View Full Version : LT+RT to surround, how?
tormento
30th May 2022, 20:05
The title tells it all.
I have found some old BD with LT+RT track, i.e. Dolby Surround.
Is there any way to recover the original surround format?
SeeMoreDigital
30th May 2022, 21:33
So.... Is it this? (https://en.wikipedia.org/wiki/Dolby_Pro_Logic#Dolby_Surround_(2014))
tormento
30th May 2022, 22:21
Is it this?
Nope, it's some version of Pro Logic, II I think. I need to invert the procedure.
junh1024
31st May 2022, 06:32
Not possible to completely reverse the procedure since downmixing loses a lot of information. Please use any surround decoder or upmixer.
tormento
31st May 2022, 10:01
Not possible to completely reverse the procedure since downmixing loses a lot of information. Please use any surround decoder or upmixer.
I know that Surcode had a piece of software that allows to recreate the mono wavs from the LT+RT audio.
I'd just like to do the same thing, with freeware software possibily, as Surcode has closed long time ago.
tebasuna51
31st May 2022, 18:59
Try with Foobar2000 and the foo_dsp_fsurround.dll (https://hydrogenaud.io/index.php?topic=52235.0) plugin.
tormento
31st May 2022, 20:09
Try with Foobar2000 and the foo_dsp_fsurround.dll (https://hydrogenaud.io/index.php?topic=52235.0) plugin.
Can it output to separate mono wav? I never used foobar.
junh1024
1st June 2022, 07:09
Can it output to separate mono wav? I never used foobar.
No need, encoding software like eac3to or Totalcode studio already accept 6ch wav.
BTW, if you play using software like PowerDVD, that can upmix 2ch to surround.
Ac3filter's expand stereo, and windows speaker fill should also do basic dematrixing.
Keep in mind you will never completely recover the original: https://i.imgur.com/0o7Rflf.png
tebasuna51
1st June 2022, 12:18
Keep in mind you will never completely recover the original
Of course if LT+RT was created with a standard downmix (LT mix of FL + SL, RT mix of FR + SR) it is not possible extract the 4 channels.
But with DplII downmix we can obtain a good aproach
tebasuna51
1st June 2022, 12:22
Can it output to separate mono wav? I never used foobar.
It is possible but better try before with wav 5.1.
Foobar method I use, ask if you need more info:
SeeMoreDigital
1st June 2022, 14:49
But with DplII downmix we can obtain a good aproach
Have I missed (another) meeting... Where does one find this?
tormento
1st June 2022, 15:51
Foobar method I use, ask if you need more info:
You are always so kind!
I think we have both different Foobar and different DSP:
https://i1.lensdump.com/i/tB488K.pnghttps://i2.lensdump.com/i/tB4AQZ.png
Where can I find your FreeSurround package?
tebasuna51
1st June 2022, 22:01
Not always the new are better.
But you can try the new one (never redirect bass to LFE please!, the AVR's make that to subwofer automatically and include filters can destroy the sound).
tebasuna51
1st June 2022, 22:11
Have I missed (another) meeting... Where does one find this?
I'm not sure what you ask.
More info in 2007 discussion in https://hydrogenaud.io/index.php?topic=52235.msg471734#msg471734
and precedents in 2003 https://forum.doom9.org/showthread.php?p=347432#post347432
tormento
2nd June 2022, 08:58
But you can try the new one (never redirect bass to LFE please!
I trust your capabilities in audio much more than mine.
When you have time, would you please do some trials with newer plugin version?
Can you please also explain me the Center Image value and Rear Phase Reconstruction options?
tormento
2nd June 2022, 09:55
Well, it seems that something is wrong with channels.
I used your configuration and:
https://i3.lensdump.com/i/tBbtwM.jpg
tebasuna51
2nd June 2022, 14:31
Well, it seems that something is wrong with channels.
Wrong is your audio editor, the channels are correct in this order: L,R,C,LFE,SL,SR and not L,C,R,Ls,Rs,LFE
EDIT: verified with your source, the wav output is correct
Of course we can't know if dplII mix method was Dolby compliant
The center channel seems too loud, maybe you can use a Center Image less than 1
tebasuna51
2nd June 2022, 16:47
When you have time, would you please do some trials with newer plugin version?
With defaults seems the same than old version using Besweet compatibility and mix coeficients 0.8145, 0.5744 (defaults).
The output over a dplII created by eac3to or BeHappy/MeGUI have a little worse channel separation and the SR channel inverted.
Can you please also explain me the Center Image value and Rear Phase Reconstruction options?
Center Image: 0 -> Center channel empty, 1 -> All common values in L and R are extracted to Center channel (recommended)
Rear Phase Reconstruction: is a topic with many opinions, read https://forum.doom9.org/showthread.php?p=347432#post347432 and https://forum.doom9.org/showthread.php?p=837667#post837667
My opinion is:
I make many test using a directshow dplII decoder from PowerDVD and, if that decoder is correct (Dolby compliant), the better downmix method to recover the original channels is:
LT = L + 0.707 C + 0.866 SL + 0.5 SR
RT = R + 0.707 C - 0.5 SL - 0.866 SR
Method used in eac3to, BeHappy and MeGUI.
The BeSweet method is:
LT = L + 0.707 C - 0.866 SL - 0.5 SR
RT = R + 0.707 C + 0.5 SL + 0.866 SR
But the PowerDVD dplII decoder invert now the SR channel
Other coeficients make worse channel separation.
vBulletin® v3.8.11, Copyright ©2000-2026, vBulletin Solutions Inc.