).

If possible, let me know if I need to worry about it... :)

Thanks and im sorry i couldnt reply earlier.

deathlord

28th June 2008, 15:03

madshi,

Great job with the new version! Thank you very much!

I have just downloaded and installed the TMT intro version on xp sp2. It seems to work flawlessly, the test is successful even when repeated several times:
ArcSoft Audio Decoder (1.3.2.49, 1.1.0.0) works fine

The version of the .exe file is 2.1.0.106, ASAudioHD.ax is 1.3.2.49.

Has anyone figured out which files are really needed, yet? I don't need the whole program, the filters alone would do.

madshi

28th June 2008, 15:27

about the 3 DTS-HD MA 7.1 samples that ArcSoft can't decode bit-perfectly using GraphEdit

1) Title: Dragon Tiger Gate, Blu-ray, HK release

2) Title: Flash Point, Blu-ray, HK release

eac3to 2.49 decodes 1) and 2) bit-perfectly

what is the eac3to 2.49 secret here, do you patch the DTS headers to achieve bit-perfect decode for sample 1) and 2) or do something else?
Not sure what the secret is, actually. I'm doing the usual eac3to magic :), but nothing special to make those 2 tracks work...

3) Title: Hi-Definition Reference Disc, Blu-ray, Japan release

eac3to 2.49 fails with 3)
From what I can see the difference is that the decoded DTS track has a lower volume compared to the TrueHD track. I see 3 possible reasons for why that could happen:

(1) Either the DTS track has more than 8 channels in it. As a result the ArcSoft decoder might mix the additional channels into the first 8 channels. I think this is rather unlikely.

(2) Or the DTS track has dialnorm activated in the DTS-HD blocks and the ArcSoft decoder applies this dialnorm. There's an additional dialnorm field in the DTS-HD blocks, which eac3to can not handle.

(3) Or the DTS encoder was actually fed with audio data which was already lowered in volume compared to the data fed to the TrueHD encoder.

Can you post the eac3to detail information for the m2ts file this problematic dts/thd combo was demuxed from?

2) "ArcSoft Audio Decoder is not working correctly" - can't find what exactly fix initially the problem, but still it is here and has random effect in one minute it is there in the next it is gone
That should be fixed by the next build (v2.50).

3) "ArcSoft Audio Decoder doesn't seem to be installed" - on my other computer, no matter what i do eac3to reports this error
If that occurs at a 2nd run when doing "eac3to -test" then this should also be fixed with the next build. However, if you never ever got the decoder to ever show anything else besides "doesn't seem to be installed" then v2.50 will probably not fix this.

What is the newest Version ? 1.0.6.29/ build 113 ?
With v2.50 the only important thing is the version of the "dtsdecoderdll.dll" file. File version 1.1.0.0 works perfectly for me.

Can i use The AudioHH Decoder for THD,EAC3 in mpc ?
Yes, but only together with the ArcSoft demuxers/splitters, I think.

it happens immediately after the conversion is complete, not until the progress bar has gone all the way across.
In that case you can safely ignore the warning.

madshi

28th June 2008, 15:33

eac3to v2.50 released

http://madshi.net/eac3to.zip

* ArcSoft DTS Decoder DLL is now directly accessed instead of using DirectShow
This should work around some of the problems with the ArcSoft DirectShow filters.

BTW, I've noticed that the ArcSoft DTS decoder is a lot faster than the Sonic DTS decoder:

Conventional DTS tracks: 50% faster
DTS-HD High Resolution tracks: 55% faster
DTS-96/24 tracks: 79% faster
DTS-HD Master Audio tracks: 175% faster

Especially the Master Audio decoding performance is in another world compared to the Sonic decoder. The Sonic takes almost 3x as much time for decoding MA tracks!

hubblec4

28th June 2008, 15:37

Hi madshi

Quote:
Originally Posted by hubblec4 View Post
hi Yraen

there is a small problem and i hope you understand what i mean.

Movie: Waterworld for example

the eac3to log shows me 14 subtitles but it is 18 subtitels.

14 normal subs and 4 forced subs. (evodemux shows me the 18 subs but can´t right demux them)

the 4 forced subs started later as the normal subs and i think thats the problem.

is there a way to fix that??? maybe by large scanning the feature.evo file or is that a problem of the eac3to.exe?

Quote:
original by Yraen

Unfortunately that is an eac3to limitation and you're right about what the problem is. Eac3to, to the best of my knowledge, only reads so much of the .evo/.m2ts files to see the streams contained within. Since they start later, eac3to never sees them. You might post in the eac3to thread and see if madshi might want a sample or there is a command line switch that will let eac3to read deeper into the files.

i hope you have time to fix that.

hubble

Thunderbolt8

28th June 2008, 15:42

can someone please enlighten me which arcsoft is advised to be used? the theatre or the extreme version? does it have to be a full version or is a(n) (re)install of the trial version fine as well?

nautilus7

28th June 2008, 15:53

@ madshi

New version works here (as before). Could you tell which .dll and .ax filters are used by eac3to for dts decoding with Arcsoft decoder (and in what paths are they are located if possible)? I am trying to make it work in my other pc without installing the whole program (just registering the filters).

And yes, Arcsoft decoder is much faster than Sonic.

@ Thunderbolt8

I use TotalMedia Theatre 15 days trial (http://www.arcsoft.com/support/downloads/downloadnow.asp?downloadid=384). This is also described as intro version in the previous posts. After expired once, I re-installed it and everything was fine. Good luck.

deathlord

28th June 2008, 15:53

can someone please enlighten me which arcsoft is advised to be used? the theatre or the extreme version? does it have to be a full version or is a(n) (re)install of the trial version fine as well?

At the moment I'm using this version (eac3to_2.49.0.0) (without registration). I don't know if it the filters will continue to work after it expires. Maybe it is also possible to just register the dtsdecoderdll.dll?

Edit: Registering the .dll only doesn't work (eac3to v2.50).

B4tm4n

28th June 2008, 15:56

eac3to v2.50 released

Thank you. :thanks:

hubblec4

28th June 2008, 16:13

xkodi

28th June 2008, 16:20

Can you post the eac3to detail information for the m2ts file this problematic dts/thd combo was demuxed from?

they are two separate m2ts file from the Qtec Hi-Definition Reference Disc made for testing purposes - the same sound in the both formats:

C:\eac3to250>eac3to.exe 00005.m2ts
M2TS, 1 video track, 1 audio track, 0:03:27
1: h264/AVC, 1080i60 /1.001 (16:9)
2: DTS Master Audio, Japanese, 7.1 channels, 24 bits, 96khz

C:\eac3to250>eac3to.exe 00004.m2ts"
M2TS, 1 video track, 1 audio track, 0:03:27
1: h264/AVC, 1080i60 /1.001 (16:9)
2: TrueHD/AC3, Japanese, 7.1 channels, 96khz

the Sonic decodes the 5.1 channels byte by byte identical to the decoded TrueHD track, but i suspect the problem is somehow related to DTS-HD MA channel order, because the decoded files has the following channel order:

sample 3)

TrueHD channel 0 = Sonic channel 0
TrueHD channel 1 = Sonic channel 1
TrueHD channel 2 = Sonic channel 2
TrueHD channel 3 = Sonic channel 3
TrueHD channel 4 = not decoded by Sonic
TrueHD channel 5 = not decoded by Sonic
TrueHD channel 6 = Sonic channel 4
TrueHD channel 7 = Sonic channel 5

for example the other TrueHD/DTS-HD MA 7.1 sample that we have from the "Dragon Tiger Gate" aka "sample 1)" has:

sample 1)

TrueHD channel 0 = Sonic channel 0
TrueHD channel 1 = Sonic channel 1
TrueHD channel 2 = Sonic channel 2
TrueHD channel 3 = Sonic channel 3
TrueHD channel 4 = Sonic channel 4
TrueHD channel 5 = Sonic channel 5
TrueHD channel 6 = not decoded by Sonic
TrueHD channel 7 = not decoded by Sonic

so, for the working file looks like Sonic decodes the BL and BR and for not working SL and SR, which is quite interesting, that is what i have found till now, i hope that will help. i am sure you will figure out what is going on.

madshi

28th June 2008, 16:20

Movie: Waterworld for example

the eac3to log shows me 14 subtitles but it is 18 subtitels.

14 normal subs and 4 forced subs. (evodemux shows me the 18 subs but can´t right demux them)

the 4 forced subs started later as the normal subs and i think thats the problem.
This is on my to do list. Will be fixed sooner or later...

can someone please enlighten me which arcsoft is advised to be used? the theatre or the extreme version?
Theatre should do.

madshi

28th June 2008, 16:21

does it have to be a full version or is a(n) (re)install of the trial version fine as well?
I use TotalMedia Theatre 15 days trial. This is also described as intro version in the previous posts. After expired once, I re-installed it
At the moment I'm using this version (eac3to_2.49.0.0) (without registration). I don't know if it the filters will continue to work after it expires. Maybe it is also possible to just register the dtsdecoderdll.dll?
Guys, in this thread please don't discuss how to avoid buying commercial software you're using.

I don't want eac3to to be connected with such topics. Besides, this thread is long enough as it is without discussion about licensing other software...

You can test the ArcSoft DTS decoder by installing the ArcSoft trial version. The decoder in the trial version should work with eac3to. But once the trial period is over and you still want to continue using the decoder, please buy the software.

Thanks!

madshi

28th June 2008, 16:30

the Sonic decodes the 5.1 channels byte by byte identical to the decoded TrueHD track
Oh, I didn't check that. You're quite right. In that case I think that the problem is most likely to be (once again) dialnorm. The funny thing is that according to the log you posted there's no dialnorm in the core. But I can only guess that there's dialnorm activated in the DTS-HD headers. <sigh>

xeonicxpression

28th June 2008, 17:15

I have a quick question about demuxing the audio. When I scan my evo it shows 2503ms at the end of each audio track (as seen below). When I demux the audio should I be adding +2503ms or -2503ms to the end of the command or should I be doing it in mkvmerge or do I ignore it all together?

3: E-AC3, English, 5.1 channels, 640kbit/s, 48khz, dialnorm: -27dB, 2503ms

Madshi, thanks for all the great work you've done. You have made this process so simple.

madshi

28th June 2008, 17:17

I have a quick question about demuxing the audio. When I scan my evo it shows 2503ms at the end of each audio track (as seen below). When I demux the audio should I be adding +2503ms or -2503ms to the end of the command or should I be doing it in mkvmerge or do I ignore it all together?

3: E-AC3, English, 5.1 channels, 640kbit/s, 48khz, dialnorm: -27dB, 2503ms
eac3to will automatically correct that audio delay when you demux the audio track. So just ignore it (as long as you're using eac3to for demuxing).

EPiPH0NE

28th June 2008, 21:49

eac3to v2.50 released

* ArcSoft DTS Decoder DLL is now directly accessed instead of using DirectShow

Works fine now :)

P:\>P:\eac3to_2.50\eac3to.exe -test
Nero Audio Decoder (Nero 7 or older) works fine
ArcSoft DTS Decoder (1.1.0.0) works fine
Sonic Audio Decoder (4.3.0.169) works fine
Haali Media Splitter (2007-11-18) is installed
Surcode DTS Encoder (1.0.23.0) is installed
MkvToolnix (v2.2.0) is installed

P:\>P:\eac3to_2.50\eac3to.exe -test
Nero Audio Decoder (Nero 7 or older) works fine
ArcSoft DTS Decoder (1.1.0.0) works fine
Sonic Audio Decoder (4.3.0.169) works fine
Haali Media Splitter (2007-11-18) is installed
Surcode DTS Encoder (1.0.23.0) is installed
MkvToolnix (v2.2.0) is installed

P:\>P:\eac3to_2.50\eac3to.exe -test
Nero Audio Decoder (Nero 7 or older) works fine
ArcSoft DTS Decoder (1.1.0.0) works fine
Sonic Audio Decoder (4.3.0.169) works fine
Haali Media Splitter (2007-11-18) is installed
Surcode DTS Encoder (1.0.23.0) is installed
MkvToolnix (v2.2.0) is installed

Thnx!

madshi

28th June 2008, 21:56

Works fine now :)
You're Haali splitter is outdated... ;)

EPiPH0NE

29th June 2008, 00:36

You're Haali splitter is outdated... ;)

I'm an 'outdated' kind of guy :p

All fixed up now with 2008-3-29.

Thunderbolt8

29th June 2008, 02:02

madshi, the DTS-HD 6.1 track from the basic instinct blu-ray is still changed to a 7.1 FLAC track with 1 channel being doubled. is this the way it is meant to work correctly?

G:\eac3to>eac3to H:\instinct 1) 3: H:\instinct.flac
M2TS, 1 video track, 4 audio tracks, 2:08:15
1: Chapters, 17 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: DTS Hi-Res, English, 6.1 channels, 24 bits, 3093kbit/s, 48khz, dialnorm: -4dB
4: AC3, English, 5.1 channels, 640kbit/s, 48khz, dialnorm: -27dB
5: AC3, English, 2.0 channels, 192kbit/s, 48khz, dialnorm: -27dB
6: AC3, English, 2.0 channels, 192kbit/s, 48khz, dialnorm: -27dB
[a03] Doubling 7th channel...
[a03] Extracting audio track number 3...
[a03] Removing dialog normalization...
[a03] Decoding with ArcSoft DTS Decoder...
[a03] Creating file "H:\instinct.flac"...

madshi

29th June 2008, 08:01

madshi, the DTS-HD 6.1 track from the basic instinct blu-ray is still changed to a 7.1 FLAC track with 1 channel being doubled. is this the way it is meant to work correctly?
That is how it is meant to work. The problem with 6.1 tracks is that it's not totally clear which channel is which. With 7.1 there can be no confusion about which channel is which speaker.

xkodi

29th June 2008, 10:13

Oh, I didn't check that. You're quite right. In that case I think that the problem is most likely to be (once again) dialnorm. The funny thing is that according to the log you posted there's no dialnorm in the core. But I can only guess that there's dialnorm activated in the DTS-HD headers. <sigh>

madshi, maybe it is dialnorm, but more interesting is which channels Sonic decodes, because on all other DTS-HD MA 7.1 samples that i have:

Dragon Tiger Gate (DTS-HD MA, 7.1, 16bit, 48kHz)
Flash Point (DTS-HD MA, 7.1, 16bit, 48kHz)

Rush Hour 3 (DTS-HD MA, 7.1, 24bit, 48kHz)
Rambo 4 (DTS-HD MA, 7.1, 24bit, 48kHz)
The Orphanage (DTS-HD MA, 7.1, 24bit, 48kHz)
Hairspray (DTS-HD MA, 7.1, 24bit, 48kHz)
BT - Dynamic Symmetry (DTS-HD MA, 7.1, 24bit, 48kHz)
Chieli Minucci - Daybreak (DTS-HD MA, 7.1, 24bit, 48kHz)

SFX Demo (DTS-HD MA, 7.1, 24bit, 96kHz)
Orchestra Demo (DTS-HD MA, 7.1, 24bit, 96kHz)

the decodes are bit perfect and Sonic decodes the following channels:

decoded by ArcSoft channel 0 = decoded by Sonic channel 0
decoded by ArcSoft channel 1 = decoded by Sonic channel 1
decoded by ArcSoft channel 2 = decoded by Sonic channel 2
decoded by ArcSoft channel 3 = decoded by Sonic channel 3
decoded by ArcSoft channel 4 = decoded by Sonic channel 4
decoded by ArcSoft channel 5 = decoded by Sonic channel 5
decoded by ArcSoft channel 6 = not decoded by Sonic
decoded by ArcSoft channel 7 = not decoded by Sonic

but on "Qtec Hi-Definition Reference Disc" not only the Arcsoft decode is not bit-perfect, but Sonic decodes the following channels:

decoded by ArcSoft channel 0 <--> decoded by Sonic channel 0
decoded by ArcSoft channel 1 <--> decoded by Sonic channel 1
decoded by ArcSoft channel 2 <--> decoded by Sonic channel 2
decoded by ArcSoft channel 3 <--> decoded by Sonic channel 3
decoded by ArcSoft channel 4 <--> not decoded by Sonic
decoded by ArcSoft channel 5 <--> not decoded by Sonic
decoded by ArcSoft channel 6 <--> decoded by Sonic channel 4
decoded by ArcSoft channel 7 <--> decoded by Sonic channel 5

we can't deny that this is very interesting and strange. so, what cause this: on 10 samples Sonic decodes one subset of 5.1 channels out of 7.1 and only in 1 sample it decodes different subset of 5.1 channels out of 7.1 and on the same sample Arcsoft fails?

p.s.

"<-->" denotes how channels from one decode map to the other and that both decodes are not identical, i.e. not bit-perfect

"=" denotes how channels from one decode map to the other and that both decodes are identical, i.e. bit-perfect

deathlord

29th June 2008, 10:48

That is how it is meant to work. The problem with 6.1 tracks is that it's not totally clear which channel is which. With 7.1 there can be no confusion about which channel is which speaker.

Don't you think the volume of the two back channels should be lowered when doubling a channel? I believe simply doubling the 6th channel from a 6.1 track will make it too loud. If both back channels are attenuated by 3dB they will add up to the original volume.
I do agree that doubling this channel is the best solution, though.
I have asked that question before, you might have overlooked it.

nautilus7

29th June 2008, 11:24

xkodi

29th June 2008, 11:35

Could you explain a bit further? Why it's not clear which channel is which?

at first i was thinking that this is because Microsoft don't define channel mapping mask for 6.1 in the WAVEX header, but actually they do define such mask:

channels: 6.1, order: FL, FR, C, LFE, BL, BR, BC, mask: 0x13F

so, now i am confused too.

Don't you think the volume of the two back channels should be lowered when doubling a channel? I believe simply doubling the 6th channel from a 6.1 track will make it too loud. If both back channels are attenuated by 3dB they will add up to the original volume.
I do agree that doubling this channel is the best solution, though.
I have asked that question before, you might have overlooked it.

here:

http://forum.doom9.org/showthread.php?p=1144930#post1144930

are some calculations for 5.1 -> 7.1 conversion, maybe 6.1 -> 7.1 is best to be done with similar calculations, but probably it is better to leave it as 6.1, after all Microsoft define channel mapping mask for 6.1.

nautilus7

29th June 2008, 12:21

And according to DTS there can be only one 6.1 speaker configuration, while there's a variety of 7.1: http://www.dts.com/DTS_In_Consumer_Products/HD_Home_Theater/Speaker_Options.aspx

If I recall correctly, madshi has chosen doubling the rear channel because all 7.1 PCM tracks he had found (at that time) had 2 identical back channels and there were no 6.1 channel pcm tracks (maybe I am wrong though). Anyway, I have 2-3 7.1 PCM tracks with all channels unique and also have seen 6.1 pcm here (http://bluray.highdefdigest.com/463/descent.html) and here (http://bluray.highdefdigest.com/dirtydancing.html).

xkodi

29th June 2008, 12:45

If I recall correctly, madshi has chosen doubling the rear channel because all 7.1 PCM tracks he had found (at that time) had 2 identical back channels and there were no 6.1 channel pcm tracks (maybe I am wrong though). Anyway, I have 2-3 7.1 PCM tracks with all channels unique and also have seen 6.1 pcm here (http://bluray.highdefdigest.com/463/descent.html) and here (http://bluray.highdefdigest.com/dirtydancing.html).

plus PS3 outputs all DTS-HD HR and MA 6.1 track as LPCM 6.1:

http://www.avsforum.com/avs-vb/showthread.php?p=13605209#post13605209

tebasuna51

29th June 2008, 12:54

at first i was thinking that this is because Microsoft don't define channel mapping mask for 6.1 in the WAVEX header, but actually they do define such mask:

channels: 6.1, order: FL, FR, C, LFE, BL, BR, BC, mask: 0x13F

so, now i am confused too.

For me the problem is know if all decoder do your channel order or can use this other:

channels: 6.1, order: FL, FR, C, LFE, BC, SL, SR mask: 0x70F

@madshi
Like 7.1 is clearly: FL, FR, C, LFE, BL, BR, SL, SR
When you double the 6th channel to 7th channel:

1) if you have FL, FR, C, LFE, BL, BR, BC you obtain
FL, FR, C, LFE, BL, BR, BC, BC'
Then BC, BC' is send to SL, SR, something is wrong.

2) if you have FL, FR, C, LFE, BC, SL, SR you obtain
FL, FR, C, LFE, BC, SL, SR, SR'
Then BC, SL, SR, SR' is send to BL, BR, SL, SR, now is worse.

Thunderbolt8

29th June 2008, 13:56

That is how it is meant to work. The problem with 6.1 tracks is that it's not totally clear which channel is which. With 7.1 there can be no confusion about which channel is which speaker.
is this a temporary problem, meaning it could be fixed in the future with other decoders, more information or samples or is it just the way that studios sometimes you one, sometimes the other channel as 6th channel so that it can vary each time?

madshi

29th June 2008, 18:08

madshi, maybe it is dialnorm
No, it's not. In the meanwhile I'm able to read out the dialnorm information of the DTS-HD blocks, too. And dialnorm is not active in that problem track.

but more interesting is which channels Sonic decodes, because on all other DTS-HD MA 7.1 samples that i have
You're right. I think the channel configuration is causing ArcSoft to lower the volume of this specific track somehow. My latest eac3to build (not uploaded yet) can now output the detailed speaker configuration stored in the DTS-HD header. Here's what I found (please ignore the order of the channels):

All DTS 5.1 movies: C L R Ls Rs LFE
All DTS 6.1 movies: C L R Ls Rs LFE Cs
Most DTS 7.1 movies: C L R LFE Lsr Rsr Lss Rss
Shoot 'em Up movie: C L R Ls Rs LFE Lw Rw
Hi-Def 7.1 demo: C L R Ls Rs LFE Lsr Rsr

I think that the ArcSoft decoder is trying to remap "Ls Rs" to "Lss Rss" and lowers the volume to avoid clipping. Probably there is no need to worry, because for most movies the problem does not exist. The Hi-Def demo is just encoded strangely.

Don't you think the volume of the two back channels should be lowered when doubling a channel? I believe simply doubling the 6th channel from a 6.1 track will make it too loud. If both back channels are attenuated by 3dB they will add up to the original volume.
I'm not sure about that. I wish I knew what receiver manufacturers are doing, or what DTS/Dolby recommend. Anyway, I've chosen this solution because of the way 6.1 PCM is stored in Blu-Ray: It's stored as 7.1 PCM where the back surround channels have identical data. This is why I implemented the same logic in eac3to.

Could you explain a bit further? Why it's not clear which channel is which?
at first i was thinking that this is because Microsoft don't define channel mapping mask for 6.1 in the WAVEX header, but actually they do define such mask:

channels: 6.1, order: FL, FR, C, LFE, BL, BR, BC, mask: 0x13F

so, now i am confused too.
As tebasuna51 has already hinted, there's some confusion (even at Microsoft!) about "SL <-> BL" and "SR <-> BR". E.g. for 5.1 content Microsoft originally recommended the use of the channel mask "0x3f" which uses BL/BR. But with XPSP2 Microsoft suddenly changed their mind and now recommends "0x60f" which is SL/SR. Still most software uses 0x3f, but I think it's actually not correct. Now for 6.1 the same problem exists. Do we use 0x13f or do we use 0x70f? The latter is the correct solution IMO, but not everyone agrees. There have already been complains about madFlac playing audio on the wrong speakers due to this channel mask problem.

There's one channel combination where there's no doubt and that is 7.1. That is why I think converting 6.1 to 7.1 works around the channel mask problems nicely.

Like 7.1 is clearly: FL, FR, C, LFE, BL, BR, SL, SR
Yep.

When you double the 6th channel to 7th channel:

1) if you have FL, FR, C, LFE, BL, BR, BC you obtain
FL, FR, C, LFE, BL, BR, BC, BC'
Then BC, BC' is send to SL, SR, something is wrong.
I think this is a wrong channel configuration, anyway. When there's a BC channel (or BL/BR channels) the surround channels should be SL/SR and not BL/BR. Having BL/BR/BC in combination is all wrong IMHO. There can only be either BL+BR or BC, but not both at the same time. At least that's my opinion...

2) if you have FL, FR, C, LFE, BC, SL, SR you obtain
FL, FR, C, LFE, BC, SL, SR, SR'
Then BC, SL, SR, SR' is send to BL, BR, SL, SR, now is worse.
I'm basing my channel doubling on what DTS decoders give me for 6.1 tracks. I'm remapping the channels so that I get this:

FL, FR, C, LFE, BC, BC', SL, SR

That should be correct, right?

xkodi

29th June 2008, 18:34

My latest eac3to build (not uploaded yet) can now output the detailed speaker configuration stored in the DTS-HD header. Here's what I found (please ignore the order of the channels):

All DTS 5.1 movies: C L R Ls Rs LFE
All DTS 6.1 movies: C L R Ls Rs LFE Cs
Most DTS 7.1 movies: C L R LFE Lsr Rsr Lss Rss
Shoot 'em Up movie: C L R Ls Rs LFE Lw Rw
Hi-Def 7.1 demo: C L R Ls Rs LFE Lsr Rsr

I think that the ArcSoft decoder is trying to remap "Ls Rs" to "Lss Rss" and lowers the volume to avoid clipping. Probably there is no need to worry, because for most movies the problem does not exist. The Hi-Def demo is just encoded strangely.

is there any chance to patch that info in the header and for example make "C L R Ls Rs LFE Lsr Rsr" of the Hi-Def 7.1 demo to "C L R LFE Lsr Rsr Lss Rss", so it decodes correctly?

btw meanwhile i'm doing tests of Dolby Digital Plus decoding with Arcsoft and here is what i have found out until now:

- Dolby Digital Plus 7.1 channel decoding is working (tested on "7.1 Dolby Digital Plus Channel Check" sample)
- on my samples Nero outputs 24bit, but i think this is wrong because if set Arcsoft to 24bit there is noise, so i believe these samples are actually 16bit and Arcsoft is correct
- the same EAC3 files decoded with Nero (and converted to 16bit) in GraphEdit and with Arcsoft are byte by byte identical, so Arcsoft is using the same reference decoder as Nero
- Dolby Digital Plus 7.1 in M2TS is not detected by eac3to

i am preparing Dolby Digital Plus 7.1 samples and will upload shortly, so you can do your magic :)

rickardk

29th June 2008, 18:38

All DTS 5.1 movies: C L R Ls Rs LFE
All DTS 6.1 movies: C L R Ls Rs LFE Cs
Most DTS 7.1 movies: C L R LFE Lsr Rsr Lss Rss
Shoot 'em Up movie: C L R Ls Rs LFE Lw Rw
Hi-Def 7.1 demo: C L R Ls Rs LFE Lsr Rsr

Does this mean that Shoot 'em Up is handled incorrect with v2.49?

xkodi

29th June 2008, 18:41

Does this mean that Shoot 'em Up is handled incorrect with v2.49?

no, it is handled correctly i tested it before a minute. only 7.1 channel tracks with order "C L R Ls Rs LFE Lsr Rsr" are handled incorrectly and there is only one such track from "Qtec Hi-Definition Reference Disc". so, all known movie track are handled perfectly correctly.

p.s.
you can always do the following test to check:

decode the track with Sonic (only 5.1 channels) to wavfile1
decode the track with Arcsoft (all 7.1 channels) to wavfile2

split the channels of the wavfile1 and wavfile2 with Wavewizard. compare wavfile1_ch0 with wavfile2_ch0 and if they are byte by byte identical then eac3to handles the track correctly.

nautilus7

29th June 2008, 18:44

- Dolby Digital Plus 7.1 in M2TS is not detected by eac3to

i am preparing Dolby Digital Plus 7.1 samples and will upload shortly, so you can do your magic :)I have sent these samples (we have the same disc) to madshi a long time ago, but he has forgot to add support for them. :eek: If he doesn't have them anymore, i have samples cut and ready from the first time, so you can count on me for that.

madshi

29th June 2008, 18:44

is there any chance to patch that info in the header and for example make "C L R Ls Rs LFE Lsr Rsr" of the Hi-Def 7.1 demo to "C L R LFE Lsr Rsr Lss Rss", so it decodes correctly?
Possible, but difficult. I think I'll rather post a warning if such a track is decoded. If the problem really occurs with a real movie (and not just a demo track) I might look into that.

- on my samples Nero outputs 24bit, but i think this is wrong because if set Arcsoft to 24bit there is noise, so i believe these samples are actually 16bit and Arcsoft is correct
No. AC3 and E-AC3 tracks do not have a fixed bitdepth. If there is noise when decoding the tracks to 24bit with ArcSoft then this is a clear bug in ArcSoft. The native output of any AC3/E-AC3 decoder is floating point. Which is much more than 24bit.

- Dolby Digital Plus 7.1 in M2TS is not detected by eac3to
I know. I don't think it's important because there's not a single movie in the world (at least as far as I know) which uses 7.1 E-AC3. There's only a Dolby demo. Or am I wrong?

Does this mean that Shoot 'em Up is handled incorrect with v2.49?
I think it's decoded correctly. At least the volume is identical between Sonic and ArcSoft. It seems that ArcSoft gets along with the speaker setup used by Shoot 'em Up, although it's not the one used by most other movies. The only track where I've seen a volume difference between Sonic and ArcSoft is that one demo track.

I have sent these samples (we have the same disc) to madshi a long time ago, but he has forgot to add support for them. :eek:
No, I didn't... :) I just decided that there are more important things to do.

nautilus7

29th June 2008, 18:53

split the channels of the wavfile1 and wavfile2 with Wavewizard
You do know that eac3to can output mono wavs, right? :p

I don't think it's important because there's not a single movie in the world (at least as far as I know) which uses 7.1 E-AC3. There's only a Dolby demo

No, I didn't... :) I just decided that there are more important things to do.
That's why I didn't bugged you with this earlier. :)

tebasuna51

29th June 2008, 18:54

I'm basing my channel doubling on what DTS decoders give me for 6.1 tracks. I'm remapping the channels so that I get this:

FL, FR, C, LFE, BC, BC', SL, SR

That should be correct, right?

Yep, but this is for next version? Or only for dts-hd/eac3?

Now with:

eac3to seven_channel.wav output.wav -extensible

the 7th channel is doubled to 8th channel without remapping then is always wrong.

xkodi

29th June 2008, 18:58

Possible, but difficult. I think I'll rather post a warning if such a track is decoded. If the problem really occurs with a real movie (and not just a demo track) I might look into that.

i agree, it doesn't worth the effort until there is only one demo track with that channel order and warning is perfectly good, after all the same track is available in TrueHD too.

No. AC3 and E-AC3 tracks do not have a fixed bitdepth. If there is noise when decoding the tracks to 24bit with ArcSoft then this is a clear bug in ArcSoft. The native output of any AC3/E-AC3 decoder is floating point. Which is much more than 24bit.

ok, but if the E-AC3 track is made from 16bit WAV file then it is a waste of space to output to 24bit or am i wrong?

i am still testing and not sure, but i think that on one of the sample ArcSoft do not output noise when is set to 24bit. is there any chance that ArcSoft can detect from what source WAV file the E-AC3 track is made - 16bit or 24bit?

I know. I don't think it's important because there's not a single movie in the world (at least as far as I know) which uses 7.1 E-AC3. There's only a Dolby demo. Or am I wrong?

yes, i also have only demo files.

You do know that eac3to can output mono wavs, right? :p

:) yes, i know, but to be honest totally forgot about that, thanks to remind me.

deathlord

29th June 2008, 19:11

madshi,

I've chosen this solution because of the way 6.1 PCM is stored in Blu-Ray: It's stored as 7.1 PCM where the back surround channels have identical data.
I'm not sure if I understand correctly. Are you sure these are tracks were originally recorded in 6.1 and the back channel was doubled without applying any attenuation?

I wish I knew what receiver manufacturers are doing, or what DTS/Dolby recommend.
I doubt there is an agreement on the 6.1 to 7.1 conversion.
However, I find the idea proposed by tebasuna51 (here (http://forum.doom9.org/showthread.php?p=1144930#post1144930)), based on the distribution of acoustic power, very convenient.
For 6.1-->7.1 it would mean:
Back_new(l,r) * cos(30) = 0.5 * Back_old
==> Back_new(l,r) = 0.5 * Back_old / cos(30)
taking sqrt:
Back_new(l,r) should be 0.76 * Back_old

In theory we should also subtract the newly added x-component from the side channel, but I think this can be neglected.

xkodi

29th June 2008, 19:21

i am still testing and not sure, but i think that on one of the sample ArcSoft do not output noise when is set to 24bit. is there any chance that ArcSoft can detect from what source WAV file the E-AC3 track is made - 16bit or 24bit?

i confirm that there are E-AC3 tracks that ArcSoft output without noise when set to 24bit and that there are E-AC3 tracks that ArcSoft output without noise only when set to 16bit. ArcSoft decodes the DTS-HD MA correctly only when the right parameters are set, so maybe it is the same with the E-AC3 track and somehow ArcSoft finds the resolution of the source WAV file from which the E-AC3 track is made. probably, the best is madshi to look at that, when there are no more important things to do, i will be happy to provide the samples when the time comes.