[linux-audio-dev] XAP status : incomplete draft

David Olofson david at olofson.net
Sat Dec 14 12:55:00 UTC 2002


On Saturday 14 December 2002 16.13, Steve Harris wrote:
> On Sat, Dec 14, 2002 at 03:35:41 +0100, David Olofson wrote:
> > > Er, well, most people will just let the host do the wiring for
> > > them. So it will all work fine.
> >
> > ...as long as they put the plugins in the right order.
>
> Well you will almost allways use a external device -> pitch
> converter, so you cant get it into the wrong order. Anyone whos
> capable of comosing in more that one scale at once is alos capable
> of placing a few modules.

I don't get it. If you're supposed to place the scale converter 
*first*, then how are you supposed to be able to apply anything like 
traditional music theory, rather than pure, continous pitch based 
theory? You will have to know the *exact* temperament of the scale 
(to decode the input, and to generate output in the same scale), even 
if you're only worried about notes.


> > Running linear pitch with a scale applied into a plugin that
> > expects <something>/note is not a mismatch? So, how is that
> > plugin going to figure out what pitch in the input corresponds to
> > which note in the scale?
>
> No plugins expect something per note. All of the expect to receive
> pitch, if ther are designed for ET they can calculate the note fomr
> the pitch trivially. You have to tell them what scale youre using
> (or they could be 12tET only).

Why design a plugin for one specific temperament of one scale, when 
you could support a range of similar temperaments for that scale...? 
Why worry about temperament at all, in every plugin of this kind?

This seems to me like a totally backwards way of implementing note 
based theory.


> > > It wont need explaining, its blatatly obvious, unlike if you
> > > have pitch and note pitch when its not obvious if they will be
> > > compatible or not (even to the host).
> >
> > I don't see how it's blatantly obvious that things will work if
> > you put the plugins in one order, whereas it will not work at all
> > if you put them in some other order.
>
> The order is irrelevant.

Yes, if every plugin is aware of the exact temperament of the scale.


> Realisticly you only need to convert from notes to pitch at the
> input stage, once it in the system you will be fine just processing
> the pitch data.

Yes, but only if you work only with continous pitch based theory, or 
ET scales only.


> If you really, really want to convert from one source of note
> numbering to two seperate scales you do the equivalent function
> with pitch mappers (we discussed this a few days back, I think you
> agreed that it was easier to do the processing on pitch data,
> rather than skewed scales anyway).

That's not what I'm talking about. I'm talking about doing any 
musically interesting processing at all, while playing the result 
with a non-ET scale. It has nothing to do with whether you're using 
one scale, or multiple scales.


> Any modules that want to do note based processing for ET scales can
> do it just by being told now many notes per octave there are (just
> like with note representations) and note based processing for
> non-ET scales is still hard, but probably not neccesary.

Well, that might be for *very* non-ET scales. However, I suspect that 
*subtly* different temperaments of the 12t scale used in some 
classical music are a lot more interesting to people in general - 
especially those who would ever think about using any event 
processors based on traditional theory.


> I'm not
> away of any non-ET scale where you could, eg. arpegiate without
> knowing a lot more than just the number of the note in the scale.

So, if you're playing a keyboard instrument, you have to play 
different notes if you're using a slightly different temperament than 
you would if the instrument was tuned to 12tET?

Well, the scales that were used for keyboard instruments before the 
ET scale became popular (thanks to Bach) aren't very useful if you 
get too far from the intended key signature. Is that reason enough to 
dismiss them as non-existent? Indeed, they're hardly ever used these 
days - but OTOH, nor is Mercator's 53tET scale, which is as close to 
perfect you get with ET, without using hundreds of tones/octave.

That said, it's easy enough to change the tuning of synths, even 
dynamically, so I'm still not convinced that working with an 
approximat 12t scale, and then converting to the right temperament, 
is useless and irrelevant.


//David Olofson - Programmer, Composer, Open Source Advocate

.- The Return of Audiality! --------------------------------.
| Free/Open Source Audio Engine for use in Games or Studio. |
| RT and off-line synth. Scripting. Sample accurate timing. |
`---------------------------> http://olofson.net/audiality -'
.- M A I A -------------------------------------------------.
|    The Multimedia Application Integration Architecture    |
`----------------------------> http://www.linuxdj.com/maia -'
   --- http://olofson.net --- http://www.reologica.se ---



More information about the Linux-audio-dev mailing list