Using compression for punch effect

I have this flute sample that I'm trying to give more punch. I've been experimenting with the compressor on the mpc1000.

At the moment my settings on the compressor effect are like this:
Threshold: -60
Gain: +7
Ratio: 25:1
Attack: 99
Release: 0
The sample is punching pretty much the way I want it to now.

Do these compressor settings (numbers) look odd?
Some people say that they only use ratios up to 10:1 and threshold values down to -20.
However, my sample was low in gain to start with, so maybe that explains the numbers.

Has anyone else used compressor settings similar to these numbers?
It would be nice to know if I'm doing something wrong. Thanks in advance!

If it sounds good & clean without obvious "pumping and breathing"don't change it.
I have the same thing when compressing the output of one of my racksamplers....The output is so low that i have to use stupid settings to make it work sometimes. There's not much you can do about that sometimes, but it's never a good thing. if your sample is too low in volume, it won't compress at all, and even if its a little low you'll be raising the noise floor while you're compressing it.
Ideally you'll want as much gain in the front of the signalchain as possible (without clipping of course).

In your case you could also normalize the samples first before compressing them, this will most likely add less noise and allow you to use "normal" compressor settings. I'm not sure if the mpc1000 has a normalize function, you may have to do it on a computer.

Thanks man!
I'll try increasing the gain before compressing next time!

Anders wrote:Do these compressor settings (numbers) look odd?

Yes they do. But if it sounds right, it is right. Although Ieism is right about applying more gain before compressing.

Maybe I can give you a little more insight in what the numbers mean:

Threshold: -60
The level above which the compression starts taking place. With your sample it's real low, because there's hardly anything there for the compressor to work on. Hence Ieism's comment. On more typical material this figure could be anything from say, -3 to -20, although it all depends on the source material and there is no 'correct' setting.

Gain: +7
This is the 'make-up gain', applied at the end of the compressor's signal chain. It is used to 'make-up' for the gain lost by compression. So typically it would be a similar figure to what's showing on the gain reduction meter at peaks.

Ratio: 25:1
The compression ratio. 25:1 is HUGE, and it makes the compressor act more like a limiter. X:1 means the source signal has to rise X above the threshold level for the output signal to rise 1 above the threshold level. Many compressors have settings from 1.2:1 (very subtle) to about 10:1 and after that they call it limiting. The higher settings are often used with higher threshold to just squash the loudest peaks. Very low settings work well with low threshold to rise the overall level without causing any artifacts. These examples apply more to busy material like (sub)mixes - on solo instruments you can get more creative.

Attack: 99
The time it takes for the compressor to react to a peak. This provides a way to let the initial attack of a sound pass through unaffected, keeping the 'punch' intact, before the compressor does it's work of raising the sustain. This figure is usually given in milliseconds. I don't know how it works on the 1000, but '99' sounds like the highest setting, which would make the compressor take a _very_ long time to start working.

Release: 0
The time it takes for the compression to stop working after the sound drops beneath the threshold level. A setting of '0' might cause pumping, though on a flute sound you might get away with it. For more percussive sounds a higher release setting is required to avoid a very hectic pumping sound.

Damn. I probably could have just given you a link to a 'compression explained' webpage. Oh well.

Lo-fi, thanks a lot for taking the time to answer my question.

The parameter I was most concerned about was the ratio. I know that 25:1 is a big ratio, but I just didn't get any "punch" when I left it at 2:1 or even 8:1.
Maybe the compressor in the mpc1000 is just crappy?!

To this day, I still only understand the very basics of compression as far as the numbers go......I just twist and turn until I like what I hear.

Knoxximus wrote:To this day, I still only understand the very basics of compression as far as the numbers go......I just twist and turn until I like what I hear.

Yeah it can be very different depending on what sounds/samples you are using so the easiest way I think is to proceed with what sounds good to your ears.

yup use your ears but it sounds like it is too quiet as it is so try boosting the gain....if after the compression you look at the sample and it looks like a squarewave you know its been compressed to absolute buggery- which isnt always a bad thing...just ask dre...

greg