Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!elroy.jpl.nasa.gov!sdd.hp.com!wuarchive!m.cs.uiuc.edu!ux1.cso.uiuc.edu!usenet
From: tmkk@uiuc.edu (K. Khan)
Newsgroups: comp.sys.ibm.pc.hardware
Subject: Re: Accurate timing on AT-computer
Message-ID: <1991May23.215001.13280@ux1.cso.uiuc.edu>
Date: 23 May 91 21:50:01 GMT
References: <1991May23.152030.7710@ugle.unit.no>
Sender: usenet@ux1.cso.uiuc.edu (News)
Organization: University of Illinois at Urbana
Lines: 19


In article <1991May23.152030.7710@ugle.unit.no> gunnarkn@itk.unit.no (Gunnar Knutson) writes:
>
>-- 
>	I need to time the interval between two interrupts with
>accuracy in the milli-seconds range. The maximum resolution on the AT
>real time clock is one second, so I can't use that. Does anyone out
>there have a better idea ?

Here's one: get a better set of documentation on the AT's real time
clock - the docs which told you its maximum resolution was only 1 second
are quite thoroughly wrong.

Byte magazine published code for a timer routine which would accurately
measure events of less than 1 millisecond (e.g. something like 53 uS, if
I recall correctly). Whatever the exact number, such event timing CAN be done
with the AT's 8253 programmable timer chip. A good reference is Robert
Jourdain's "Programmer's Problem Solver" book (note: he now has a new one
out, co-authored by Peter Norton, which may replace the earlier book).