Tim Padrick wrote on Sun, 27 June 2010 20:01 
I disagree. Case in point: In one club at which we play regularly, our stacks (including subs) have to set about 4 1/2' from the side walls (owing to the DJ system). Usually I cut a bit of 60Hz from the subs, but here, owing to boundary cancellation, not only do I not need the cut, EQ at 60Hz makes very little discernable difference. http://www.padrick.net/LiveSound/CancellationMode.htm

I'm not sure what you are disagreeing with. The discussion was about boundary or space loading where the concept is that the source is close enough to the boundary to have coherent summation of the direct and reflected sound at all locations and for all frequencies of interest. You seem to be discussing almost the opposite, where the path length difference from a reflection is sufficient that it does affect the response within the frequencies of interest.
The interaction of the direct and reflected signals is based on the path length difference between the two signals at the receiver (listeners). If the difference is one half wavelength then destructive interference or cancellation occurs, if it is one quarter wavelength or less then some level of constructive interference or summation occurs and if there is no difference (both signals in phase) then there is full summation. The idea of boundary loading is that the source is physically at the boundary, thus any reflection has no path length difference from the direct sound and sums for all frequencies. The same applies to a receiver, if the receiver is located at the boundary then that also causes the reflections from that boundary to be coherent with the direct sound. So whether the source or receiver is at the boundary, you get the same resulting summation, but only for one or the other.
It gets more complicated when the source is close to but not at a boundary. In that case you have to look at the relative paths for the direct and reflected sound and how those relate to the wavelengths of the frequencies of interest. With subwoofers you have a limited frequency of interest and one that relates to longer wavelengths.
If a listener was at the subwoofer then any boundary interaction would be be based solely on the distance from the subwoofer to the boundary. However, the listeners are not usually at the subwoofers, thus you have to consider the path length differences for the listeners. Look closer at the typical floor boundary or 'floor bounce' condition and you will find that for a subwoofer one usually does not have to be that far from the sub for the resulting path length difference between the direct and reflected sound to be less than one quarter wavelength. Basically, even though their ears are not at the floor, the listener does not have to be that far from the sub to start sharing the effects of the floor boundary. But that does not necessarily hold true for more distant surfaces, those may create reflections with a much greater path length difference, or for higher frequencies with much smaller wavelengths.