[sdiy] "Synth design questions" or "Learning from Dave Smith"

[Roy J. Tellason]

On Thursday 18 October 2007 09:12, Tom Wiltshire wrote:
> I did some experiments trying to control a parameter value with a
> rotary encoder using the rotation speed to control the increment
> size. This sounds plausible at first: you turn quicker, it goes up
> more, you turn slower, it slows down, you turn really slowly, it
> steps up one at a time. I tried both linear and various non-linear
> relationships between rotation speed and increment size.
> Unfortunately I found that I couldn't control it at all. A single
> twisting movement starts off from rest (e.g. really really slow) and
> then speeds up to some maximum value, before slowing down to rest
> again. This makes it hard to get a 'feel' for how much or how fast
> you need to turn for a given result. When I changed it to two or
> three bands (e.g. single-step, fast, faster or just single-step and
> fast) I did better, since I learned to turn at the right speed to get
> the effect I wanted. Still, I wasn't everso impressed. A simple pot
> is much easier to get to where you want it accurately. Perhaps with
> cleverer software than I managed to write rotary encoders might be as
> good, but I couldn't find a way.

Perhaps it would be useful to examine the code used to handle mouse movement?  
I know that even back in the DOS days there were parameters you could feed 
the driver to control "acceleration" as well as other parameters.  I 
currently only run linux here,  and there seems to be a definite correlation 
between how far it goes and how fast I'm moving it.

Just a thought...

-- 
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space,  a critter that can
be killed but can't be tamed.  --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James 
M Dakin