Path: utzoo!censor!geac!lethe!yunexus!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!cs.utexas.edu!usc!samsung!uunet!mcsun!ukc!cam-cl!news
From: bdb@cl.cam.ac.uk (Brian Brunswick)
Newsgroups: comp.sys.acorn
Subject: Re: Archimedes keyboard
Message-ID: <1991Feb25.161054.6477@cl.cam.ac.uk>
Date: 25 Feb 91 16:10:54 GMT
Article-I.D.: cl.1991Feb25.161054.6477
References: <2313@cybaswan.UUCP> <1991Feb21.143445.22959@fwi.uva.nl> <1991Feb22.154034.7845@vax1.tcd.ie> <$0`&80+@warwick.ac.uk> <1991Feb23.211436.7855@vax1.tcd.ie>
Reply-To: bdb@cl.cam.ac.uk (Brian Brunswick)
Organization: U of Cambridge Comp Lab, UK
Lines: 40

[Many people muttering about mice, using other mice]

The problem is that most PC mice are serial ones - they actually send
coordinates back to the system via an RS232 interface!

If you do get a straight wired mouse, its fairly easy to hook up though.
I've interchanged BBC, Amiga, Achimedes mice with no problem, just fixing
up a weird cable.
X or Y direction can be reversed by swapping around the two signals
that give the quadrature... ok, explanation coming up.

Mice tend to work by having an X-roller, say, with a slotted disk
on it and TWO detectors for the slots. These are arranged so that they
produce square waves a quarter cycle out of step:

--+     +-----+     +-----+     +-----
  |     |     |     |     |     |
  +-----+     +-----+     +-----+

-----+     +-----+     +-----+
     |     |     |     |     |
     +-----+     +-----+     +-----

Thus. (plotting signal against X position of mouse)

All that is necessary then is to wait for say a rising edge of one
signal, and look at the level of the other signal to see which way the mouse
is going. (Look at pic from left to right, and from right to left )
For 200 dpi resolution, look at rising and falling edges and eor them.

The thing that most irritates me is that the design of having a ball and
internal rollers is completely stupid.

I saw a completely WONDERFUL DESIGN on a mouse on a Decstation 3100...
It had two entirely separate rollers underneath, each angled with its
axis just off vertical, arranged so that one had an edge touching the
desk up/down and one across. Bingo! No gunge, Only ONE place to slip
instead of two..  Pity the mouse itself is circular... (Yuch)

Brian.Brunswick@uk.ac.cam.cl  Disclaimer.  Short sig rules!