We will need to buy more speakers to get all the answers, but here's what I found so far :
1. Two M2s in parallel seemed to provide more detail than a single M2. I did equalize the SPL as best as I could without using test disk and meter -- ran most of the "2xM2" tests with the center taken down 3dB from the single M2 setting, and went right up to 4dB difference "just in case". There wasn't a huge difference on regular dialog but on the helicopter crash scene from the Matrix there was a surprising difference in the "crunching glass" noises as the helicopter goes in, to the point that I was clearly hearing sounds I hadn't noticed before. Hardly scientific, of course, but I did try real hard to compensate for the higher output of the two speakers.
Just got the tiniest bit of fringing with the M2s touching the sides of the 27" 4:3 direct view TV ($300 Sanyo). The first round of tests were run with the M2s sitting on the same surface as the TV, ie with the speakers below the horizontal centerline of the CRT. I found this pretty distracting since the sound seemed to be a bit below the image -- much more so than when the speaker was on top of the TV (even further off center). My guess is that since the mains are a bit below the centerline of the TV having the center a bit above the centerline sort of balances things out as long as the distances aren't too big.
Moving the speakers up so their center was a bit above of the centerline of the CRT (woofers down) put the center channel sound right in the middle of the CRT, or at least close enough. The effect was good; the sound seemed closer to the image (eg. dialog vs. lips, explosion vs. helicopter) than with any other configuration tested to date.
There were two minor downsides with this configuration. The first was that the sound didn't seem to be as sharply centered as with a single inverted M2 on top. I'm not sure this is really a problem -- 99% of my experience is with stereo so I always tend to associate sharper imaging with better speakers, but in the surround world that isn't always the case. By the end of the testing I still preferred the sharper image but I realize that might be the wrong judgement. Meditate on this I will...
The second problem was that using a pair of speakers seemed to let the center sound move from side to side a bit as the listener moved off center. It wasn't a huge problem but I didn't like it (see previous problem and possible "old stereo fart" factor). With mains 8 feet apart, center(s) 3 feet apart, and standing 10 feet from the mains, moving 4 feet off axis (end cushion on a big 3-seat sofa) made the dialog seem to move 12-15" in the same direction. All the above distances are rough estimates, none were measured, but they should all be proportionally wrong
Next experiment was two inverted M2s side by side on top of the TV, vs. 1 inverted M2 in the same position but centered. In hindsight I screwed this one up -- didn't have two mazazines or coasters to protect the tops of the M2s so used a couple of paperbacks instead. The resulting height was enough to make the center sound seem to be noticeably "above" the TV image for the first time. I need to re-run this test with the inverted speakers sitting right on the top of the TV, no paperbacks.
Finally, I spent some time trying to isolate a mild distortion that showed up intermittently throughout the tests. It turned out that one of my dogs had gone to sleep on the far side of an M60 and was snoring away happily, quite unaware of my efforts to locate "that funny sound".
So... where does this leave me ? Single M2, inverted on top of the TV, wondering if I should change my "single M2" order to a VP100 before the nice folks at Axiom process it in the morning.
