Haha etc... funny how hardcore audio DIY people look at a preamp, a decoder, or whatever box & see the potential for what it
can be instead. One of the more successful & fun projects I did was to modify a Sansui QS-500B decoder for an input/output control center.
View attachment 48581
The 3 knobs on the left were used for 2 ch select, level, level balance & phase balance. The big decode knob on the right was used to select the 4 ch decoded output from Photolume QS, Ambisonic, Tate. And of course the balance & main level knob served original purpose. I removed the top plastic panel & replaced it with new plastic, kodalith windows for meters & new text for the knobs. The bottom panel was painted black & used press type lettering to denote new functions for the controls. Interestingly this project came about also because of the smell of smoke. The power amps fried & since I didn't use them anyway I found a new purpose for the unit. I gave it away to a friend when I moved to 5.1. Wish I had it today.
dts Neo 6 did it's thing with 12 bands in early units & 19 in later implementations. Can you imagine doing this in analog? DSP would be the only way but I disagree it adds a slight harshness if done right. Or maybe at my age I only have 14 bit ears.
A case can be made both for & against multi-band decoding. Bass from early audio days right up to now is generally mixed "center front." Meanwhile all fundamental notes in music are actually quite low with only pipe organs & synth/midis pumping out really high notes. Therefore the high frequencies would be upper harmonics of the fundamental notes & wouldn't they both be coming from the same direction? The benefit of tri-band decoding seem obvious but also it can be challenged.
I have often thought great approach would be 2 band decoding with simple decoding for bass, say <200Hz and variomatrix for the upper bands. That way the bass would have no concern of time constant control voltage filtering, simpler circuit of extremely low distortion & the bass would still be coming from the intended direction. It has the potential to put a bit more bass in the rears but that could be controlled & might be of benefit. Meanwhile the high pass signal could be treated to high separation variomatrix with well chosen time constants & no worries of ripple on the control voltages from the bass signal.
I agreee with you on the output phase shifting which was token at best in Sansui units. The QSD-1000 had a considerably better output phase shift. But doesn't matter much if playing stereo.
I think it's extra cool you are still using & liking the Photolume DIY kit. Think how improved it could be tuned up if the fixed resistors were replaced with pots for adjusting the control voltages?