EQing. Conference of the Fuzz.

[D.o.S.]

no one loves Rumi.

I'm not sure a 250 is what you want if you're going to be using a guitar for this -- also take into consideration that a big part of dude's tone on the first couple of records is his attack (you have to bash the shit out of your strings to get the DOD to do those Variations/Conference things when you're using a bass) which is where a lot of that super-saturated tone comes from.

[PeteeBee]

Outside of phasing, which is totally a thing, a big part is trial and error in that you have to get a guitar tone you dig, faux bass tone you dig, then master them together. Each step takes a little time and the mastering time might make you rethink the other. When I've tried to do this I've found I like having one or the other cleaner than I started with.

[crochambeau]

You need something that does something like this:

The illustrated phase coherence between two sources is only an important aspect when you're collecting from a single point (the listener in that image), and that example is aligning the players stack (at stage backline) with the house PA. So in essence the FOH is delayed enough that it fires when the wavefront coming from your amps reaches the mains - so it reinforces as opposed to subtracts at some acoustic determined frequency that will shit on your sound.

For recording purposes, they are usually applied to two microphones at different points of collection so that when the signals are summed they are in alignment as opposed to cancelling each other out at a frequency that is determined by wavelength/distance apart.

That's the key though, time delay tricks are usually employed to maintain the integrity of a signal, and inverting the application will probably do more to blur your content or introduce EQ-esque type effects from cancellation/reinforcement than it will split a signal.

Historically there have been many gimmicks aimed towards extracting a stereophonic sound from monophonic source, by and large, they all have pronounced artefacts and are probably not the best solution to the puzzle, in addition to the millisecond delay trick, there's also a sort of comb filter - imagine feeding your signal into a pair of graphic EQ and then killing all the odd bands on one side and killing all the even bands on the other. You'll wind up with two separate signals, but they will be heavily processed and I feel that goes against the grain of the original goal.

What about mid/side (sum/difference) processing of your instrument, either via collecting the additional signal with a contact microphone (or tapping your signal post dirt/whatever) or electrical modification that allows you to treat the bridge/neck pickup as two different sources?

[Chankgeez]

See? This is what I was talking about when I said someone who knows what they're talking about will come here and say something. ^^^