How to use a Mix Bus Compressor and EQ / Compress Bass Frequencies when Mixing Audio

How to Use a Mix Bus Compressor, and How to EQ and Compress Bass Frequencies when Mixing Audio

Your Questions:

1) I'm getting my mixes close to being done (I hope) and I'm interested in the issue of what to put on the master busses. For example, I know some people like to start mixing with an SSL compressor or whatever on the mastering bus. Some people put it at the end (which is what I did since I found out about this in the middle of mixing this project). I notice it gels the sound together so that's cool. Now what is your preference on this?

2) And second thing, this is really bugging me and I need an answer lol what do you do to enhance the bass in recordings? you know how metal and rock recordings have that big bass sound, how do you do that? I watched a video that talked about multi-band compressors to boost the low frequencies, which definitely works. I put one on the master bus, BUT is that something that is generally left to the mastering engineer? the multi-band compressing I mean. Or is it okay for me to send mixes in with that already on? I want to make sure he gets the best mix possible.

Thanks Steven, definitely looking forward to your answers.

My Response:

Thanks for your inquiry, I love conversations about production and mixing.  Here are my thoughts based on my rudimentary knowledge and experience with recording:

1. It's my understanding that many mix engineers use some bus compression during the mix stage, before mastering.  There are two schools of thought on when to put the compressor on:

Putting Bus Compression on at the Beginning

People that favor putting it on from the get-go think it helps them mix, set levels, etc. better with all their tracks since the processing is already there, already affecting the sound.  This translates into different fader moves, EQ settings, etc. than you would have without the compressor to achieve the desired sound.  It saves some time and means less finagling with faders later than if you had waited to put the compressor on at the end.  Plus, once you're done mixing, your mix bus compressor is already set and done, too.  No more work to be done.

BF 76 Peak Limiter

Putting Bus Compression on at the End

People that favor putting it on at the end (once the mix is as good as it's going to get) think this is a more correct process flow and think it yields better end results, since you are "forced" to get good levels, EQ, settings, etc. without the compressor on.  In other words, it makes you work a little harder to get the unprocessed tracks up to par.  Then, as the theory goes, when you slap the compressor on at the end, it will enhance the mix and bring it to a higher level.  Plus, you will have more freedom to adjust the settings of the compressor by putting it on at the end since your base mix was not dependent upon the compressor's settings.  In example (a) above, you can't change the compressor settings too much throughout the mixing process because your fader moves, EQ, etc. were all prejudiced by the settings of the compressor at the time those moves were made.  Change the comp settings too much and you have to remix everything again.

I personally favor putting the mix bus compressor (and any other mix-altering effects) on at the end for the reasons described in Method (b) above.  Also, even though it might take longer to get unprocessed tracks "up to par", I think this is technically better since having the compressor on the whole time is sort of cheating, and even if your levels end up about the same, it's going to be more heavily processed, including more processing artifacts.  (That said, some guys with really sweet analog gear might want those processing artifacts... but not me, and I'll save that for mastering.)

Get those settings right!

Method (a) can still be used to produce great results, though, especially when you already know the material, artist, and genre well enough to settle on your bus comp settings from the get-go.  If you already know that you're just going to just absolutely crush the mix, maybe you just crush it from the beginning and save yourself the time.  Or if it's light compression, or a certain attack/release time you're going for that won't fundamentally change based on tempo or whatnot, maybe you decide to dial it in at the beginning (set and forget).  The last thing you want though is to start with Method (a), spend hours and hours on the mix, then decide you don't like the compressor settings.

All in all, I think Method (a) requires more experience and better gear than Method (b), and even if you have the experience and gear, I like the practical step-by-step mentality and process flow of Method (b) better.

[PS - There is an option (c) that we haven't discussed, which is not using any bus compression and simply leaving it for the mastering engineer, since they're going to apply compression anyways.  But I'd rather have my mastering engineer focus his time (and my money) on key multi-band comp enhancements at that stage, among other things, rather than correcting basic leveling and inter-mix balance problems.  That said, I generally don't compress too much on my mix bus because I know the mastering engineer will have a better compressor, a better ear, a better monitoring environment, etc., and he/she should handle the majority of final compression.  BUT I don't want to give him a mix with parts jumping out too loud or disappearing here and there because the levels aren't even, so a moderate amount of compression is still appropriate.]

Shadow Hills Mastering Compressor

2. Enhancing Bass Frequencies

(Whew, there are a lot of angles and aspects to discuss here!)

Recording Quality

Testing Mics/Pres at Capricorn Studios

Before we even talk about the mixing process, it's important to make sure things are done right during tracking.  Garbage In, Garbage Out.  Use decent gear, mics, preamps, etc., and make sure you're getting a good DI track out of the bass guitar, too, to blend with the mic'd sound.  And FYI, phase issues are really important with bass, so if you do end up blending DI with mic'd tracks, you'll need to compensate and most likely manually line up the tracks after they're recorded (or use a phase rotation plugin) so you don't end up with frequency cancellation or comb filtering.  This can be detrimental to bass sound and lead to weak, thin sounding bass.  Some amateurs stay away from blending multiple bass tracks together at all for just this reason, only using one recorded track or one quality DI to ensure nothing's being phased out.

Monitoring Issues

Once you do have a decent track (or tracks) recorded and lined up without any phase cancellation problems, you need a good monitoring environment to get the bass right.  If you're sitting in the corner of a box-shaped bedroom with KRK 5" monitors, which is a very common home recording situation, you aren't going to hear things properly and it will bias your fader moves and other adjustments.  At least not without a thousand dollars worth of bass trapping that would likely take up all the space in a common, small home studio area.

I trust my DT-880 Pros and so
does Mike Senior (pro mix engineer).

One of the first things I've noticed about every mixing book I've ever read is that they first recommend getting your monitoring setup fixed up properly.  Don't spend all your cash on fancy preamps and outboard gear until you've invested the time, energy, and money designing and implementing a decent nearfield monitoring setup.  High-quality mixing headphones like the BeyerDynamic DT880 Pro's (which I have and love) can be a great second reference, but almost no one does all of their mixing that way and for good reason.  Mixes usually don't turn out well when done on headphones, even good ones designed for mixing.

When it comes to bass, getting a proper monitoring setup is all the more difficult, because even with ample 2" (dampens down to roughly 300 Hz) to 4" (dampens down to about 125 Hz) panels, you aren't controlling any bass at all.  You need thick, expensive bass traps positioned in the corners of the room and along the crevices where walls meet to eliminate the reverberation and resulting room modes that are created by bass frequencies.

The Mastering Room at Capricorn Studios

And this is assuming you even have monitors that accurately produce bass frequencies.  My 5" JBL's roll off at 42 Hz, which isn't bad, but here's the thing -- as they "roll off", they produce progressively less and less volume even before you enter and decline through that range, and even if the output volume technically remains the same.  Say your WAV file is telling your speakers to produce a 100 Hz frequency at say 90 dB, then a 60 Hz frequency at 90 dB, then a 40 Hz frequency at 90 dB, then a 35 Hz frequency at 90 dB, then say a 28 Hz frequency at 90 dB.... the speaker itself might get pretty close to 90 dB for 100 Hz, but then because of rolloff, maybe only reproduces 80 dB at 60 Hz, 60 dB at 40 Hz, 40 dB at 35 Hz, and next to nothing at 28 Hz.  8" monitors will typically improve this response, and of course adding a matched sub will help, too.  But without good acoustic treatment, this can all be for not.

Then there's also the fact that human hearing perceives the balance of bass / mid / treble different at different volumes.  As you turn the whole mix down, the bass and treble start disappearing and the mids stick their heads out a little more.  As you turn the whole mix up, the bass and treble start sounding "louder" and more even in comparison to the mids, and eventually maybe even overpowering the mids, because of this well-studied effect.

Supposedly it's an evolutionary thing that we hear mids better because it just so happens to match up exactly to the pitch of a baby's cry/scream (See Mixing Audio by Roey Izhaki [first edition], Pgs. 12-13).  According to Roey, "[The Fletcher-Munson Curves] teach us that our frequency perception has a bump around 3.5 kHz - this is due to the resonant frequency of our ear canal.  Interestingly, the center frequency of a baby's cry falls within this bump."

Thus, you want to make sure you're varying the volume at which you mix, because if you mix really quiet, you'll tend to mix in too much bass/treble, whereas if you mix really loud, you'll tend not to put in enough.  Then it just won't translate well when the average end-user puts the CD in their car, or when it's coming through whatever crappy sound system most people use to listen to music.  Find a good balance, mix at different volumes, and make good use of reference material to get around this.
One final note on monitoring -- especially given all of these issues, many mix engineers and super-nerd home studio guys like me will use frequency/spectrum analyzers to visually "see" the bass (and other frequencies) for mixing purposes.  After you do this enough while listening and mixing, you'll start knowing how to visually judge these regions of your tracks and mixes, and even in poor monitoring environments, etc., you can do a decent job of getting levels right.

This is a great free Frequency Analyzer from Blue Cat.

Lo-cut on other interfering instruments

Okay, so you have your recording done and you've figured out a way to decently monitor your mix.  Now you need to apply a lo-cut filter (EQ) in varying degrees to your guitars, vocals, drums (except perhaps the kick, maybe toms), and basically anything/everything that could possibly be producing bass frequencies.  For certain instruments an EQ shelf might work better, in other cases a 6-12 dB rolloff might be best... you just need to experiment and see where it needs to be for each track.

Using a higher high-pass filter on my Haas Guitar Delay.

Usually most people start rolling the bass off on all these other interfering tracks in the 200-400 Hz range.  Realistically, almost nothing belongs below 125 Hz besides bass and kick; maybe a little snare if that's what you're going for.  And between 125 - 300 Hz, you're still dealing with bass harmonics and the lowest of low mids.  Lows/low mids tend to build up very quickly and become very fatiguing to the ear, messy sounding, and create all kinds of weird phase issues as you stack the tracks on top of each other.  It ends up sounding not very clean and this is why it's good to "cut the grass" and get this crap well out of the way in the very beginning of any mix.  Get your bass and kick in place, then adjust that upper boundary to taste depending on the genre, what other instruments you're dealing with, etc.

Bass EQ

Now that you've cleared or reduced the bass frequencies of your other tracks, you can boost the EQ a bit in the bass region on your actual bass guitar (and kick).  We'll talk more about this in a minute when we discuss EQ / Kick balancing techniques, but the concept is pretty easy.  Start with a decently wide Q and boost at least a few dB or more in the 30-125 Hz range.

Sub-Bass Enhancers / Harmonics Processing

 

Waves LoAir Plugin

There are certain plugins that you can use and/or buy that will read your bass track and add in additional sub-bass or harmonics.  The sub-bass enhancers can do a really good job of adding the low, low end to the mix and filling out the bottom end.  Harmonics processing has a similar effect and may add some slight distortion and harmonics above and below the fundamental note, giving the impression and sound of a "bigger" bass sound, more in your face, more full, etc.

Mike Senior (pro mix engineer) recommends Waves LoAir and Apple Logic's bundled SubBass plugins.

EQ / Kick balance techniques

There is one instrument that necessarily competes with your Bass for room in the bass region of your mix, all the time: the Kick drum!  The kick gets more deep and boomy around 40-60 Hz, whereas it has more "punch" and thump in the 100-125 Hz range.  One technique I particularly like is to decide which area I want the kick to dominate, then do the opposite with the bass.  So if my kick is going to be punchy, I boost the kick around 115 Hz or so, while I boost the bass guitar around 50 Hz.  The Q's can be pretty wide as long as these target frequencies are a good deal apart, such as in this example.  And on the other hand, you could boost the bass around 115 Hz while boosting the kick around 60 Hz instead.

Kick Drum and Bass
must jive together.

This isn't the only way to deal with the frequency competition problem, though, and it can even be combined with other techniques.  Some people will delay the bass or kick track by a few milliseconds to move the "attack" of each instrument out of the way from the other.  What's better in my opinion is to set up a multi-band compressor triggered by the kick drum (from a side-chain) that lowers the volume of the bass guitar by about 5-10 dB every time the kick hits.  You use fast attack and release times so the volume drop is quick and relatively transparent; essentially the bass invisibly moves out of the way and allows a very clean kick sound every time the kick hits, just for long enough for that kick sound to ring out for its short little pop.  Similarly, you can play with expanders, gates, and other tools to essentially accomplish the same thing -- moving the bass out of the way partially, or completely, in one frequency area or across the whole bass track, whenever the kick hits.

Another approach to getting a really solid bass and kick rhythm is using your good ole' frequency analyzer to look at where the fundamentals are peaking for each instrument.  This is easiest with the kick, because you'll see where most of the frequency energy happens.  You'll have the most "natural" sounding boost, and perhaps the fullest sounding kick, if you can boost the EQ on the kick at this exact frequency.  Then you can build your bass guitar sound around it (above and below it, even), but notch your bass EQ a little bit at that kick area.  And of course, all of these techniques can be combined if you really want to go crazy.

The exact EQ settings used on the kick drum when mixing my song, "Economy".

In fact, the combination approach is usually best, because you can get away with more subtle and gradual changes for each adjustment that build on one another, leading to a more natural end-sound than if you just brutally abuse one.  I.e., instead of using an extreme side-chain compression technique to get the bass out of the way when the kick hits, use a moderate one, plus a moderate EQ notch, plus some moderate separation of EQ enhancement.  On an unrelated note, this idea works well on kick drums, too -- in the 4-6k range where most people put their "snap" in their kick sound, it's usually best to boost a little at the main frequency, then a little bit at all of the natural harmonics of that frequency.  Maybe you boost 5-6 dB at the main frequency and its harmonics, as opposed to 13+ dB at the main frequency only.  This will result in a more natural sound that corresponds to harmonics heard in real life, as opposed to a synthetic EQ click that results from such a crazy, narrow Q boost.

Compression / Multi-Band Compression

I suppose this section could have gone above the bass/kick portion of this discussion, but either way... One of the first things I do to my bass track(s) once I have it in is add some pretty crushing compression.  You really need a decent amount of compression and/or limiting to get an even bass sound, and you'll find even at high settings you'll still preserve a decent amount of dynamics.  The last thing you want is for the bass to be spotty and getting too loud or dropping out at times.  It will be very noticeable on any sound system where bass is represented at all, including your monitors.

The exact Bass Compression settings I used on Economy.

It's usually sufficient to simply compress the whole bass track a good deal, but if you want really fine tune control of what parts of the sound are being squashed vs. not squashed, really good results can be achieved with a multi-band comp.  You may even consider parallel comp methods as discussed in this post.

Maybe you absolutely crush the uber-low end to keep the sub-bass really tight, use similarly high compression on the bass fundamentals in the 40-80 Hz range, and then perhaps you leave the low mids and high mids alone, leaving 100% of the picking dynamics and rising/falling harmonics.  Kind of a best-of-both-worlds situation, actually.  But this is just one approach -- the point is that multi-band comp can be very effective. You just have to be careful because you'll typically wind up with phase issues at the crossover frequencies, which can be major or minor depending on your plugin and how much compression you're applying.  Depending on the tuning the bass player is playing in, you might want to set the crossover frequencies at the boundaries of where he plays, or where he usually isn't playing, so the affected frequencies don't really get heard anyways.

***

That's all for today, but please feel free to send in more questions for next time.  Thanks!