Kim Lajoie's blog

Using subtractive EQ

by Kim Lajoie on January 12, 2012

Petri Suhonen recently posted some good pictures explaining the ‘boost and sweep‘ technique for finding and reducing troublesome frequency ranges. The images are a great way of describing the technique quickly and easily. I do, however, take issue with this statement:

Boosting increases the volume levels of frequencies and it will easily lead to distortion, muddy mix and cause other unwanted artifacts [sic] if you overdo it. Cutting instead leaves more room for the instruments and sounds to ‘breathe’.

There’s nothing inherently wrong with boosting.

Distortion can occur if the EQ boost increases level (i.e. the overall level isn’t turned down to compensate) and there’s a non-linear saturating device further downstream. If you’re reading this blog (and Perti’s blog), you’re probably working in your DAW and the only distortion you’ll get is deliberate (or foolish).

Muddy mixes usually occur if there’s a buildup of lows or low-mids. Yes, you’ll get a muddy mix if you’re inappropriately boosting a lot of tracks down there, but the mud isn’t from the positive gain on the EQ band. It’s from the positive gain and the centre frequency for that band. You won’t get any mud by boosting above 1kHz (you might get honky or harsh sound though).

-Kim.

8 thoughts on “Using subtractive EQ

  1. Hey Kim,

    Thanks for pointing this out – I see now that statement of mine was incomplete!

    So basically:

    Boosting can lead to distortion if its not compensated with level fader, but not inherently.

    And:

    Boosting can cause muddiness, but only if the “mud range” is boosted.

    Right?

    Best regards,
    Petri Suhonen

  2. Kim Lajoie says:

    Hi Petri, that sounds about right. Personally, I think the biggest problems with additive EQ aren’t the ones you describe, but simply the way it makes things sound less natural. Like I described in one of my linked posts, it’s because the modified part of the sound is being emphasised, whereas subtractive EQ de-emphasises the modified part of the sound.

    Thanks for dropping by!

    -Kim.

  3. Hi, I also read around that additive EQ helps increase the perception of phase shift induced by using a non linear phase EQ. Subtractive EQ should instead reduce those phase shift artefacts. This is another pro of subtractive EQ, right?

  4. Kim Lajoie says:

    If you can hear ‘phase shift artefacts’, do whatever you think works best. Personally, ‘phase shift artefacts’ fall in the same category as dither and unity gain. My artists and their listeners can’t hear it, and there are more important things for me to care about.

    You should use EQ to adjust the tone of a sound to make it work in the mix. If you’re optimising your settings for ‘phase shift artefacts’ instead, you’ll end up with sub-optimal tone for the sake of something your clients or listeners can’t hear.

    -Kim.

  5. Hi Kim,

    sorry I simply asked because I thought this kind of EQ phase artefacts reduction was one of these unaudible factors that could actually contribute to “blindly” make a better sound (helping with more consistent transients and so on).
    And as far as I know this is one more reason, even though not necessarily related in a way you can hear and feel, to prefer subtractive EQ over additive.

    Thanks,
    Mario

  6. Kim Lajoie says:

    Don’t be sorry! And don’t do anything ‘blindly’!

    If you can hear it, work with it. If you can’t hear it, don’t worry about it. You’ll waste time and energy that you could otherwise spend dealing with things you *can* hear.

    The only reason to prefer subtractive EQ over additive EQ is because it sounds right for *that* sound in *that* mix.

    -Kim.

  7. cuejosh says:

    Petri Suhonen says: “Boosting can lead to distortion if its not compensated with level fader”

    Reading Bob Katz book and he quotes Roger Nichols as saying that you should away’s cheak you plugins are not clipping as this can lead to unwanted pop’s and click’s the you may not hear untill it’s mastered. Not so important if your working with floating point but still best practice.

  8. Kim Lajoie says:

    What you’re referring to is ‘gain staging’. It’s basically about making sure the level is high enough to reduce the background noise, but not so high that it distorts. It’s critical in the analogue world, but it’s still important when working digitally – especially as many plugins are designed to emulate analogue characteristics.

    I’ve written more about gain staging here: http://blog.kimlajoie.com/gain-staging/

    -Kim.

Leave a Reply

Your email address will not be published. Required fields are marked *