Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!sdd.hp.com!spool.mu.edu!news.cs.indiana.edu!noose.ecn.purdue.edu!dirac!gibbs.physics.purdue.edu!murphy
From: murphy@gibbs.physics.purdue.edu (William J. Murphy)
Newsgroups: comp.sys.amiga.advocacy
Subject: Re: AMIGA
Message-ID: <5054@dirac.physics.purdue.edu>
Date: 6 May 91 20:14:27 GMT
Article-I.D.: dirac.5054
References: <1991Jan10.194127.20625@rice.edu> <17564@cbmvax.commodore.com> <.v1G&v0*1@cs.psu.edu>
Sender: news@dirac.physics.purdue.edu
Distribution: usa
Organization: Purdue Univ. Physics Dept, W.Lafayette, IN
Lines: 29

>In article <17564@cbmvax.commodore.com> daveh@cbmvax.commodore.com (Dave Haynie) writes:
>
>   that would have made far more impact for the same price than the DSP.  
>   Especially a DSP that can't be used for mathematics; something most 
>   Workstations just about require these days is at least 1.5-2 MFLOPS or better.

While the Amiga with the GVP A3000/50 accelerator board may have placed the 
Amiga at the top end a few months ago, I certainly wouldn't buy and Amiga
if I wanted to do serious gronking of calculations.  The reason is I don't 
like the added worry of errant tasks writing into the low memory and crashing
the machine.  This is not to say that it can't happen on the NeXT, You can crash
the window server by trying to run bad PostScript code.  

There is of course another solution, that is to write out the intermediate
results of a simulation or job.  For most of the thousands of CPU hours we
have logged on our NeXTs, I don't think that it is worth the extra effort to
write intermediate results when a job lasts 0.3 to 3.0 hours.  When the jobs
get in to the range of days and weeks, then it is worth the trouble.

So what's the point.  When I developed software on my Amiga, I regularly
crashed because I am not a great C programmer.  I didn't appreciate rebooting
once for every compile and test cycle.  On the NeXT or any UN*X system with
memory protection (I suppose we have to include OS-2 and VMS) it takes something
more drastic than "int *data; data=0;" to crash the machine.

Bill Murphy
murphy@physics.purdue.edu
Anything above the line beneath the line below is false.
________________The Line Beneath________________________