There are a lot of engineers out there who only employ subtractive EQ. For a lot of reasons, additive EQ has some pitfalls, but so does subtractive.
For the purposes of this chapter, I am not going to address issues like phase and other problems that can occur with EQ manipulation. If you would like a little more information about additive EQ techniques, see Mastery in EQ: Additive EQ
MODERN EQs
My primary concern when dealing with additive and subtractive EQ, I usually incorporate both on just about every track, is the EQ curves you create when boosts and cuts are combined on the same plugin.
In the old days, we had an EQ on every channel, but they weren’t by any means nearly as powerful as the multi-channel EQs we have come to love like Pro-Q3, Logic’s Channel EQ, Slate Infinity, and so on. Today, we can make scores of surgical moves on a channel, the options are generally unlimited.
But there is a kind of beauty to a four channel parametric EQ. It forces you to use your ears and be very selective about your moves. More complex consoles like the SSL and Neve had high-pass shelves and low-pass filters as well, which made them even more powerful. All of this is now captured in the digital plugin world as well.
But now we can throw just about anything against a modern digital EQ, and even manipulate different features like L/R, Mid/Side, Notch, Band Pass, Brick Wall Filters, etc., the list goes on and on.
EQ TROUBLE
With modern multichannel EQs, the trouble really starts when they become overly complex. Case in point, I usually place 2-4 subtractive points on my EQ before I start adding anything. When I am done, I usually have 4-8 points on my EQ. Here is a common one for a Kick.
Given these are all pretty standard moves for a kick, this would likely sound great, but there is one problem: the Curve.
Before I make any EQ moves, I generally look at EQ as a flat line. Of course most multichannel EQs always start with a flat line at 0dB, but if everything sounded good, the flat line means nothing needs to be boosted and nothing needs to be cut. Everything is at unity, and all is good. So a flat line is obviously desired, but not always achieved.
But we wouldn’t be here if everything was always good, flat, there is usually something to improve.
Now start crowding the spectrum with boosts and cuts. Your curve is starting to look like a bunch of real curves now. and if they are dynamic, you’ve got all kinds of curves happening.
And now look at your curve! You went from a completely organic flat line to something that resembles a hilly landscape. Not really natural at all.
Worst of all, you are cutting very near areas where you are boosting, and the opposite. Those curves are overlapping each other. When this happens, the cross fade between them will bias the strongest Q. For example, a notch filter next to a boost will drag the curve of the boost down. If the boost has enough gain, it may actually boost the frequencies around any cuts.
Now you are starting to make EQ moves that do t really fit with your design. When this scenario occurs, or to simply prevent it from occurring in the first place, you should consider separating your subtractive and additive EQs.
SEPARATE FUNCTIONS, SEPARATE EQs
I am not a big fan of overly complex EQs. And I am also a big advocate of trying to maintain the most natural, frequency curve given the sound I have to work with.
Commingling a bunch of boosts and cuts on the same EQ is both moving away from my natural curve rule and potentially diminishing the effectiveness of any moves I am trying to accomplish.
Subtractive EQ - So I will first start off with a subtractive EQ and make all of the cuts I need, and only cuts in this EQ. Typically cuts are not right on top of each other and I will generally use tighter Qs when subtracting anyway, so my subtractive EQs don’t become crowded.
Notice the cuts are very narrow, and I am not pulling a lot of gain from them; normally I am cutting 1-3dB. If I have to get really surgical, I may incorporate some notch filters as well:
But again, all of these cuts are occurring on a single EQ plugin, and they aren’t overlapping which diminishes the frequencies I am controlling.
Additive EQ - Now I will add another EQ plugin after the subtractive EQ, and use this for my boosts. As with the subtractive EQ, the frequency bands I usually manipulate will not be on top of each other, so I don’t expect to see a lot of overlap between them.
Yes, I will also generally give boosts a much wider Q than my cuts, they sound more natural that way, but even then, you won’t generally see one boost biasing another.
If I am working with the Slate or Waves SSL or Neve EQs, I may just incorporate two different 4-band EQs to achieve the same effect. One subtractive and the other additive. Either way, I am usually not cutting and boosting on the same EQ.
Comentários