Path: utzoo!attcan!uunet!world!jamesp
From: jamesp@world.std.com (james M peterson)
Newsgroups: comp.lang.pascal
Subject: Re: Concurrent processing in DOS??
Keywords: concurrent processes, parallel, data acquisition
Message-ID: <1990Oct3.122141.28245@world.std.com>
Date: 3 Oct 90 12:21:41 GMT
Expires: 10/5/90
References: <5184@mit-caf.MIT.EDU>
Sender: jamesp@world.std.com
Distribution: na
Organization: The World
Lines: 25

In article <5184@mit-caf.MIT.EDU> gavin@mit-caf.UUCP (Gavin C. H. Zau) writes:
>
>	I am using a PC XT to control and acquire data from my 
>experimental aparatus.  The acquisition program is written in Turbo 
>Pascal 4.0.  I need to collect data that is 
>	(A) continuous over the time of the experiment. 
>	(B) data that is collected over a short period. 
>Can anyone please advice me on how to achieve the background data
>acquisition needed when I am collecting (B) type data, so that I would not 
>lose the (A) type data.  So kind of multiasking or concurrent processing
>would be great.  I would also like to run mutliple (B) type acquisition
>at the same time, which is not possible under our current data acquisition
>scheme.
>****************************************************************
>Gavin Zau	Dept of Chemical Engineering, MIT, Cambridge, MA
>		gavin@caf.mit.edu	
>****************************************************************

You mean something like a circular queue, one entry for each process?
Can you read your instruments and get a code back saying no data is
ready to be sent (so you don't read garbage).  If so, set up a scheduler
that has times you need and away you go..

Jamesp@world.std.com Reality is not an industry standard
                     I can't spell, but APL never cared.