Right from the word go, I’d like to stress one thing very heavily; I’d rather receive stuff that sounds right, and doesn’t need me to do anything. I’d rather there wasn’t any doubt about how the artist wanted it to sound. If I don’t have to do anything then that’s a bonus – at least I can’t make a pig’s ear of it.
With that said, the low end is hard to get right when mixing. Physics (acoustics) means compromised and problematic monitoring in the low end is, unfortunately, the norm. A small room with terrible time domain and frequency responses, with ported speakers and a single sub only adding to the problems, mean that the typical bedroom producer has to rely on a lot of cross-referencing on a multitude of systems. The car, headphones, mate’s places etc – each of these has their own foibles and tells you a different set of half-truths. This reads as critical or snobby – “just build a mastering room, for goodness’ sake” – but it’s not meant that way. It’s reality, a fact of making music in the real world. This is a whole different topic to the matter of adding/cutting bass in mastering, but bears mentioning.
[At some point I very much want to write on kicks vs low end monitoring vs time domain vs freq domain vs compression/sidechaining, but I’m usually working hard on addressing this stuff and therefore short of time to write about it.]
Anyway, to the point. Very generally speaking, if I had to choose, I’d rather add bass than cut bass when mastering. There are two main reasons for this; one subjective, and one very solid technical reason. Remember here that the essential idea behind mastering is to aim for consistency, sonically. Bringing everything into line as needed. But the direction you go to get there makes a difference, and I’ll try to explain my point of view on it, as far as the low end goes.
The subjective point is simple and straightforward to explain; assuming a good quality final master, nobody ever complains if their tune sounds fatter. If you add low end, even at frequencies “below” where things are supposed to have any content, everything tends to sound fatter. Other elements will very often come along for the ride, so to speak, and that’s a recipe for a fuller, meatier, more satisfying result (again, too much and it’s Mudsville, but I’m assuming we have a good result).
There’s a perfectly valid counter to this; nobody ever complains if their tune sounds clearer, either. And that’s a very valid point. Again, that’s kind of the point of mastering, from a certain point of view – bringing things into line. You’re not going to add bass to a too-bassy mix. However, it can be delicate work to cut low end while keeping things satisfyingly meaty, or at least not sounding thin. I might need more fiddly/fancy processing to keep the kick satisfying while effectively turning down the bass line, and I try to minimise fiddly/fancy stuff wherever I can.
But the stuff above is not the main reason I prefer to add bass rather than cut, if I were forced to choose. That reason is as follows;
Cutting low end is practically identical to adding distortion.
I thought about getting a ton of analyser charts to demonstrate this, but it’s pretty straightforward really.
Consider a sine wave at, say, 50 Hz. You distort it a little bit, and you get harmonics start to pop up at 100Hz, 150Hz, 200Hz, and so on. The exact strength and distribution of the various harmonics depends on the type of distortion, but that doesn’t really matter a great deal here. You get harmonics, that’s the point.
Now you make two copies of that result. With one copy, you distort it more. The energy of the 50Hz is increasingly shared out between all the harmonics, reducing the level of the fundamental tone (50Hz) relative to the harmonics.
With the other copy, you make an eq cut at 50Hz, reducing the level of the fundamental tone (50Hz) relative to the harmonics.
Haaannng on a minute….
A lot of this stuff should now be pretty apparent, but I’ll highlight some concerns anyway. In an era where, apparently, you have to compress/process the arse off everything all the time, and everything has to have colourvibemojosaturationwhatever, there can be a lot of “hidden” distortion in a mix that has too much bass. Your poor, starving mastering engineer makes a 3dB eq cut in the bottom of a very compressed, very bass heavy mix, and all of a sudden the distortion has been turned up by 3dB (not exactly, but it makes the point). So if you’re going to compress stuff, hard, you’d better be absolutely 100% sure that your mix is nailed in all other departments, particularly frequency balance.
Ignoring buss compression for a moment; say you have a comparatively well balanced mix using a warm/fuzzy/fluffy reese-type bass with prominent first harmonic. The mix is good, but that bassline is just a bit too loud. Now we have a problem. The bass is made of harmonics, that’s why it sounds warm/full etc. But that first harmonic now sits at exactly 100Hz, the second at 150Hz (for example). That’s generally where your kick might sit in a style like that. But if the bass is too loud and we have to cut sub, that upper stuff is always going to get more prominent, changing the tone and feel of the bassline. Your ME can cut the 100/150 stuff, but there goes your kick. Therefore we have to look at more complex solutions, effectively expanding that area to keep the kick but reduce the fuzzy stuff. It can get very tricky to de-fuzz and de-bass a mix like that without killing the kick and therefore killing the track. You know, cos there’s music and somewhere behind all these technical issues.
And even without any of that, cutting low end will inherently make a mix sharper/clearer. So you might not have compressed anything at all, and have a clean bassline. But if your tonal balance is off in your wobbly synth pad epic masterpiece, you might find that when it comes back from mastering, it feels a little more ‘edgy’ from distortion in addition to just being clearer. That distortion was there all the time, it’s just been revealed.
There’s lots more of course, but this makes the point pretty well. This is why IF I have to do one or the other, I’d rather add bass – effectively reducing that stuff – than cut bass and reveal it. Yes, you run the risk of muddying things, but you can do the addition in a much more controlled way; you are in full control of what’s being added.
One final note I really wanted to make. I’ve seen videos of highly respected (and rightly so) engineers saying the exact opposite; they’d prefer a bit more bass as it allows them to sculpt the low end. It’s an absolutely valid approach, of course, and much better engineers than me have said this is their preference (again, if they had to choose).
However, for what it’s worth, I’d sugget that people these days are doing much more buss processing, compression, saturation etc than ever before, and often in poor rooms. This tends to mean that people overdo it, all while generally having poor low end monitoring. This means there is rather more distortion lurking behind all that bass, which will inevitably come out when the bass is cut, as described above.
So there you go – my take on it. In reality, I’d suggest that the take away point is that if you’re unsure, ease back your bass by a dB or three. Provide your mastering engineer with two versions if need be – nobody thinks twice about doing this with vocals. So why not bass?