since post #49 I'm doing 3 measurements (center of the listening position, 10cm to the left and 10cm to the right) then EQing on the average, should be good enough, right?
There are differing opinions here as to how many measurements is good enough, but this is also depends on what range of frequencies you are trying to "correct". Since you have only a single desk and seat setup using (anechoically) well known measuring speakers, a single-point technique may be enough to primarily equalize only the bass response akin to how Gelenec's GLM does things. Some people think a single-point is good enough if you take into careful consideration the wavelength sizes and use
frequency dependent windowing.
If you are trying to do more than that, essentially making adjustments above the bass and transition zone well beyond the simple generalized filters already found in the internal DSP of your monitors e.g. HF shelving, you certainly would want more data points. Averaging multiple measurements like what Dirac and Sonarworks does may be warranted...
Charles Sprinkle from Kali Audio has a different approach altogether and has given talks about using the moving microphone method (MMM) to correct up to ~700Hz. Everything above that is expected to be either left alone or adjusted tonally (depending on your ear's preference) with shelving filters.
Also still thinking to the question/image in post #50 if I should for example EQ at 37 Hz and in the bass region in general to have it be smoother, let me know on that guys, please ^_^
Personally, I'd use MMM to possibly smooth it out. IMO, a boost of no more than 5 or 6 dB at most, if desired. Check the decay behavior before and after EQ.
Question: how many filters is too many filters? Right now my EQ is using 19, but on a youtube comment someone said the less you use the better because they add distortion in the time domain... what does that means?
is 19 acceptable?
I know that this may sound contradictory from what I said before, but you can use more... as long as you are aware of the range affected by your filters -- so you might want to think about the specific frequency or range of frequencies affected and the size of the wavelength, the Q as well as the gain. When one is equalizing above Schroeder, the likelihood of mangling the pristine "direct sound" increases. So ask yourself: do I really want to alter the
"direct sound" of the monitors? In the case of more automated algorithms that use FIR corrections, there may be more than a hundred points used for all we know -- though, again, we are not primarily doing detailed speaker equalization here... the safest and simplest goal is to EQ problematic room modes only in the bass that actually stand out to the ears.
Distortion in the time domain has to do with the fact that regular filters can alter the phase response further away from what's considered as the expected "minimum phase" behavior of the system. This generally is not an issue unless if one is using high order high-pass filters, for instance. But if your room's bass decay is really bad, a boosting EQ applied into a problematic area can potentially add more ringing -- so examine the decay behavior . In such cases where it causes additional unwanted ringing, keeping the dip mostly untouched may be preferable for critical listening.
