Getting good measurements of large dipole loudspeakers is not trivial.
By way of example, John Atkinson uses a fairly sophisticated quasi-anechoic measurement technique and you can see his frequency response measurement of the Magnepan LSR on this page, Figure 2:
https://www.stereophile.com/content/magnepan-lrs-loudspeaker-measurements
Compare that with the frequency response Amir Majidmehr of Audio Science Review recorded using a state-of-the-art, hundred-thousand dollar computer-controlled Klippel measurement system, second graph in the first post on this page:
https://www.audiosciencereview.com/forum/index.php?threads/magnepan-lrs-speaker-review.16068/
There is so little resemblance between those two curves that you’d never suspect they were the same loudspeaker.
And imo conspicuous by its absence is the "in room" curve that John Atkinson normally runs, perhaps because he was unable to get good data using his normal in-room-curve measurement technique.
My point being, it is extremely unlikely that casual non-gated in-room measurements made with a spectrum analyzer are "accurate" when there is major inconsistency between far more sophisticated, industry-standard measurement techniques. This is a situation where I would say, trust your ears far more than you trust your measurements.
Duke
By way of example, John Atkinson uses a fairly sophisticated quasi-anechoic measurement technique and you can see his frequency response measurement of the Magnepan LSR on this page, Figure 2:
https://www.stereophile.com/content/magnepan-lrs-loudspeaker-measurements
Compare that with the frequency response Amir Majidmehr of Audio Science Review recorded using a state-of-the-art, hundred-thousand dollar computer-controlled Klippel measurement system, second graph in the first post on this page:
https://www.audiosciencereview.com/forum/index.php?threads/magnepan-lrs-speaker-review.16068/
There is so little resemblance between those two curves that you’d never suspect they were the same loudspeaker.
And imo conspicuous by its absence is the "in room" curve that John Atkinson normally runs, perhaps because he was unable to get good data using his normal in-room-curve measurement technique.
My point being, it is extremely unlikely that casual non-gated in-room measurements made with a spectrum analyzer are "accurate" when there is major inconsistency between far more sophisticated, industry-standard measurement techniques. This is a situation where I would say, trust your ears far more than you trust your measurements.
Duke