Mixing Tips

Many times musicians and engineers will work for long hours on a mix only to find the next day that critical balances are off, i.e., vocals too loud / too low, certain instruments too loud / too low, the mix overall too bright / too dull, too much reverb / too little reverb, etc. The way they usually discover this is by playing them in a more familiar environment (their home or car), and comparing them to other CDs in their collection. Much better to bring those CDs to the studio and compare there, while mixing. Bring along some of your favorite CDs, and do an A/B test periodically. For example, if you find that you have more bass than the reference CDs, turn down your bass! Does your vocal seem low by comparison? Adjust it accordingly. If your mix doesn't pass this comparison test in the studio, it won't at home either.

If you are using a computer-based recording system (Pro Tools, etc.), try importing the reference songs right into your session file. Play the CD tracks through the same outputs as your song. Now you have a totally level playing field. Just remember to turn DOWN the CD tracks, as they will be much louder than your mix, due to the CDs having already been mastered. Also, if you have any plugins on your master fader, you will need to avoid your CD reference passing through them, so you may need to rout them to another set of outputs. But once you get this set up properly, consistency will be yours. There's no need to reinvent the wheel at every mix session.

1 - Don't try to match your mix the volume of the reference CD. It's already been mastered. The mix that THEY used for mastering was not as loud as their final CD, and neither should yours.

2- Don't try to match their compression. Additional compression was likely added in mastering. If you squash your mix, you will be stuck with whatever you did.. Mastering is usually a better time and place for that.

Sometimes mixes fall apart in mono, for a variety of reasons. Why is this important? Who listens in mono anyway? We all do, all the time. If you are not sitting right in front of your speakers, you are hearing some form of mono. If are are listening from way off to the side, or from another room, it's now mono. Most club sound systems are run in mono. Unless you are right in front of the jukebox, you are listening in mono. The majority of TVs are mono. If you are listening to a boombox from more then a dozen feet away, it might as well be mono.

That said, here are the pitfalls to be careful of. Any instruments that have been recorded with more than one mic, such as drum overheads, acoustic piano or guitar, choirs, etc., AND panned wide apart, need to be checked in mono, as there may be phase cancellations that dramatically reduced their perceived level in mono. I have seen cymbals and pianos utterly DISAPPEAR in mono. This can lead to severe inconsistencies when your mix is played in different places.

There are several ways to watch out for this. The best place is when you are doing the initial recording. Some of the more expensive consoles have phase meters. If you are using Pro Tools, your can download the free Bomb Factory Phase Correlation Meter, which will do the same thing. Put it on the master fader, and solo your stereo track. The phase meter should swing to the right when there is sound. If it stays in the middle, or swings to the left, try moving the mics and checking again. You can also check with your ears by switching to mono for a moment, but the metering is much easier and more reliable. For "after-the-fact" phase canceling issues, simply narrowing the panning and raising the level a bit usually fixes it.

I see many mixes where the bass drum (and/or bass) are quite loud and deep on large speakers, but are non-existant on smaller speakers. Yet on some CDs, one can hear the bass drum CLEARLY on even laptop speakers. How is this possible?

Low-frequency instruments (kick, bass guitar, string bass, etc.) exhibit two different aspects of perceived "loudness". One is volume, and the other is bass response. You can make these instruments appear louder by either raising the fader level (volume), OR by increasing the low end response by making eq adjustments (bass response). If you boost the low end, the bass (or kick) will now appear louder, but ONLY on large speakers. If you mix only on one set of speakers, and these speakers have a good degree of low end, then you will be somewhat in the dark as to how this will "translate" to other (smaller) speaker systems. The solution is to have a small set of speakers to switch to periodically while mixing. Speakers so small that they do not have "warmth", speakers that are devoid of "boom". In other words, speakers that do not have any real degree of "low end". Can you still hear the kick? Can you make out the bass? If not, you either need to raise the level of these instruments, or raise some of the mids and highs of these instruments. Then switch back to the larger speakers? Bass and/or kick now too loud? Then roll off some of the low end. Go back and forth until the kick and bass are the same apparent level on both sets of speakers. Try this test with some of your favorite CDs while you're at it. One can hear the kick on Led Zeppelin CDs quite clearly on laptop speakers. Can your mix pass that test?

Is your bass guitar disappearing into an abyss of muck? Hard to make out the notes? Hard to even know the bass is even there? There are many factors, sometimes the main ones being the instrument and the player, but that's another topic... Another very common factor is that there may be too much low end swimming around in the mix for the bass to make it though. Fix that by rolling off the low end on EVERYTHING that is not the bass or the kick (or whatever low end instrument in your mix that needs it's own space). The first tool to grab is the high-pass filter, common to most eqs. High-pass filters remove low-end rumble that has no business being there. Guitars, vocals, and most of the drum mics don't need to be passing such low frequencies. Roll off the low end until you hear it start to negatively impact the vocal or instrument, and then back it off just a bit. You may also need to do additional low end adjustments in addition to the high pass filter.

Not rolling off the low end on these tracks results in an out of control high-bass/low midrange buildup. Aside from making the mix murky, they make it very difficult for the bass to sound good. Proper cleanup of low end on every track will go a long way to solve this.

Below, a worthwhile article on how the volume level of CDs has gotten way out of control.

HOW RECORDING LEVELS EFFECT THE FINAL SOUND

Most people record at levels that are way too hot. Overly hot recording does audible damage on many levels, even if you don't hear distortion while you are doing it. Here they are, in order of appearance.

But first, a quick primer on digital vs. analog metering.

On a digital machine, the top of the meter is just that. Try to go above it, and you will distort the signal. But on an analog machine, it's not as easy to know where the top really is. There is a 0dB marking on the meter, and then there is another 3dB of markings beyond that, with the knowledge that one can maybe go even higher than that. How do these two different metering systems compare with each other? Depending on the individual unit's calibration, the "top" of the digital machine is 14-18dB higher than the 0dB marking on the analog machine. Does this mean you could (or should) using the apparent "extra" range? No, and this leads us to #1:

#1 - MIC PREAMP DISTORTION
If you are playing "pack the meter" on a digital machine (in other words, recording as close to the top as you can without going over), then you are running your mic preamps, compressors, and any other gear you are using PRIOR to the record inputs of your system, at maybe 10-15dB hotter than is was designed to do. You are certainly moving past it's comfort zone, which potentially impacts transients and adds distortion. It's like driving a car at ninety miles per hour all the time.

#2 - DIGITAL INTERFACE DISTORTION
Inputting overly hot signals will also assault the analog electronics in your convertor box, as well as the A-to-D convertors themselves. Some convertors do okay with this, but they tend to be the expensive ones.

#3 - PLUGIN DISTORTION
Now you're in the system. What happens next? Your hot signal will now likely go through some plugins. Some plugins may well show clips lights lighting up. Clip lights indicate overload (of course). Plugin designers DO NOT intend for you to do this. For example, you can hit an LA2A plugin hard and get some kind of emulation of "tube saturation". That is built into the modeling of the plugin itself. But pushing it until the clip light comes on IS NOT part of the design. Now your are past the available headroom of the plugin, and you are clipping in the digital realm. It's not emulated LA2A distortion anymore, it's a clipped digital signal path. Even if you think you can't hear it, it's adding to the global downfall of your mix sonics.

To make matters worse, many plugins will ADD gain to your signal, will takes us to the MIX BUS.

#4 - MIX BUS DISTORTION
Here is where some serious damage gets done. Twenty or thirty channels of high gain tracks will kill the mix bus. Even if your individual channels show no clips, and your master fader shows no clips, the mix bus is still being slammed. Clipping in this place will rob your mix of transients, collapse your stereo image, and make for a small, ratty-sounding mix.

SOLUTIONS
1 - Don't record at levels higher than -10dB, and in most cases, -15dB would be better still.
2 - if your tracks have already been recorded hot, use a destructive gain plugin to permanently change the track to a lower level. In Pro Tools, the (Audiosuite) GAIN plugin works well for this. Carefully monitor the level of the track (not the fader level, but the ACTUAL level of the track itself) until you have it down to a reasonable level. In Pro Tools, you can monitor the actual level by switching the fader reading to show "peaks" instead of "fader level".
3 - Be careful to not add gain with plugins, and if you are using a plugin where that is unavoidable, then gain down the track an additional amount to compensate for it.
4 - Keep your individual faders low enough so that you don't get clips on the master fader. But you were already doing that, right?
5 - If you want to make a LOUD mix for a CD, put a limiter plugin on the master fader and go to town. That is the place to get your hot level. That place and ONLY that place. Record and process your tracks at a reasonable levels, and then crank it THERE. With your tracks clean and transients intact, it's not so hard to get it loud and still have it sound good.

NOTE: If you are planning on doing ANY additional work on the final mix (like mastering), skip the limiter. Limiting must be used LAST, and if you mix is going to be mastered, then your mix is not the last processing step. Failure to do this will yield a noticeably inferior end product.

HOW TO CREATE A LEVEL PLAYING FIELD IN ITUNES FOR THE PURPOSE OF COMPARING YOUR MIXES TO COMMERCIAL CDS

It can be really helpful and enlightening to hear your own mixes in a shootout with your favorite CDs. But iTunes has some preference settings that will attempt to "alter" spacing, general eq, and volume settings. You will need to temporally turn these off in order to make legitimate comparisons. Here's how:

Go into iTunes PREFERENCES, and DISABLE these three items:

1 - Crossfade Playback
- Or else the space between your songs will be shorter than on a CD

2 - Sound Enhancer
- This boosts both the bass and treble response from the original CD, you don't want this!

3 - Sound Check
- iTunes attempts to make all of your songs play back at the same volume, thereby altering their individual playback volumes. Great for casual listening, but not good for judging mixes and masters.

Make sure these three items are UNCHECKED, as below.

You can turn any or all of these items back on when you are done.