I would divide it up into a lot finer pieces than that. This process has been going on for a very long time.
Clippers/Limiters. These are the shock troups of audio processing. Everything over a certain level gets cut off. Not reduced, but brutally whacked off. You can achieve this effect by amplifying a sound until it smashes into "0" and stops increasing. You can get both Normalize and Amplify to do this by defeating their reluctance to go beyond zero. There is a filter called the Leveler which does something very similar.
All these effects are in Audacity 1.3. 1.2 has very crude, old effects. You can hear this effect in any two-way radio, fuzz guitar, and aircraft control tower. Frequently called "overmodulated." Audio people called it distortion and plug their ears.
Compressors. Limiters suck for music, so at the next level were the compressors. These machines reduce the level of the show according to how loud the show is getting. The effect is somebody very rapidly turning the volume control up and down during the show. This is better, but the loudest part of the show always takes over. The loudest part of a song is always the bass notes with an occasional drum sneaking in. There is a popular song out where they make use of this effect. The singer vanishes at each bass note.
Dynamic-ish Compressors. The first really successful music and broadcast compressors were dynamic. They made a stab at figuring out what the actual content was before applying compression. The advertising was they simulated somebody with good taste turning the volume up and down. They featured attack and release times. They would rapidly reduce the volume of a loud part, but then bring up the volume again very slowly in hopes that nobody would notice what they did. Depending on how old you are, you listened to CBS Laboratories Volumax and Audimax on every radio station. They semi-successfully combined peak limiting and slow compression.
You can hear these working even today. Listen to Dr Laura when she stops for a second for a response from the distressed housewife.
"Why do you think he left you?" .... s s sss sssss ssssss SSSSSSS SHSHSHSHSHSHSHSHSH!
"What did you do to tick him off" .... s s sss sssss ssssss SSSSSSS SHSHSHSHSHSHSHSHSH!
I can tell you with a stopwatch what the release time of her compressor is.
So close, but still no points. This kind of compressor is available natively in Audacity, It's settings depend a great deal on what the content is and it features incomprehensible control panels. It also must frequently be applied more than once or in combination with other tools.
Dynamic Compression. Fully Dynamic Compression takes into account what the content of the show is and changes compression accordingly. These are the ones that produce a loud show but you can't tell what they did or how they did it. These are enormously complicated internally because they feature response shaping, look ahead processes, and other stuff that normal men and women were not meant to know. Complicated internally, but not necessarily externally. Because they take into account show content, you never have to change them except for a tweak here and there. The compression character descriptions contain words like "Musical" and "Transparent."
This is why Chris's Dynamic Compressor was such a hit.
http://pdf23ds.net/software/dynamic-compressor/
I change the top compression number from .5 to .77 and it produces a show almost exactly like my local NPR FM station. PodCasts tend to not have any compression at all and so it's impossible to listen in the car or jogging on a busy street or on the beach. One trip through DC Bias Removal and Chris's Dynamic Compressor and I'm good to go.
Chris comes up in a Google search even if you spell his name wrong. You can still use the other compressors for special effects or other specialized work, but if you're just interested in making your show louder and even with the least damage, this is the way to go.
Koz
).