The fast and easy way to understand compression and reaComp accessibility
The fast and easy way to understand compression and reaComp
lets say you have made a recording, and you are largely satisfied with the results, however, the voice or instrument you recorded is not quite OK because at some points, you find it being overpowered and at other times you can find it is just too quiet and you have to strain to hear it. So you might either divide the track into items, adjust their volume, or you might be a super fancy kind of person who uses her super brain powers and a controller and does a very awful lot of dedicated automation on the volume for the track.... and then, one day, the entire mix changes and you have to do them all over again. Boy! Turns out that you can have someone inside your computer do that on command, automagically, being respectful of the volume changes so they are just right, and you can compare the results immediately instead of pressing undo, redo, undo, undo, redo. ETC.
enter the world of compression and reaComp
Rather than explaining the entire effect controls and window, we can do an explanation from a "lets use this thing!" perspective. So, get to work.
1. Find a quiet section or a section that sounds good to you and find out by using your peak watcher. You can set up the peak watcher by pressing Control+Shift+W, and setting the first track to, "follow Current Track." Then use Alt+F9 and Alt+F10, to report the left and right peaks, for the current track. These keys will report the loudest peak since last the peak watcher was reset. Use Alt+F8 to reset the peak watcher. You can also simply press the J and K keys whilst playing the project and have the instantaneous peak. for left and right channel respectively. As these keys report the current peak rather than holding the highest value as the peak watcher does, you will need to press them a number of times to gage the average. 2. Open the effects on the track by pressing F after selecting it, and insert any compressor you would like; ReaComp will be used for this example. In your compressor, set your threshold value to about 3 db less than the average level reported by the Peak Watcher, and set the ratio to 2:1 or similar. Setting the ratio in this way means that the compressor will allow 2 db to pass if the sound exceeds 4 db over what you have set in the threshold. 3. Finally, you can see how many db's the compressor is pulling back next to the rms size box. NVDA will read them, first the left and then the right channel. So you might want to select a part and play it as a loop over and over. 4. If you find that compression is pulling the voice too far back or that the volume is abruptly changing, you can increment the knee parameter, or the attack, or also play with the preComp setting to get a more subtle compression. So experiment with all of those values and settings to find something that is good for you. You can also increase the wet output to try and bring things back up to the volume they were.
the box labelled RMS, is the one that tells you how hard the compressor is working. You should Leave the rms size alone, and instead try increasing the threshold more towards the negative values, and also increase the ratio. You should hear differences in the sound if you are compressing things too hard because a compressor indeed makes the dynamic range smaller and "compresses" the volume of whatever you are putting it on. which is the idea, because then you would want to slowly adjust the values to get a more decent sound. To hear what the compressor is doing, you can toggle bypass on the currently focused effect with Control+B. When the effect is bypassed, you won't hear what it is doing. This is good so that you can hear whether the changes you have made with this effect are improving the sound or not.
Ok, so what about the other controls?
There's also a makeup gain checkbox, which will put the compressor to the top of it's range, with the parameters you've set of course, but it's good to use when you have really squashed a signal and need to well, make up the gain. There is also a limiter, a brick wall limiter box you can check to make sure you don't over-blow the limit of the plug-in itself, a safety brick wall limiter.
There is also the over sampling slider, but it sets the quality of the processing, but it can also use more computer resources though you will get the best quality by ramping that all the way up to 64.
lastly, some useful presets
one of the presets, called "stock - Fader Ride Vocal - aim for 8dB GR" uses the out-put of the compressor to help determine how the compressor works, how much it compresses dynamically based on the output.
The advantage of driving a compressor based on it's output is that you can really do some deep compression without your original signal sounding so squashed, good for a podcast where you have widely differing levels of speech and you don't want to have to slice and dice and adjust volume levels. You'll still have to adjust threshhold and ratio and such but try this setting to start with if you have a lot of wide dynamics involved.