ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 [All]   Go Down

Author Topic: Parametric EQ frequency resolution  (Read 3107 times)

Weogo Reed

  • Full Member
  • ***
  • Offline Offline
  • Posts: 231
  • Western NC,
    • LiveEdge
Parametric EQ frequency resolution
« on: February 27, 2018, 09:52:57 pm »

Hi Folks,

I work with some Old Time bands that will sometimes wail away in the key of A.
When you get Bass, Guitar, Banjo, Mandolin and Fiddle all sitting on A, it can start to roar a bit, even on good systems.
So I go to the offending A(usually 220, sometimes 110 or 440) and make a narrow cut in the monitor mixes, and occasionally the house mix.

I used to work on a couple different analog mixers with infinitely fine frequency adjustment.
In reality, the knobs were small and a few degrees rotation probably moved up or down the frequency scale quite a bit.

The first digital mixer I worked on, an 01V96, had 1/12 octave steps for the parametric EQs.
For A-440, 450 is the nearest frequency option.

The Midas MR18 has 1/20 octave steps.
432.7 or 447.9 are your choices.

Allen & Heath QU goes even closer, to 439 and 441.

All of these options are far better than a 1/3 octave graphic EQ, at 400 or 500, and have worked fine for me in a variety of situations.

Some mixers, like the Mackie DLs, have one Hertz steps, except down low where they have 1/10 Hertz resolution, and way up high where they round off a bit.
I can go right to 440.
Maybe not a big deal, but nice just the same.

When looking at new mixers, is this a consideration for you?

Thanks and good health,  Weogo
Logged

Luke Geis

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1864
    • Owner of Endever Music Production's
Re: Parametric EQ frequency resolution
« Reply #1 on: February 27, 2018, 11:48:21 pm »

I can't say that I have ever really considered it an issue or concern. I typically mix with my ears and while I am sensitive to frequency within a few Hz, I am not so sensitive that I have ever said darn, I wish this EQ would allow me to center on 433.789hz. The biggest reason I guess it has never been an issue is that when I center around the offending frequency I also utilize the Q to shape the tonal quality of the filter. So while the exact center of the filter may be 5hz or more off of the actual offending frequncy, the Q can be used to dial in something that I can be happy with.

The question may really be is the resolution problematic at the higher or the lower frequencies? If the resolution of the filter is 5hz, then at lower frequencies it may be more obvious? Higher frequencies are much more difficult to determine a 5hz shift with wide band audio signals. Again with adjustable Q it can pretty much a moot issue.

What you might consider a narrow Q, I may not? Typically with musical / tonal changes you use a wide Q filter that is perhaps 1 or wider. While for problem frequencies you will typically go with a filter Q of 1 or narrower. A filter Q of 1 is usually a good starting point. This is a rather wide filter in the grand scheme of things being ( usually ) around 1-1/3 octaves wide. A few db in cut will usually sweeps out a large range of frequencies both musically and with adequate precision starting with a Q factor of 1. Need more precision, narrow the filter Q. Need more musicality widen it.

The only test I have ever really done with the EQ's is to see if the numbers they say are accurate. If I center the filter on 400hz and raise or lower it, does the RTA reflect that change at 400hz and with the degree of cut or boost I present.  My finding has been that there is no short coming of modern digital EQ's and they will pretty much perform at a degree of precision higher than our ears can audibly detect under practical conditions. Obviously if you boost +15db with a very narrow filter you will hear the change if you shift the frequncy up or down even a couple hz. That would be unpractical though.

I might agree that a finer resolution would be better because in theory and practice it is, but I have never been behind a mixer and said this EQ is inaccurate and not fine enough to acquire a desired result. I think in reality it is a non issue, but your values may be different. I say if it is good, then it is good.
Logged
I don't understand how you can't hear yourself

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8866
  • Atlanta GA
Re: Parametric EQ frequency resolution
« Reply #2 on: February 28, 2018, 07:29:09 am »

You could always "ask" them to change their tuning to match your console-----

YEAH, good luck with THAT!
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Caleb Dueck

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 995
Re: Parametric EQ frequency resolution
« Reply #3 on: February 28, 2018, 07:43:25 am »

It's something I assume is relegated to older/cheaper consoles and thus no longer an issue. 

Sent from my VS980 4G using Tapatalk

Logged
Experience is something you get right after you need it.

Dave Garoutte

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1847
  • San Rafael, CA
Re: Parametric EQ frequency resolution
« Reply #4 on: February 28, 2018, 01:15:55 pm »

The instruments described, when on A, are probably playing three different octaves, so the 440 cut may not be the ultimate fix.  Plus the mando is out of tune anyway.
Logged
Nothing can be made idiot-proof; only idiot resistant.

Events.  Stage, PA, Lighting and Backline rentals.
Chauvet dealer.  Home of the Angler.
Inventor.

Nathan Riddle

  • Hero Member
  • *****
  • Online Online
  • Posts: 1752
  • Niceville, FL
    • Nailed Productions
Re: Parametric EQ frequency resolution
« Reply #5 on: February 28, 2018, 01:41:31 pm »

I mostly just have pet peeves with non rounded numbers. Even if the freq of choice was 438.0972745 Hz I'd rather put the EQ right at 440 or 438 NOT 439 or 437.

The signal gen on the X32/M32 bothered me. QU is better, SQ is even better.

QU ALWAYS bothers me that I can't get the EQ back to zero dB gain (without reset) UGH, my OCD is kicking in.

Also, the pan L/R bothers me, the lights are in the center, but idk if the value is really zero, or it is slightly to the left or right.

It's stuff that can't be heard I know that. Still bothers me for some reason.
Logged
I'm just a guy trying to do the next right thing.

This business is for people with too much energy for desk jobs and too much brain for labor jobs. - Scott Helmke

lindsay Dean

  • Classic LAB
  • Hero Member
  • *
  • Offline Offline
  • Posts: 638
Re: Parametric EQ frequency resolution
« Reply #6 on: February 28, 2018, 02:27:42 pm »

Sounds like an acute case of splitting hairs😕
Logged
"A mans got to know his limitations"
     and Pray for higher guidance

Hanno Meingast

  • Full Member
  • ***
  • Offline Offline
  • Posts: 124
Re: Parametric EQ frequency resolution
« Reply #7 on: February 28, 2018, 05:56:04 pm »

Hi Folks,

I work with some Old Time bands that will sometimes wail away in the key of A.
When you get Bass, Guitar, Banjo, Mandolin and Fiddle all sitting on A, it can start to roar a bit, even on good systems.
So I go to the offending A(usually 220, sometimes 110 or 440) and make a narrow cut in the monitor mixes, and occasionally the house mix.

I used to work on a couple different analog mixers with infinitely fine frequency adjustment.
In reality, the knobs were small and a few degrees rotation probably moved up or down the frequency scale quite a bit.

The first digital mixer I worked on, an 01V96, had 1/12 octave steps for the parametric EQs.
For A-440, 450 is the nearest frequency option.

The Midas MR18 has 1/20 octave steps.
432.7 or 447.9 are your choices.

Allen & Heath QU goes even closer, to 439 and 441.

All of these options are far better than a 1/3 octave graphic EQ, at 400 or 500, and have worked fine for me in a variety of situations.

Some mixers, like the Mackie DLs, have one Hertz steps, except down low where they have 1/10 Hertz resolution, and way up high where they round off a bit.
I can go right to 440.
Maybe not a big deal, but nice just the same.

When looking at new mixers, is this a consideration for you?

Thanks and good health,  Weogo
Do you readjust the eq when they switch keys?


Sent from my iPhone using Tapatalk
Logged

Josh Millward

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 713
  • Meridian, MS
Re: Parametric EQ frequency resolution
« Reply #8 on: February 28, 2018, 05:58:39 pm »

As far as frequency goes, I generally consider 3 digits of resolution to be accurate enough.

For example, I would prefer to see:
25.3Hz   This gives you 0.1Hz resolution from 20Hz through 99Hz.
543Hz   Then this gives you 1Hz resolution from 100Hz through 999Hz.
1.25KHz   Followed by 10Hz resolution from 1000Hz through 9,999Hz.
14.3KHz   And finally 100Hz resolution from 10,000Hz through 20,000Hz, or however high you want to go.

I prefer to see gain settings accurate to at least two decimal places, but often the three digit rule can apply there, too.
As an example, I would prefer to see 5.55dB, but when you start dealing with larger number this can be a pain so -55.5dB would be fine too, though that extra decimal place isn't as problematic with gain settings (-55.55dB) as it is with frequency.

Delays require microsecond accuracy. So, 5.555mS is where I would like to see it. You could argue that different delays are required for different places. For example applying delay to a whole loudspeaker for delay alignment needs about half millisecond accuracy whilst driver alignment should be microsecond accuracy. So, it could be reasonable to assume 25.6mS would be fine for loudspeaker delay, whereas 0.054mS may be required for driver alignment.

I love the features on some devices where double clicking on a control sets it to the "nominal" position. This is especially true with pans and EQ gain where I can double click on the control and it sets it back to zero.
« Last Edit: February 28, 2018, 06:02:17 pm by Josh Millward »
Logged
Josh Millward
Danley Sound Labs

Weogo Reed

  • Full Member
  • ***
  • Offline Offline
  • Posts: 231
  • Western NC,
    • LiveEdge
Re: Parametric EQ frequency resolution
« Reply #9 on: February 28, 2018, 10:53:34 pm »

Hi Folks,

Thanks for all the replies!

Ivan, ever work with Medieval instruments tuned to A432?

Caleb, there are $600 mixers that conform to Josh's rule below.
And there are some fairly expensive ones that don't.

Dave, you are correct, as noted, sometimes it is A220, occasionally A110.

Nathan, I hear ya on rounding numbers.

Lindsay, you might have me pegged.

Hanno, Generally I just leave the cuts in.  They are fairly narrow and nobody has ever complained.

Josh, I like your 3 digit rule.

Good health,  Weogo
Logged

Nathan Salt

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
Re: Parametric EQ frequency resolution
« Reply #10 on: March 01, 2018, 01:43:22 am »

QU ALWAYS bothers me that I can't get the EQ back to zero dB gain (without reset) UGH, my OCD is kicking in.

Also, the pan L/R bothers me, the lights are in the center, but idk if the value is really zero, or it is slightly to the left or right.
Cool I'm not the only one haha
Logged
Nathan Salt

Tim McCulloch

  • SR Forums
  • Hero Member
  • *
  • Online Online
  • Posts: 20456
  • Wichita, Kansas USA
Re: Parametric EQ frequency resolution
« Reply #11 on: March 01, 2018, 01:19:22 pm »

Cool I'm not the only one haha

Quit mixing with your eyes.  Seriously.

Back in Ye Olde Analogue Dayz« the click detent of a rotary control probably was not in the exact electrical center of the element by at least 0.25dB.  You couldn't see it, didn't hear it and therefore it didn't matter.

That hasn't changed.  Pay attention to the sound, not the display.
Logged
"Practicing an art, no matter how well or badly, is a way to make your soul grow, for heaven's sake. Sing in the shower. Dance to the radio. Tell stories. Write a poem to a friend, even a lousy poem. Do it as well as you possible can. You will get an enormous reward. You will have created something."  - Kurt Vonnegut

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 16397
  • Hickory, Mississippi, USA
    • Resotune
Re: Parametric EQ frequency resolution
« Reply #12 on: March 01, 2018, 02:45:23 pm »

Quit mixing with your eyes.  Seriously.

Back in Ye Olde Analogue Dayz« the click detent of a rotary control probably was not in the exact electrical center of the element by at least 0.25dB.  You couldn't see it, didn't hear it and therefore it didn't matter.

That hasn't changed.  Pay attention to the sound, not the display.
Back in the day I recall our european reps complaining about a german powered top box that they said was kicking our butt with stronger bass response. We brought one into the lab and checked it out. It turns out they were using a knob with splined shaft that was designed for horizontal orientation on a vertically oriented pot. The net result of this incorrect pot/knob combination was that the pointer when centered on the panel legend was actually commanding a small amount of bass boost.

Not enough to be an obvious problem but enough to phool some reps with low expectations.  :o

JR
Logged
Don't tune your drums half-ass. Listen to what a properly "cleared" drum sounds like.   http://circularscience.com/

Art Welter

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1571
Re: Parametric EQ frequency resolution
« Reply #13 on: March 01, 2018, 02:51:02 pm »

Dave, you are correct, as noted, sometimes it is A220, occasionally A110.
Weogo,

A110, A220, A440, plus the octaves of the notes B, C#, D, E, F#, and G# all used in the key of A, 18 fundamental frequencies for just one key if we limit that to just 3 octaves and no harmonics!
 
I have seldom heard Bass, Guitar, Banjo, Mandolin and Fiddle all exclusively play only one note of the key they were in, so never have felt the need for using specific notes to set an equalizer to reduce a "roar"...

Art
Logged

John L Nobile

  • Hero Member
  • *****
  • Online Online
  • Posts: 1777
Re: Parametric EQ frequency resolution
« Reply #14 on: March 02, 2018, 11:05:15 am »

Weogo,

A110, A220, A440, plus the octaves of the notes B, C#, D, E, F#, and G# all used in the key of A, 18 fundamental frequencies for just one key if we limit that to just 3 octaves and no harmonics!
 
I have seldom heard Bass, Guitar, Banjo, Mandolin and Fiddle all exclusively play only one note of the key they were in, so never have felt the need for using specific notes to set an equalizer to reduce a "roar"...

Art

I would think that would be a problem in rap music. Not sure about them having a key but some songs seem to have a "note". If you eq that out, you're left with no song at all.
Logged

Scott Helmke

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1400
Re: Parametric EQ frequency resolution
« Reply #15 on: March 02, 2018, 12:27:08 pm »

I would think that would be a problem in rap music. Not sure about them having a key but some songs seem to have a "note". If you eq that out, you're left with no song at all.

Then there's the time I had a feedback eliminator on the system and the bagpiper started up.
Logged

ProSoundWeb Community

Re: Parametric EQ frequency resolution
┬ź Reply #15 on: March 02, 2018, 12:27:08 pm ┬╗


Pages: 1 2 [All]   Go Up
 



Page created in 0.059 seconds with 25 queries.