In my experience, omnis and other polydirectionals (such as bipoles and dipoles) that bounce a lot of energy off the wall behind the speakers do not work well up against the wall. Those reflections need a much longer time interval before they arrive in order to be beneficial rather than detrimental.
On the other hand, the additional reverberant energy (if spectrally correct) adds a great deal to soundstaging and the overall sense of realism if it arrives fairly late. And in this context, the fuzzy line where "fairly late" begins is about ten milliseconds. This is the sort of time delay the rear bounce of a dipole speaker will have if it's about five feet out in front of the wall. That's not a hard and fast cut-off point, but in general owners of dipole speakers find they sound better with a fair amount of space behind them.
Okay, back to the problems of up-against-the-wall speakers. The soundstage depth is naturally pancaked by the wall being so close... I think the ear/brain system can hear how far away the wall is from the sound source by the reflections off the wall, and generally won't let the image depth extend any deeper beyond the wall than the speakers are out in front of it. So if the front of the speaker cabinet is 12 inches in front of the wall, on a good recording I think you might get the soundstage to extend 12 inches behind the wall. If the front of the speaker is 5 feet in front of the wall, I think you might get the soundstage to extend another 5 feet behind the wall.
But, what if we do a polydirectional speaker that ends up delivering a much longer time delay, by firing the additional reverberant energy off in another direction? That mask the relatively short distance to the wall behind the speakers, and result in greater soundstage depth.
We could fire the additional energy to the side, if the sidewalls are far enough away. I built such a system in 2010, but was not the first - ESP did it before me.
Or we could fire the additional energy up, so that it takes a long bounce off the ceiling. Richard Shahinian did it first, but imo Jim Romeyn did it best, and I've licensed his configuration, and showed it at RMAF 2013.
Finally, we could use an enclosure that allows the woofer and midrange backwave energy to bounce around inside the box a bit and emerge with the spectrum well-preserved after a decent time delay, thus simulating to a certain extent the presentation of a dipole out in the room several feet. I showed a speaker that does that at RMAF 2013 (yes, I was crazy enough to show two systems in one modest-sized room).
Of course getting decent soundstage depth is not the only challenge facing up-against-the-wall placement; the bass response is boosted, and if that isn't taken into account by the designer, can easily be overbearing (Audio Note and Zu and the Gradient Revolution mentioned above all address this issue well, and ime Audio Note in particular does a superb job of taking advantage of the resulting boundary reinforcement). We can end up with over-emphasis in the lower treble region if the radiation pattern has a flare in that region, in which case the likely result is eventual listening fatigue. We can get a comb-filter notch from the bounce off the wall behind the speakers, which the ear/brain system would normally tend to ignore, but if it coincides with the floor-bounce notch, it could be an audible problem.
I didn't set out to get deep into up-against-the-wall speaker design, but in the course of pursuing what I believe to be room-friendly concepts, it just sort of happened.
In a good used up-against-the-wall speaker, I'd suggest the Shahinian Diapason or the Snell Acoustics Type A.
You might be able to modify conventional ported speakers for up-against-the-wall placement by lowering their tuning frequency. This can be accomplished by either reducing the port area (throughout the entire length of the port), or lengthening the port, or some combination thereof. You can also reduce the port's contribution by stuffing it with polyfill or somesuch, or even seal it off entirely, resulting in an (probably low-Q) sealed box. User-adjustable port tuning is something else I like to include in my ported designs, specifically to accomodate a wide range of room acoustic situations and amplifier types.
Speaking of which, I'm a big fan of OTL amps, and imo you might not want to trade away OTL friendliness if you don't have to.
Best of luck with your quest,
Duke LeJeune
dealer/manufacturer
