EQ bandwidth question...

Farview said:
Why would you want to equate a parametric to the controls on a graphic?

It doesn't make any difference what the numbers say. Twist the knobs until it sounds right, then look at the numbers and write them down. Not the other way around.
Click to expand...
Why not - He was asking - that's all. I assume for some sort of aural experiment. :D

But when you try to use the same settings on another EQ you can bump into the thought - "gee was I crazy when I wrote these down?" - that's why I mentioned the calibration experiment or some other way to get familiar with the range of the controls and the sound. Maybe flopsy wants to know how every octave sounds or something - or is going from the familiar to the unfamiliar? I guess simply chopping up your favorite reference with peaks & cuts works too if you have the ear and monitoring for it.

I guess a DQ does sound better now that you mention it!
 
Ford Van said:
Well personally, I think the bandwidth should be represented like you say. .33 sure looks like 1/3 octave to me, and 5 could be 5 octaves. Fair enough!
Click to expand...

That's similar to how Klark-Teknik designed their DN410 parametric. Each Q control has a range of 1/12th of an octave to 2 octaves. So you can eq individual notes or broadly over 2 octaves. The marking are from .08 to 2 on the Q knob. The next markings up from .08 are: .3 for a third of an octave, .5 for half an octave, 1 for an octave, and 2 for two octaves. A really well set up eq if you ask me, and it sounds good too.
 
SonicAlbert said:
That's similar to how Klark-Teknik designed their DN410 parametric. Each Q control has a range of 1/12th of an octave to 2 octaves. So you can eq individual notes or broadly over 2 octaves. The marking are from .08 to 2 on the Q knob. The next markings up from .08 are: .3 for a third of an octave, .5 for half an octave, 1 for an octave, and 2 for two octaves. A really well set up eq if you ask me, and it sounds good too.
Click to expand...
The Sonic Foundry parametric plug is set up the same way where they don't use "Q" values, but rather spit it out straight as measuring the width in decimal increments of an octave. It sure is a more intuitive way of looking at it than "Q" is; I'd actually prefer reading that value by octave, because that directly relates to my ear better than an abstract number like "Q".

But the concept of "Q" is so imbeded in the history of electronics technology from radio broadcasting to circuit design that it was natural for the engineers to take it down this street as well. Unfortunately for those that grew up without an ARRL membership, Q is perhaps a bit on the counter-intuitive side.:D

G.
 
thanks guys, i'm glad we had this little talk.

it would be nice actually if my eq's bandwidth gave octave-based bandwidth parameter options after all but i think i have learned enough about Q values to at least guess my way through the aural soup i'm stuck in.

thanks a lot for all the info!

f. bunny
 
Flops,

Jay gave you the best advice of all. Don't get too hung up on the numbers. If the sonic soup starts getting too thck, turn off the brain and turn up the ears.

Use the old parametric sweep to ID the center frequency you want to pick out of the soup. Then just start expanding the Q (lowering it's value) until it sounds like you are boosting/cutting more than you need or want to, and back off a bit. The actual numbers are not so relevant.

G.
 
Last edited:
That's the old trick for finding your trouble frequencies.

Boost the band volume really high like at least 10 dB, narrow the Q, and then sweep around until you find the offending frequency. Once you've zoned in on where the trouble is, then you play around with the bandwidth, and finally turn the band volume down into negative numbers. It's much easier to find problem frequencies this way than by sweeping around with the band volume turned down.
 
You must log in or register to reply here.
Back
Top