Originally Posted by JOHNnDENVER
Let be real here... Are you guys really doing any sort of room EQ at those high freqs?
For most people, I try to get a decent sub curve, for a few really picky people I will try to get a good curve into the upper ends of the dialog freqs.
I'm just not how much if any practical impact there is with the type and degree of roll off the somewhat lowly regarded RS meter has is all.
You bring up a couple of interesting questions of a general nature.
I will answer your last question first. If the tweeter has a peak or valley, or general roll off of 1/3 octave or more, you will notice a sonic difference. If the RS meter drops that much on "top", then one will adjust the tweeter way too high.
To now answer your first sentence, let me expand on how one can enhance specs and graphs etc.
With electronic components one could extend the frequency response of an amp by using a non-linear signal generator to enhance either the lows or highs or both specifications.
I suppose one could also lower the distortion spec by feeding a small signal of opposite polarity from, say the second or third harmonic.
I knew a new company that fabricated nearly all the specs of an amplifier; they had no test equipment except a digital voltmeter.
Generally, with room/speaker measurements, one could place the mic so that the front to back, side to side, and floor to ceiling nodes would be near each other on the graph and at their peaks. This means mic placement is crucial. The room dimensions would also influence these measurements, how close each peak is to the others.
So while one room may have peaks and valley spread apart, just the right room dimensions could produce peaks (and nulls) from each room dimension adjacent to each other.
Other methods include rotating the speakers away from the mic, or placing the mic facing away from the speakers (There is a difference). One could also aim the mic at the tweeters, midrange, or at the woofer to manipulate the response.
Another way is to use an equalizer and set it to boost or reduce adjacent frequencies. See this link.www.rane.com/ppt/allaboutequalizers.ppt
On click 23 or so (or for that matter there are several pages so read closely) the page will show a peak but flattened on top like a table top. Then notice the middle sagging. In otherwards, there are actually two peaks. One could raise or lower two or more equalizer adjustments, and with the desired Q, one could produce just about any graph one wants with whatever peaks one wants, at whatever frequency one desires.
Another way to influence the measurement/graph is to reverse polarity one of the two speakers. This will cause a near doubling of the amplitude of the peaks and nulls one measures (esp at low frequencies), thus exaggerating the measurement/graph.
Hope this helps John.
Not because it makes it easier to read graphs, which is true, but because our ears respond to both frequency ranges and volume levels logarithmically.
With all due respect, we are not talking about listening/hearing, but how two mics correlate from 250hz to 20khz, which is 98% of the frequency range. Are you sure you had the RS mic pointing directly at the front of the cones/speakers as the RS meter is very directional and being off a little will cause a huge HF rolloff.
Hope this helps.