With a threshold of -10, the audio above -10dB is made less by the ratio - so 3x means that to reach 1dB above -10, requires 3 dB of input. An input of -1dB with these settings will output at -7dB.

To add +11 to the abpve will invariably cause clipping (unless your headroom is ludicrous) and would require a pretty large ratio.

So yes, you can. But it's not advisable. Like giving phantom power to a dynamic mic.

While Codemonkey is correct, I think the short answer to your question is yes, you can make up the gain. Many compressors actually do this automatically. And in fact, it's the main purpose of using a compressor. Lower the peaks so you can raise the average level.

Attack times - My rule of thumb is: The quicker the attack of the source, the faster the attack of the compressor must be to catch it.

You can also use a compressor to increase the attack of an instrument by using longer attack times. It's best to play with it. Put up a snare track, and dial in some ridiculous compression and threshold. The release should short for this experiment. Start with 0ms, and you will hear the whole thing squashed, and possibly some distortion on the transient. As you bring up the attack time you will hear the transient "snap" start to come through more. Continue to bring up the attack time and the "snap" will start to go away as more of the body of the sound makes it though the compressor.

This will leave you with a nasty transient, and a smashed snare sound common in pop rock. To make this mixable I run it through a soft clipper to tame the peak. The added harmonics of the soft clipper allow it to still sound loud, but not clip your master bus.

What you end up with for numbers is going to depend entirely on your track and your ears.

natural wrote: And in fact, it's the main purpose of using a compressor. Lower the peaks so you can raise the average level.

While, in principle, I agree that this is the use to which compression is put today, I would tend not to say that in plain language, as it can be easily misunderstood.

I would tend to say that the main use of compression is to positively affect the dynamic range of the performance, in order to make your mix decisions more simple. Maybe I'm just being pedantic, but your definition, while technically correct, can be construed as pandering to the loudness war. YMMV, of course.

Thank you Hackenslash, I guess to the novice it could be construed that way, but I find there's no need to sugar coat the definition of a device.

Compressors can make the program louder or too loud.
EQ can make the program sound Equal or hideous.
Delay can be used to recreate the sound of an acoustic space or something that doesn't exist in the 'Real' world.

Any effect or tool can be used to improve and solve problems. To make things sound more 'Natural'.

They can also be used to the extremes to make things sound unNatural.

The loudness wars are not new nor have they always been fought with limiters.
The loudness wars of the last generation were fought with EQ's. In order for mixes on jukeboxes to cut through louder than the competition, The EQ was taken out of it's 'normal' usage of making music sound 'Equal' and forced a midrange curve that could be heard through a crowd.

We also did some horrendous un-natural things with reverb back in the 80's too.
All these abominations will run their course and listeners will dictate when they've had enough.

Makeup gain is just one more fader in the gain structure of the track you are compressing. Generally you want to apply enough makeup gain so that the peaks of the output of the compressor are at the level that you would like your raw tracks to be for mixing. To some degree that's a matter of taste. If you add too much makeup gain you have to deal with the result the same way you deal with a track recorded close to full saturation, e.g. if you add eq you will probably have to back the gain off so you don't clip, etc.

My standard recommendation on how to experiment with setting: take a very short clip of a single instrument (couple of snare or kick hits, a two bar bass line, a short vocal phrase) and loop it. Put your other settings in some sort of default mode that you are comfortable with but make sure the threshold is set so that the compression isn't too subtle - for this experiment you want to hear it clearly. Now sweep the attack back and forth slowly from its lowest to highest values. Get it to the value that you think sounds best, then tweak the other parameters before going back and repeating the experiment with the attack. Then raise the threshold to something more subtle and do the sweep again and see if you can hear the difference the attack makes.

natural wrote: Thank you Hackenslash, I guess to the novice it could be construed that way, but I find there's no need to sugar coat the definition of a device.

-I must be spending too much time at future producers. There, you have to be really careful what you say, because it gets misunderstood, and then hundreds of newbs are spouting the mistranslation as gospel, even to the point of arguing with the poor schmo who gave them the info that they misunderstood in the first place. :lol:

It's a battlefield.