Originally Posted by AdvancedTheater
I would put that comment from Mark under the marketing department even though he's partially true. Obviously Mark makes coaxial designs and it would behoove him to express the postives of that design and not express so much of the negatives.
There's a good thread on another forum where they talk about the Harman spins and how they look for coax designs. They didn't use Mark's speaker but they did test and talk about a KEF which is extremely well regarded. In the midrange and on very sensitive material the issues where detectable by trained listeners in blind testing. However it's still a rather excellent speaker and the point that's made best is that coax designs can be excellent speakers they just have their own challenges. The point source is a good thing, but it comes at a price. Mark explaining getting the off axis sound good makes perfect sense as you explained, that's the reason for the focus on directivity index and achieving a good off axis response and behavior. But I don't think you can't do that with a traditional 2 way as good or better than a coax.
That's a rather cynical analysis of exact claims relative to a reference that was framed as the following:
Originally Posted by VisionxOrb
In the thread that spawned this one ( or heck might be early in this one ) Didn't Mark Seaton say something to the effect ...
I'm not going to waste time trying to find the original. I vaguely recall the comment as a response to discussion of why many find coaxials to subjectively sound very good despite some jagged on-axis response. I suspect the comment VisonxOrb recalled was related to the observed response at the listening position being just as important as the direct sound, as the direct sound doesn't matter much if you can't deliver it to the listener, and in real rooms that requires good off axis behavior. The same exact issues exist for any perfectly round waveguide, and are the reason some squiggle in a horn's direct response can sound and measure more accurate at the listening position than a dome that looks perfectly smooth near field. All designs have their hurdles. A single driver horn/waveguide will usually have bandwidth limitations that require some creativity and *compromises* when blending to the next lower frequency driver. Coaxial drivers often have geometric discontinuities which result in irregular response directly on axis. How significant those issues are and how wide an angle they cover depends greatly on how the coaxial is executed. Similarly any 2 way horn/waveguide design has one axis you can get very good off axis performance, and you must then make a relative compromise in the other axis, which is most often the vertical axis. These are all compromises with varying degrees of impact. That impact is greatly dependent on the quirks and balance of compromises in each implementation.
Rather than the KEF, I'd look to something like the TAD reference series for a better grasp of what is possible.
Originally Posted by tuxedocivic
Ive listened to the Kef. The problem it has doesnt have to do with coax. The designers didnt whack the woofer break up down enough. I forget the details, but stereophile has measurements and the XO on the woofer is too whimpy. Unfortunately, a $5 XO part does matter in mass production. They chose to leave those parts out because in a show room its very hard to hear the problem. If it were fixed, gosh that speaker would be amazing.
That's a common issue in some coaxials, and I agree it is often not dealt with aggressively enough. Of course that's easy to do with DSP and sometimes with a more complex crossover design. Ultimately I've found the biggest battle with coaxials is to get a good grasp of their response on and off axis and using that to decide what set of measurements to correct and design with. The tricky part is that the answer isn't the same for all coaxial drivers.