input Level determines how much of the incoming signal gets fed to the delay. It doesn't affect the dry signal, which just passes through to the Mix control.
Level works the same as it does in all blocks. It sets the output level of the entire block, after the dry signal has been Mixed back in.
Okay, it makes sense, but it also raises more questions. These are the things I struggle with, since there are endless parameters to tweak.
Why wouldn't every input gain (is this what you mean by input level above) be set at 100%? If I set it for less than 100%, and I hear a substantial volume drop, do I simply compensate by turning up the effect's level, or do I go back and set the input gain at 100% again?
Is the incoming signal whatever is coming from the output of the previous effect's output block (assuming series connections)? If there is no effect before the DMM, then the incoming signal is the dry signal, right?
In this preset, my chain is simple--an OD into the DMM and out to my amp. If I find that what I'm hearing coming out of my amp is too loud, I can: 1. Turn down my amp (I tend to have a pretty good idea of where I need to have it set, though); 2. Turn down the Level of the OD; 3. Turn down the Level of the DMM. (Am I missing anything else? Let's assume my signal levels to the FX8 are fine...) Why would it matter whether I turn down the level of the OD vs. the DMM? My instincts tell me to turn down the OD first.
Let's say when I do that, I feel like the character of my OD changes--so maybe I compensate by during up the drive knob a little.
Then I'm not sure my repeats are clear enough--so I tweak the mix knob a little bit on the DMM. And then maybe I bring up the DMM Level.
And then something else isn't quite right...and so on. I feel like I can go around in circles endlessly chasing optimum levels relative from one effect to another--and how that, in turn, affects the overall output/sound.
It's also easy to say "F it...this is good enough!" and move on. But I still wonder if my tone could be better sometimes.