Your 24Hz peak from the sub is a symptom of room boundary reinforcement. I recommend you pull it away from the floor / wall boundary. If you have it near a corner, put it closer to the middle of the front wall (1/3), or smack in the middle.
Also, can you please tell us at what Frequencies you are crossing over the Summits to the Descent. Is there a possibility that they are both trying to reproduce the low frequencies (<60Hz) and therefore reinforcing each other?
The other thing to double and triple check is phase between the Summits and the Descent. As being out of phase at say 60Hz and in-phase at 30Hz will of course skew the numbers.
And do re-check the phase *after* the SMS-1 corrections have been applied to the sub. Most EQ’s will skew phase as they correct, so ensure that once the sub is dialed in, it is still in correct phase to the mains.
Another item that impacts this is the Speaker Delay settings in your pre-processor. Make sure they reflect your current speaker locations, and for the sub, you will need to factor in the fact that the SMS-1 adds latency (not sure how much latency it adds, but at least 1ms). So if the processor setting for the sub based on distance says 4ms of delay, change it to 3ms (or whatever, based on calculations above) to reflect the added delay of the SMS-1.
[/UNQUOTE]
Thanks for your input, Jonathan.
The Summits are 4ft from the backwall and 3ft from the sidewalls, 6ft apart (inner panel edge). The sub is currently just off-center of the room and also 4ft from the backwall. But like I said before, I have been shifting it all over the room with little effect.
Descent is only used for LFE (processor crossed over at 80Hz) when playing movie. My measurements were taken individually, not both working together.
That's how I can tell they exhibit similar characteristics in the room.
I am not sure at the moment about the phase aspect, since I have not taken measurements of them working together. Am just trying to get them performing right individually before taking a look at how they do together. By the way, when measuring, the speakers are not connected to the preamp/processor. For source, they are only fed the test signal from the SMS-1; the response picked up by the mic at listening position and fed back to the SMS-1 to generate the graph on the screen.
On deeper reflection, Jon, I think I am lucky the peak and dip points are at such low frequencies with the Summits that many respectable speakers can only dream about. The good part is the output is surprisingly flat from around 40Hz up. Of course, I can't tell with the SMS-1 how the upper bass fare/low mid, but it sounds good to my pair of tin ears. Still, I wonder how badly the problem is affecting the overall sound and am anxious to get it right. You'll agree that a +-10dB off is a lot to accept.
Ben