Single-Point Digital Acoustic Correction for desktop & HiFi speaker systems Using REW and rePhase P4

Part 4
This tutorial is similar to my first except that it generates a filter using only one measurement position. It also presents a different approach to creating corrective EQ banks.
The filters created here will be slightly more accurate due to the fact that only a single measurement point is used. This accuracy is tailored specifically to the main listening position. Although multiple positions are not considered, the sweet spot is large enough to accommodate a listening zone around the main position of reasonable size.
Alternate target slope settings:
LF Rise: 0.5 / HF Fall: 0.4
This set of values produces a very well balanced tone with crisp highs and full lows. The high frequency slope is ideal to my ears, and the slight nudge to the lows helps fill out the punch of deep bass transients.
LF Rise: 0.4 / HF Fall: 0.4
This was my baseline while developing the method. Keeping the two slopes the same gives you a ruler-flat linear target.
Remember to keep these values the same for all EQ windows, including the magnitudes and vectors. Don't change them once you've started generating filter banks. Configuring these options in the preferences at the beginning ensures that each EQ window will have the same settings when you come to them.

[@l.s.1709]

When checking the excess phase copies, the phase line is completely flat, phase is 0 for all frequencies, I don't think that is correct.

[@davidbrancato]

Whether it is mathematically correct or not, the filters generated from this data "sound" great. Granted, this is not the most optimized procedure I could have made, but still I am more interested in how the final filters sound than whether the graphs look "correct" or not.

[@l.s.1709]

@@davidbrancato but if the excess phase is 0 for all frequencies and you use the magnitude measurements for frequency gain filters, I fail to understand what it does add to the filter. Can you please explain?

[@davidbrancato]

@@l.s.1709 it’s not really zero though. When you pull the excess phase text files into rephase, there’s usually a slight rise in the top end so it’s not 100% zero. What this actually does technically is a little bit beyond my understanding but I do know what I hear when I listen to music through the filters and it is a vast improvement. Perhaps one day I will fully understand how it’s working, but for now, I just enjoy the fact that these filters are very close to the ones generated by other professional packages.

[@l.s.1709]

@@davidbrancato thanks, I've been using the minimum phase version of your filter for a week now and I like it. I also saw OCR's version who suggests a mixed power filter instead of the cutoff at 500 hz. Have you tried that already?