Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uwm.edu!linac!att!news.cs.indiana.edu!houghton@iuvax.cs.indiana.edu
From: houghton@iuvax.cs.indiana.edu (Ric Houghton)
Newsgroups: comp.sys.next
Subject: real-time applications on 040 cube
Message-ID: <1991Mar27.082322.27952@news.cs.indiana.edu>
Date: 27 Mar 91 08:23:05 GMT
Sender: daemon@news.cs.indiana.edu
Organization: Computer Science, Indiana University
Lines: 26


  We are thinking of purchasing a color NeXT station for research
purposes (040 cube, with 40 megs of ram, and a 600 meg harddisk).  The
only problem is that the required task is timing sensitive.  That is,
we need to be able to reliably present a visual stimuli at some time
x, after another stimuli, auditory (or maybe another visual stimuli)
has been started.  (Where x is measured in milleseconds.)


  Does anyone have a feel for how accurate or reliably we might be
able to do this task?  +/- 1 millesecond is acceptable, but more than
that, and we have to start thinking about PC's or Macs. (YUCK!)


  Intuitively, it seems almost impossible to assure the experimenter
that the task will be carried out reliably, but I've not worked with
MACH, so I have no clue on this one.

  Thanks for what ever input, 
        I'll summarize, of course....
  
    Ric Houghton
    houghton@cs.indiana.edu

  btw, Any chance there exists a real-time unix for the NeXT?  
       Probably not?.....