Re: omega race drawings

From: Clay Cowgill <clay_at_supra.com>
Date: Tue Aug 05 1997 - 13:54:27 EDT

>actually, what I was thinking of was a zero center sort of thing, with
>a little dead-band near the center,

Right, I was thinking of treating it like a "broken" analog joystick pot
that doesn't auto-center. (Turning a little to the left outputs some pulse
train on the "left" button line, turning more to the left pulls the line
low constantly-- remember you can't "turn" faster than holding the button
down all the time.)

Zonn's right about losing the wheel position though. How about just going
on the direction of the knob, but with some hysteresis? Something like:

If the knob ever moves 6 "ticks" left in a row, "push" left button and latch
If the knob ever moves 6 "ticks" right in a row, "push" right button and latch
If the knob ever moves less than 6 but more than 3 "ticks" in the opposite
direction (from the current "latched" direction), unlatch current
direction.

So, if you want to turn right, turn the knob right. You can spin it right
all day long and it doesn't matter. You can give it 1/8th turn right and
leave it and it still turns right. If you turn it a "little" left it still
turns right, if you turn is a "little more" left it stops turning, if you
turn it "yet a little more" left it starts turning left.

Does that make sense? Basically try to tune it so that the change of
direction decides the right/center/left choice, but build in some "jitter"
factor so a little twitch the wrong way doesn't turn the ship accidentally,
but a little turn the other direction *does* turn the ship...

-Clay

Clayton N. Cowgill Engineering Manager
_______________________________________________________________________
/\ Diamond Multimedia Systems, Inc. clay@supra.com
\/ Communications Division http://www.supra.com/
Received on Tue Aug 5 09:48:21 1997

This archive was generated by hypermail 2.1.8 : Thu Jul 31 2003 - 23:00:53 EDT