Originally Posted by Mfusick
I'm mostly talking speakers, the room dominates the discussion for bass. I don't see a problem with sub measuements as much because with subs you don't worry about off axis performance or total sound power. Speaker measurements and room design andtreatment kind of go hand in hand
Originally Posted by Mfusick
Since you’re so interested in “off axis performance or total sound power”, I’m surprised you posted this data but I’ll go ahead and use this data as an example to explain some basic physics as it applies to audio.
If you look at the Beamwidth graph (top, right, mislabeled “Bandwidth”) than you’ll see there are some interesting things going on. The vertical directivity goes all the way down to 45ish degrees at about 630hz (crossover point of the active versions).This is because of the distance between the acoustical centers of the horn and the woofers. The lower woofer’s center is probably near 36 inches away from the center of the horn. Passive this will cause a downward tilting lobe. Plus loss in intelligibility due to smearing that occurs from the multiple sound arrivals, sound from the top woofer followed by sound from the lower woofer. If you were to put a microphone 1m in front of the speaker and centered on the horn than this is easy to see in a “group delay” measurement, woofers being milliseconds behind the horn. You could tri amp the cabinet and use DSP to delay the drivers and get a good on axis response however the off axis isn’t fixable.
Here is the group delay of the 2014 Noesis 228HT (2015 even better yet) with the microphone 1m away and centered on the horn:
Back to the beamwidth graph. Then within the horn’s frequency range the directivity in the vertical shoots up to 80 degrees while the horizontal remains normal. This is because this horn losses its directivity in the vertical axis much sooner than in the horizontal. This happens because this is an asymmetrical horn (rectangle). If you measure the distance from this horn’s throat to the center of the month on the vertical (top or bottom) and then measure the distance from the horn’s throat to the center of the month on the horizontal (left or right) than you’ll notice the that the horn is longer in the horizontal plane. This causes the horn to have more directivity in the horizontal.
The horn’s we use have symmetrical months plus we use much lower crossover points for the best off axis and power response.
Originally Posted by Mark Seaton
With speaker measurements it isn't usually a matter of fabricated curves, but rather omission of measurement conditions and details. Measurement axis, distance, acoustic conditions, gating and of course smoothing all are factors. Was any processing/EQ used in the measurement? It's actually fairly tedious to simply "generate" a curve that looks real apart from manually editing 1-2peaks or dips. Most of the graphic manipulation comes from a fat marker/lineweight and compression of the vertical axis. Beware those 10-20dB/division graphs!
Where the microphone is relative to the speaker is extremely important (centered on the horn or the cabinet or ???) but never listed.