T' & non-vocal compression

[PaulG]

Thanks to all that responded to my related vocal post. Here's another q, though...

If I only need hardware compression for vocals, and all my other instruments are either plug-in, or sample-based, then...

1) How much is compression really necessary for a dynamic range (error-type correction) issue, and

2) How much is compression useful for audible effect?

My main app, thus far, is rock-oriented instrumentals, but I'm going for a combination of sampled and generated bass, as well as drum samples.

If I tweak the volume on each drum sound, why would I need compression if the sounds are sample-based? There's certainly no "humanizing" in that respect to correct...

I guess a tangential question would be - does compression do anything pragmatic/practical besides fixing dynamic range issues? And how does Tracktion's Compressor do, IYHO?

Thanks, and cheers to all,

- Paul

BTW - I just went through a major hurricane here in Orlando, Florida - do they get anythng like taht in the UK for those residing there currently?

[valley]

BTW - I just went through a major hurricane here in Orlando, Florida - do they get anythng like taht in the UK for those residing there currently?

Not very frequently. Flash flooding is becoming a problem for much of the country[1]. The last 'hurricane' to hit the UK was around '88 if memory serves, and by the standards of those that tend to hit Florida, it was only moderately severe. As for tornados, I still remember wandering into my in-law's lounge proclaiming that America is truly a f**ked up country 'cos the sky outside was the colour of grass, and being bemused as frenzied outbreak of window shuttering, and flashlight hunting broke out around me.


[1] obRant: who'd have thunk that filling in all the f**king drainage ditches on subsidence area flatland would lead to serious run-off problems and flooding. f**king idiots.

(sorry none of this was in any way relevant to your actual question, but its' late, I'm drunk, and I need to get up in four hours to run some electric conduit to my studio room, so i'm not all that together right now.

[jens]

depends on the samples actually - I prefer drum-samples which are completely unprocessed (despite being normalized) because that gives me the biggest freedom to work with them. But a lot of the sample libraries out there are already heavily processed and that sadly often includes a fair amount of compression. So in this case you can't compress them anymore that much.

Synthies don't actually need any compression (only limiting on very few occasions).

All in all rock tends to use a lot of compression and often as an effect (pumping&breathing).

hth

[Lunch Money]

1) How much is compression really necessary for a dynamic range (error-type correction) issue, and[/url]

If it's just a line that was recorded at too low a level, you don't necessarily need compression at all. Use a gain/volume/level (whatever you choose to call it, same thing!) filter. I have to admit, when I DO use compression I've never gone by the numbers, mostly 'cause I'm lazy (though I'm sure I'd do a better job if I was more attentive!). I twiddle knobs until it sounds good.

2) How much is compression useful for audible effect?

Depends on if you LIKE compression artifacts. I've never actually confirmed this, but I suspect that the "pumping" that people refer to is a compressor set to VERY aggressive settings so that as the signal fades out it's still jacked up, along with the ambient noise and hiss that you don't notice at low levels. On guitar, I've found that aggressive compression isn't all that useful, taking out the immediacy of the sound.

Also note that not all compressors behave the same way, and some WON'T produce much for audible effect even if you want it. In the digital world (unlike the analog), programmers usually have to make an effort to specifically program a natural-sounding "pumping"-type compression.

My main app, thus far, is rock-oriente ... correct...

Again, I'm just a schuck with no scientific explanation, but I've found that compression will make drums sound "tougher" than simply raising gain. Possibly because even though it's only for a short period (ie a snare crack), it's more of a constant "loud" sound than a natural snare which will peak and fade very quickly (keep in mind, even with compression it's 'quick', I'm only talking relative-- a compressed snare vs. an uncompressed one). That's just a guess. I just know that with rock, I almost always add compression to my drum samples to toughen them up.

I guess a tangential question would be ... do, IYHO?

Sure. It actually changes the entire shape of your waveform in varying degrees, depending on how it's set. If you compare a compressed soundwave to its compressed version, it will look very different. That difference has a greater impact than mere 'level' management. When people refer to "transparent" compression, though, they want those changes to be as unnoticeable to the human ear as possible. Ie. even when the compressor is changing the waveform, the illusion of a completely natural sound is maintained. As an example, a vocalist who isn't good at controlling the mic, after 'transparent' compression will (ideally) sound as though they were much better at using a microphone properly, and will still sound natural and unprocessed.

Disclaimer: I'm no expert by any stretch... and although I like to share what I know (in this case, very little!), I'm making no claims to being an authority on the subject.

Greg