They SHOULD be used on single tracks (IMHO). Running an entire mix through them (which has become awfully popular recently) should be saved for mixes that have problems that require it.

That being said, if there isn't a specific reason to use it, don't. Just like with anything else - If the track "needs" it, go for it. If it's being thrown on "because it's there," it's probably the wrong application.

i recently downloaded a couple of multiband compressor plugins from the 'free plugins' page of this website. honestly, i don't notice a difference between "compressing" certain frequencies, and just applying eq boosts and reductions to them. maybe a hardware multiband compressor would be more noticeable a difference? why would one use a multiband compressor instead of eq?

Used properly, they can be used like they were in a virtual side-chain...

If a guitar track sounds fine except for some overtones in heavily muted "crunch" parts, a MBC could compress the signal that reaches above a particular threshold in that particular frequency range.

Same with vocal sibilance and plosives (where they really come in handy). They can essentially leave everything alone until a giant "ssss" comes in - It can clamp down on the sibilant part of the spectrum while leaving the rest untouched for the most part.

The problem I keep seeing over the last few years is that they're becoming the most abused things on the planet - A lot of the plugs have "multiband maximizer" settings that will just crush the life out of everything in a not-so-subtle manner...

I know a "mastering guy" who puts one on basically everything he does - Obviously, he can't hear what he's doing, because all of his mixes wind up with horrific sounding high end and flabby lows. Lots of both, but not pretty at all.

IMO, there are better ways to ruin a mix... :shock:

There was this one time...

A friend of mine told me about this concept, and I tried it. He said it made the vocals sound more like the person is right there sort of whispering in your ear. Forgive me if this is the oldest trick in the book and I'm the last guy no earth to hear of it.

I recorded the vox track, and then made a copy of it, so now I have the vox on both tracks one and two.

On track one, I took out the high end, and applied little or no compression to it.

On track two, I took out the low end. I then applied a pile of compression to that track(8:1 or so if memory serves).

I can't remember where the 'crossover' point was between the 2 tracks, but I think it might have been about 2-3K.

I was able to mix and match them just by using the faders.

Bingo. Instant multi band compression. Now this may not be as 'good' as a read multi band, but it did the trick for me. Also you can change the crossover point, have it overlap, add or subtrack effects to only one frequency, bust your track into as many pieces as you want based on frequency and have complete control over each one, and so on, and so on...

You can then also assign your tracks to a 'group track' where you can have even more control over the EQ and such. It's a pretty cool tool.8)

And for what it's worth, great to hear Massive talk about compression(and other effects I'd imagine) on each track instead of globally. Maybe it's a big error on my part, but I deal with each track individually and then don't run any compression or EQ at all on final mixdown. I once got a mix done by someone and there was a fair bit of dynamics. The bass was getting louder and softer as the other stuff kicked in and out. YUK!

I haven't tried it, but maybe one could deal with acoustic guitar string squeaks with multi band compression? Maybe Massive can shed more light on these topics if he comes back? There may be a major flaw to my method.

My 2 pesos
Keith

Yeah, I too would love to hear some tips on proper use of multiband compression. Anyone up for it?

I use multiband compression [or expansion] often on bass and kick. Compress the snare-rattle or kick-click and leave the lows intackt... works pretty well..