Here is the response I received from Piero when I queried him on the low volume issue, the upper end sub garbage, and the upper end roll-off.
Thanks for the infoFist things first, what does ARC sound like with the new processor and are you using the latest ARC beta 2.4.17? You certainly have the levels set correctly, What I do see in your results (of course not listening to your system) is quite a bit of correction. I'll bet that some boominess has disappeared. The stuff above 500z in the sub measurement could be resonances (a grille vibrating for ex.) If you notice when the sweep tone runs on the sub channel, it still sweeps higher up in the frequency response, just that ARC does not calculate anything above about 500hz (see flat line in the chart). The roll off you see is quite natural, this is the proper response of your room & speakers. There is less energy in the higher frequency spectrum, and some have come to think that this is a loss of some kind. It's not. ARC will not mess with the speakers natural dispersion and response. Given that we are using an omni directional mic, it's most effective to 5k, where correction is most detectable.
Hope this helps, but if you have more thoughts please let me know.
Here was my response:
The system sounds good.
I used 2.4.17
You might be interested to know that if I take an old ARC run with the same speaker array and position, the bass garbage, not present in the original graphs, is created by the newer versions of ARC using the old data. That would point to a quirk in ARC.
I doubt if the Paradigm Sub1's would have a loose grill.
The important question to me is the low volume levels across the board in this ARC run. The actual output level, based on my preferred setting of -15.0db is the same to my senses as it always has been with either processor. So why did ARC graph the levels so low?
FYI...I am outputting all via the balanced outputs.