You'd be surprised. This made me write a somewhat lengthy post myself, detailing some system design quirks we have to deal with in the current surround (AVR) market situation.
First some background in my music listening and system integration preference evolution.
As soon as computers started to have descent affordable multichannel soundcards and signal processing tools, with some knowledge of small-room acoustics and psychoacoustics, a good 2ch solution for small and narrow rooms looked like this: a modular 3-way system with woofers in a corner, crossed as high as possible (~250Hz?) to a monitor speaker in correct elevation. An Allison-type system, if you will, with room correction dealing with residual problems - modal peaks and SBIR effects. As the speakers were close to boundaries, a tunable room correction was quite robust (as in effective in a large spatial region) getting rid of unwanted very-early reflections up to a very high frequency (think almost up to 1kHz/34cm wavelenght). What remained was too early high frequency reflections, but that was easily taken care of with very thin absorption. A definite no-no in acoustics, but as they say - Exceptio Probat Regulam...
That system worked, because the music source was a computer and I could somewhat tolerate the quirks of 2ch listening. Well not any more! I like to turn my head, listen at different spots in-room, face different directions, enjoy music socially etc. and I still demand a believable illusion. (regarding usability, don't get me started on black-box appliance vs 'computer' subject...) Also, even with head-in-a-vice situation, there are problems with 2 channel stereo, first identified almost 90 years ago by Steinberg and Snow. Sadly 3ch is not standard medium today, so we have to come up with 'something' to deal with the idiosyncrasies of 2ch and unstable phantom sources.
Enter upmixing. First some background. You could think of 2 distinct methods getting rid of 2ch problems - make the whole room a sweet spot (think of an infinite large equilateral triangle) using WFS techniques or use ambisonics-inspired techniques, popularized by Michael Gerzon. The idea is to map phantom sources to arbitrary number of physical speakers. Luck would have it, all-in-one boxes had excellent, TUNABLE variant integrated. Either Griesingers Logic7 (Lexicon/HK receivers) or Dolby PL2 Music (everyone else). Or you could use a passive matrix, but I still prefer all-in-one 'black box' solutions to custom 'tinkerware'.
Sadly all the all-in-one solutions (AVR-s
) in the market make some (stupid?) assumptions about small room acoustics and their users competence setting them up, so they cater to lowest-common-denominator. So no flexible bass management schemes and customizable room correction filters AND most don't have pre-outs, even if you would like to come up with a solution yourself. But you can get excellent upmixing and with some external multi-sub scheme the modal region is taken good care of.
Now enter 'immersive' era. For the first time (yes, lets discount quad, multich SACD/DVD-A for laughs...), the music industry is on board with immersive mixes. Like with early stereo, most of them are stupid gimmicks, but there are folks who know how to correctly capture and render ambiance. BUT - all the legacy material is left in the dust. No more tunable upmixers, only DSU and NeuralX with full-on effect (yes, Auro is bankrupt and who knows what happens to Auromatic upmixer, that some TOTL AVR-s can update to with an add-on licence - but for how long?).
So for legacy music listening, outside custom solutions, a 20 year old Lexicon MC-12 still remains the pinnacle, SOTA if you will. Solely due to customizable Logic 7. Paradoxically, I would consider in today's market the most basic, cheap AVR-s, that still come with 'PL2 Music' SOTA
. If only they had pre-outs...
Did I mention that music upmixing, to be good, also requires identical speakers above modal region, in same elevation and in similar acoustic settings? You wouldn't use dissimilar L-R speakers with different elevation, would you? I didn't even touch the ambiance extraction side of upmixing, that is a large topic in itself.