2022-level tuning features.
Andreya-Autumn opened this issue · 3 comments
Hello again! First of all, I know there are many bigger fish to fry, and don't expect anything here, but I had written most of this out for other reasons and wanted to drop this here anyway in hopes that it's helpful at some point.
So, all the Madrona synths allow re-tuning, which is great! But they are missing some of the features that make a software instrument truly appreciated by us who work extensively with alternative tuning systems. I thought it might be helpful to share my perception of the state of this art nowadays.
There are two synth re-tuning methods that matter nowadays. MTS-ESP and scl/kbm.
As discussed on the forums, MTS-ESP is a centralized tuning system, where any compatible softsynth connects behind-the-scenes to a master tuning plug, which controls the tuning. It's great because you don't have to load tuning info into every single instance, and because tuning can be changed dynamically during playback.
Scl/kbm is the older workflow, in which little plaintext files are loaded with information about frequencies and keyboard mapping. It lacks the advantages MTS-ESP has, but some folks have used it for years and built lots of workflow round it. It can also retune different synths separately which some consider an advantage.
The community seems to be in agreement now that in order to be fully adequate tuning-wise, a software instrument should support both methods.
Madrona synths presently support either workflow, but only partially.
MTS-ESP can be utilized via MPE. But that's some extra hoops to jump through, and because of event timing ambiguities and a bunch of other reasons it doesn't always work reliably. Pitch bend in mono patches can also be used, but these are the kinds of limitations and workarounds we want to get away from.
And as for scl/kbm, all the synths load .scl files which is great. But the very important .kbm files aren't supported, which severely limits the usefulness of the scl re-tuning. I'm happy to share specific examples of why that is, if needed.
Good news is, both of these tuning methods themselves are available as open source libraries (0BSD and MIT licence respectively), so implementation is easier than ever!
https://github.com/ODDSound/MTS-ESP
https://github.com/surge-synthesizer/tuning-library
Each of those repos has more detailed instructions for how to implement either. Both methods need some menu options to work as well as they can. MTS-ESP strictly can function without them, but it's very useful to include at the very least the one that chooses between continuous retuning and note-on-only retuning. And bonus points from all Lumatone owners if including the multi-channel tuning table function!
It's my perception that with both those implementations done right, no microtonally-inclined musician will have anything to complain about. :) Surge XT, Dexed and Pianoteq are cornerstones of the scene for this reason. Quite simply, they reliably do everything we need.
Again, hope that's helpful somehow. Thanks for all the work so far, and good luck continuing it! :)
Hello.
Great to hear about the .kbm loading! I never knew, sorry for not researching that further. I'll be sure to let some of my friends know that's possible.
Thanks, take care. <3
MTS-ESP has long been on my radar. It’s on my list of things to add.
Good to know! Your plugins would make a significant contribution to that scene.