Two choices - either you record this and work with the recordings, or you accept that no two performances will ever be the same and make it part of the appeal.

I'd say most performances on these are accompanied by a DAW. Then you just record it and re-use sounds between performances. For some instruments/tracks you probably want the freedom to play it live and accept that it sounds different between performances. In fact that's the whole appeal of it in my opinion.

I like to take photos and sometimes write things down! In a sense you're assembling a sound sculpture. But once you've found a great sound and lost it, sometimes you end up finding something very similar but not quite 100%! You see this a lot with live performances from any electronic musician to be honest.

Still, most of the time you find something incredible and it's gone as quickly as it appeared!

> In a sense you're assembling a sound sculpture.

Or, in the case of the Medieval, a sound scripture ;)

There was recently a post and discussion on this very topic here on HN: https://news.ycombinator.com/item?id=40954679

Seems complicated, indeed.

As people also pointed out in that thread, some classical instruments like church organs have exactly the same problem.

Look at this one for instance: https://en.m.wikipedia.org/wiki/Organ_console#/media/File%3A... It has 522 draw knobs and 796 total controls for the musician to use. How do you play something on this organ to sound exactly like another organ?

Video (audio) demo; these are different and simpler consoles, but this still gives an overview of the issue:

- https://www.youtube.com/watch?v=LMLWLM_RbNs

- https://www.youtube.com/watch?v=wvfawKQVw04

- https://www.youtube.com/watch?v=rkmn4AO85L4

Many performers use a semi-permanent patch. You effectively build a synthesizer with exactly the affordances you want, and carry the thing to the gig patched up. (Those setups tend to focus a lot of cable management too, lol, so that you can actually reach the controls)

Or, some do improvised live patching, in which the goal is NOT to replicate sounds.

You don't (mostly, except by ear) and that's part of the allure of it.

That’s up to you how you do it.

You can sample certain sounds and perform with that instead.

You can patch it once and never touch again.

You can learn how to patch it exactly the same, even with 100s of cables.

You can keep parts of the patch the same. Some modules can help, saving their internal state or communicating with other modules on their own.

Etc.