I think some of us have quite different ways of interpreting measurements. I'm honestly a little surprised at all the people claiming these measurements are that bad. I know they wouldn't score very highly on the preference score, for various reasons, but I think it's a stretch to suggest these don't generally follow the olive/toole school.
What counts as "a lot of resonances?" the only thing I'd really call a resonance is around 1khz. Something like the JBL 305P is a speaker I'd consider to have a lot of resonances.
The overall tonality is neutral, save for the slightly boosted bass and the bump at 1khz. The boosted bass around 100khz is often shown to be inoffensive in bookshelf speakers, and harman does that on all of its bookshelf speakers, presumably for a reason.
The dip at 1.8khz is likely to be inoffensive as it's fairly narrow and dips are less audible than bumps.
I generally ignore anything above 10kHz unless it is very elevated.
The horizontal directivity is as good as any of the best neumann and genelecs -- I'd personally argue even better than most of those as it maintains something closer to constant directivity from roughly 2-8 khz where it matters most for soundstage. Subjectively, I find this type of behavior leads to more for neutral tonality.
Vertical directivity is better than most non-coaxials.
As noted earlier, the HDI-1600 is basically the 'clean' version of this speaker. I shared my comparison earlier but here it is again with Amir's measurements.
View attachment 158692
However, I preferred the 4309. I do prefer how it looks too, so maybe that has something to do with it, or some other variable, but I was pretty confident in this preference.
I tested it immediately after the HDI-1600 ( I think it arrived the day after I shipped the HDI-1600 back) and the first thing I noticed was "this has a better soundstage," not "this sounds more jagged/different in tonality" I'm not sure I could definitively tell a difference in tonality without a side by side comparison. Imo the only meaningful difference is the elevation at 1kHz, which wasn't immediately obvious to me. Perhaps because it's surrounded by two dips? idk. I tried to listen for it even after I saw the data., but even doing a sine sweep it was hard to convincingly hear.
Maybe the bit at 4k would be audible, although it's maybe worth noting my own measurements didn't show a difference here.
View attachment 158694
Just some food for thought.