Subtractive vs Additive EQ – What Should I Use?

Many beginner and intermediate home recordists ask themselves this very question, and there’s a whole lot of misinformation about this subject to be found online – especially from blogs and YouTube channels geared towards beginners.

The short answer is YES. You should use both. When and what depends on what you’re trying to achieve. Don’t believe anyone who is telling you otherwise. And yes, it’s perfectly fine to boost with a stock digital equalizer in the box.

SO WHEN SHOULD I USE SUBTRACTIVE EQ?

Remember, there are no hard rules when it comes to mixing, only guidelines. One such guideline is that when you have to much of certain frequencies, you cut. And this will happen with pretty much every track you’re ever going to record.

A very typical thing is you’ll have a buildup of room resonances present in most of your tracks, especially if like most of us you’re recording in rooms with little to no acoustic treatment. Then there’s proximity effect on vocals, and low mid buildup on drum tracks in particular.

How wide you cut really depends on the track though. For low mids on toms and kick drums you can normally do a fairly wide cut, while room resonances or frequencies that poke out on guitars and vocals may only need a very narrow notch. Best tip here is if you notice you’re losing some of the good sounding stuff in the track – stop. Try adjusting the width of the cut and then the amount, and normally you should be fine.

This is where the stock eqs in your DAW tends to shine. Others I really like for this purpose are the FabFillter Pro-Q3 as well as Waves Renaissance EQ.

Tip: don’t pay too much attention to the visual representation of the cuts you’re making in the plugin. On some tracks you really need some quite drastic cuts. It may look «wrong», but can absolutely sound very right. Don’t mix with your eyes!

ADDITIVE EQ AND ITS CHALLENGES

When cuts have been done I’ll usually get to doing boosts, as by that time it’s a lot easier to hear what each track needs. And very often I’ll use different plugins for different tracks. Doing some experimenting with this is very wise.

And boost as much as you need for a track to work in the mix you’re working on. Do not pay attention to the so-called 3db rule that a lot of «experts» like to tell beginners. It’s nonsense.

One legit criticism of stock/clean eqs though is that with big boosts in the high end in particular things can quickly sound a bit harsh, tinny or abrasive. That’s mainly because they are incrediby clean and impart virtually no colouration to the sound, more than being in some way inferior or poor. Our ears are just more accustomed to the colouration from analog equipment that were used decades ago, we just find that more pleasing to listen to.

Remember when CDs became a thing in the 80s? A lot of bands and artists reissued their albums and released them on CD, most of them direct transfers of albums that were mixed to tape. And without the limitations of tape and vinyl, those albums sounded absolutely horrible on CD. Thin, harsh and unbelievably bright.

Back to additive eq in a mix, what many like to do when boosting (including myself) is use emulations of analog eqs – SSL, API, Neve, Pultec and similar. Most of the ones out there are well enough modelled that they get in the right ball park quite easily, and they do sound more pleasing to the ear than a lot of the stock clean ones.

If all you have is the stock eq with your daw, try adding some saturation or tape emulation after your eq to take the edges off. This really works well. A little goes a long way!

TDR Nova

USE YOUR FILTERS!

In addition to normal subtractive and additive eq, using high and lowpass filters is also crucial. Cleaning up and tightening the low end in particular – and yes, even on both the kick drum and bass. The amount of clarity you get from doing that right is quite spectacular. Sounds counterintuitive, but a kick drum will often sound bigger and louder just be using a high pass filter (or low cut as some call it) at about 30 Hz. This is because you no longer have low sub frequencies covering up the fundamental frequencies of the instrument.

What many really overlook though is the lowpass/high cut. In a rock mix, as an example, most instruments in a mix don’t really need all that high end up to 20+ kHz. Bass and guitars especially, but most of the drums as well apart from the overheads.

Try taking out excessive high end on the instruments that don’t need it, and you’ll suddenly notice how much better your vocal sounds, as it’s no longer fighting against the fizz from electric guitars, string noises of the acoustics and overly splashy cymbals. A huge part of eqing is making room for the important elements of the mix. Keep that in mind. 😉

THE PHASE ARGUMENT

«Eqs screw with the phase» is one thing that will often get posted in various forums or videos online. And yes, they do. But they will do that whether they’re analog or digital. There will pretty much always be some phase shift involved, especially when eqing multi-miced instruments. As you can guess this frequently happens on drum kits. This is why it’s important not to spend too much time eqing in solo. When hearing the whole kit it’s way easier to tell when you’ve gone too far.

It’s a game of compromises where you just need to find something that works. In most cases you should be fine though. Most modern eqs are good, and some have options for natural/minimum phase – the FabFilter eq is a great one for this.

Be very careful when eqing parallel buses or duplicated tracks though. Phase shifts can get very obvious when doing that. In those cases a linear phase (or minimum phase) eq could be handy – but avoid huge narrow cuts with those! You can get what is called pre-ringing that can quickly screw up your transients. Can sound sort of like a backwards echo and is quite noticable on snares.

What I prefer to do though is send to parallel buses post-plugins, meaning I normally won’t have to do eq on the parallel bus. You can always group them together with the individual tracks later and process them as a group there.

CONCLUSION

In short, do what you need to do to get the results that you want. Sometimes that means you’ll have some really funky looking eq curves. But never stop what you’re doing because it looks weird. There isn’t really any wrong or right way, no matter what some dude in the Gearslutz forum might say. Trust your ears. And a little mono check for phase issues occasionally doesn’t hurt.

Sharing is caring!