Path: utzoo!attcan!uunet!cbmvax!grr
From: grr@cbmvax.UUCP (George Robbins)
Newsgroups: comp.sys.amiga.tech
Subject: Re: Analog Joysticks
Keywords: Analog, Joystick, Full-range
Message-ID: <5204@cbmvax.UUCP>
Date: 9 Nov 88 07:13:40 GMT
References: <12539@steinmetz.ge.com> <8@sdcc10.ucsd.EDU>
Reply-To: grr@cbmvax.UUCP (George Robbins)
Distribution: na
Organization: Commodore Technology, West Chester, PA
Lines: 47

There seems to be a lot of excitement about joystick these days...

Here are some notions you might want to think about or comment on:

1) The pinout for the analog joysticks is supposed to compatible with
   traditional game machine game paddles / analog joysticks - you
   shouldn't need any sort of adapter for these.

2) Most of these things seem to come in around 470K - give or take
   a factor of two.  You can diddle the pot values or add some more
   timing capacitance, but there are some trade-offs.  Increasing
   resistance increase noise level, increasing capacitance may
   decrease repetability.

3) Noise level varies from machine to machine.  The A500 is not so
   good, others somewhat better.  Much depends on where the timing
   capacitor are located.  If they are near the Paula chip, they
   tend to filter out noise spikes, if far away induced noise tends
   to randomly trip the comparators.  I don't know the effect of
   locating them in the joystick assembly itself.

4) The best way to evaluate raw performance is to write a little
   program that does a simple point plot display of the x-y location
   read from the pot registers.  Idealy, you have an etch-a-sketch,
   in reality you get some kind of fuzz-ball, relativly small in
   the low resistance quadrant, larger at the high resistance
   extreme.

5) Software using the pot input should be pretty defensive.  Toss
   bogus values, do some kind of averaging to reduce the effect of
   noise.  Provide some kind of training/calibration mode so that
   the program can adapter to whatever the person has plugged in.

6) Consider the application - there just isn't enough accuracy for
   your paint/precise point location thing, but that doesn't mean
   it won't do for a continuous control of some sort or even a
   selection function if the gadgets are big enough.

7) Somone mentioned external hardware to get more accuracy repeatability -
   it's been done before, though I don't know what the results were.  I'm
   not sure how much that will really help...  You can probably do about
   as well by moving the timing caps and/or messing with RC values...

-- 
George Robbins - now working for,	uucp: {uunet|pyramid|rutgers}!cbmvax!grr
but no way officially representing	arpa: cbmvax!grr@uunet.uu.net
Commodore, Engineering Department	fone: 215-431-9255 (only by moonlite)