Home Recording on a Budget

Masters of the Universe: Sweetening the Mix

this shiznit is dedicated to my homey gfg

The Basics

So you've got a mixed track. It probably sounds pretty good. Hopefully, it's something you wouldn't mind showing to your friends and family and saying, "Look ma, I make music about killing hookers!" Or maybe that's just me. Whatever the case, it's likely that the mix you have now sounds good, but still not professional, still not polished. Ideally, this is the part where you throw some bread at the deal and send it off to a professional. Most of us don't have the golden ears or the know-how to make a record sound really good. But, since we're cheapskates (or just flat broke), let's talk about DIY.

If you're listening to your tracks side-by-side with commercial tracks, you're probably noticing a few things. First, your track sounds flat and dull. No sparkle, maybe even a little muddy, comparatively. You have done nothing wrong. This is exactly the sort of thing we correct now. You are probably also noticing that it is very, very quiet. Commercial records are getting louder by the minute, and while the hyper-compressed sound of Nu-Metal may not cut it for your style, you had better give, at least, a deferential tip of the hat to the gods of volume.

The Equipment

There are lots of software packages that might be useful for sweetening a track. On a basic level, you're looking to do two things: change the EQ to bring out the good sounds and get rid of the bad ones, and (usually) compress the dynamic range. Many audio editors (Sound Forge, for example) have built-in effects that fit the bill. There are also plug-in packs specifically designed to let you master your own tracks (my favorite is the Steinberg Mastering Edition; it's available in both DirectX and VST flavor). While the former saves you money (piracy excluded), the plugin packs will be great effort-savers, and certainly easier for the novice. You may also want to consider a BBE-like device -- one that enhances the upper harmonics of your audio.


I've said this before, but I'll repeat it, because it's very important: the primary function of EQ is to de-emphasize the bad sounds. It is not like air freshener, used to cover up the stuff that stinks (thatbadmetaphor, fortunately, is not mine). Listen to your track. Get a good feeling for where it's strong and where it's weak. Put it on your stereo; put it on your headphones. Listen to it under a variety of conditions; at about 90dB almost anything can sound good (it's a fact), but precious few people listen to their music that loud. Then put on a commercial recording in a similar style that you really like. Note the differences. Do you notice the bass more in one recording? Is one brighter? While you're not trying to duplicate the sound of another recording, using one as a guideline is a good idea. You might even want to compare the spectral graphs of the two recordings to see what these different sounds look like and get an idea of exactly which frequencies to cut or boost.

Then, you just have to decide which parts about the other recording you preferred to yours and get to tweaking your EQ (30 bands or more, preferably). If you have the Steinberg Mastering Edition pack, you've got a plugin called "FreeFilter" which will do all this stuff for you. Just feed it a source and a destination, click "match" and it will take care of the EQ for you, and give you 30 bands of sliders to tweak with. Possibly the most useful plugin for mastering. An important thing to watch here is your volume: don't forget that you'd be better off cutting frequencies than boosting them. Digital clipping is not pretty.

Don't be discouraged. This is, by far, the most difficult part of making a professional-sounding record. You might come out with something brilliant on the first try, but you're just as likely to go through three, four, maybe 10 revisions before you're satisfied with the sound.


Once you're satisfied with the EQ of your track, you're going to want to make it louder. Most likely, you're dealing with a waveform that doesn't quite hit 0dB. If it does, then you might have some clipping; have a very close listen before continuing. If your style of music doesn't need to be very, very loud, then you might just need to normalize your wave from here. Bring it up so the maximum volume is 0dB (or maybe -0.03dB or so), and you're done. Read on, though, because even those of you who play wussy music might be able to benefit from a little compression.

Although we lose some dynamic range when compressing our audio (duh, that's why it's called "compression"), it can be quite helpful in many cases. Let's face it: most rock music doesn't have much dynamic range anyway. Even in the "quiet" parts, you've got a sea of guitars trying to fill up the sonic space. So by compressing the dynamic range -- making the loud parts a little quieter and making the quiet parts louder to compensate -- we do ourselves a tremendous service: we can squeeze those extra few decibels of perceived volume out of the mix.

The most basic way to do this is with your plain, vanilla compressor. Take a look at your waveform: where does the bulk of it peak? Let's say you've got a pretty consistent high point of about -6dB, with occasional spikes that send it up to 0dB. You will probably want to set your compressor at -5dB or so with a pretty high ratio. This way, you'll clip some of the loud stuff -- which was probably just plosives anyhow -- and let the meat of the mix be louder. You could, of course, do more complex things with a graphic or parametric compressor, but most of that is beyond what we need here. It certainly doesn't hurt to experiment, though; you can get some interesting effects with different-shaped compression, untraditional as they may be.

Another option for compression that you may want to try is a loudness maximizer. It's a bit like a normal compressor, but does some black magic to increase the perceived volume without clipping. I have no idea how these work; I'm sure somebody does and can enlighten us all. In any case, they are tremendously useful in making your recording sound as loud as commerical recordings. The one shipped with the Steinberg Mastering Edition does a crackerjack job of boosting the perceived volume without making the recording sound harsh. Some of the cheaper maximizers tend to lend an overly bright, fuzzy sound to the recording. You might want to watch out for these.

Conclusion; Words of Warning

Probably the most important part of sweetening your mix is knowing when to hold back. Overdoing it will just give the mix an overprocessed, fake feel, which (in most cases) is highly undesirable. Overdoing it, at least in the digital realm, is also likely to introduce all sorts of artifacts you definitely don't want. Stay at the highest possible resolution for your sound card at all times. If you can do 24-bit editing, do it. You will use more disk space and memory overhead, but it will be worth it. Also, take frequent breaks to rest your ears. Ear fatigue is the scientifically-proven, kid-tested, mother-approved cause of bad mixes. Tired ears gradually lose their ability to hear high frequencies; you will wind up making mixes that sound brittle and tinny.

Mastering takes time and trained ears, and training your ears takes time and energy, and energy takes bananas. So eat your bananas, kids, and you'll make better music. I promise.


Log in or register to write something here or to contact authors.