How do you use a limiter ?

[ScumBum]

I know a limiter is just a compressor with a high ratio . But I've never learned how to use one properly , or been satisfied .

Lets say your compressing a bass track that has a little too much dynamics and I want to limit the track a few dbs .

Do you treat the limiter like a compressor and use the same attack and release settings ? Or is it different with a limiter , and you need faster attack times for the peaks , to not let anything through ?

Who do you approach using a limiter and make it transparent ?

[jdjustice]

sounds like you need compression and not really limiting.
you can of course call compression limiting and limiting compression but limiters are mostly used to finalize mixes and create squarewave recordings nowadays.

Originally Posted by ScumBum

I know a limiter is just a compressor with a high ratio . But I've never learned how to use one properly , or been satisfied .

Lets say your compressing a bass track that has a little too much dynamics and I want to limit the track a few dbs .

Do you treat the limiter like a compressor and use the same attack and release settings ? Or is it different with a limiter , and you need faster attack times for the peaks , to not let anything through ?

Who do you approach using a limiter and make it transparent ?

[ScumBum]

Originally Posted by jdjustice

sounds like you need compression and not really limiting.
you can of course call compression limiting and limiting compression but limiters are mostly used to finalize mixes and create squarewave recordings nowadays.

I know .....

I was taught not to use limiters when mixing but I've read some people do limit sometimes during mixing , kick , snare and bass ,

[shaneoconnor]

if i am using a limiter before a compressor, it is usually to level out a signal so the compressor can "see" a similar dynamic range. so then i would use fast attack fast release.

if i am using a limiter to pump and snap back, i will slow down the release and the attack, and pull the threshold down.

examples ITB, sonnox limiter, stillwell audio limiter.

examples OTB, LA2A, API 2500

[Empire Prod]

I hate that this is the case, but it is pretty common for people to hard limit tracks with an L1 before mixing. All the peak transients are gone, and then new envelopes are created with compression. It makes things very uniform, but to me sucks the life out of the tracks. Like I said this is all too common.

[firby]

Hi,

I don't pretend to have the skill that many on this board commit to a mix. But I do use a limiter on bass (your application) alot. In my case it is a hardware unit, it is a MXR dual limiter. I use it on bass sometimes pretty heavy handidly 10 dBs or so into it. In this case it is almost used to effect. The bass gets dirtier and more full at the same time and a pleasing grime appears on it. I rarely freeze a VU like you can with some limiters but if I like the effect I go for it. Sometimes I like the effect of a clean track and a limited track bussed together and mixed ala parallel compression. With that particular box though I really like it as a stereo insert on a buss, I don't use it on the master buss because I think it has too much grunge for that.

In my experience, software based limiters don't really do the same thing for me. That could be me though, I rarely reach for software. Anyways, sometimes it is just the thing to stress the audio a little and add drama.

[deeper]

Limiting has a different goal than compressing. When limiting, the device should only come on occasionally, in order to "limit" how loud the "occasional" peaks are. So, yes, faster attack times are necessary to catch and "limit" the peaks. Then, because you've knocked a few decibels off the loudest peaks, you can turn up the entire track. Limiting should be transparent. If you can hear it, it's bad. If you are familiar with the four parts of a wave (attack, decay, sustain, and release), the perfect limiting action will be fast enough to bring down the attack on each overly spikey transient without knocking down the level of the decay, sustain, or release (because that would diminish the fullness and body of the sound). Limiting is quieting down the occasional spikes. Only. As invisibly as possible.

Compressing is different from limiting because the goal is to "smooth out" the entire track. Therefore the device is in operation a lot more than when limiting. And because the compressor is "on" so much of the time, lower ratios are called for, or else the life will get sucked out of the track. Compression is also used as a tone-shaping tool because, by leveling the entire track (not just the peaks) the signal seems fatter and bolder. So sometimes compression is not meant to be transparent. To limit effectively, the attack has to be quite fast, but compression can have a very slow attack, and usually sounds more natural when the attack is moderate-to-slow. And the onset of compression often sounds more natural when it kicks in smooth and slow, that's why some compressors have "soft knee" or "over-easy" modes. It wouldn't make sense to have a soft knee on a limiter -- that would let the spike through!

Example. Suppose you have an Acoustic Lead Guitar part that needs to have more "authority" and needs to stand up in the mix. You zero in on the waveform and see that every time a note is struck, the picked string has a huge spike, and the "body" of the note (decay and sustain) is rather wimpy, with a too-abrupt release. I would use a limiter to tame that huge spike first. (If you try to compress first, that spike is going to cause the compressor to "over-shoot" and further weaken the body of the note.) Once you have the transients set correctly (relative to the rest of the note), then I would go in with a compressor set to improve the sustain and fullness of the entire note (this is where the tone-shaping occurs.)

Are you starting to see the difference between compression and limiting? Some tracks need both. And some don't need either. But it's a good idea to know, before you patch in a compressor-limiter, whether your track needs compression, or just a little limiting.

[Doc Mixwell]

Originally Posted by ScumBum

How do you approach using a limiter and make it transparent ?

Not hitting its threshold with average information. It all depends on what you trying to do with the box in question [and how that box was designed] It sounds like you want a high threshold, and a tunable attack [onslaught of compression determined by the threshold] and release [when the compressor "lets go" of source material], set so that it is simply taming the peak information gently determined by your enveloping settings. Everyone will use these things differently, but limiters are usually good for pushing things into saturation and keeping them in one place.

[skiroy]

For limiting use fast attack and release.Usually the highest ration the compressor offers.But the often undiscussed part is the threshold.With a limiter the threshold is set high so that only the top transients are being affected and everything else is mostly left alone.Opposed to using it as a compressor the threshold is set lower to affect more of the dynamics of the track.

[Sonic Lush]

hey I this is a great thread and I enjoy how different people write about this topic.
happy holidays y'all.
walter

[ScumBum]

Originally Posted by deeper

Limiting has a different goal than compressing. When limiting, the device should only come on occasionally, in order to "limit" how loud the "occasional" peaks are. So, yes, faster attack times are necessary to catch and "limit" the peaks. Then, because you've knocked a few decibels off the loudest peaks, you can turn up the entire track. Limiting should be transparent. If you can hear it, it's bad. If you are familiar with the four parts of a wave (attack, decay, sustain, and release), the perfect limiting action will be fast enough to bring down the attack on each overly spikey transient without knocking down the level of the decay, sustain, or release (because that would diminish the fullness and body of the sound). Limiting is quieting down the occasional spikes. Only. As invisibly as possible.

Compressing is different from limiting because the goal is to "smooth out" the entire track. Therefore the device is in operation a lot more than when limiting. And because the compressor is "on" so much of the time, lower ratios are called for, or else the life will get sucked out of the track. Compression is also used as a tone-shaping tool because, by leveling the entire track (not just the peaks) the signal seems fatter and bolder. So sometimes compression is not meant to be transparent. To limit effectively, the attack has to be quite fast, but compression can have a very slow attack, and usually sounds more natural when the attack is moderate-to-slow. And the onset of compression often sounds more natural when it kicks in smooth and slow, that's why some compressors have "soft knee" or "over-easy" modes. It wouldn't make sense to have a soft knee on a limiter -- that would let the spike through!

Example. Suppose you have an Acoustic Lead Guitar part that needs to have more "authority" and needs to stand up in the mix. You zero in on the waveform and see that every time a note is struck, the picked string has a huge spike, and the "body" of the note (decay and sustain) is rather wimpy, with a too-abrupt release. I would use a limiter to tame that huge spike first. (If you try to compress first, that spike is going to cause the compressor to "over-shoot" and further weaken the body of the note.) Once you have the transients set correctly (relative to the rest of the note), then I would go in with a compressor set to improve the sustain and fullness of the entire note (this is where the tone-shaping occurs.)

Are you starting to see the difference between compression and limiting? Some tracks need both. And some don't need either. But it's a good idea to know, before you patch in a compressor-limiter, whether your track needs compression, or just a little limiting.

Thanks for the info....

Now if limiters are are meant to catch and stop the transient , and never accentuate the transient like a compressor can , why is there a need to adjust the attack ? Shouldn't it just be set to the fastest setting there is so no transients get through ?

[FreshSkweez]

Originally Posted by ScumBum

Thanks for the info....

Now if limiters are are meant to catch and stop the transient , and never accentuate the transient like a compressor can , why is there a need to adjust the attack ? Shouldn't it just be set to the fastest setting there is so no transients get through ?

Good point. A lot of people just do that and get away with it. Some compressors don't even have the attack parameter (dbx 903s aka 160, Neve 2254s etc). There's a serious danger of sucking the punch out of the mix though. Most pros use compressors to CREATE transients and limiters to control them. That way you have all the punch you want while maintaining total control over it. Listen to a few Chris Lord-Alge mixes - this guy is like God in creation.

In general the terminology referring to dynamic control is the most confusing thing in pro audio theory IMO. Some say the difference between compressors and limiters is in ratio, some say in how often the unit performs gain reduction in your programme. It took me years to figure out what the hell people were talking about. For me the easiest way to look at it is:

COMPRESSORS have to do with RMS. They control your programme envelope, without touching the transients. Now how often you want this unit to engage throughout your mix totally depends on you programme and your taste. Most people have the units on low ratios working constantly in order to maintain constant control over the sustain part of the sound. Another term for this type of device is a Leveling Amplifier which to me really summs it up. A lot of classic units don't even have a ratio control but rather a 'soft knee' or 'OverEasy' response to input signal as in the more you put in not only the more GR you get, but also the higher the compression ratio becomes. But still it's all RMS.

LIMITERS perform transient compression. In order to catch transients you are likely to need a fast attack (but totally dependant on the nature of programme as in a handclap has a very fast transient whereas vocal transients are much slower for the most part). How high the ratio is is still up to you. If 3:1 is all that's needed to keep your transients in check - go with god. In most cases it isn't though, hence the stereotype of limiters having high ratios. Bottom line - limiters are for controlling transients.

More often then not you end up with both a compressor and a limiter on a given instrument troughout the recording/mixing process. But there are no rules as to how high the ratios should be or how hard the device should be working - when you reach for the tool though you should know what you wanna use it for - sustain or transients.

Peace, A

[GuitarRuss]

I'm really digging the EMI limiter. It sounds great on vocals. I use this or a compressor on most vocal tracks in addition to fader rides for larger dynamics to avoid having the compressor or limiter kick in too much.

[vernier]

Probably better to avoid running your music completely through a whole set of electronics that will slightly shrink your sound. All processing shrinks/compromises. If you have to limit, use a world-class limiter a few dbs ..but a good bass player doesn't need any.
'

[dubrichie]

Originally Posted by vernier

All processing shrinks/compromises.

not where i'm from.

some processing enhances and em-biggens. this is the kind of processing that i tend to dig. this the kind of gear that i use.

harmonics?

saturations?

mojoifications?!

and some even without tubes, for shame!

[wannaberocker]

So it would seem an 1176 (for limiting) then go into an LA2a ( for decay envelope control or "compression") is a good thing to do.

Dynamics processing the right way has been the biggest hurdle for me, it seems that dynamics processing is what "unlocks" the greatness in a well recorded sound.....

[Mike Caffrey]

Originally Posted by ScumBum

Thanks for the info....

Now if limiters are are meant to catch and stop the transient , and never accentuate the transient like a compressor can , why is there a need to adjust the attack ? Shouldn't it just be set to the fastest setting there is so no transients get through ?

Based on the questions you're asking, I think you'd benefit from approaching things as if the answer was yes. It will help you wrap your head around the concepts and you'll soon find that you don't think about the differences much and do everything by ear.


The real world answer is that there are times where people go for a sound where the limiter is not set as fast at is it can be set, but is faster than a compressor. It's usually not a sound that I like.

Another thing that people do a lot of is putting compressors and limiters in parallel with the unprocessed signal and use the balance between the two to control the attack. That might be an arguement for always using the fastest attack.

[uptoolate]

This thread has been very helpful so far.

I have a question about the Neve 33609 (UAD-1).

It has limiter controls on the left, and compressor controls on the right.

Does the signal hitting this plugin go through the limiter first, and then into the compressor?

If so, this seems like a great piece for limiting first and then compressing.

Am I thinking in the right direction on this?

[uptoolate]

Originally Posted by uptoolate

This thread has been very helpful so far.

I have a question about the Neve 33609 (UAD-1).

It has limiter controls on the left, and compressor controls on the right.

Does the signal hitting this plugin go through the limiter first, and then into the compressor?

If so, this seems like a great piece for limiting first and then compressing.

Am I thinking in the right direction on this?

Any thoughts?

[Phelan Kane]

In Stav's book, mixing with your mind, he states that the processor with the highest ratio should more often than not be upstream (i.e. the limiter).

It seems to work for me often in a mastering environment. It seems to give the track more air/width. Some songs need it -some don't.

Check his book out. It fantastic.

Mixing with your Mind

Peace

P

[Mike Caffrey]

Originally Posted by Phelan Kane

In Stav's book, mixing with your mind, he states that the processor with the highest ratio should more often than not be upstream (i.e. the limiter).

It seems to work for me often in a mastering environment. It seems to give the track more air/width. Some songs need it -some don't.

Check his book out. It fantastic.

Mixing with your Mind

Peace

P

While that's pretty much what I always do, it doesn't have to be that way.

I'm going to make up some numbers for illustrative purposes....


Let's say you've got a 5ms transient - meaning your limiter has to be set to 5ms to catch it.


If you put your limiter first it will see the full volume signal and catch every transient peak..

If you have your compressor first with a 50ms attack, those 5ms transients will pass through and be caught by the limiter in the exact same way, but with one exception.

If your compressor release time is slow enough the gain will stay reduced which may keep the level of that transient low enough that the limiter doesn't catch it or catches it, but reduces it less.


I think you could generalize as follows: If you're using the peak limiter for envelope shaping, put it first. If you're using it solely to prevent downstream clipping, put it second.

[Renegade Prod]

Pull the threshold down and RUN!!!!!!!

[Phelan Kane]

Envelopes are certainly gonna change things. I try and stay away from brickwall...

[Grahamdwc]

Originally Posted by Phelan Kane

Envelopes are certainly gonna change things. I try and stay away from brickwall...

Just bumping this thread.

It's a couple of years on and i'm just wondering if peoples opinion about where and when to use limiters has changed much in 2011.

Graham

[AstralPStudios]

There's nothing wrong with using a limiter in mixing at all! I sometimes use it on a kick or bass as it helps to keep things a bit more solid. As for how to use it, it's dependent upon the situation.

I've got a PDF for you that may help you understand a little more of how to utilize limiters in a mix. It's a Waves PDF. If you'd be interested in taking a peek, shoot me a PM.

[Bassmankr]

What this thread didn't touch on but some other older threads do is using limiters for color/an effect, ala using distorsion/overdriving via soft limiting. Look up LEX PCM42 tricks.

[Aaron Miller]

If it's a reasonably small number of transient peaks that I don't want, I almost always automate. If there are too many to automate, I may throw on a limiter. I'll do this every now and then with drums but only when there's a need to squeeze more overall volume out of the kit. Usually with a good drummer I don't have to do anything more than compress the drum buss. With inconsistent drummers playing modern rock, I may put a limiter on snare, for example, just doing 2-3 dbs of reduction on every other hit or so. This is pretty transparent and evens out the snare so it's not triggering the drum buss compressor so wildly and inconsistently. It allows me to hit the drum buss a little harder and still remain somewhat transparent.