upwards expansion

[Mostess]

Why doesn't the plugin scan the whole sound file and make decisions based on the entire thing rather than taking it as it comes? Seems like a no-brainer.

I've noticed that a lot of good compression allows the attack to be louder than the (compressed) sustain. The delayed reaction seems to maintain the sound's integrity. A poor-sounding compressor kicks in too early (makes the sound flat and makes instruments all sound the same) to too late (just sounds fake) after the attack, and all the hardware and software compressors I've seen let you adjust that attack time. Given a delay time greater than a few microseconds, analyzing the whole file, or even looking very far ahead, wouldn't make any differences in the output.

Now, on a larger time-scale, I see your point. I'd like to see some sort of normalizer/hard-limiter/compressor that tries to make the RMS as constant as possible through an entire 3 minute file. Right now, I try to control that sort of thing at recording time. But when my song has a soft verse and loud chorus, I have to do some tricky things during mastering to keep it balanced. A full-file analyzer would be helpful.

[king_arthur]

Thanks for all the comments... my "half a second" or whatever was just a wild guess; the idea of variable lookahead makes sense. Maybe even two or three levels of lookahead (or running multiple passes?) - if there was a soft ten second passage followed by a big kaboom for dramatic effect, it might make sense to gradually fade the soft section over a few seconds so that you don't lose the dynamic change, but, yeah, you don't want the whole mix fading out just because there's one big snare hit coming, either.

Thanks,
Charles