The manual I read for the Orville suggested how-to connect it. If you have a '68 Fender amp, you love the sounds, but have no effects loop. Their recommendation was mic the speaker and send that signal thru the effects (which sounded like a major pain). It took years for me to try it, but when I did...
I learned that Eventide designed their units around the source signal. Also that the source signal plays a major role in the processed sound. Unlike pedals. Pedals work with guitar signals, they work exceedingly well. The slap backs sound immense. But with a rack mounted unit, the echos and verbs are thin and tinny. Listen to what a guitar preamp outputs, and ah ha! Preamp signal is brittle, hot, and is this way because a poweramp thrives on it.
Always a skeptic, I was unsure of the digital preamps available built into recording interfaces. I was afraid because I had tried many times to make use of the MP-1 output signal... and failed to impress. So I figured I should invest in an analog mic pre. Well, I was wrong again, both analog and digital are good at hearing the speaker. Both sound believable. The analog let's you wooly up the sound with its transformers, but the effect is mild, almost indistinguishable.
Routings. I have three routers, and what complexities those bring. Router 1 is just for the guitar rig. You can patch things together in any way and store it. Router 2 is the interface. Eight ins & eight outs. Four of the ins are microphones. The other four ins are Orville signals. Outputs go back to router 1. Router 3 is the software channel mixer.
In software you will setup your channels. Channel 1 logically assigns to mic input one. Its a stereo channel, so I initially assumed it would feed it mic 1 & 2. But let me stop right here and say how important it is to avoid stereo channels at first. Instead, setup ch1 as mono, with only 1 mic on it. 2, 3, 4 are the same way. They could all be stereo, but skip stereo for now. Mono, delivers a huge (both L&R) sound of one speaker. Channels 5 & 6 are stereo setups. To get an ISO (solo) of one speaker, you have to hard pan the channel. My software has a feature 'stereo separation' which needed analysis. It has the ability to put the mic 1 signal into L & R... not what I wanted. I wanted mic 1 on the left only, mic 2 on the right. Stereo channels need to be checked to see that this is happening proper, otherwise hard panning is ineffective. Adjusting 'stereo separation' feature to remove this effect was crucial.
(To explain further, duplicate paths are bad news and phase cancel themselves, drop volume, and skip certain frequencies. For example the D note is usually nice and loud, can disappear and become the weakest note.)
Using the stereo channels, I can pair mic 1 with mic 4, mic 2 with mic 3 as well as 1&2, 3&4. But to solo a speaker, mono ins work much better. If I solo a stereo channel, it comes out of Left or Right (and not both). It has half the volume, and doesn't translate near as well as a mono source playing out of both listen back monitors.
Now that the software is able to select microphones and sort them out properly, mono & stereo I next looked at the interface router. The outputs of the interface get sent back to router 1. But you can be selective here. By choosing the software source to be sent out, which ever channel is selected in software goes out the interface (router 2) outputs to router 1 input. This means the mic 1&4, or 1&2, or 2&3, or 3&4, or mono 1, mono 2, or mono 3, or mono 4 can be 'sent'.
Router 1 receives the signal as selected in software (router 3). This is my guitar rig. This is patched into the effects inputs. Now the effects are based off of microphone source signals. Not only that, but each mic is on a different speaker, and effects are based on that specific speaker. This let me listen to a 6L6 tube/Vintage 30 as paired with EL84 tube/Alnico Blue thru a stereo delay. Each slap back accurately represents the V/Blue combination. It is shockingly accurate. Plus, signal degradation is present (this is what I want from a stereo delay, each repeat should degrade slightly, to simulate analog tape echo degradation). The digital world differs from analog, because things do not degrade. So nicely built delay algorithms will have filters on the repeats so they progressively degrade, simulating analog. Of course it can get seriously complex, modulators, pitch shifters are also possible, but speaking of basic digital delays, you really want to hear degradation.
How about the rest of the patches? Does everything improve? Well, this is subjective territory. Reverbs are now based off very warm speakers, and they sound thick. So Reverb can instead develop a thunderous ominous hue, dark and dense. The near/far feature of Reverb can be a bit harder to detect audibly (where am I?) but is completely realistic in feel. So often I listen to Johan recording his Plexi amp at full throttle. A close mic speaker doesn't really sound like it is cranked, at all. It still just sounds like a Marshall. The Omni mic he blends in is what catches the whole 'in room' feel. Since I am a low volume recordist, I have developed a crutch for delays and reverbs to add this 'in room' feel. It is a simulation based off of the real speaker, however, and has a much more believable final impression, studio quality to it.
My initial recommendation was Bricasti/Eventide Reverb 2016 (or software plugins) because they have a selection of rooms, spaces, and positions that recreate what a room mic is capturing. It hears a lot of garbage, and is not usuable alone, you still need the close mic to define the 'in your face' guitar.