I'm late again.
Like perhaps others, I enjoy analyzing and critiquing the speakers Amir measures.
Since I, like others, have certainly wondered about the very unsteady radiation in the range 4-8kHz, this time it should be more about the need for compromise in speaker design, than simply criticizing.
What were the designers thinking and did they think anything at all?
There are certainly, unfortunately, commercial speaker manufacturers where the driver placement is more or less random and even measurements up to +-90° are considered irrelevant.
But one should also not judge too hastily.
To show this, I have created a BEM simulation for the Solo6 and transferred the resulting frequency responses of the tweeter from +-180° to VCAD.
Of course, when the amateur developer (like me) looks at the design of the Solo6, he immediately knows everything better and notes that the woofer is vertically centered and the tweeter is arranged so that the distance to the top edge is exactly the same as the distance to the woofer surround.
My God, it's so obvious!
All we have to do is move the tweeter and woofer down two centimeters on the baffle and everything will be better!
Then let's see what happens (only tweeter frequency responses without any filter/crossover):
1. Sketch
2. on-axis FR, LW, SP, PIR, SPDI
View attachment 145046
3. Horizontal frequency response +-90°
4. Vertikal frequency response, normalized to on-axis FR, +-90°
View attachment 145045
Ohh, the amateur designer moved the problem area, at 5-6kHz, down so now the area around 3-4kHz is unsteady and also the frequency range around the crossover frequency (2-3kHz) fits less well

So the developers at Focal did have something in mind when it came to the driver positioning.
So there is no way to improve the design anymore?
Of course, only a listening test can judge whether a change leads to a sound improvement.
But one could, for example, move the woofer a bit further away from the tweeter and round off the upper baffle edge as well.
Then you get the following:
View attachment 145049
In the 3-4kHz range the axis frequency response (vertical too) worsens slightly (better not move the woofer after all?) and in the 2-3kHz range a bit more "sound energy" is radiated, but radiation is much more even overall and the 5-6kHz range is now inconspicuous.
The production of the speaker would be more costly.
Most of the time it is more complicated than it seems at first sight