Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!zaphod.mps.ohio-state.edu!van-bc!ubc-cs!alberta!ccu!umhild11
From: umhild11@ccu.umanitoba.ca (Jeff Hildebrand)
Newsgroups: comp.sys.atari.8bit
Subject: Re: mouse
Message-ID: <1990May30.024523.4608@ccu.umanitoba.ca>
Date: 30 May 90 02:45:23 GMT
References: <1369@lectroid.sw.stratus.com> <1990May22.161958.11370@watdragon.waterloo.edu> <30251@cup.portal.com>
Distribution: na
Organization: University of Manitoba, Winnipeg, Canada
Lines: 37

In article <30251@cup.portal.com> Chris_F_Chiesa@cup.portal.com writes:
>In a recent article, rbharding@trillium.uwaterloo.edu (Ron Harding) writes:
> 
>>     As a matter of fact, the Commodore Amiga mouse will plug into the
>> Atari (or C64) joystick ports.  I wrote a program a while back to read it.
>> Unfortunately, the program uses a tight polling loop.  I haven't come up with
>> a hardware-free way to put a mouse driver in the background on a VBI or
>> anything nice like that.
>
>Now THAT's INTERESTING!  I'm curious as to the nature of the difficulty
>you're having putting the mouse driver in a VBI.  If you're able to read it 
>at all, I'd think putting it into a VBI would be only a side issue.  If 
>you simply require help setting this up, perhaps I could help -- I KNOW I'd
>be INTERESTED in the project.  On the other hand, if there's something about
>the hardware (as you seem to be saying), tell us what it is and perhaps 
>someone here can brainstorm a solution for you!
>
Actually, I wrote a driver for the Atari Trakball once. The routine would work
perfectly if it was a polling loop (check the port, move, check the port, etc),
but if I put it in a VBI, it just didn't check often enough to get the changes.

Hmm, from what I read in another message, the Trakball may work somewhat
differently from the mice. I didn't get any docs (to speak of) with my Trakball,
so I had to figure out what the values meant on my own. As near as I could
determine (and this method worked great) one bit gave direction (up/down, or
left/right) and another gave its frequency. Every time the frequency bit
changed, I knew that I had to move the marker in the direction specified by the
direction bit. (that may not be very clear, there are two pairs, one for
vertical movement, and another for horizontal.) From what I understood of the
workings of the mice (mouses? what DO they call them in the computer world?)
this is a slightly different method. Perhaps periodic polling could work with
the Amiga mouse.

For anyone who is interested, I tried my driver with an ST mouse, and it didn't
work worth a damn, so either the pins or the philosophy is different.

Jeff Hildebrand