On 2/14/06, Hudson Lacerda <[EMAIL PROTECTED]> wrote: [...] > Yep, I should be written that the *final* _quantised_ pitchbend is > 1/4096 semitone resolution. Thus one can use (1 to 8192) / (1 to 8192) > in 0-2 semitones range to access the full range step by step.
Um, don't you mean (1 to 8192) / 8192? If the denominator isn't a power of two it can't be represented exactly. [...] > Compatibility between applications. ABC standard should say the values > normally a program can accept. > > Is 'unsigned short int' your suggestion for ``NUM''? That sounds reasonable, but currently there are no restrictions on any of the number ranges. In the BNF specification (http://www.norbeck.nu/abc/bnf/abc20bnf.htm) they're all just "1*DIGIT", which means 1 or more digits. [...] > What about score generators like abcm2ps? The number of accidental > glyphs is limited, should it quantise also the fractions? Currently, > abcm2ps accepts (1 to 256) / (1 to 256) because it identifies the used > accidentals with a single unsigned 16 bit integer. Of course, if we > define that numerator and denominator are both 16 (or 32) bits length, > that code shall be changed accordingly. What does it currently do with weird note lengths? > Hudson Keenan Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/abcusers/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/