Here's a thing I just did: in attempting to make a sound like the chorus from "In The End" on Dookie by Green Day, I thought I'd take a methodical approach to obtaining similar tones:
1) listen to their sound
2) set up my amp, trying to get the sound as close as I could by memory
3) record it
4) listen again and compare
5) tweak the sound w/EQ in the DAW to make it sound more like the original
Granted, mine will never sound like theirs...but, by following that approach, I did make mine sound better.
Then the obvious occurred to me: wouldn't it sound better if I improved the EQ on the amp first, and recorded that sound, rather than messing around with EQ in the DAW?
The following is the result of *that* experiment.
http://eastofcleveland.com/m/moarGeetar ... riment.mp3
0:00-0:17 - take 1, raw
0:18-0:34 - take 2, raw
(I left out the part where you hear take 1 + EQ in the DAW, since the point of this exercise was to get it sounding as good "at the source" as it sounded with post-processing)
To continue the experiment, even though I could hear a difference, I wanted to *see* the difference to better understand what was happening. I slapped a spectrum analyzer on the master fader and grabbed this screenshot:
(click there ->
http://eastofcleveland.com/i/spectrumAnalyzer.jpg <-for the full version)
The left side of the spectrum meter covered "take 1", the right "take 2". You can see the concentration of lower frequencies on the left, and see more evenness in the spread on the right. Sure, we've all heard that you shouldn't "mix with your eyes", but it is interesting to see the effect of tweaking EQ settings to "spread out" the frequencies. even better that it was EQ on the amp.
btw, here are the guitar/amp settings from the two takes (Dean Vendetta - cheapo guitar, bridge humbucker, tone/vol 100%; Vox AD30VT, early-gen modeling amp, in "UK 70's" model, with upgraded speaker (10" Eminence Ragin Cajun); into an SM57 1-1.5" from grille, on-axis, a little inside the cone/dustcap seam):
take 1
brighter tone (right): gain=35-40%, vol=100%, treble=70-80%, mid=70-75%, bass=15%
darker tone (left): gain=70%+, vol=100, treble=15, mid=25, bass=10-ish
take 2
brighter tone (right): gain=35-40%, vol=100%, treble=80%?, mid=35-40, bass=15
darker tone (left): gain=70%+, vol=100, treble=20, mid=20, bass=5
Tn general, and keeping the gain/vol about the same, the "improvement" I made was to thin out the mids and lows, and push up the highs...which means, given the same relative volume, you'll hear more of the higher frequencies and less of the lower ones. That's exactly what you see in the spectrum analyzer. It makes sense, though; that mic position is relatively close, so the lows are probably boosted by proximity effect, and even though I didn't hear it as being "woofy" in the room when I first started, once I started toggling between the Green Day track, take 1 raw track, and take 1 + post-processing EQ, I knew exactly what was "wrong" with the amp settings in the first take.
Anyway, long-winded and maybe explained in a confusing way, but there it is.
I actually got bummed out. 
