Start a hosting plan from $3.92/mo and get a free year on Tuts+ (normally $180)
Just like everything else in music and recording, it’s now possible to master your own material. The tools are readily available and are very inexpensive compared to previous audio generations, but just because you own a hammer doesn’t mean that you know how to swing it.
Also available in this series:
- Mastering: You Can Do It Yourself (With a Little Caution)
- Mastering: You Can Do It Yourself (Part Two)
Mastering is a basically simple process, but like all simple processes, it’s a lot more involved than it seems. As long as you know a few tricks and don’t have beyond-reality expectations for the end result, it can improve your program material by varying degrees, or just as easily make it a lot worse than what you started with.
Before we get into the mechanics of mastering, let’s take a look at what this process called “mastering” really is.
What is Mastering?
Technically speaking, mastering is the intermediate step between mixing the audio and preparing it to be replicated or distributed. But it’s really much more than that.
Mastering is the process of turning a collection of songs into a record by making them sound like they belong together in tone, volume, and timing (spacing between songs).
Mastering is not a set of tools or a device that music is run through and automatically comes out mastered (despite what the adverts for these types of devices say). It’s an art form that, when done conscientiously, mostly relies on an individual’s skill, experience with various genres of music, and good taste.
Why Master Anyway?
Mastering should be considered the final step in the creative process since this is our last chance to polish and fix our project. A project that has been mastered simply sounds better. It sounds complete, polished, and finished. The project that might have sounded like a demo before now sounds like a “record”. Here’s why:
- The mastering engineer has added judicious amounts of EQ and compression to make the project bigger, fatter, richer, and louder.
- He’s matched the levels of each song so they all have the same apparent level.
- He’s fixed the fades so that they’re smooth.
- He’s edited out distorted parts or glitches so well you didn’t even notice.
- He’s made all the songs blend together into a cohesive unit.
- In the case of mastering for CD or vinyl, he’s inserted the spreads (the time between each song) so the songs flow seamlessly together.
- He’s sequenced the songs so they fall in the correct order.
- He’s proofed your master before it’s sent to the replicator to make sure it’s free of any glitches or noise.
- He’s also made and stored a back-up clone in case anything should happen to your cherished master.
- He’s taken care of all of the shipping to the desired replication facility if you’re using one.
And all this happened so quickly and smoothly that you hardly knew it was happening.
The Difference Between You and a Pro
If we really break it down, a mastering pro usually has 3 things over what you do at home.
The Gear: A real pro mastering house has many things available that you probably won’t find in a simple home or small studio DAW room such as high end A/D and D/A converters, ultra-smooth outboard compressors and equalizers, multiple tweaked 1/2” and 1/4” two tracks tape machines (if needed), a great sounding listening environment, and an exceptional monitoring system.
The monitor systems of these facilities sometimes cost far more than many entire home studios. Cost isn’t the point here but quality is, since you can rarely hear what you need to hear in order to make the adjustments that you need to make on the commonly used nearfield monitors that most recording studios use. The vast majority of monitors, and the rooms in which they reside, are just not precise enough.
The Ears: The mastering engineer is the real key to the process. This is all he does day in and day out. He has “big ears” because he masters at least 8 hours every day and knows his monitors the way you know your favorite pair of sneakers. Plus, his reference point of what constitutes a good sounding mix is finely honed thanks to working hours and hours on the best and worst sounding mixes of each genre of music.
A Backup: I don’t know who said it, but it’s true. “The difference between a pro and an amateur is that a pro always has a backup.” Good advice for any part of recording, but especially for mastering. You wouldn’t believe the number of times masters get lost. This is the one thing that you can do just as well as a pro can with no trouble at all!
But you don’t have the money to use a pro, right? Here’s how to do it yourself.
The Mastering Technique
What you’re trying to accomplish in mastering is the following:
- Raise the level of the songs so that they’re competitive with others on the market
- Make them all sound the same in relative level and tonal quality
- Finish them by editing out count-offs and glitches, fixing fades, adding PQ and ISRC codes, and creating spreads for CDs and vinyl records
- Export them as MP3s, AIFF or WAV files
Let’s look at audio level first.
Perceived Audio Level
The amount of perceived audio volume, or level, without distortion (either on an audio file, CD, vinyl record, or any other audio delivery method yet to be created) is one of the things that many top mastering engineers pride themselves on. Notice the qualifying words “without distortion” since that is indeed the trick; to the make the music as loud as possible (and thereby competitive with other products on the market) while still sounding natural. Be aware that this generally applies to modern Pop/Rock/R&B/Urban genres and not as often to Classical or Jazz, whose listeners much prefer a wider dynamic range where maximum level is not a factor.
The volume/level wars really began way back in the vinyl era of the 50s when it was discovered that if a record played louder than the others on the radio, the listeners would perceive it to be “better” sounding and therefore make it a hit. Since then it has been the charge of the mastering engineer to make any song intended for radio as loud as possible in whatever way they can.
And of course, this also applies to situations other than the radio as well. Take for instance the iPod, CD changer or, in the very old days, record jukebox. Most artists, producers and labels certainly don’t want one of their releases to play softer than their competitors because of the perception (and not necessarily the truth) that it doesn’t sound as good if it’s not as loud.
But the limitation of how loud a “record” (we’ll use this term generically) can actually sound is determined by the delivery medium to the consumer. In the days of vinyl records, if a mix was too loud the stylus would vibrate so much that it would lift right out of the grooves and the record would skip. When mixing too hot to analog tape, the sound would begin to softly distort and the high frequencies would disappear (although many engineers and artists actually like this effect). When digital audio and CDs came along, any attempt to mix beyond 0dB Full Scale would result in terrible distortion as a result of digital “overs” (nobody likes this effect).
So trying to squeeze every ounce of level out of the track is a lot harder than it seems and that’s where the art of mastering comes in.
Hypercompression – Don’t Go There!
That being said, over the years it has become easier and easier to get a record that’s hotter and hotter in perceived level, mostly because of new digital technology that has resulted in better and better limiters. Today’s digital “look ahead” limiters make it easy to set a maximum level (usually at -.1 or -.2dB FS) and never worry about digital overs and distortion again, but this can come at a great cost in audio quality if not careful.
Too much buss compression or over-limiting either when mixing or mastering results in what has become known as “hypercompression”. Hypercompression is to be avoided at all costs because:
- It can’t be undone later
- It can suck the life out of a song, making it weaker sounding instead of punchier
- Lossy codecs like MP3 have a hard time encoding hypercompressed material and insert unwanted side effects as a result
- It’s known to cause listener fatigue, so the consumer won’t listen to your record as long or as many times
- A hypercompressed track can actually sound worse over the radio (if you care about airplay at all) because of the behavior of broadcast processors at the station.
A hypercompressed track has no dynamics, leaving it loud but lifeless and unexciting. On a DAW, it’s a constant waveform that fills up the DAW region in the timeline. Here’s how the levels have changed on recordings over the years, using this hit recording from the 80s and its subsequent reissues as an example.
But getting the most level onto the disc or file is not the only level adjustment that the mastering engineer must practice. Just as important is the fact that every song on the disc must be perceived to be just as loud as the next. Once again, perceived is the key word since this is something that can’t be directly measured and must be done by ear.
How to Get Hot Levels
The bulk of the audio level work today is done by a combination of two of the mastering engineer’s primary tools: the compressor and the limiter, which, contrary to recording practices where there’s one box that can do either job (depending on the settings), are actually two different boxes or plug-ins during mastering. The compressor is used to increase the small and medium level signals while the limiter controls the instantaneous peaks. Remember, though, that the setup and sound of the compressor and limiter will have an affect on the final audio quality, maybe for the worse, especially if you push them hard.
The Signal Chain
Although the equalizer position might change from before the compressor to after, the usual signal chain looks like this.
EQ > Compressor > Limiter
The limiter is ALWAYS the last in the chain, no matter how many other devices you add and in which order, because that’s what adds any additional level and stops any overs from happening.
The compressor will give you the apparent level and is equally as important as the limiter to the mastering process. If you want to master like the pros, you must have both.
In order to understand how a limiter works in mastering, you have to understand the composition of a typical music program first. In general, the highest peak of the source program determines the maximum level that can be achieved from a digital signal. But because many of these upper peaks are of very short duration, they can usually be reduced in level by several dB with minimal audible side effects. By controlling these peaks, the entire level of the program can be raised several dB, resulting in a higher average signal level.
Most digital limiters used in mastering are set as “Brick Wall” limiters. This means that no matter what happens, the signal will not exceed a certain predetermined level and there will be no digital “overs”. Thanks to the latest generation of digital hardware or software limiters, louder levels are easier to achieve than ever before because of more efficient peak control. This is thanks to the “look-ahead” function that just about all digital limiters now employ. Look-ahead delays the signal a small amount (about 2 milliseconds or so) so that the limiter can anticipate the peaks in such a way that it catches them before they get by. Analog limiters don’t work nearly as well since an analog input can’t predict its input like a digital limiter with “look-ahead” can. Since there is no possibility of overshoot, the limiter then becomes known as a “Brick Wall” limiter.
By setting a digital limiter correctly, the mastering engineer can gain at least several dB of apparent level just by the simple fact that the peaks in the program are now controlled.
As the names imply, compression actually increases the lower level signals while a limiter decreases the loud ones.
The key to getting the most out of a compressor are the Attack and Release (sometimes called Recovery) controls, which have a tremendous overall effect on a mix and therefore are important to understand. Generally speaking, transient response and percussive sounds are affected by the Attack control setting. Release is the time it takes for the gain to return to normal or zero gain reduction.
In a typical Pop style mix, a fast Attack setting will react to the drums and reduce the overall gain. If the Release is set very fast then the gain will return to normal quickly but can have an audible effect of reducing some of the overall program level and attack of the drums in the mix. As the Release is set slower the gain changes that the drums cause might be heard as “pumping”, which means that the level of the mix will increase then decrease noticeably. Each time the dominant instrument starts or stops, it “pumps” the level of the mix up and down. Compressors that work best on full program material generally have very smooth release curves and slow release times to minimize this pumping effect.
4 Rules For Hot Levels
Setting the levels may be the most important part of mastering, since it will determine not only the competitive and perceived level, but also if all the songs being mastered sound like a cohesive collection. Here’s how you do it:
1. Set the master level on the limiter to -.1 or -.2 dB to contain the peaks and avoid digital overs. (Setting to -1 or -2 might actually be better, as we’ll see in the next post)
2. Set a compressor at a ratio of 1.5:1 or 2:1 to gain apparent level. Generally speaking, the trick with compression in mastering is to use a slow Release and less (usually way less) than 5 dB of compression.
3. Increase the level of the program to the desired level by increasing the Output control of the compressor.
4. Compare the song with other songs in the mastering session until they’re all at the same level (use your ears, not the meters).
Mastering Compressor Tips and Tricks
- Adjusting the Attack and Release controls on the compressor and/or limiter can have a surprising effect on the program sound.
- Slower Release settings will usually make the gain changes less audible but will also lower the perceived volume.
- A slow Attack setting will tend to ignore drums and other fast signals but will still react to the vocals and bass.
- A slow Attack setting might also allow a transient to overload the next piece of equipment in the chain.
- If the source is too percussive or has loud drums in the mix, try adjusting the Attack and Release controls.
- Sometimes fast Attack and medium Release helps tame drums.
- Gain changes on the compressor caused by the drum hits can pull down the level of the vocals and bass and cause overall volume changes in the program.
- Fast Attack and Release settings tend to reduce transients.
- Usually only the fastest Attack and Release settings can make the sound “pump.”
- Slower Release settings tend to be the least audible.
- The more bouncy the level meter, the more likely that the compression will be audible.
- Quiet passages that are too loud and noisy are usually a giveaway that you are seriously over-compressing.
This is usually the place that gets engineers who are mastering their own mixes in trouble. There’s a tendency to over-compensate with the EQ, adding huge amounts (usually of bottom end) that wrecks the frequency balance completely.
The first rule to avoid this is:
1. Listen to other CDs (no MP3s) that you like first before you touch an EQ parameter. The more CDs, the better.
You need a reference point to compare to or you’ll surely over-compensate.
The 2nd rule is:
2. A little goes a long way. If you feel that you need to add more than 2 or 3 dB, you’re better off to remix!
Where in recording you might use large amounts of EQ (+/- 3 to 15 dB) at a certain frequency, mastering is almost always in very small increments (usually in 1/10ths of a dB to 2 or 3 at the very most in rare cases). What you will see is a lot of small shots of EQ along the audio frequency band, but in very small amounts.
For example, you might see something like -1 at 30hz, +.5 at 60Hz, .2 at 120Hz, -.5 at 800Hz, -.7 at 2500, +.6 at 8kHz and +1 at 12. Notice that there’s a little happening at a lot of places.
Seriously though, if you have to add a lot of EQ, go back and remix. That’s what the pros do. It’s not uncommon at all for a pro mastering engineer to call up a mixer and tell him where he’s off and ask him to do it again.
Rule #3 is equally important.
3. Keep comparing the EQ’d version with the original version as well as other songs that you’re mastering.
The idea of mastering, first of all, is to make the song or program sound better with EQ, not worse. Don’t fall into the trap where you think it sounds better just because it sounds louder. The only way to do this well is to have the levels pretty much the same between the EQ’d and pre-EQ’d tracks. That’s why I like to use IK Media’s T-Racks for mastering. It has an A/B function that allows you to compensate for the increased levels so that you can really tell if you’re making it sound better or not.
Rule #4 is:
4. You have to keep comparing the song you’re currently working on to all the other songs you’re working on.
The idea is to get them to all sound the same. It’s pretty common for mixes to sound different from song to song even if they’re done by the same mixer with the same gear, but it’s your job to make the listener think that the songs were all done on the same day in the same way. They’ve got to sound as close as possible to each other as you can get them, or at least reasonably close as to not stand out.
As you can see, mastering isn’t that difficult as long as you keep in mind exactly what you’re trying to do, which is to make a group of songs sound like they belong with each other.
Remember: even if you can’t get the songs to sound just like your best sounding CD, you’re mastering job will still be considered “pro” if you can get all the songs to sound the same in tone and volume!
Next time: mastering finishing touches and exporting secrets.