@ Holmz - take off your car audio ball cap and put on your home theater slippers
You make some valid points, but they aren't really relevant to the OP's question/situation. Sometimes taking a third route is good, but he's asking between two only.
@nstain - good on you for using WinISD to find the answer. Indeed, in this situation, from an SPL, space saving, and cost situation, it appears you are better off to use all of the available power to you by wiring your two subs as described for the amplifier to see a nominal 4 ohm load.
What you would gain from using three subs for a total impedance of 6 ohms would (and could) be lower distortion at SPL and more over-excursion protection. These may or may not be important to you. IMO, at 118 db in a room, 1% THD in the low bass region isn't going to be any more discernible than .5% THD or even 10% for that matter. My walls and lights and couch and soda can and fillings in my teeth are usually making more noise and distortion at those levels. But splitting 750 watts three ways vs 1000 watts two ways pushes half the input power to each sub which, assuming the drivers are operating in their linear pistonic range (this is an assumption we all make and WinISD makes, but is almost never true), means the excursion of each driver is halved. Thus - lower distortion (which grows with excursion) and more protection for the drivers.
As for running two separate subwoofers on two separate channels as has been suggested - this is true that it can correct room nulls and create a more even bass response throughout the room. But it isn't always necessary with proper placement of the single sub to begin with. Right now, I am fighting this in my basement; we rearranged some things to improve other aspects of the room and it put a giant hole in my room response from 100-200 hz at the main listening position. A hole over 12 db deep. EEk. But by moving the sub 1 foot to the side, tweaking the delay on the receiver, then clocking the sub 20 degrees back the other direction, I was able to add 8 db back in to the response. Still not perfect but I'm still working on it. And I'm prolly going overboard on the solution but it will be fun
Interestingly, nothing really changed from 20 to 70 hz with those tweaks. I took the brute force approach and ran 30 different iterations of placement, box clocking, and delay and crossover settings on the AVR and that region remained pretty constant. This also tells me that much of my problem is actually coming from the interaction between the sub and the mains. But that's another story.
