Originally Posted by SBuger
^^^ I agree, it's a trade-off, and one may be better than the other if you need much correction.
Good questions and ones that I've thought about as well.
As far as if the subs or MA's are going to suffer, which will it impact more? Man, I don't know, hard to say, but I'm leaning towards the subs like you say. The subs are so important, but so are the MA's when you combine them. What little bit I've messed with PEQing (peak, not LS) on my MA's (with just the MiniDSP), they just don't seem as sharp to me. Now my BK LFE's I'm fine with PEQing (still feel great), but for whatever reason I don't just love the effect it has on the MA's. Maybe that's why I prefer them ran though my Oppo (untouched completely) compared to my AVR subout, even if Audy's correction should be very minimal with coolrda's approach. Running though the Oppo has it's drawbacks as well, but at least not with any form of room correction (PEQ, phase, or anything else that Audyssey may do, even if very minimally).
Also, very good point about the suspended floor resonances combined with the phase interaction between the subs and MA's. I think I may be dealing with a little bit of that now that I'm on a suspended floor, whereas before, I was on concrete with my previous setup. Or it could just be my different seating now in the new room (old single recliner) compared to my couch in my previous setup ..we'll see soon as I've got some new HT seating on the way.
Your comment about PEQ making your MAs seem less sharp isn't entirely surprising, considering I feel a difference with only 2.5ms of delay with mine. Perhaps we are much more sensitive to group delay with TR rather than ULF sound. This may mean that getting the MAs the most untouched signal is most important, which may make coolrda's approach a better trade-off. However, in my alternative approach, when we auto EQ the Crowson channel to undo the Audyssey filters, I believe we should be canceling out most or all of the group delay as well, since inverted filters also have inverted group delay. I'm just not sure if "unEQing" FIR filters with IIR filters will leave some residual group delay that will be detrimental. I guess testing is the only way to find out, unless someone who knows a lot more than me about this stuff can comment.
As for the phase interaction with the suspended floor resonances, they don't seem to be an issue for me at all. I believe that is because in the frequency range of floor resonance (something like 15-20Hz for me) you'd have to badly mess up your delay in order to create serious phase issues. Additionally, for me at least, the TR from the MAs is so much more powerful than the TR from my floor resonance, I'm not sure I'd notice even if they were interfering destructively.
As a side topic, I don't yet own a MiniDSP and am very intrigued by the USB streaming ability of the 2x4 HD. Being able to send the output from a MiniDSP channel digitally back to REW for analysis seems to be extremely useful, considering it is capturing the entire signal chain up to and including the MiniDSP, without any soundcard influence or calibration necessary. It would allow us to see exactly what is being fed to our MA amp, and flatten the MA response perfectly if we know the rolloff of the amp, which is the single downstream device in the signal path. My only concern is that on other MiniDSP devices, the 96kHz MiniDSP plugins have really funky ULF filter response due rounding errors with their fixed-point filter coefficients. The 2x4 HD has 32-bit floating-point coefficients, so it may not have the issue, but I can't find the information online and haven't yet received a response from MiniDSP about it. I may just have to be the guinea pig and return it if it doesn't work properly. The good thing is the USB streaming capability makes the resulting response really easy to test.