Instruments with microtonal support

[ReverendLove]

Iannis Xenakis reborn as digital lifeform
https://www.iannix.org/en/

[monomaker]

Doesn't Alchemy (now owned by Apple) have it?

AFAIK they removed full keyboard tuning from the Apple version, so now it's 12 note retune only. Really backwards move, deeply disappointing.

[monomaker]

NI Reaktor

How do you retune Reaktor outside of 12 notes or EDO's?

[Cinebient]

Doesn't Alchemy (now owned by Apple) have it?

AFAIK they removed full keyboard tuning from the Apple version, so now it's 12 note retune only. Really backwards move, deeply disappointing.

Yes, but works still great in v1.55 and i imported all the great .tun files from Alchemy into all my U-he synths, Omnisphere and all other tools which support .tun files. They were the best for me.
And also Omnisphere comes with some great .tun files which i also import in all my other tools.
I have at least a good bunch of micro tunable tools which works nice together now.
But yeah, i was really dissapointed that they removed that feature.
I still wonder if there would be no way for a simple midi FX to play all tools in every tuning.
I think the Wilsonic app f.e. get a huge update soon (maybe finally midi out) and this would be great to control my plug-ins with. But does that even work and/or translate to tools which doesn´t support tuning at all?

[aMUSEd]

Doesn't Alchemy (now owned by Apple) have it?

AFAIK they removed full keyboard tuning from the Apple version, so now it's 12 note retune only. Really backwards move, deeply disappointing.

Micro and alternate tuning is built into Logic itself though including Scala import, the only problem with this would be if you wanted different tunings per instance, but the advantage is that the whole project understandings that tuning, including and especially the midi recording/editing. Very few DAWS have this level of microtonal and alternate tunings (because it is really about alternate tunings, these are not always microtonal) support.

[Cinebient]

Doesn't Alchemy (now owned by Apple) have it?

AFAIK they removed full keyboard tuning from the Apple version, so now it's 12 note retune only. Really backwards move, deeply disappointing.

Micro and alternate tuning is built into Logic itself though including Scala import, the only problem with this would be if you wanted different tunings per instance, but the advantage is that the whole project understandings that tuning, including and especially the midi recording/editing. Very few DAWS have this level of microtonal and alternate tunings (because it is really about alternate tunings, these are not always microtonal) support.

Isn´t that bound to 12 notes per octave in Logic always. You can´t use 13 or more if you want/need?

[monomaker]

I'm still using Alchemy 1.55 too, love it. One of the best microtunable vsti's ever until Apple ruined it.

The problem with MIDI FX methods is they use MIDI pitch bend, so polyphony can only be achieved by spanning multiple channels. They're calling that method MPE (Multidimensional Polyphonic Expression) now, but it's been around for a lot longer than the term. When you measure the resulting notes they're often inaccurate compared to decent TUN/SCL+KBM implementation (which can be right to fractions of a cent if fractional tuning is supported - major turn on). The Roli stuff (and many other microtonal instruments) use it.

It would be amazing if vsti creators could get behind a standard that gave realtime global tuning with single note changes (so you can bend tuning over all instruments simultaneously). You can do that with MTS Realtime, but hardly anything has ever supported it. Since many DAWs filter Sysex, it's probably not an appropriate solution for today. At the moment none of this stuff is standardised, so if a vsti supports TUN files you have to assess to what level it's using the data. Is it using the basic tuning, the exact tuning or the fractional tuning? Does it have a system for assessing the best data in the TUN. These problems aren't such a big deal for SCL support.

Thanks to this thread I'm going to get an iPad (or steal my wife's), the Wilsonic app looks excellent! Cinebient, I imagine if that app gets MIDI out it'll be via the MPE method, so I'm not too excited by that. Scala can do realtime translation (which you can use a virtual MIDI port to route in and out) but it's pretty terrible and the resulting accuracy (for me at least) in both pitch and time has always been disappointing. Plenty of things mentioned here that aren't on the Xenharmonics list, hopefully he/she can update it with the suggestions here. Bigup to Martin for this thread.

If everyone on this thread hit every new vsti manufacturer with the feature request we'd make this standard in no time. Maybe 2018 is the year...

[ReverendLove]

NI Reaktor

How do you retune Reaktor outside of 12 notes or EDO's?

Well you got the MIDI in. You catch the notes and translate them with a table macro to the desired pitch. There are macros and blocks for that in the user library already. As Reaktor supports OSC it would be possible to send it outside. With MIDI OUT that would be more complicated and would work only monophonic as MIDI does not support polyphonic pitches.

Found that in seconds

https://www.native-instruments.com/de/r ... how/11297/
https://www.native-instruments.com/de/r ... show/2971/

[monomaker]

Isn´t that bound to 12 notes per octave in Logic always. You can´t use 13 or more if you want/need?

That was my understanding, but I don't use Logic so would be good to hear from someone that does full keyboard retuning in it.

[monomaker]

Found that in seconds

I've tried all of these over and over again. The first just does EDO's, the second I can't import tuning from elsewhere, it's a bunch of preset values in a table and incredibly cumbersome to do anything with, when you look at the resulting accuracy it's really disappointing. Plus, I'm pretty sure when I tried it in R6 it just doesn't work any more.

Anyone have an accurate working method for microtuning in Reaktor outside of 12 note's and EDOs?

[ReverendLove]

Sorry for not being a native English speaker. What does EDO mean?

[monomaker]

Since I couldn't find one anywhere else online, here's my Halion 6 script for full keyboard retuning. Sadly there's no way of importing the tun files in Halion, so I open the tun file, strip everything but the exact tuning, then turn it into a 128 value comma delimited list (search and replace using a wildcard for the note value). Once you've got that, paste the values into tun={ } and you should be good to go. It's not perfect, but it's the best method I've got at the moment, I'd be really happy if anyone could improve on it.

Code: Select all

tun={ }
function onNote(event)
	local id = postEvent(event)
	local tunout = tun[event.note+1]*0.01
	changeTune(id, tunout-event.note, true, true) 
end

Hope this helps someone.

[monomaker]

Sorry for not being a native English speaker. What does EDO mean?

Equal Divisions of the Octave. Basically, dividing the doubling of a frequency by an integer.

[ReverendLove]

Ah, I see.

That Halion script is an equal distributing function dividing an octave to 100 tones, right?

I am more (like you?) interested in not equal distributed tunings like Indian micro tones or those used by Partch or Danielou.
https://www.semantic-danielou.com/seman ... nielou-53/

Here is a forum entry in the reaktor community explaining using a table for tuning more general.
https://www.native-instruments.com/foru ... ic.317350/

As I understand (and have done it with another issues but not tuning) you create a table with 128 entries and read it accordingly the incoming note.

[monomaker]

That Halion script is an equal distributing function dividing an octave to 100 tones, right?

No, it takes the values from a tun file and translates them per MIDI note, so it can handle any 128 note scale. The 0.01 bit is just changing the value from the tun stored one to the format Halion's changeTune requires. Why they couldn't have built that into Halion I have no idea, but that script lines up with most other synths that accept tun files. One thing that surprised me was the accuracy, it accepts a load of decimal places, so the exact tuning values are useful (as opposed to the simplified list, usually at the start of a tun file).

I am more (like you?) interested in not equal distributed tunings like Indian micro tones or those used by Partch or Danielou.

I like it all, but I am more interested in non-EDO scales, I'm certainly a fan of Partch & Danielou.

As I understand (and have done it with another issues but not tuning) you create a table with 128 entries and read it accordingly the incoming note.

Yes, I read the link you provided before (I think it's how I wrote my Reaktor retuner), but it's really not very satisfying to make music with. I can't find an easy way of updating the tables and then the accuracy is disappointing, it seems to only be able to get to the nearest cent (as opposed to that changeTune function in H6, for example). It's really the same method as I've provided for H6 (2D table of values - note# and cents/frequency, pick entry by MIDI note number), but it takes so much longer to get a scale into it and it's less accurate - a distinct lack of fun.