Not him but I work in music so I can give my input on some of these
Twiggyquestions as i'm reading
-what makes mastering so difficult that you (and lots of others) outsource it to some expensive studiodude? According to Crypto's definition you take your audio files and then put some analyzers, audio meters and whatnot on it and tweak a compressor/eq until all the tracks look like you want it.
Basically, mixing a track is taking the record "stems" (basically just the raw audio files) of each instrument, voice and whatever and adjusting them so that the track sounds as good as it can do, which are then bounced down into a single two channel stereo mix. You then master this stereo mix to make sure it sounds balanced across formats and playback devices, as well as making sure the album, EP or whatever sounds consistent. This is compression, very tiny EQ tweaks, shit like that. I've also seen people just master on the L/R output of the mixing session if they're doing it all themselves.
I think the main reason people don't do it themselves is it's a very particular skill and is also finicky as all fuck.
Twiggy-So if you play electric guitar, you put it through a "standard" guitar amp, then put a mic in front of the amp? Why not ditch the mic and plug the amp signal directly to your recording device?
You can record guitars directly into recording desks, most famously Nile Rodgers from Chic did it to keep his sound very clean. However the amp itself has a certain "character" to it's tone, and even beyond that the speakers add their own character to it. Then depending on where you place the mic on the speaker cone, and what the speaker is, it can sound even more specific. A lot of people mark a "Sweet Spot" on their guitar amp speaker, which is basically where the amp sounds the best.
You can use Amp Simulators now to recreate a lot of that, or even take a line from just the Amp Head itself without the speakers and use a Speaker Sim, but for a lot of rock stuff the sound of a real amp with real speakers just sounds a bit better and more authentic.
Twiggy- I thought compression was about putting a max threshold on a signal's amplitude. I should probably read more about this because I seem to understand it backwards (ie compression = reduced amplitude -> less pronounced attack)
Compression is basically putting a threshold on volume, but you can change how much is ducks it below the volume, or can set it to raise the volume, or "squash" everything together to sound flat. You can also adjust how quickly the compressor acts on the signal with it's attack, how quickly it stops, you can set a "knee" in the graph of how it adjusts, a lot of things.
I just fiddle with the knobs til good sounds happen