I really don't meant to pick on you but... this is extremely anecdotal. The only effective research we have into what people actually find sounds "good" shows that people prefer flattish and smoothish directivity. It's not just Harman, studies from Klippel and others show this too. In the past few decades it doesn't seem anyone has been able to find evidence for otherwise. Yes, we could use even more research, but this is the best we have.
Keep in mind this research doesn't only look for what people find most 'neutral,' but also what people find most
enjoyable. And time and time again it is determined these are one and the same.
Doesn't mean individuals don't have their own preferences, and that's always fine. But certainly, when it comes to
making recommendations for someone else, unless you know that exact person's specific tastes, it makes sense to go with what the science says.
I've been on audio forums nearly as long and my own anecdotal observation is very much that the vast majority of people buy into hype and do not get to try all the potential contenders. Even if they do, there's a good chance their own tastes could be influenced by what prominent forum members and reviewers say; "they say it's good, so it must be good!" I've certainly been guilty of that myself.
I think part of the problem is that audio enthusiasts tend to think upgrades are almost limitless. There's this assumption that if you spend a little more, do a little more research, or find something with more exotic materials, you can always find a speaker that's 'better.' But I truly believe diminishing returns sets in rapidly past a few thousand bucks, and when it comes to sound alone, you can only do "different," not better. Problem is, at first "different" often does seem to be "better," until you get bored and want something new again. It's a vicious cycle.
....
Now, all that being said, returning to the topic of the Amphions... I'll play a bit of a devil's advocate. While I'm definitely more in the flat is best camp, I do think it's important to keep in mind diminishing returns for 'flatness,' and the fact that most measurements out there don't create a full spinorama the way we see here from Amir, Erin, and me.
For example, we often see a speaker with a messy on-axis, that looks quite a bit better in the listening window and especially PIR. I suspect the Amphion monitors are a bit in this camp. And yes, the PIR was created for home listening in an untreated room, but I still find it useful for monitoring situations too.
My spin of the Amphion Argon1 (a home speaker, but measures very similarly to the One18) shows some obvious flaws, for example:
View attachment 100728
In this home design, the speakers are likely meant to be heard slightly off axis, rectifying the peaking above 7kHz. The real problem is the mess in the sub-2kHz mids.
These are mainly caused by a resonance escaping from the port:
View attachment 100730
But I have often found these to not sound nearly as bad as they look. Less than ideal, but still, not the obvious distraction it appears to be on a graph. In any case, on many speakers that are EQ-d flat on-axis, the resonance shows up in the off-axis curves.
Taking the spin as whole, the PIR is still fairly smooth and trends quite close to the harman target. And though we all know the preference score has its problems, I think it can sometimes be useful for keeping certain deviations in context. The w/sub score only puts the Argon1 only 0.5 away from the KEF R3 in my measurements, for instance, 0.3 if we're looking at the listening window.
There are other things that I like as well, such as the almost-coaxial-like vertical directivity, and the very smooth directivity after 2kHz (I consider 2-10khzish to be the prime zone for soundstage performance).
One thing I've considered to be of greater importance these days, for example, is the smoothness of the step changes in the individual off-axis curves, not just the average. It's something I've noticed in some of my favorite speakers. To pick a bit more on the R3, for example:
View attachment 100733
We can see a sharp dropoff from the 3-4kHz, about 4dB at 60 degrees. It's a small but potentially noticeable directivity mismatch in a rather crucial region for imaging.
Meanwhile the Argon1 has what I read as smoother changes between angles beyond 2kHz:
View attachment 100734
So while I personally would like to see the measurements tidied up, I also think it's possible the Argon monitors are not
perceptually as far off from the more objectivist-friendly speakers as a few measurements might make them seem. They are not pristine measurements, but they
do still generally satisfy the criteria of flat-ish on-axis and smooth-ish directivity.
For example, let's look at my PIR of both the Argon1 and Neumann KH80:
View attachment 100744
Yeah, the KH80 is smoother, but it's hard to imagine the Neumann would sound
that much smoother.
Comparing it against Neumanns and some the non-coaxial Genelecs, it also wouldn't surprised if, in some setups, the Amphions sounded better by virtue of having less offensive desk and ceiling bounces. I think theres a possibility these would be more audible than squiggles in the on-axis curves.
All of this is not to say I wouldn't like to see more modern design, but rather to provide an explanation for why Amphions can be so popular in a world of flatter speakers. I think they are likely in many cases "flat enough" and "good directivity" enough that it would be difficult to quickly tell their flaws compared to flatter speakers, and in some cases it's possible vertical directivity balances out flaws in other areas.