Ahh, good point irt the dual sub outs. But here's the thing: that's so SubEQHT can measure the level and distance of each sub and adjust accordingly. Then XT32 EQs them firing simultaneously. There has been discussion here about an Onkyo AVR model with XT32 but without SubEQHT. The advice has been that if your subs are equidistant from MLP, are identical and have the sub vol controls set the same, SubEQHT is really just a convenience. If that applies to you, you are free to use a single sub out and run it to the SMS if you want to play with it to tweak the bass. Also, in your case, if the SMS allows distance and trim individual settings those restrictiions wouldn't even apply.
What I'd do is run XT32 the usual way (no SMS) and listen to the result. Chances are it'll be pretty darn good.
If I thought the bass needed further tweaks, I'd use a measurement system like OmniMic, REW, XTZ, etc to get a more objective view of what's going on in the room. Then I'd come back and ask for advice on what might work. IIRC some folks are getting smoother low freq graphs by simply tweaking distance, others have been using PEQ. So you could consider playing with the SMS PEQ. It's YMMV-whether the PEQ will help really depends on your room and current sub placements and how well XT32 can EQ it on its own.
Excellent advise from SoM, as always.
On a second note, when you arrive to the conclusion that you need to take measurements with any of the above mentioned software it's always recommended to take multiple mic position scans and average them coz 1 mic position in the audio world won't tell you the "truth". A mic always "hears" differently than our ears, the results of 1 measurement may thus be easily misleading. You may end up with tweaks that look very nice on a single graph, yet have nothing to do with real world results. It's not just Audyssey that requires multiple mic position measurements, but other Room Correction systems work just the same, so for best results it worth to follow these principles.
Hope this helps.
If you only care about one seat, as I do, for example, then using the mic for REW measurements located at just the MLP is fine. Many (most?) of those who use REW do it this way, after using several Audyssey mic positions to set the system up as well as Audyssey is capable of doing so. So I don't think it is "always recommended to take multiple mic positions" when using REW (but it definitely is when using Audyssey).
I have also found that using REW and making adjustments based on measurements from the MLP only has also benefited the rest of the room in general - eg with sub placement, sub distance/delay tweak for optimisation of the XO frequency area (which Audyssey seems universally to get wrong) and so on. Have you not also found the same when using REW and measuring your post-Audyssey results and then optimising the system based on those REW results? Or do you always use several mic positions and average them?