Simply put, compression is evening out the decibel (dB) level in a signal path, making quiet sounds louder and louder sounds quieter. It can enhance any single track or stereo recording to professional status, or completely suck the life out of it. Understanding compression is difficult, but we're here to help engineers take control of their tracks.
Controls
Most compressors have the multiple variable controls that allow limitless forms of compression and tone-shaping. Typically, these include, a threshold, ratio, attack and release. Each are equally important in compressing a signal to achieve optimum results. 
The threshold is the point at which a signal starts to be compressed. It is like a ceiling with a variable height. The lower you bring your threshold or the ceiling, the more compressed it will be. Some compressors have a fixed ceiling. This means that the threshold will stay in the same place, however, you can raise the signal (or the floor) into the ceiling to dial in the amount of compression you like. Ratio is the how many dB are being taken out to compress the signal. For example, a ratio of 2:1 would compress a signal that was 6dB over the threshold (ceiling) to 3dB instead. Likewise, a 3:1 ratio on the same source would compress further to yield an outgoing signal of 2dB. Ratios that are between 20:1 and ∞:1 are considered "limiting" and ensure that a signal will not go much over the threshold, if at all. Attack and release determine when the compressor will kick in after the signal reaches the threshold (attack) and falls below it (release). On the specific compressor above, the attack is read in milliseconds (ms) and the release in seconds (s). Those measurements are standard, but there are a few here and there that shy away from the norm. 
