Well, we have two Klippel measurements against your word of mouth so I'd say if you didn't publish your measurements it didn't happen.
This is ASR where negative inuendo needs to be backed up by evidence.
Whatever one thinks of his approach there's no reason provided here to doubt
@mitchco's recollection or veracity. If you understand his measurement methods, which deviate materially from the mainstream approach* in that he not only measures from a "listening position" rather than anechoic or quasi-anechoic, but also only captures a
single point as measured by an omnidirectional capsule, his result is plausible. Remember we are talking about a loudspeaker with a relatively tall tweeter. Amir's anechoic data clearly show the tweeter rolls off quickly off axis vertically:
So, given the measurement approach and the speaker's vertical off axis response, assuming the speaker was working normally, Occam's Razor suggests the microphone was simply placed too high or too low relative to the tweeter. Assume that, and the anechoic data at about 30 degrees off vertical axis tends to corroborate
@mitchco's narrative!
The variance between this measurement approach and the mainstream approach also fully explains Mitch's narrative that a Neumann speaker has more extended top octave response under the same measurement conditions. Neumanns have a wider vertical listening window due to the round tweeter and a highly performant contoured waveguide.
*While Mitch is wont to
name-check certain titans of the field, the practice of applying equalization in the statistical region to fit in-a-given-room measurements to a target curve is at odds with their collective body of work, and IMO is an approach grounded in the logical fallacy
if Y then X; X, therefore Y. See, e.g. "
Again, a room curve is NOT a target, it is a result," and
"When all you can measure is a steady-state room curve (some algorithms can do more, but most do not) a "target" room curve has to be the recipe of the day. This is in spite of the extensive knowledge that it is not a reliable means of delivering neutral, uncolored, sound."(And incidentally
here is the counter to the "but headphones" argument Mitch made in the "name check" post linked at the top of this endnote.)