Path: utzoo!attcan!uunet!cs.utexas.edu!samsung!usc!apple!netcom!ergo
From: ergo@netcom.UUCP (Isaac Rabinovitch)
Newsgroups: sci.electronics
Subject: Re: Is PC parallel port latched?
Message-ID: <12169@netcom.UUCP>
Date: 15 Jul 90 16:54:46 GMT
References: <33965@ut-emx.UUCP> <1990Jul15.014354.12657@ux1.cso.uiuc.edu> <15568@ucsd.Edu>
Organization: UESPA
Lines: 34

In <15568@ucsd.Edu> brian@ucsd.Edu (Brian Kantor) writes:

>The IBM technical reference manual shows an LS374 in there, which is an
>8-bit latch.  The data lines are latched, but there's an LS244 reading
>them back in when you read from the port so that you can perform a
>self-test by reading back the byte you had just output.

>Thus you can make the port bi-directional if you want.  Just chop the
>wire from the latch's output enable pin, and run it out to a spare pin
>on the connector.  The external device would control the direction.

>There are other strobe and control lines, some in each direction.  Best
>you check a schematic if you're planning to hack it.

This subject has come up in the computing conferences, and one little
detail came up that's worth mentioning here.  Many "IBM-compatible"
machines differ from the PC in various minor ways (even IBM's own
laptop machine was notorious for this) and the printer port is an
obvious place to cut corners -- most printers don't really *need* a
bi-directional channel, do they?  The upshot is that one shouldn't
assume that a given clone's parallel port's bidirectional, and (say
some pundits) you can even damage some machines by trying to use them
that way.

I should emphasize that I'm not qualified to defend or even amplify
the above.
-- 

ergo@netcom.uucp			Isaac Rabinovitch
atina!pyramid!apple!netcom!ergo		Silicon Valley, CA
uunet!mimsy!ames!claris!netcom!ergo

	"I hate quotations.  Tell me what you know!"
			-- Ralph Waldo Emerson