If you've been mixing for a while, you probably know what the headline refers to. Simply put, the mix you're working on isn't working. Maybe you don't know exactly why, but it's obvious the mix isn't good. Or you may know perfectly well what's wrong with the mix but not how to fix it: the drums have no punch, you can't find the right balance between instruments, it lacks clarity and definition... When mixing, you have to deal with a great many issues, and having fresh ideas for solving problems is important. Below is a collection of techniques used to resolve some common situations. Recording
During recording you can prepare “backup” resources that can be used if something comes up during the mix.

A common practice is to place one or more extra mics in the setup, used solely to contribute unusual effects. They can be set up in odd locations, or run through some extravagant processing, to be added to the original tracks or even replace them.

Similarly, using a polar pattern different from the usual one can open up new ways of approaching the mix.

A very versatile mic technique is MS (Mid-Side), whether on vocals or instruments. For example, you can use only the mid channel during the verses and add the side channel during the choruses to bring size, or conversely, use only the side channel and automate the mid channel's entry to emphasize words, phrases, riffs, etc. You can also use automation to subtly move from mono to stereo across a bridge...

If ideas for the mix occur to us during recording, it's worth writing them down, because several days (or weeks) later we may not be able to remember them.

Any interesting sound that comes up at this stage should also be recorded or saved as a preset. If that isn't possible, a screenshot (Print Screen key in Windows, Command+Shift+3 on a Mac) lets you capture a given setup as an image.

During the Mix
If possible, it's best to rest for at least a day before starting, rather than beginning the mix right after a long recording session. Having fresh ears can work wonders when mixing. However, don't forget that a rough mix can capture the essence of the song. If the project has rough mixes like this, listening to them can give you ideas about where to take the song.

It's also important to listen through the material you're going to mix, even if we recorded it ourselves, to decide which elements are the important ones throughout the track. Maybe what makes the song great is the chorus vocal melody, a guitar riff in the bridge, the drum groove...

Having these elements clear in advance lets you build the mix around them to give them the importance they deserve. Likewise, knowing what's important lets you avoid dwelling too long on what isn't.

Nobody wants to be working on the vocal at three in the morning. If the vocal is important (and it usually is), it's better to start on it early, so you can be sure it sounds good when it gets late and your ears are tired.

Similarly, if it's a demo of a few songs, some are probably better than others. In that case, starting with the weaker ones will let you experiment, define the sound, and even find effects or plug-ins that work well with certain instruments, so that when it's time to mix “the Hit,” you have more tools at your disposal.

Another common problem is fighting with a particular instrument that just won't sit in the mix, whether with EQ, delay, reverb, etc. It may simply be that the arrangement isn't right and the part needs to be removed or swapped for another.

Monitoring
The best way to make the mix translate well on different systems is to check it in as many environments as possible: on different monitors, PC speakers, hi-fi speakers, an MP3 player, in the car... Having several pairs of speakers available is easy with a monitor controller.

One way to emulate the behavior of PC speakers is to put an EQ on the master channel that cuts below 130 Hz and above 7 kHz (approximately). If you also insert a plug-in that sums the signal to mono, you can switch these plug-ins on at any time for a quick check of the mix in a different (and “hostile”) environment.

Headphones are useful for checking for small noises or details that go unnoticed on the monitors, and for checking panning.

Similarly, listening to the mix from the control room doorway offers a fresh perspective, in which it's sometimes easier to judge the level relationships in the mix, or whether all the instruments sound the way they should.

Listening with the screen off helps you focus on the sound, not on the waveforms in the DAW. A quick way to do this is to set the screensaver to a blank screen and put a shortcut to it on the desktop or the toolbar, so you can activate it whenever you want.

In Windows XP you can do this by setting the screensaver to “Blank” and creating a shortcut to the scrnsave.scr file in the Windows\System 32 directory. In Mac OS X, you can use one of the Active Screen Corners, or drag and drop ScreenSaverEngine.app to the toolbar, the Dock, etc.

In general, it's advisable to mix at low levels (60-80 dB), and only raise the level occasionally to check certain aspects, such as how bass and kick interact at high levels.

Similarly, listening to the mix at very low levels is useful for checking level relationships, for example whether the vocal or snare is too loud. This can also be done by listening to the mix with your ears covered (with your fingers), which is especially useful when you've lost a bit of objectivity after long hours of mixing, since it gives you a different perspective.

Another advisable practice is to analyze mixes from commercial records (including genres we don't usually listen to) to check instrument placement, level, EQ, etc.

For this exercise it's useful to import the mix into your favorite DAW and sweep through the spectrum with an EQ to see exactly where the kick, bass, guitars, vocals are... Often that kick that impresses you so much isn't EQ'd the way it seemed.

Likewise, summing the mix to mono with the polarity of one of the two channels inverted will let you hear only the sides of the mix, with everything panned to the center disappearing.

This makes it easier to hear all the stereo effects used (reverbs, delays, etc.), and it usually reveals small arrangement touches that had gone unnoticed.

Processing
An ounce of prevention is worth a pound of cure, but sometimes you have to mix material you didn't record yourself, or a problem shows up that had gone unnoticed until then.

Drums
If hi-hat bleed into the snare mic is a problem, and a noise gate doesn't solve it, putting a de-esser on the snare track can help minimize the hi-hat sound.

You may like the sound of the room mics because they add weight to the snare, but turning them up too much muddies the sound of the rest of the kit. One solution is to put a gate on them triggered by the snare, so the room sound is only heard when that instrument is hit.

One aspect that can make a big difference in the drum sound, and is often neglected, is properly aligning the tracks:

If there are two mics on the kick, the idea is for the transients of both to coincide, in both time and polarity. This can be done by slightly shifting one of the tracks and flipping the polarity if necessary.

Time difference between two kick mics. Depending on the physical distance, this difference will be around 1 ms. On the right is the result of aligning the second track (it wasn't necessary to flip the polarity). This practice should be carried out on every element with multiple mics (the snare will probably be on this list too), and the next step is to adjust the overheads, generally using the snare as a reference.

This element is usually chosen because it bleeds into the rest of the mics the most and because the overheads are often high-passed, so the misalignment caused with the other important element (the kick) will be less of a problem.

The result of this process is usually a sound with more punch and brightness.

Guitar
A high-frequency noise problem from an amp can be reduced by filtering those frequencies to cut the noise, and then using a harmonic exciter to restore some of the lost highs.

A guitar amp can be very useful when the sound of something doesn't convince us and re-recording it isn't possible. Reamping consists of sending whatever signal you want to alter to the amp and recording the result the way we'd record the amp. Very useful on vocals, guitars, snares...

Vocals
If some distortion shows up in the vocal at some point and there's no other take to replace that part, one solution can be to distort the whole vocal during that section.

Compressors let you even out a vocal track, but good vocal riding can take it to another level. This technique consists of drawing the level automation of the vocal track to bring forward the ends of words or to accentuate small details and nuances.

Reverb
A common mistake is too much reverb, which translates into a lack of clarity and definition. One way to avoid this is to get the mix working as well as possible without this effect, and once you've achieved that, add reverb where it's needed.

The reverb time and EQ of this effect should also be chosen carefully. In general, the faster the song, the shorter the reverb should be, and the less low end it should have.

Along with reverb time and EQ, the reverb's density helps create a sense of distance: the denser it is, the closer the object seems. Similarly, adding some brightness to the reverb can make the instrument it's applied to seem closer, without needing to EQ the instrument itself.

Sometimes the mono compatibility of certain reverb processors isn't great, especially on the low-end ones. One way to fix this problem is to use two different processors, one for the left channel and another for the right, instead of the L and R channels of a single processor.

Although a little reverb on each instrument can make everything sound better together, too much can cause the effect to be lost. In a mix with a lot of reverb, the instrument that stands out is the one without any, and vice versa.

Delay
To create space in the mix, the first effect that comes to mind is reverb, but using short delays can create similar effects, and since they tend to be less dense, they help create a sense of space without the instrument losing too much punch.

Subtle, almost subliminal, delays synced to the tempo of the track can help an instrument sit comfortably in the mix. Along these lines, to create some air around the vocal you can use a quarter-note delay. The idea is to EQ it a little bright and raise the level until it starts to be audible, then lower it slightly until it's felt more than heard.

Sync can be very useful, but sometimes you need to turn it off to keep a certain repeat from disappearing, for example behind a snare hit. Also, setting the time manually can bring unexpected results.

Similarly, using less usual note values, such as quarter-note triplets or dotted eighth notes, can inspire new ideas.

Although today's delays can create exact copies, degrading the repeats (generally by filtering the high frequencies) helps the delay sound more natural.

One way to keep reverb or delay from taking up too much space is to use a gate or expander triggered by the instrument the effect is applied to. That way, while the instrument is playing there is barely any effect, and when it goes quiet, the repeats appear to fill its space.

Modulation and Distortion
To close this collection of “possibilities,” a couple of suggestions for using these well-known types of effects:

Modulation effects like chorus and flanger help thicken an instrument's sound, but they can also “dilute” it in the mix, so they should be applied carefully to the main elements.

However, if you apply these effects to the instrument's reverb or delay, the sound gets bigger without the element losing focus. Combining reverb, delay, chorus and flanger in different ways usually leads to interesting results.

As for distortion, it's an effect that can be applied to more than just electric guitars. In small doses it can give character and add interest to any instrument, such as vocals, drums, etc. A slight distortion often brings presence and energy.