The limiter is an essential tool in the mastering phase, and has a role at other stages of music production as well.
Increasing overall loudness and preventing overload of a signal are the two main functions of the limiter, but how do you dial it in?
If you are already familiar with compression, then you’ll see that the limiter is like the ultra-buff big brother of the compressor.
Simple as that.
A Compressor on Steroids
It’s actually essential that you understand compression before approaching the limiter.
Read my in-depth explanation of compression here: Music Production 101: Compression
Once you’re comfortable with the principles of compression, limiting is really a breeze!
That’s because the limiter is just a juiced-up compressor.
With a compressor, if a sound approaches the threshold, there is some negotiating and compromise.
“Oh, you want to cross the threshold? Ok, well, ummmm … how about by just a few dB but then I get to bring you back this way just a tiny little bit. Is that alright?”
But when it’s a limiter you’re talking about, there is no quarter given.
“You wanna cross this threshold? I’ll MURDER you!”
The Final Boss
In the context of mastering, the limiter is inserted as the last active plug-in on the Master Stereo Output track.
I say “active” because you will most likely place a metering plug-in of some sort after the limiter to gauge the final dynamics and loudness of the mix.
Otherwise, the limiter is the final stop in actually affecting or sculpting your track.
Hey, make sure you read that article I did on compression: Music Production 101: Compression.
Ok, then, move on.
Anatomy of the Limiter
Output Level / Ceiling
This is where the limiter gets its name; the Output Level represents the absolute limit for how loud a sound can get after passing through the limiter.
This limit is important because it allows you to keep your sound levels from cracking the maximum sound level of your DAW (or any audio system) which is 0 dBFS.
Anything peaking over 0 dBFS would be clipping—your DAW will snip off any signal that exceeds 0 dBFS because it would otherwise overload the system.
When the peak is clipped off, a small amount of distortion is introduced. Cumulatively, clipping can destroy your track.
The space between your loudest sound levels in your song (your peaks) and 0 dB called the headroom.
The Output Level is set to 0 dB by default, which is called the ceiling.
But you can adjust the Output Level to essentially lower the ceiling, so that you are creating a cushion between your song’s loudest peaks and that 0 dB clipping point.
And this is where the limiting happens—if you lower your Output Level ceiling to, say, -0.2 dB (just a bit below 0 dB), the incoming signal will peak and get stopped at -0.2 dB rather than hitting or exceeding the 0 dB clipping point.
Some people argue for a -0.2 dB maximum Output Level in mastering, but I’ve seen other engineers say you can get away with -0.1 dB.
Listen to your track at several values and if you can take it to -0.1 dB without it sounding harsh, do it!
Gain Reduction / Attenuation Meter
Gain reduction tells us how much of the signal has been clipped—or limited—by the limiter.
Some limiters get all fancy and instead use the word “attenuation” for gain reduction. Both mean the same thing in limiting.
If the GR meter shows values like -4 dB to -6 dB, that means a signal tried to pass the Output Level’s limit/ceiling by -4 dB to -6 dB, and was stamped out.
Once your GR level is passing the -4 dB mark, there is a strong possibility that your track is being limited too much and driven too hard by the input Gain.
Gain / Threshold
The gain slider or dial is going to add gain to the incoming sound to boost it.
It’s really tempting to drive the gain up until the mix is screaming loud, but be careful—you’ll quickly slam the mix into the ceiling.
What happens when your loudest peak in the track smashes into the ceiling?
You get a form of clipping, of course! And that sucks!
If you keep on driving the gain up, the rest of the music—the softer or quieter parts that make up the body—get driven into the ceiling as well.
The whole mix gets smashed, and all of the dynamics disappear and unpleasant distortion artifacts begin to saturate the track.
If I find myself wanting to add about +4 dB to +6 dB of gain on the limiter, I know I need to revisit the mixing session and see why my average loudness is being held back.
That Other Stuff
Most mastering limiters have deeper parameters and settings you can tweak. Some of these settings seem mysterious or far too geeky to approach for many home-studio artists. Let’s see if we can demystify some of these for the sake of better understanding.
Although the files you’re working with in your DAW are 24-bit resolution, your final destination is digital and/or CD release. When you create your final bounce of the song, you’ll be creating an MP3 or WAV which has a resolution of 16-bits.
Going from 24-bit resolution to 16-bit resolution audio is called down-converting.
“Wait! What the hell does 24-bit resolution mean?”, you ask.
For a 24-bit audio file, it means there are “24-bits” worth of data (1’s and 0’s in your computer’s brain) for each “sample” of the source signal (e.g. a recorded guitar). A single WAV file recording of a guitar part, for instance, contains thousands of these “samples”.
The real sound is represented by, or resolved at, 24-bits worth of computer juice.
A higher resolution means there are more digital bits available to represent/recreate the recorded sound in your DAW.
When you reduce the resolution, some of these bits of data get rounded up or down, which can create a nasty artifact called “truncation distortion” that doesn’t sound very pleasant.
Dither adds nearly inaudible amounts of noise or saturation to the audio to fill in or smooth over the gaps created by down-converting, which effectively masks the unpleasant truncation distortions.
How much of this noise should be applied can be set in the Dither option (Type I, Type II, etc.). Read your limiter’s user manual to understand what types of dithering it offers.
The character or shape of the noise is defined by the Noise Shaping setting. Again, the user manual for your limiter will offer a better explanation of the Noise Shaping algorithms available to you.
All that aside, let your ears judge how things are effected by tweaking!
This setting relates to the output resolution of the limiter.
By default, a limiter’s Quantize value is usually set to 24-bits to match the WAV files you are working with in your DAW environment.
Setting the Quantize to match the intended output resolution (16-bits in most cases), allows you to hear how the final bounce will sound in real time as you adjust and tweak the settings.
So, although your tracks are 24-bit, setting the limiter’s Quantize to 16-bit lets you “hear in 16-bits” so that you can anticipate any strange audio distortions or artifacts before you commit to the bounce.
If your limiter has a “soft knee” parameter (like Logic’s Limiter), you’ll need to disengage this or switch to a “hard” knee so that you can achieve the type of limiting critical for mastering—brickwall limiting.
The concept of the brickwall limiter is simple—nothing gets through it.
Some types of limiters are called “soft” limiters because they compress with a specific Ratio, like 1:10 or 1:20 by default. These can find use in tracking or recording, but are not impenetrable!
Some compressors working at a high-ratio, like 1:10 or 1:12 can perform similarly, but again peaks can slip through if loud enough.
The other variety of limiters are the “hard” limiters, better known as the brickwall limiters which are designed for mastering. These are also sometimes called maximizers.
These work with a Ratio of 1:∞, so it doesn’t matter how loud a peak gets because it is pulled infinitely down to the Output level you set (Calculus class, anyone?).
Absolutely no signal goes over the threshold, not even 0.0000000001+ dB.
Put a brickwall limiter on your Master Output, and it doesn’t matter how hard the drums smash, how loud the vocalist screams, or how brutally the guitars chug—the threshold stays firm.
Setting the Attack/Release Times on a Limiter
Even though the limiter stops the signal dead in its tracks, it still has Attack and Release settings—how quickly it smashes the signal and how soon it readies itself for the next poor bastard that dares to cross depends on these Attack and Release settings.
Finding the right Attack and Release settings is a subtle art. Each track will require a different balance between Attack and Release times.
Make sure you actually did read my post on Compression, because the same Attack and Release principles apply here!
I’ll even have a special e-Book in the future giving you some cool tips I’ve learned along the way for dialing these parameters in!
For now, experiment with conservative attack times and super short release times, and find the settings which allow the song to still move and groove without any strange “breathing” or “pumping” artifact, and without making the song’s dynamics feel “flattened” or hits sound “blunted”.
Transparency is Key
When the sound approaches the set Output Level, the limiter throws up the brickwall and smacks in right in the face.
You’ll need to set the Attack and Release on the limiter so that you maintain the “pulse” or “breath” of the song, and it remains “open” and “full”.
Fabfilter’s Pro-L Limiter lets you dial in Attack and Release times independently for ultimate transparency.
Many limiters will only have a “Release” setting that you can manipulate, like the famed Waves L2:
If signals are smashing into the brickwall constantly, you’ll get a smeared, flattened, harsh sound with no life or breath.
If that’s happening, your Gain is too high and/or the Attack and Release settings are not dialed in right.
Automating the Limiter
The same threshold setting or Attack/Release times may not work for different segments within a single song, especially if the song is some sort of mercurial progressive-metal madness with fast-changing tempos and time signatures.
Automating the limiter’s different parameters across the timespan of the song is an often employed technique to help sections transition better while maintaining a fullness or presence.
Because automation is a bit more of an advanced topic, I’ll be making it the subject of a dedicated article in the near future!
Choose Your Weapon
There about as many limiters out there as any other plug-in software device.
The most legendary of limiters is probably Waves L2 Ultramaximizer limiter.
A very capable and FREE alternative that mimics the Waves L-series is Yohng’s W1 Limiter (PC/Mac).
Having personally used both Yohng’s free alternative and the real deal from Waves, I didn’t hear any difference that would be obvious to your casual listener.
Here is the short list of other great limiters finding widespread use in home and pro studios:
- Fab Filter Pro-L
- iZotope Ozone
- PSP Xenon
- Sonnox Oxford Limiter
- IK Multimedia T-RackS Brickwall Limiter
- Slate Digital FG-X Mastering Processor
Oh, and don’t forget your DAW’s stock limiters! You may have a plug-in that is of a brickwall class, like Logic’s bundled Limiter plug-in.
As always, knowing how to use the tool is more important than the tool itself!
Was this article helpful to you? Do you have any burning questions? Let me know in the comments!