Limiter plus compressor = good idea?

[GarretGraves]

I'm working on a metal track at the moment and (as per FUCKING always) I'm having difficulty in the loudness department. I've been fooling around with volume levels left and right and I get the volume at a decent level with only a clip here and there. Of course the goal is ZERO clipping.

So messing around with things, I tried adding both a limiter AND a compressor on the master track. I never thought to do this before because I always thought that these two tools did the same thing. Even though I got better results than before, I still feel like what I'm doing is redundant.

I know full well about limiters now. Simple plugin. Gain, Ceiling Threshold. But compressors have so many more nooks and crannies in them that I tend to get lost.

So in lamen's terms, what's the difference between these two plugins?

[Rozovian]

A limiter is a kind of compressor. The compressor is a more general tool that can be used for a variety of different effects, while limiters are built specifically to prevent clipping. Using both might not be smart. At least if the limiter isn't last.

Technically, the limiter has iirc an infinite compression ratio, that's the main difference. Basically, this means that the compressor can change the dynamics of a track by x dB, while the limiter has no such limit and just prevents the track from getting louder than x (which should be set to just under 0dB).

So it's generally a good idea to use a limiter at the end of your master effects chain, but whatever compressor, EQ, exciters and stuff you use before that is up to your own preference and whatever wisdoms you can find from other ppl.

[GarretGraves]

Thanks for the info! While I'm on the subject, can anyone recommend a good compressor plugin? or do they all just come with whatever DAW? (forget about price range for now.)

[Cerrax]

Thanks for the info! While I'm on the subject, can anyone recommend a good compressor plugin? or do they all just come with whatever DAW? (forget about price range for now.)

Most DAWs come with decent compressors, though I'm not terribly certain what third-party compressors/limiters are out there. I just use my DAW's (Logic) compressor and limiter.

Basically, their names explain their function.

A compressor uses a ratio to reduce the difference between the loudest and softest parts of a wave. It COMPRESSES the sound.

A limiter simply reduces a wave's loudness below a certain threshold. It LIMITS how loud the wave can get.

Just be careful how much compression and limiting you use. The last thing you want is a brickwall/wall-of-sound effect where the wave just looks like one giant rectangle on your screen. On that note, it's not a bad idea to use a multi-band compressor along with a limiter for a master track. Some light compression with a multi-band can keep erratic frequencies (such as a punchy snare) from ruining the mix.

[Fishy]

Technically, the limiter has iirc an infinite compression ratio, that's the main difference. Basically, this means that the compressor can change the dynamics of a track by x dB, while the limiter has no such limit and just prevents the track from getting louder than x (which should be set to just under 0dB).

If it had an infinite ratio it would just clip the signal anyway. They're typically around 20:1 ratio (ie the input signal has to be over 3 times louder then the allowed peak signal to still clip it). Anything more then that has too much distortion, anything less may as well be a regular compressor.

Key point though, limiter should ALWAYS be the last thing in your master effects. Nothing wrong with a compressor before it.

[Yoozer]

I'm working on a metal track at the moment and (as per FUCKING always) I'm having difficulty in the loudness department.

EQ, EQ, EQ. If you haven't been cutting away anything to make things fit, no amount of compression is going to help you out. Get rid of that mud first - mercilessly. Cut narrow, boost wide - cut first, boost later.

[GarretGraves]

EQ, EQ, EQ. If you haven't been cutting away anything to make things fit, no amount of compression is going to help you out. Get rid of that mud first - mercilessly. Cut narrow, boost wide - cut first, boost later.

ok. so you're saying cut any frequencies that spike? like with the multiband comp?

[Cerrax]

He's talking about cutting any frequencies that aren't useful to the sound in the full mix. As with a metal mix, the bass and kick drum are going to need a lot of room in the low end of the sound spectrum, so there's no need for the guitar to have any frequencies down in that range. Use a parametric EQ to cut out the frequencies that aren't necessary for that instrument. If you don't, it will sound muddy and it will be very hard to make it loud without entirely squishing and distorting the sound. It's like a giant puzzle of sound waves. You don't want to just cram any old wave into the mix, you have to shape it to make it fit nicely.

Each track should have it's own spot in the frequency spectrum and stereo spectrum. You don't want tracks fighting each other because they're trying to squeeze into the same space. If two tracks have same frequency range, move them to different parts of the stereo spectrum (with a metal song, that would mean the guitars should be panned to different sides of the stereo mix, so they don't blend together). The kick drum and the bass are usually in the center of the stereo mix, but they should be in different frequency ranges, so that they don't clash with each other. Obviously, some crossover will occur, but the idea is to try separate the tracks as best you can so that they don't fight for dominance in the mix and blend together and get all muddy.

[Yoozer]

Exactly. Time to quote myself.

Basically you can abstract everything in a mix into something that takes up a part of the frequency spectrum. To use the old and tired car analogy:

See it as a big mover's truck that you load up with household stuff. The height of the truck is the volume; the length of the truck is the frequency range.

You have a cabinet in your living room. It goes at the left of the truck. It's pretty high, so stacking anything on top of it - well, you can't do that, because you'll reach the ceiling of the truck. If you persist, you punch through the ceiling - not good. So what you stack on top of it should be low, or you should find a way to make it low, or you should find a way to make the cabinet lower.

That tool is simply volume; by decreasing the volume you squash the cabinet's height. By using a compressor, you squash the height of everything that's stacked on top of eachother. By using a multiband compressor, you squash the height of everything that's stacked on top of eachother at a certain place.

You have a couch. It's pretty big and hard to move, and since you know you've got too much stuff in your household and can't make it fit otherwise, it actually has to overlap with the cabinet you put there earlier. Since overlapping can't be done physically - you can't push one object through another - you'd have a problem - if this was in real life.

The solution is an equalizer; instead of squashing, it saws off a part of what you want to put there. You can either saw off the part of the couch that's in the way, saw off the part of the cabinet that's in the way, or saw off a bit of both so they fit properly.

This is why just throwing a compressor on everything is not going to help.

[Cerrax]

I like that analogy Yoozer. It really helps someone visualize what is happening when mixing. It's a cliché analogy because it works.

[GarretGraves]

Ok i think I get it. So to make room for the kicks and bass in the low end, i should cut the low end on the guitar, correct? Since the guitar is mostly going to be in the mid to high range i should keep it as such. And cut any mid to high frequencies in the kicks and bass to make room for the guitars.

right?

[zircon]

You're half-right (on paper anyway, it depends on the song.) Most likely, the kick and bass are not going to be producing much mid/high output, so muting them may just muffle those instruments too much. Sometimes, you WANT mid/high freqs from the bass so you can hear pick noises or whatever.

[GarretGraves]

I see. so is this where the spectrum analyzer comes in handy? (i saw the post for the free one earlier and downloaded it but havent had a chance to use it cause for some reason none of my DAWs can recognize it even when i put the .dll file in the correct plugin folder. GRRRR!)

[Moseph]

I see. so is this where the spectrum analyzer comes in handy? (i saw the post for the free one earlier and downloaded it but havent had a chance to use it cause for some reason none of my DAWs can recognize it even when i put the .dll file in the correct plugin folder. GRRRR!)

I'd say it's more a situation where you just need to know what to listen for. Like Zircon said, depending on the bass sound, you may want some mid-/high-end to catch the pick/finger noise. Same may be true of the kick -- you might want mid-/high-end to catch the noise of the beater. All of this is situational, though. A spectrum analyzer might help you balance the frequencies, but it'll tell you neither which parts of the spectrum to emphasize in which instruments nor whether the balance you've achieved "on paper" is actually compelling as a mix. I don't generally use a spectrum analyzer to mix, although I'll sometimes check mostly finished mixes with one to get a different look at how my mix which already sounds pretty good at that point sits in the total frequency spectrum and may make minor tweaks depending on what I see.

EDIT: The point I'm trying to make is that ideally you shouldn't need to consult a spectrum analyzer to tell whether the frequency spectrum is balanced. If you have decent monitors/headphones and are very familiar with how things sound on them, you should be able to hear spectrum balance issues just by listening to the mix. (Obviously, this is something that comes only with practice, and consulting an analyzer may help you to develop an ear for frequency balance and to catch things that you didn't notice or that were masked by deficiencies in your monitoring setup.)

EDIT 2: Also, just to be clear, I have something of an anti- visual mixing bias, so don't let me dissuade you from using a spectrum analyzer if you think it will be helpful to you.

[GarretGraves]

Hmm ok. So far I've managed to make the track cleaner but not louder. it may also have something to do with how loud I've recorded everything. for example, the guitars' recording level peaks at around -12 dB (without effects, limiter etc.). Could something like this effect the headroom i have?

EDIT:http://soundcloud.com/metaldevin/garretmetal1

Here's what I have right now.

EDIT2: lol come to think of it, now that I listen to it on soundcloud it sounds really bass heavy. hmmm

EDIT3:http://soundcloud.com/metaldevin/boab2-mixdown

Ok I completely forgot about a factor or two. The first link (my current working demo) was recorded using the Line 6 Gearbox set. (Toneport to plug the guitar into via USB for those who don't know) The second link is a demo I recorded when I first got my Mackie 820i mixer and Line 6 Spider IV amp setup (miked with an SM57 with rhythm guitar recorded twice, panned left and right). The older demo sounds much better than my new one. And LOUDER. Where the hell did I go wrong? Was it cause of the hardware? It sounds like everything is balanced just fine. Ugh I need consistency in my mixing!

[Moseph]

On the first one (garretmetal1), I'd say make sure your drums aren't peaking too high. It looks like a lot of the spikes on the waveform are coming from the drums, so you want to try to bring those down if you can. The snare drum and toms in particular sound to me like they could stand to be compressed (or be compressed more if you're already compressing them).

I'm not sure if this will help the overall loudness issue, but I think the lower frequencies (~20 Hz-800 Hz) are too dominant (as you've noticed). I'd say roll off everything below ~80 Hz, cut everything below ~800 Hz by at least 4 dB, and see if that lets you pull the overall level up a bit.

[Cerrax]

Yeah I felt like the snare was quite loud and probably caused a lot of spikes. As well as the low end on the whole song. Pretty much exactly what Moseph said

[GarretGraves]

tried it. helped a little but i think im gonna scrap the whole thing and start over. i think im gonna try i different guitar tone cause im starting to not be happy with it. im gonna take a rest and start again when my ears arent fatgued to all hell.

thanks for the help tho guys!

[Moseph]

Any chance that you could zip the unmixed audio files for the individual tracks and upload the zip file to Mediafire or some similar host for me to download? I'd kind of like to take a shot at mixing/mastering it, and if I can come up with something high-loudness that works well, I can screencap my plugin parameters and write a few paragraphs about what I did to make the mix sound like that.

[GarretGraves]

that can be done. give me a bit to sort em out and take care of a few other things (errands and so on) and i'll send them to ya.

[GarretGraves]

Ok, here the wav files.

http://www.megaupload.com/?d=MI6D3UK5

The drums are on one wav since i mix them down from FL and import it into Audition (or Cubase).

al wavs were mixed down WITHOUT any effects (no compressors, multiband or anything) I hope I learn something from this. And again thanks!

[Legion303]

Before you even think of putting a compressor or limiter on your master output, get your mix levels sorted. Once you're happy with them, then start adding effects to the output chain.

-steve

[Moseph]

The drums are on one wav since i mix them down from FL and import it into Audition (or Cubase).

I can attempt a mix with the single-track drums, but it may not turn out well and I won't be able to give any meaningful commentary on drum mixing. I need multi-track drums to do a proper mix.

When you mix, do you do the compression and such on the individual drum tracks in FL before mixing them down to one track and bringing them into Audition? Because if you're doing the compression etc. in Audition after the drums have been mixed down to a single track, the drums will be really, really difficult to deal with. You need to be able to get at the individual parts of the drumkit and put effects on them separately, so you really need individual tracks to do a good mix.

Things don't necessarily need to be entirely split up -- you can usually get away with having all the toms on one track (or two tracks divided between high and low toms), all the cymbals/hats on one track (or maybe the hi-hat on its own track), the snare on its own track, and the kick on its own track.

[GarretGraves]

I can attempt a mix with the single-track drums, but it may not turn out well and I won't be able to give any meaningful commentary on drum mixing. I need multi-track drums to do a proper mix.

When you mix, do you do the compression and such on the individual drum tracks in FL before mixing them down to one track and bringing them into Audition? Because if you're doing the compression etc. in Audition after the drums have been mixed down to a single track, the drums will be really, really difficult to deal with. You need to be able to get at the individual parts of the drumkit and put effects on them separately, so you really need individual tracks to do a good mix.

Things don't necessarily need to be entirely split up -- you can usually get away with having all the toms on one track (or two tracks divided between high and low toms), all the cymbals/hats on one track (or maybe the hi-hat on its own track), the snare on its own track, and the kick on its own track.

I normally dont compress the drums or anything until after i mix them down into Audition. (Im using Superior Drummer Metal Foundry by the way) I COULD mix them down on their seperate tracks but for me it saves time if I mix them at once then import them into audition and work from there. Yeah I would need to mix them down again if I needed to adjust levels or something.

Is this wrong to do?

[Moseph]

I normally dont compress the drums or anything until after i mix them down into Audition. (Im using Superior Drummer Metal Foundry by the way) I COULD mix them down on their seperate tracks but for me it saves time if I mix them at once then import them into audition and work from there. Yeah I would need to mix them down again if I needed to adjust levels or something.

Is this wrong to do?

You definitely should keep the drums separated. Mixing the drums down to a single track before putting effects on them may actually be one of the reasons that you're having trouble getting a good overall mix. The issue is that when the drums are all on one track, you can't apply effects to specific elements of the drums; for example, you can't use separate compression on the snare and the kick, you can't apply more reverb to the cymbals than to the kick, etc. Any effect you put on the drum track affects all of the the drums, and that means that you can't fine-tune the sound of the drums to the extent you ought to be able to. It's best to think of the kick, the snare, the toms, and the cymbals as being four completely separate instruments and to deal with each separately instead of mixing them down to one track.

Separating the drums before putting effects on them should also make it easier to control the waveform spikes that are messing up the loudness of your mix because it will let you compress each type of sound individually -- it will let you, for example, reign in the snare drum spikes without affecting the way the other drums sound.