by Dave Labrecque » Thu Aug 14, 2025 1:05 pm
Hi Dell,
It strikes me that you're still not quite getting some of the details of this particular EQ parameter. Others have answered your specif question. I'll try to fill that out with the "why" aspect.
As has been said the W or width or bandwidth or Q setting determines the range of the boost (or cut) being applied. (note that shape is really a different property that emerges from a combination of the range, the gain, and the EQ's own character) This range can be described in terms of Hz (Hertz), i.e. the higher frequency minus the lower frequency (the difference); this can also be described by a Q (Quality Factor) number, i.e., typically, the center frequency divided by the bandwidth (the ratio of the former to the latter), or in the way Bob does it: the bandwidth (Bob calls it "Q") expressed in octaves of the center frequency. Note that the latter two approaches have the benefit of "accommodating" the moving target that the center frequency is for an audio engineer; width expressed in Hertz alone may be less useful as you hop around the audio spectrum with your EQ tools. A Q number or an octave number better address practical reality for people with ears. A boost or cut of 500 Hz bandwidth is a different thing at 400Hz than it is at 15KHz; whereas a Q of 0.8 (i.e. 1.4 octaves) would likely yield more-expected results as you move from one center frequency to the next. At least that's the idea, I think.
I know when I set up a de-esser using Bob's channel strip EQ, I find it intuitively useful to "back engineer" the range affected using specific Q octave settings. At a setting of 1: double the center for the top number and halve it for the bottom. Work from there. I really don't want to get too far below 3 or 4KHz if I can help it (lest the process audibly impact non-ess content). Even so, as has been said,
use those ears!
Understanding that both Q and bandwidth in octaves each have a corresponding mathematical formula, I was going to figure out the relationship between the two and present the resulting equation. Then it occurred to me that AI could do that for me (if not a good, old-fashioned Google search). Boy am I glad I didn't try to take it on myself.
When I saw in a Q/octaves equivalence table that one octave is equal to a Q of about 1.414, I recognized what appeared to be the square root of two emerging. Right then I knew it was getting deeper than I'd anticipated.
I digress. Does this make any more intuitive sense to you?
Hi Dell,
It strikes me that you're still not quite getting some of the details of this particular EQ parameter. Others have answered your specif question. I'll try to fill that out with the "why" aspect.
As has been said the W or width or bandwidth or Q setting determines the range of the boost (or cut) being applied. (note that shape is really a different property that emerges from a combination of the range, the gain, and the EQ's own character) This range can be described in terms of Hz (Hertz), i.e. the higher frequency minus the lower frequency (the difference); this can also be described by a Q (Quality Factor) number, i.e., typically, the center frequency divided by the bandwidth (the ratio of the former to the latter), or in the way Bob does it: the bandwidth (Bob calls it "Q") expressed in octaves of the center frequency. Note that the latter two approaches have the benefit of "accommodating" the moving target that the center frequency is for an audio engineer; width expressed in Hertz alone may be less useful as you hop around the audio spectrum with your EQ tools. A Q number or an octave number better address practical reality for people with ears. A boost or cut of 500 Hz bandwidth is a different thing at 400Hz than it is at 15KHz; whereas a Q of 0.8 (i.e. 1.4 octaves) would likely yield more-expected results as you move from one center frequency to the next. At least that's the idea, I think.
I know when I set up a de-esser using Bob's channel strip EQ, I find it intuitively useful to "back engineer" the range affected using specific Q octave settings. At a setting of 1: double the center for the top number and halve it for the bottom. Work from there. I really don't want to get too far below 3 or 4KHz if I can help it (lest the process audibly impact non-ess content). Even so, as has been said, [I]use those ears![/I]
Understanding that both Q and bandwidth in octaves each have a corresponding mathematical formula, I was going to figure out the relationship between the two and present the resulting equation. Then it occurred to me that AI could do that for me (if not a good, old-fashioned Google search). Boy am I glad I didn't try to take it on myself.
[img]https://i.imgur.com/k7TeuVM.png[/img]
When I saw in a Q/octaves equivalence table that one octave is equal to a Q of about 1.414, I recognized what appeared to be the square root of two emerging. Right then I knew it was getting deeper than I'd anticipated. :p
I digress. Does this make any more intuitive sense to you?
