jim, thanks for that, i thought it might be to do with phase and polar
patterns.

i guess a wide dispersion pattern flat over frequency, same signal coming
from 2 speakers would mimic similar sound waves to a single actual source
located in the middle. thinking back to o'level physics, i guess you could
get a tray full of water and compare two pulsating bobs (two omnidirectional
speakers) with one pulsating bob in the middle and compare the waves
arriving at an equidistant point (the listener).


Fairly flat response with no significant time/phase anomolies.

Controlled dispersion pattern that is fairly uniform as a function of
frequency. (See KH's article on Polar Patterns in this month's HFN.)

Relative absence of colourations or distortions.

Decent listening room acoustics.

Careful placement of speakers and listening location

Appropriately recorded source material.