[linux-audio-dev] XAP: a polemic

David Olofson david at olofson.net
Mon Dec 16 22:57:00 UTC 2002


On Tuesday 17 December 2002 04.25, Tim Goetze wrote:
> Paul Davis wrote:
> >play this kind of music as part of their own culture. i've also
> > heard and watched both zakir hussain and v. vikaryam count talas,
> > and believe me, they don't do what you propose a musician would
> > do.
>
> i believe they do more or less the same. you only hear the
> beats they emphasize, and that's why they call it three and
> a half, not seven. it's a little like in western music, you
> say it's 3/4 but you might as well say 6/8 -- they only
> differ in emphasis, or groove.

Try that with 4:5, 5:6, etc, and you'll see why it doesn't work. At 
some point, you must start thinking in some other way.


> >hmm. i'm not sure of this. i'm not entirely clear what you mean by
> >these nomenclatures. do you mean "1 beat per measure,
> > beat-note-value = a third note" and "1 beat per measure,
> > beat-note-value = a quarter note"? or something else.
>
> yes, 1/3 and 1/4 time but it's just one of many examples. if
> you don't implement running tick/beat counters in integer,
> there are a lot of pitfalls you have to be very careful to
> avoid if precision is an objective.

First of all, fp *is* equivalent to integers for all practical 
matters, as long as you don't do anything that results in values that 
don't fit in the mantissa. Adds and multiplications with integer 
values are perfectly safe, as long as the result fits.

That said, precision *is* an objective - but having 32+ actual bits 
of accuracy per sample is not. Again, as long as you lock to the 
timeline, there's no real problem.


//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 -'
   --- http://olofson.net --- http://www.reologica.se ---



More information about the Linux-audio-dev mailing list