microtonal notation (?s and requests for the future)

Discussions, advice, bug reports and much more about the "bach" environment.
Post Reply
joelgtaylor
Posts: 3
Joined: Wed Nov 15, 2017 12:55 am

microtonal notation (?s and requests for the future)

Post by joelgtaylor » Mon Dec 04, 2017 1:04 am

Hi,
Being a new user I'm going through the tutorials and browsing through the help files on various and sundry objects...As a composer who is very wrapped up in using microtonal resources I've got a few questions and comments on the way microtonal notation is implemented in Bach so far...Please forgive me if I've overlooked methods that already exist, or misunderstood something obvious. I want to say at the outset that I appreciate and admire the work that you have done here. Bach is a great package, and I imagine Cage is too (I've not yet had time to look at Cage, and figure I should really understand Bach first). Having rationals and lllls supported in Max, and a good notation facility, is going to make a huge difference to many of the composers who use Max.

1) The entire idea of tone divisions is odd from a microtonalist's pov. If I type 22 into the number of divisions (per octave) in your help patch I get a division size of 54 cents, which is correct, except that 54.55 cents would be better. And apparently this translates in 3 tone divisions, but 4 tone divisions would be closer...apparently there is truncation instead of rounding involved here. So, thinking about this, it seems that rounding up would almost always be better, in order to better accomodate enharmonics in microtonal temperaments. Better too many possibilities than too few. Finally, there are many tunings where the "period" of the tuning is not an octave, for example the Bohlen-Pierce tuning, which divides the Tri-tave (3/1), into 13 equal parts...How that tuning, or Wendy Carlos' alpha, beta, and gamma tunings, which are non-octave repeating, and involve dividing the P5th (3/2) into 9, 11, and 20 equal parts, respectively, are to be dealt with by a system that thinks in terms of steps per tone, is difficult to figure out. Every octave will have to have different notes in it, I guess. Can Bach figure out which notes are closest in that situation? If not, how can that ability be acquired? Is there a way for the composer to simply define a table for this?

2) The ability to add a cents indication is key to any adequate notation system. But it should be able to be specified relative to not just the diatonic, but also chromatic and 1/4 tone scales, possibly 1/8 tone scales, yes? Having this limited only to diatonic alterations in inadequate. What I, and the few others I know who actually try to get microtonal pieces played by acoustic musicians have found, is that a cents indication above or beside the note in question which shows a deviation from 1/4 tones is usually adequate. The other thing about this is that often once a few pages have gone by and the players have absorbed the inevitable table of the tuning or tunings at the beginning of the score, you can often dispense with the "complimentary cents" indications for a section of music, and just give the nearest 1/4, 1/3, 1/8 of a tone symbol, like a regular sharp or flat. So, to summarize, I think that the ability to add a "plus or minus cents" indicator for all of the various ways of notating intervals is what is needed.

3) I haven't been able to quite figure out how alternative font systems are being supported by Bach. That is, how the system goes about learning what accidentals to use given a new microtonal font. Maybe I've just missed this so far. There is now a font called Saggital that is an attempt to create a system that can handle *any* microtonal situation. Of course, as would be expected, it has turned out to be too complicated for most performance purposes, but quite a few composers are actively using it now nevertheless. The problem is, that a given Saggital accidental may mean one thing in one tuning and another thing in another tuning. There is a computer Saggital font available, and it is being used with Finale, Sibelius and a few other notation programs...There are of course, as your tutorials suggest, many microtonal notations floating around, looking for a home...Since I tend to use multiple tunings simultaneously in my music, I've stuck with the 1/4 or 1/8 of a tone accidentals plus or minus cents notation...but in the long run, providing support for more alternatives rather than fewer is desirable for most microtonal composers.

4) Is there a way to support Extended Just Intonation in Bach? I would think that if a composer used a scale that was a sequence of JI intervals that should be fairly easy to support, (like Harry Partch's 43 tone per octave scale) but what about what Lou Harrison described as "Free Style Just Intonation"? When composing in free-style JI, one can specify the ratio (which ultimately, could be any ratio, including non-JI ratios) between successive notes, or between vertically aligned notes... Can I do this in Bach? Can it be notated without great pain?

5) You may have already done this for Bach Roll, I don't know. It would be great if it was possible to notate ongoing pitch inflections in sustained notes accurately. A zoom function that would let you zoom in on a staff and draw exactly the pitch-height fluctuations desired would be a great boon when writing for electronic instruments...

I expect that the bottom line is that anything can be done as far as finding a way to tune pitches and/or pitch-classes, just as in Max, only easier because of the support for Rationals, but that notation is still problematic in a variety of ways. Which is what is expected...I hope you have plans for extending your current approach in the future...

By the way, I'd be more than happy to help with this part of the package in any way that I can...

All best,
Joel

andreaagostini
Posts: 209
Joined: Fri Dec 03, 2010 1:51 pm

Re: microtonal notation (?s and requests for the future)

Post by andreaagostini » Mon Dec 11, 2017 11:27 am

Hi Joel,

thank you for sharing your interesting thoughts.

You are right: support for alternative tunings in bach is quite limited, and the further away you stray for 12- or 24- or 48-TET, the worst it becomes.

The new pitch data type is meant to be a first step towards a more general implementation of musical pitch, but the problem here is that there are so many different approaches, each with its own standard notation and meaningful operations, that we haven't managed yet how to abstract them in one reasonably coherent and efficient system. I hope that at some point we will...

andrea

andreaagostini
Posts: 209
Joined: Fri Dec 03, 2010 1:51 pm

Re: microtonal notation (?s and requests for the future)

Post by andreaagostini » Mon Dec 11, 2017 9:04 pm

... to answer your questions more directly:

I agree that being limited to an integer microtonal division is not ideal, but it's practical from a graphical point of view. We can improve on that, but here's what I would do for a start: you can calculate your non-octaviating tuning in terms of midicents; then you choose a "base" microtonal division (say, quartertones) for display — I'd pick one capable of representing each step of your tuning with a different graphical symbol; finally, you can have a "lambda" system in bach.roll or bach.score writing automatically in a slot the deviation in cents of the actual pitch from the graphically represented one. Have a look at the “Slot machines” and "Lambda scores" tutorials for more info.

As for the Sagittal music font, if I'm not wrong (Daniele, please correct me if I am) there is currently no support for custom fonts — you're limited to November, Maestro and little more. Allowing users to load their own fonts and font annotations would be interesting, but also extremely complicated, both on our side and the user's: it's something we happen to talk about, though, and it is possible that at some point in the future we implement it.

Expressing precise pitch inflections is definitely possible — the standard way to do it is using the duration line, and feed it its values to your sound generation module... or am I missing something?

And Just Intonation, well, you’re not the first person asking us to support it. The problem is that, once again, this would require a massive rethinking of how pitch is represented, or some sort of open syntax for pitches, and I’m afraid we’re not ready for it. Consider that it would not just be about graphically displaying the accidentals, but also find ways to deal with their semantics, represent them as text, operating arithmetically upon them, and so on. Anyway, I’d like to have some good references on the subject (I’m definitely not an expert) — one never knows ;)

Cheers,
andrea

Post Reply