Audyssey will almost certainly give you the runaround and deny that their technology could damage the sub by boosting it below its capabilities. During the measurement they calculate the lower limit (F3) of the sub and the filters taper off below that point to avoid boosting the sub below its lower limit. However, the subwoofer can still be boosted below this frequency due to the "normalization" problem that has been amply discussed in this thread (also something else which Feri will deny exists despite the large body of evidence).
Well, on a slow Tuesday here's some seed for thought I'm throwing in for further discussions:
A recent Q&A I had with Chris K. on the subject:
Q: Hi Chris et al,
I still occasionally come across forum threads where people claim Audyssey boosts low frequencies below the measured -3dB point of their subs and are worried about a possible damage. Some even show graphs taken off the pre-outs of their AVR’s sub-output as evidence showing huge boosts below 10-15 Hz even going down to 2 Hz. Others claim that the issue has been solved in newer AVRs and that is a problem of the past. While some think it’s related to normalization issues Audyssey implements as a final stage of auto-setup thus having such a hiccup at the very low end of the spectrum. Chris, is there any way to put an end to this “urban” legend” or there is something behind in the technology that you could share with us for better understanding? Thank you in advance as always.
A: I've seen these as well. MultEQ (all versions) is limited to a 9 dB maximum boost. Nothing more than that is possible. In the very early versions (many years ago) it was just left to do that and the natural roll off of the speaker that was much steeper would prevent any over boosting. In later versions there is a tapering method applied to stop the boost gradually as you go below the roll off point.
Calling all experts to chime in for further discussions, no "Chris is biased " comments please. Also awaiting members to the discussion who had such issues that they have finally and permanently solved with explanation on how they solved it.
Chris says the boost is maxed at 9dB. That's 8 times more amp power. IMO he doesn’t answer the question. Boosting by 9dB below the F3 could certainly cause problems for many subs, as has been reported here time and again. Chris dodges the issue of normalisation causing the problem and offers no comment at all on that. I don't think it's an issue of Chris being 'biased' - he is doing what he always does - answering the part of the question that suits him and ignoring the part that doesn’t. I am not criticising him for this - he works under commercial restrictions - but it does mean that his answers are often less than useful IMO, especially when the question is linked to the 'secret sauce' as it is here.
Perhaps you'd care to go back to Chris and ask him to comment specifically on normalisation and how that might cause excessive boosts to be applied in, for example, a circumstance where a huge peak has been tamed, thus removing so much energy that a normalising boost has to be applied to 'normalise' the response in dB terms. By 'normalising' in this way, it is entirely possible that the overall boost applied is too great for the sub in question to handle cleanly. Ask Chris to specifically comment on normalisation and let's see what he says...
Edited by kbarnes701 - 5/8/13 at 3:12am