| Geberally speaking:
First things first: a compressor will generally make quiet things louder, and louder things quieter in a mix, bringing them more in balance. So depending on how you use the ratios of a compressor will determine the outcome of the final sound source. To much compression will squash and flatten out any recording you make if not used correctly rendering it flat and lifeless.
Second: you've listed amplify as your second option. Why? You'll be guessing at what level the audio signal is likely to peak, so you aren't necessarily getting the maximum output you can. You can also likely include clipping if you're not carefully. This is something I would do before trimming but after normalising. See below.
Normalise: this is usually your safest bet. This will bring all audio up to a predetermined loudest point without clipping. This is determined by the loudest points within the audio. So if you compress first and even out your sound, your likely to have louder normalisation than you would without.
What I would do:
I would first run a gentle compression to iron out any peaks and troughs, to help acheive a more even keel and balance. I would then normalise. If at that point you feel it needs to be louder, run any amplification, be mind full of clipping. Then do your trimming such as fades as you see fit last. (There's no point doing fades and then running any amplify is there, as you'll introduce unnecessary noise etc?)
__________________
[ Derren Lee Poole ]
The audience simply doesn't know, or care what bass you play. If it feels good, play it!
| 
