I've been mixing in Logic for years, and it just hit me this evening that I don't know how Logic's standard compressor responds on a stereo channel when the right and left side volumes are different. Apparently, when the compressor is instantiated (as they say) on a stereo track/channel, the stereo version of this plugin is automatically selected. Interestingly, I couldn't find any mention of this topic in the Logic 9 Help menu or anywhere on Google. The reason why I assumed there was a mono and stereo version is because the plugin insert menu indicates 'Stereo' when selecting the main Logic compressor on a stereo channel. So my question is: How does the compressor behave when the incoming gain on each side is different? For example, if I inserted a Direction Mixer in front of the compressor (I never do this, but let's say hypothetically), such that incoming gain was much stronger on one side, do I get the same absolute gain reduction from the compressor on both sides (depending on the Threshold, Ratio, etc.), or does it discriminate and apply more gain reduction to the stronger signal, thereby effectively negating some of the Direction Mixer function? The question is important to me because I use a lot of loops that are stereo in nature and automatically create stereo tracks when they are imported. But I can see from the waveforms that some of these are shifted right or left or, even more commonly, bounce around in the stereo field. I usually don't try to compress such files, but sometimes, to economize the track count, I load these into a track that has another instrument in place for which compression is desirable, and then insert the compressor on the stereo track. I just don't have a clue how that compressor is dealing with the audio region that is bouncing around in the stereo field. Thanks in advance.

Don't confuse Dynamic Range, which a compressor controls, with Volume, which can change drastically within the dynamic range allowed, with Gain which are specific decibel (db) values that are used to measure both dynamic range and volume. Example, I can tell a compressor to give me a dynamic range between values of C and W db's. Compressors work for the most part with absolute threshold values. My audio can still change in volume between the values of D through Y db's. And I never have to talk about "gain" thus far. Gain is simply a measuring stick by which we talk about the audio level parameters. Think of it this way; A compressor only determines the absolute maximum and minimum volume settings, measured in db's, that a channel's volume values, are allowed to exist within. It will adjust the volume of anything outside of those max/min db values, lowering or raising as necessary. The volume changes between those max/min values are still dynamic and unchanged. Again, gain is simply a measuring stick by which we talk about the audio level parameters.

Please replace all instances of the word "volume" in that first sentence, with the term "Audio Level". At this moment it is too late for me to edit it directly. AND, even in stereo, you set the min/max values for both sides together, or in some compressors, individually. But best practice is to set them equally, unless you are targeting a very specific playback system in a very specific environment.

Ben, Thanks for your comments. Sorry about my sloppy terminology, but I do know how compressors work (I wrote the question at ~ 2:00 AM). I think it's too late for me to substitute 'audio level' for 'volume'. I'm not entirely clear about your answer (i.e., in your second messge about how, in stereo, the max/min values for both sides are set together). Does this mean that the same absolute Threshold and other settings on most compressors are applied in the same manner to each side of a stereo channel and apply independent reduction to each side? I understand that some compressors have a stereo linking button that will allow you to choose whether they will respond identically to both sides of a stereo channel (linked) or permit different degrees of compression for each side. As far as I can tell, there is no setting on Logic's standard compressor to specify different compression on the right and left sides of a stereo channel. So should we assume that this is 'linked' processing? Perhaps this explains why, when I apply Logic's standard compressor to a bus channel receiving inputs from multiple tracks, if there is an instrument (e.g., a drum track) that is equally balanced on right and left sides, the appearance of a new instrument in the bus channel that is highly off center (e,g., a loud synth that is directed largely to the right) will seem to shift the drum sound strongly to the opposite side. Or is that some kind of psychoacoustic phenomenon?