From the original SynthMaker forum (2004–2014), with its members’ names and dates.
rocko ·
Dear all,
I'm looking into building a simple MS Matrix controller. I plan it to input a normal L,R signal encode it to M,S, provide a volume slider for each (M,S) and then decode it back to L,R.
The 'naive' way to do it is:
//encode M=L+R; S=L-R;
//Decode L=(M+S)/2; R=(M-S)/2;
Is their a more advanced way to separate M/S?
Thanks, Rocko
KG_is_back ·
To maintain consistent gain-staging you may reduce input by 3dB both on encoding and decoding, not just 6dB on decoding. simply multiply inputs with sqrt(0.5) (I recommend precomputing it and adding it as a constant for best cpu preformance).
tester ·
In terms of practical use of M/S - you could focus on using it with filters. You can split the spectra into bands (there was somewhere here or on FS forum such splitter by Martin), and perform M/S to each band individually.
Wrote something here? To have a post removed or your name hidden, write to [email protected].
SynthMaker