Message 30/32
Date: 26-Sep-02 @ 07:08 PM -
RE: virus filter
one thing (ok, this will be pedantic influx, but at least it is potentially interesting :--) i hope that doesn't become a yellow circle. (i hope i get this right, a lot of it is from memory)...
anyways, Q is often interchanged with resonance, because they're about the same thing; just usually, for those 'out there' not in the music world, they're not designing filters to sound good with the resonance up, actually they're trying to design filters with the maximum cutoff and no ringing whatsoever. most practical non-musical reasons for filtering (and there are many) have totally different goals. Q=quality comes from the internal gain structure; as you increase feedback in the system, you can make the cutoff steeper, in exchange for some 'ripple' or 'ringing' at the point of cutoff.
so, in a real filter, it's not really a strict '12 db' or '24 db' response. the poles do have a large influence on the filter response, but so does Q - when Q is low, we have no ringing, but the frequency slope isn't as steep. as we bring Q up, the filter slope gets steeper, but we start 'ringing'. ringing equals that traditional boost around the cutoff. the higher Q, the steeper the cutoff still but the boost is higher. so there really is more going on than just a boost.
another interesting feature of this is that in a real analog filter, you don't just get one boost right at the cutoff. it's more like a ripple in a pond.. a large boost right at the cutoff, a smaller boost a little lower in frequency, an even smaller boost farther down, fading away (these 'ripples' as they are called also go above the cutoff frequency, but we are generally so far down at this point they are irrelevant). the smaller ripples do not affect the signal near as much as the main ripple, but they could have some effect...