Although telling you when a compressor is set right via text and not being able to hear the audio track is tough, there are something’s that you need to consider.
There isn't really a right or wrong ratio setting generally speaking; it becomes good or bad based on how it makes your source audio sound. First of all the ratio determines how much actual gain reduction is going to be applied to you source audio. For instance is your ratio is set at 3:1 that means that for every 3dB that goes in the compressor only 1dB is coming out, therefore you have a -2dB in gain reduction. This is where the actual compression takes place by reducing the dynamic range.
Simply put a compressor makes loud sounds soft and soft sounds loud. If you find yourself struggling to hear certain things in a mix (like a soft vocal part or some of the ghost stokes of a snare roll) or if you want to over emphasize something (like the breathing of a singer or the string scratches of a guitar player) you would want to use the ratio to determine how much compression to apply or increase/decrease of the targeted audios dynamic performance. So it only becomes good or bad based on if you achieved your goal with the reason you chose to use the compressor to begin with.
However in certain genres of music compression becomes more of a desired effect than a tool to fix dynamic performances. For instance any type of hard rock, hip hop, and other "in your face" genres of music require "heavy compression" in order to get that "in your face" type of sound. When you highly compression something it means that you are reducing your threshold level much lower than the peak volumes of the audio source and are apply a much larger compression ratio, probably somewhere from 4:1 and up. The effectively reduces all the dynamic range out of a performance and makes everything on the audio track the exact same volume. The spit smacking in the singers mouth becomes as loud as the singer screaming at the top of their lungs. This is known as "heavy compression". There are other examples such as what happens to music when it's broadcast from a radio station, very heavy compression.
Attack and release times need to be used together in order to keep what's called "pumping and breathing" out of your audio. The way you will use the attack and release will also depend on what element you are using the compressor on. Is it a very fast attack instrument like a drum? Or is it something that changes attack based on performance like a stringed instrument? Or is it a vocal?
This can be debated from now until the end of the earth and I'm sure this will stir up some conversation but generally speaking you'll normally want to set a fast attack and release when compressing percussive instruments and a bit slower on both when compressing guitars and vocals. Again this is a very generic assumption.
However when you move into some advanced techniques such as compressing reverb tails on a snare drum you'll change this approach completely. In this situation you'd want to have a slow attack on the output of the reverb that is on the snare drum and add some moderately heavy compression to the tail of the reverb in order to keep the reverb tails from getting lost in the mix.
Compression on vocals is different based on the genre of music. In hard rock I will typically compress the snot out of the vocal track. In country it will depend on the nature of the song. In hip hop again I will use heavy compression. In jazz i will not use any compression at all.
Here is a simple method to add some moderate compression to any vocal track. First start with setting your ratio at 3:1, and then move your attack time from the "extremely fast" position and decrease it about 10%. Then increase your release time to about the same. Then as the audio track is playing back, start reducing your threshold dB levels until you start getting some gain reduction on the compression meters.
If you are still
developing your year, this will be a good way to start as you'll begin to know when to start listening for the effect of the compressor on your vocal once the gain reduction on the meter starts lighting up.
My guess is that this thread is going to grow.....and grow....and grow....as many people have lots of opinions about this topic.
On my audio engineering educational site I have some more details about this. Please feel free to let me know if you have any more questions.
Make Happy Music
