Mail Archives: djgpp/1994/09/30/23:29:21

Date: Fri, 30 Sep 1994 14:15:16 -0400 (EDT)
From: Edwin Phillips <flaregun AT strauss DOT udel DOT edu>
To: "Victor K. Chu" <victor AT euler DOT Berkeley DOT EDU>
Cc: DJGPP AT sun DOT soe DOT clarkson DOT edu
Subject: Re: accuracy of clock()

On Thu, 29 Sep 1994, Victor K. Chu wrote:

> Dear DJGPP Users,
> It seems to me that the accuracy we can get from the time() 
> function is only 1/18.2 second (about 55 mili-seconds), even 
> though CLOCKS_PER_SEC is defined as 1000000 in time.h.
> I wonder if anyone knows that there is any way to get a more
> accurate time?
> Thanks,
> Victor Chu

	Check the archives of the list... there were two posts of a timer 
reading function.  The C version is called hrtimer.c, which is based on 
the original ASM routines called timer16() and timer32().  timer32() reads 
the time-clock chip at a resolution of roughly 1 microsecond.  I recently 
compiled the C version with v1.12m2 and they work.  I'll dig up the 
original post(s) if you'd like.


/* Ed Phillips  flaregun AT strauss DOT udel DOT edu      University of Delaware       */
/* Jr Systems Programmer  (302) 831-6082    IT/Network and Systems Services */

- Raw text -

  webmaster     delorie software   privacy  
  Copyright 2019   by DJ Delorie     Updated Jul 2019