From: Shawn Hargreaves <Shawn AT talula DOT demon DOT co DOT uk>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Exact timming II
Date: Mon, 17 Feb 1997 20:14:18 +0000
Organization: None
Distribution: world
Message-ID: <u15Y6JAawLCzEwTX@talula.demon.co.uk>
References: <Pine DOT SUN DOT 3 DOT 91 DOT 970216085938 DOT 14977A-100000 AT is>
 <Pine DOT LNX DOT 3 DOT 91 DOT 970217141522 DOT 32046C-100000 AT tabor DOT ta DOT jcu DOT cz>
NNTP-Posting-Host: talula.demon.co.uk
MIME-Version: 1.0
Lines: 22
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

Jan Hubicka writes:
>You should do following
>allegro_init
>horo works uclock well
>install_timer
>and here uclock crashes
>allegro_exit

You can't mix the timer routines from Allegro with calls to uclock():
they use the timer in totally different and incompatible ways...

It will work better if you use the non-dynamic version of the Allegro
timer code (this is new in v2.2: to select it either run your app under
win95 or set the global variable i_love_bill to TRUE). On my machine
that stops uclock() from crashing, but still screws up the measurements.
The BIOS tick value will be correct, but the low bits will be wrong, so
uclock() will only be accurate to an 18.2'th of a second.

/*
 *  Shawn Hargreaves - shawn AT talula DOT demon DOT co DOT uk - http://www.talula.demon.co.uk/
 *  Ghoti: 'gh' as in 'enough', 'o' as in 'women', and 'ti' as in 'nation'.
 */