Tilt Equalizer

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
Gold
Posts: 435
Joined: Sat Dec 01, 2007 6:20 pm

Re: Tilt Equalizer

Post by Gold » Sun Apr 19, 2015 2:38 pm

JR. wrote:I have used tilt EQ in the side chain of noise gates where they were pretty effective at improving the selectivity of the gate helping key on or ignore frequency spectra.
That sounds like an ideal application. I often do what a tilt EQ does but a tilt EQ is not flexible enough for daily use. I want a four band shelving EQ. Two high and two low bands with frequency selection. A common problem is an apparent lack of bass because whoever mixed it put in a ton of sub bass but left out from 200Hz to 800Hz. A tilt EQ deals with the lack of mid bass/low mid but does damage by increasing sub bass.

I think the crowd over at GS is a young one. The GS bias seems to be for equipment that is very gentle and stuff that when you put it in line makes things sound "better" without doing anything. With power comes responsibility.

User avatar
mediatechnology
Posts: 3276
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Tilt Equalizer

Post by mediatechnology » Sun Apr 19, 2015 3:50 pm

I want a four band shelving EQ. Two high and two low bands with frequency selection.
Thanks for that Paul.
What particular frequencies and boost/cut range?
I know you said the problem is often 200-800 Hz.
What top end ranges?

Gold
Posts: 435
Joined: Sat Dec 01, 2007 6:20 pm

Re: Tilt Equalizer

Post by Gold » Mon Apr 20, 2015 1:29 pm

For the two low bands I'd want band one to be about 30hz to 250Hz and band two to be 200hz to 800hz. I would only need +/- 3dB but others would probably want more. Band one would be mostly for cutting. I think a medium narrow bandwidth for this. Band two would be the broad strokes band so this should have a very wide bandwidth. The High bands would be the opposite. Band Three would be wide bandwidth from about 700hz to 8K Hz., wide bandwidth and band four would be from 7k-18k medium bandwidth.

User avatar
JR.
Posts: 2418
Joined: Sat May 24, 2008 7:21 pm

Re: Tilt Equalizer

Post by JR. » Mon Apr 20, 2015 2:20 pm

If we forget about the the "less is more" theme, I made a nice console input strip EQ for LOFT in the late '70s that was 4 band full parametric. As i recall the top and bottom bands were switchable between shelving and peaking, while the low-mid and high mid were fixed peaking.

I made the shelving response by summing the band pass and High pass (or low pass) of the parametric SVF together... Not simple and not a conventional Baxadall shelving response as there was also some q adjustment/interaction in the shelving mode. IIRC high Q settings gave an underdamped peaking before roll-off response that could be useful (or not).

I don't recall getting any complaints but Loft did not deliver that many consoles back then. If you know what you want it seems like a pretty powerful EQ approach.

JR

PS OTOH I know a guy who put 9 band graphic eqs in every console input of his custom built console and he loved it because he could very quickly dial in EQ (He did a lot of advertising work, so speed was a good thing). More powerful EQ requires more time to dial it in. So that is a classic trade-off.

Gold
Posts: 435
Joined: Sat Dec 01, 2007 6:20 pm

Re: Tilt Equalizer

Post by Gold » Mon Apr 20, 2015 7:28 pm

I was just giving my idea for a fully featured "tilt like" EQ. I have lots of parametric bands at my disposal. I look at it as 'more than a tilt' not 'less than a full featured EQ'. The consoles at Bernie Grundman use a graphic EQ as the main EQ. I think it's a modified UREI.

User avatar
mediatechnology
Posts: 3276
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Tilt Equalizer

Post by mediatechnology » Tue Apr 21, 2015 1:12 pm

If I'm reading this correctly the desired shapes would be bandpass with a flat adjustable shelf within passbands.

An analogy might be a speaker crossover.
The "excess energy in the sub and too little LF energy in the mains that occurred in mixdown" problem you cited would seem to suggest the need for a "crossover" like device to correct it.

One approach might have four knobs, one for each bandpass, with +/- 3dB trim range.

For a second approach would an asymmetrical tilt curve like the one published in the OP that had a rotation point of 250 Hz and 8 kHz be useful?

"Tilt" in the sense that it allow the 30-250 Hz band to be oppositely increased/decreased relative to the 250-800 Hz band.
One that as you cut 30-250 Hz it would simultaneously raise 250-800.
A second stage would allow the 700-8 kHz shelf to be "tilted" relative to 8-18 kHz shelf.

Gold
Posts: 435
Joined: Sat Dec 01, 2007 6:20 pm

Re: Tilt Equalizer

Post by Gold » Tue Apr 21, 2015 1:33 pm

I guess I'm confusing matters by using the word tilt. I pictured the EQ I'm describing as each band having a knob for gain and for frequency select, like a normal EQ. When I was talking about the bandwidth of the shelf I was refering to 'slope' of the shelf. I don't like using the word slope with shelving EQ. I consider the word slope only applicable to filters with poles.

User avatar
mediatechnology
Posts: 3276
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Tilt Equalizer

Post by mediatechnology » Tue Apr 21, 2015 4:06 pm

I guess I'm confusing matters by using the word tilt.
I thought I might have been the one confusing it...
I visualize "tilt" as meaning a control where, for example, you increase the low end and there is a simultaneous decrease in the high end or vice-versa.

What you've been describing provides more control that's also non-interacting.

I'm wondering how four bandpass filters (or LP, BP, BP, HP) added or subtracted to the main path (in parallel) would sound.

Dolby had a cinema processor that a CAT-22 NR plugged into that allowed the CAT-22's bands to be skewed.
I realize that's dynamic EQ but IIRC it allowed for static changes in spectral balance using the CAT-22's filters and gain cells.

User avatar
JR.
Posts: 2418
Joined: Sat May 24, 2008 7:21 pm

Re: Tilt Equalizer

Post by JR. » Tue Apr 21, 2015 4:42 pm

mediatechnology wrote:
I'm wondering how four bandpass filters (or LP, BP, BP, HP) added or subtracted to the main path (in parallel) would sound.
Adding or subtracting the BP response of a SVF is how typical (peaking) parametric EQ works, so it sounds like EQ.

Adding BP+LP or BP+HP is similar to shelving (I've done this before) while variable Q affects the shape of the transition region before the stable shelf.

We may be over thinking this. I'm not sure the world needs a new type of EQ... While those in more of a fashion business need to respond to current trends. In an extremely mature segment (like EQ) competitors need to be alert to new ways to differentiate themselves.

Being old I am generally resistant to change, and as an old engineer I am resistant to re-inventing already very round wheels.

JR

PS: I will apologize in advance but with DSP you might actually execute an arbitrary "tilted" +/- XdB transfer function. I remain unenthusiastic.

Gold
Posts: 435
Joined: Sat Dec 01, 2007 6:20 pm

Re: Tilt Equalizer

Post by Gold » Tue Apr 21, 2015 8:38 pm

mediatechnology wrote:
I guess I'm confusing matters by using the word tilt.
I visualize "tilt" as meaning a control where, for example, you increase the low end and there is a simultaneous decrease in the high end or vice-versa.
That is a Tilt EQ. You use a tilt EQ to make very broad changes of the frequency spectrum around a center point. The EQ I'm describing is to make very broad changes around a fuzzy and moving center point with separate control of highs and lows.

It's not a tilt EQ in function but it would be used in the same situations that one would use a tilt EQ. Clear as mud?
Last edited by Gold on Wed Apr 22, 2015 12:32 am, edited 1 time in total.

Post Reply