Thanks for asking, desertdome.
I have four JTR Captivator 1400s (three original 2015 models and the fourth one is the last generation). They're placed on the front and rear walls 1/4 of from the side walls. Here's a screen shot from REW's room sim that shows the positioning:
The back right sub is the more recent Cap, and the theater door is on the side wall next to it, just in front of the sub. All four subs were gain-matched prior to testing. A Cross Spectrum calibrated UMIK-1 was placed at the primary listening position.
I do not have the subs' heights aligned, due to the height of the riser in the back of the room. One thing I thought I'd try would be to turn the front subs upside down so the port is on top, which would get them more closely aligned.
My receiver's sub output feeds into a MiniDSP 2x4HD, with each of the outputs going to one of the four subs.
From there, I time-aligned the output of the front subs, and then the rear pair. This was done by looking for the smoothest response from each set of pairs, since I didn't have the ASIO drivers working in REW at that point, so I will have to go back and align them with an impulse response.
The next step was to invert the two back subs and change the sub delays. I started with a delay of 7 ms on the front subs, and 0 ms on the back subs, and then took measurements at one ms intervals until the rear subs had a delay of about 45 ms, and also flipped the polarity each way after each change, so it was over 100 measurements. Once I got to the graphs with the fewest valleys, I adjusted the gains on the rear outputs of the MiniDSP to see if there were any improvements.
That sounds like it would take a long time, but one of the beautiful things about the MiniDSP and REW tools is that it only takes a few seconds to change settings and remeasure, so the process is pretty efficient. I was really hoping that at one of those 1 ms changes that the frequency response would have just snapped into place, but it didn't happen.
Once that was done, I set a house curve and had REW calculate EQ parameters and fed them to the MiniDSP. I ended up with a better response than before, but not a clean as some other people have gotten. Also note that I haven't tested out how they sound with any real content yet, so I'm just basing the results on REW response graphs. This may not be wise.
What I still plan to do:
- Time align the subs using REW's impulse response feature and retest
- Try flipping the front Caps upside down to get the cones closer in height to the rear subs and retest
- Take measurements from the front three chairs and the two middle chairs of the back row and plug them into the multi-sub optimizer
- Construct some bass traps, which are definitely needed for the corners of the room, then go back and redo all measurements