Jump to content

A Question about Parametric Q


YammerUK
 Share

Recommended Posts

After struggling to get the Triangle Fuzz properly heard in a keyboard-heavy mix, I decided to copy the mixer EQ onto a Helix block. I picked the "Parametric" EQ, and went about approximating the DAW EQ plugin settings.

 

The defaults for Q in each band are all 0.7, which I thought was an odd value. I'm no expert in EQ — I've understood, so far, that Q represents how pointy the frequency boost/cut is. This is pretty well covered all over the internet, but not much is said about the number itself. I just read today that it's the ratio of centre frequency to bandwidth. So that could mean that (for example) a Q of 1.0 at 440Hz means a curve "beginning" at about 220Hz and ending at about 660Hz (simplified, and not allowing for the logarithmic and dB threshold things). I'm sure it's more complicated than this, but it's a start.

 

My confusion starts with defaults, and the results I got with the Helix Parametric model. I gave up trying to copy the Reaper EQ parameters, and adjusted by ear. Reaper's EQ defaults to a Q of 2.0. Meanwhile my DAW's EQ (I use the Reaper EQ as a plugin) defaults to 1.0. My audio editor's parametric EQ uses Octaves instead of Q.

 

So, is Q the same everywhere, and different manufacturers just like different defaults for some reason? Or is Q not really standardised acrodd the industry, and everyone uses a different implementation/calculation method? And, if so, how does Line 6 calculate Q?

Link to comment
Share on other sites

3 hours ago, YammerUK said:

So that could mean that (for example) a Q of 1.0 at 440Hz means a curve "beginning" at about 220Hz and ending at about 660Hz (simplified, and not allowing for the logarithmic and dB threshold things). I'm sure it's more complicated than this, but it's a start.

  

I don't think L6 is using a different Quality factor (Q), that would be a nonsense from a technically standpoint.

 

Math is easy peasy;

  • Q=f0/BW(f2-f1)@-3dB, where BW is the Bandwidth (delta between f1 and f2)  and f1, f2 are the lower and upper Cutoff Frequencies (let's see them as the width of the f0, at -3dB from the peak)
  • Bandwidth, BW(f2-f1)=(f0/Q)

 

So, for example, using your numbers, at f0=440Hz with Q=1 we'd have (at -3dB from the peak)

 

FormulaLowerCutoff.gif.02a9f635a202da193df8da0a27636621.gif=~272Hz

 

 

FormulaUpperCutoff.gif.f5122800ddc4e91e04de93e22958c4cd.gif=~712Hz

 

Unless you are thresholding that -3dB from it's natural point, you should get pretty much the same numbers and curves in any EQ.

 

EDIT: About that 0.7 that Helix use as default, well, there are not right or wrong numbers. We can only assume they had a reason to do that... like for example to avoid overlapping. Who knows...

 

  • Like 1
Link to comment
Share on other sites

Yes, that's more or less what I read, but without the 3dB criterion. I assumed there would be a threshold for the bandwidth (much like there is when discussing frequency response). But, like frequency responses, there is more than one way of measuring it. For example, the cheapskate manufacturers often use -10dB to make their gear sound more impressive.

 

I wondered if different manufacturers use variations of definition of threshold, or even the formula above. I certainly got the impression that there was some differences between manufacturers, but without any specifics.

Link to comment
Share on other sites

The ReaEQ bandwidth control has it's units in octaves. I've been using Reaper ten years or so, and assumed that while it's the same function as 'Q', the units weren't the same. I don't know whether that's the case, but it's always bugged me - why they just couldn't standardize it.

FWIW, an octave either side of 440hz would be 220-880hz, but I assume you know that, and it's just a typo...

Link to comment
Share on other sites

3 hours ago, somebodyelse said:

The ReaEQ bandwidth control has it's units in octaves. I've been using Reaper ten years or so, and assumed that while it's the same function as 'Q', the units weren't the same. I don't know whether that's the case, but it's always bugged me - why they just couldn't standardize it.

FWIW, an octave either side of 440hz would be 220-880hz, but I assume you know that, and it's just a typo...

 

These are standards. You just need to know how to use them in their best field of application.


Q Factor it's acting on narrow bandwidths (stock 0.7 is ~2 octaves), especially at very high Qs. That's why it is mostly used for fine tuning EQ, like introducing notches, for click/noise removal, harmonics removal etc... You can't do all this stuff with a bare band/octave EQ, without overlapping frequencies you shouldn't touch. Q factor is also very useful for "on the fly" room correction, since you can quickly act on the "offender" without change color to your overall sound. And I guess that's why they used this for the Global EQ.

 

Then, you just go with one band per octave EQ for a standard overall equalization (curve shaping).

Link to comment
Share on other sites

Yeah, I know how to use them, it was the 'Reaper/rest of the industry units thing' that was confusing me. At least it keeps me from 'EQing by numbers'.

Trust yer ears, as they say.

My first amp, back in the 80s, had a parametric control for the mids (frequency and gain). Of course, I had no idea what I was doing, back then. If I'd known then what I know now... I would dearly love to get hold of one, now, but it was a cheapish tranny amp - Laney Session 40 - so most got binned when they died and are rare as rocking horse poop, today.

  • Upvote 1
Link to comment
Share on other sites

 

On 12/28/2021 at 10:24 AM, somebodyelse said:

The ReaEQ bandwidth control has it's units in octaves. I've been using Reaper ten years or so, and assumed that while it's the same function as 'Q', the units weren't the same. I don't know whether that's the case, but it's always bugged me - why they just couldn't standardize it.

FWIW, an octave either side of 440hz would be 220-880hz, but I assume you know that, and it's just a typo...

 

Doh! I didn't notice. I just assumed it was Q, but it does say "Bandwidth (oct)", and the penny didn't drop. That makes quite a difference.

 

It wasn't a typo, I was assuming a Bandwidth (Q=1) of 440Hz, and simply going 220Hz either side of 440Hz (f0), for the sake of argument (and avoiding the slide rule), even though I knew it wouldn't be symmetrical. As PierM calculated, it's more like 272Hz to 712Hz.

 

 

On 12/28/2021 at 10:41 AM, PierM said:

 

These are standards. You just need to know how to use them in their best field of application.


Q Factor it's acting on narrow bandwidths (stock 0.7 is ~2 octaves), especially at very high Qs. That's why it is mostly used for fine tuning EQ...

 

Then, you just go with one band per octave EQ for a standard overall equalization (curve shaping).

 

So, this is where my penny starts to drop: the relationship between Bandwidth measurements in Q and Octaves. I feel a spreadsheet coming on.

 

BW.png

 

And this website does all the maths: http://www.sengpielaudio.com/calculator-bandwidth.htm

  • Upvote 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...