Quicktime pitch resolution (was:: Names for MIDI units)

hi paul and Aaron,

i'm consolidating both of your posts about Quicktime,
and responding by quoting the links to the old tuning list
posts about it.

> From: "wallyesterpaulrus" <wallyesterpaulrus@yahoo.com>
> To: <tuning-math@yahoogroups.com>
> Sent: Wednesday, July 02, 2003 2:00 PM
> Subject: [tuning-math] Re: Names for MIDI units
>
>
> was the quicktime resolution verified to be 3072
> (=9-bit?) parts per octave? if so, it's a *far*
> better choice for emulating 5-limit ji harmonies --
> in fact, it lies on the "atomic" 5-limit uber-temperament
> line (and all the ets that do are multiples of 12,
> following from kirnberger's construction of 12-tone
> equal temperament by subtracting the schisma from
> the just fifth to approximate the ET fifth, the error
> being 1/12 of the "atomic comma"), as you can see on
> the deepest zoom level of the graph at:
>
> http://www.sonic-arts.org/dict/eqtemp.htm
>
> its worst error is about 0.016 cents in the 5-limit.

> From: <pitchcolor@aol.com>
> To: <tuning-math@yahoogroups.com>
> Sent: Thursday, July 03, 2003 7:07 AM
> Subject: [tuning-math] Re: Names for MIDI units
>
>
> > [me, monz]
> > so if Quicktime *does* use 9-bit resolution, then
> > i guess we have "nomu" too?
>
> I searched archives on this and didn't find anything.
> I assume it refers to apple's quicktime documentation.
> Does anyone have proof that quicktime uses 9 bits
> for pitch bend?
>
> Aaron

here is the whole history of our previous discussion
regarding Quicktime's pitch resolution:

/tuning/topicId_40904.html#40904
/tuning/topicId_40966.html#40966
/tuning/topicId_41283.html#41283
/tuning/topicId_41287.html#41287
/tuning/topicId_41287.html#41288
/tuning/topicId_41292.html#41292

be sure to read all of them, and in the correct order,
because Robert Walker contradicted himself several
times as he kept performing his experiments.

in fact, after reading them all again myself, i'm
still confused as to what the pitch resolution of
Quicktime actually is. it's either 256 (=2^8) units
per semitone (= 3072edo) or 512 (= 2^9) units per
semitone (= 6144edo).

the confusion arises because of Quicktime's use
of signed integers, so one bit represents the sign.

-monz

Hi Joe,

I read all the quicktime posts; thanks for listing them. I really think we should
ask Robert to clarify the QT resolution if possible. He states things in terms of
the semitone, which is fine, but first he says 600+, then 256, then 512, then in
an offhand way he says it's 256 again. I think we need something more
conclusive than this.

Aaron

--- In tuning-math@yahoogroups.com, "monz" <monz@a...> wrote:
> hi paul and Aaron,
>
>
> i'm consolidating both of your posts about Quicktime,
> and responding by quoting the links to the old tuning list
> posts about it.
>

> here is the whole history of our previous discussion
> regarding Quicktime's pitch resolution:
>
> /tuning/topicId_40904.html#40904
> /tuning/topicId_40966.html#40966
> /tuning/topicId_41283.html#41283
> /tuning/topicId_41287.html#41287
> /tuning/topicId_41287.html#41288
> /tuning/topicId_41292.html#41292
>
>
> be sure to read all of them, and in the correct order,
> because Robert Walker contradicted himself several
> times as he kept performing his experiments.