Xref: utzoo comp.sys.ibm.pc:17866 comp.periphs:1114
Path: utzoo!attcan!uunet!husc6!bbn!uwmcsd1!ig!agate!pasteur!ames!amdcad!weitek!dms!albaugh
From: albaugh@dms.UUCP (Mike Albaugh)
Newsgroups: comp.sys.ibm.pc,comp.periphs
Subject: Re: Using a Joystick instead of a Mouse (Kinda long)
Message-ID: <525@dms.UUCP>
Date: 8 Aug 88 17:00:10 GMT
References: <618@starfish.Convergent.COM>
Organization: Atari Games Inc., Milpitas, CA
Lines: 55

I do not read comp.sys.ibm.pc, but have cross-posted because of point
1 below. Please don't assume that if two plugs are physically the same
you can safely plug them together!

Mild (?) flaming and some info here:

From article <618@starfish.Convergent.COM>, by cdold@starfish.Convergent.COM (Clarence Dold):
>>   Have any of you plugged an Atari joystick into the serial port and used it
>>   instead of a mouse?  Doing so would make playing some games easier.

1) Has anyone considered that there is more to an interface spec than the
number of pins and the shape of a connector. The Atari game connectors,
used on 2600,7800,400/800... and also several other computers from various
manufacturers, were intened to hook up to "controllers" that had two analog
(resistance) inputs and 5 (? it's been a while) switch-closure inputs, plus
power and ground. One of the switch inputs was connected in such a way as
to be usable as a light-pen on the 400/800... Anyway, this scheme has NOTHING
but the shape/number of pins in common with a DB9 serial port. Several years
back Lafayette Radio sold intercoms with connectors that had two flat blades
the right distance apart to plug into power-strips etc. I suppose they
thought it would make cabling easier, but too many folks "accidentally"
plugged them into the wall (REAL flames!)

> 
> Assuming we're talking about an 'Atari 400/800/1200' joystick:
> The joystick only has switches, not pots.  The return value will be a bit or

2) As mentioned above, some "controllers" used the pot inputs. I agree that
the common four-position joystick was switches only. Incidentally, these
were _real_ common. "a major manufacturer of home computers" went so far
as to clone the stick, removing only the Atari logo when copying the mold.
Forgot to remove the Atari part number... hello lawyers :-)

> With the polling rate on the Atari, that was okay, but on a PC or AT, I would 
> expect to reach the limits of the screen before I could let go of the stick.

3) Some time ago I did a study of the memory systems of the Atari 400/800,
the Apple ][, and the (original) PC. Is anyone supprised to hear that the
PC is only 5% faster than an 800, and 25% faster than an Apple ][. Please
direct flames about instruction sets and memory sizes elsewhere. I was
comparing _memory_ systems. Still, as a first order guess, the data
processing rates (without cache) will be pretty similar. Anyway, many
games sampled the switches several times a frame. If you were using pots,
the hardware allowed you to sample the pots _every_scan_line_ (although
only to six bit accuracy). I find it real hard to believe that a serial
controller on a PC would do as well. In fact, having used a serial mouse
on a PC, I find it _impossible_ to believe. Often software uses really
big steps per sample to compensate for low sampling rate, so I may
just agree that the PC would "peg" when using a switch-type joystick
(but not for the reason you say :-)

| Mike Albaugh ({decwrl!turtlevax!}weitek!dms!albaugh) voice: (408)434-1709
| Atari Games Corp (Arcade Games, no relation to the makers of the ST)
| 675 Sycamore Dr. Milpitas, CA 95035
| The opinions expressed are my own (My lawyer isn't listening)