[A6] 1 ms increments on Env's

[Leary, John]

> Some quick math: A6 parameters are 14-bit.  2 raised to 14, then times 2ms
> resolution returns 32.768 seconds.  2ms to 32sec is the full range of the
EGs.
>
> To get 1ms resolution while retaining the 14-bit width, you'd chop the
range
> in half so the max EG time is 16sec.  I'd prefer the current
implementation.
>
> Once you get above 200ms, a 1ms resolution would be insignificant anyway.

The ideal would be logarithmic resolution -- very fine resolution for short
durations, increasingly coarse for long one.  For instance, say the shortest
duration was .5 milliseconds, and the second notch was 1% longer, the third
one 1% longer than the second, etc.  With the same 14 bit parameter range,
you'd cover the time range from .5 ms to... 10^60 years.

0.5 milliseconds x (1.01 ^ (2^14)) = 3.165e+70 milliseconds, which is equal
to roughly 10^60 years.

Of course, testing to see whether the Andro tracked over its whole range
would be prohibitively difficult.  Really, 11 bits would be more than
enough!

Just some thoughts,
John