Is this the common denominator?
I think most of us will agree that ideally, the front 3 should be as similar to each other as possible.
Even better if all “bed” speakers are identical.
But the main front 3 are salient.
Can the AV10’s Dirac or Audyssey improve the Phase-matching ?
No, but it sure lets you know if it is adrift.
This whole issue of speaker phasing and tilt, I think is understood by very few.
So even order filters have an acoustic axis in line with the drivers, all odd order filters have axis tilt.
So a first order filter has 6db/octave driver attenuation and 15 degrees of tilt, downwards if the tweeter is above the bass/mid. So there is lot of driver interaction. The phase shift is only 90 degrees and so are considered for all intent and purposes to be transient perfect. I actually don't think that is the case, because of driver separation.
A second order filter has 12db per octave attenuation and no tilt, but the drivers are 180 degrees out of phase at crossover, which is a significant issue.
A third order crossover has 18db per octave attenuation and 15 degrees of tilt and the phase shift is 270 degrees, so 3/4 of a cycle.
A fourth order filter has 24db per octave attenuation no tilt, and 360 degrees of phase shift, which puts the bass mid a whole cycle behind the tweeter at crossover.
So those are significant problems.
So my front three all have second order low pass and third order high pass at 2,500 Hz. That keeps them very well time aligned and phased. The left and right are MTM which corrects the phase shift tilt, and the center is coaxial and coincident. So, those speakers work as a coherent whole very well indeed with a seamless front stage.
The cross to the bass section is acoustic on the high end. The upper woofer is a mix of the sub channel and the BSC to the mids, to offload them. The lower bass drivers handle only the sub channels.
For the sides and backs I was just lucky. The sides are first order low pass, with second order high pass, with a slight downward tilt. So being mounted just above the seating position radiate to the listening position. These were my location monitors and they were on stands that tilted them to the mixing desk position.
The rears were my previous studio monitors. They are four ways, with first order pass to the lower and upper mids and the tweeters. The lower mid is third order low and high pass at 180 Hz active.
So the tilt is slightly downward, and axis right to the listening position. Being four ways they are very tall and the sound field covers the listening position well. I will never do a speaker with first order crossovers like that again. Theortically simple with good transient response. However getting a good smooth response was an absolute nightmare, that went on four years at least. As mentioned before this was a project I did jointly with Dynaudio, who designed a similar speaker. This was in the first order phase coherent speaker craze. However, they put their tweeters at the bottom with the axis tilted up to the listening position. The problem was it sounded as if the sound was coming up from the floor!
My ceiling speakers are good full range drivers. They work fine for this and don't draw attention to themselves. Most of the time they are just reproducing part of the ambient field, but if called upon in movies they sound very natural.
The front speakers sound and measure well, the waterfall plots do reveal the inevitable slight discontinuity of analog filters at 2,500 Hz, but it is well managed and something you have to accept with analog filters.
The whole issue of center speakers is the biggest problem. The horizontal MTM is just wrong. So the other options are coaxial (best) or a three way design.
For the inwall system I chose the three way design. However I ran into a problem, which I should have anticipated better. The left and right are two ways. The center was a three way, second order filters. This put the center mid out of phase with the woofers and tweeter at the crossovers. So the usual practice of reversing the phase of the mid, did not work. This put the center mid 180 degrees out of phase with the left and right speakers. This showed in measurements, and Audyssey objected strongly. So I had to put the center mid in phase. I was lucky as there was only a narrow null at the upper crossover, that is not audible, I don't think. Audyssey was happy and the measurements are better. I should have foreseen this problem from the beginning, but these speakers were designed in a hurry, when I had a lot going on. This system was done at the insistence of my wife, and had to meet her design spec. of not taking up any floor space. I am glad she pushed my on this, as we use this system a lot, and everyday.
