Corrected for Multiple Locations

[Ayreonaut]

After I relocated my sub to the corner of my room the uncorrected response looked like the plot below. I had gained 4dB in the lowest frequencies and was quite happy.

Uncorrected at Seat 1


After applying the necessary filters I managed to get this curve at the primary listening position. Looks pretty nice huh?

Original Correction at Seat 1


But I could tell there were problems in other seats in the room and so I measured four listening positions to find out what was going on. This was with the original correction designed for seat 1. Sure enough, there was a huge peak in some of the other positions. Yuck.

Original Correction Checked at All Seats


This peak had to be brought in check even though it apperas that my "primary lisening position" is in a null. Comprimises are necessary if you want everyone to be able to enjoy the sound. I think that a valley at one position is prefereable to a peak at another since an exaggeration sounds much more pronounced than a leanness. I settled on a comprimise that leaves just a little extra energy in a couple of seats, and a bit of a dip in the other two, but it still sounds very good everywhere.

Final Correction at All Seats

[JohnM]

Good work. Those don't appear to be sweep measurements, are you manually measuring at spot frequencies and importing the figures? Sweep measurements, if you can get the equipment set up to make them, give you a lot more precision which allow your corrections to be more accurately targeted - ensures they are never wider than they need to be, which can be helpful in minimising the dips in locations other than the primary position.

[Ayreonaut]

Yeah, these are from the old 1/6 octave test tones. I'm planning to build an IB in the fall, so the EQ on my current sub is only temporary. Once I get the IB installed I'll suck it up and transplant my PC from the bedroom into the living room and figure out how to do the sweeps.

I recently read the paper "Loudspeakers and Rooms for Multichannel Audio Reproduction, Part 3 – Getting the Bass Right" by Floyd E. Toole. He concludes that "To see what is happening at low frequencies, analyzers must have resolution of 1/10 octave or better." Still, for a temporary solution, I'm glad that I tamed the big peaks, even if it is a bit blunt.