No longer (Re: [rescue] Re: [geeks] THIS. MAKES. ME. SICK.)

[Joshua D. Boyd]

On Fri, 15 Jun 2001, Matthew SAMS wrote:

> Ok. I'll bite. I'm looking for
>   Mathematica and other simulation software for NeXT and SunOS

Hey.  I've found two nifty free packages.  The first is YACAS.  I used the
windows port of this package for a bunch of homework that I was supposed
to do in Mathematica ().  It is darn simple to use.  Of course, not nearly
as powerful.  I couldn't figure out how to row reduce a matrix in it.  It
did everything else imaginable to matrices, except row reduce them, which
was all I needed.  See http://www.xs4all.nl/~apinkus/yacas.html.  I mainly
used this program for Calc 3 stuff because of the lack of row reduction
capabilities.  

The other program to check out is called Maxima.  Awhile ago, so people
petitioned whoever it was that was incharge, and so the completely
government funded 1982 version of Macsyma (apparently later versions were
privately funded) was released public domain, and then extended into the
GPLed project Maxima.  Maxima is significantly more difficult to use.
According to some published tests, it is also significantly more
powerfull.  Maxima doesn't particularly look like Mathematica at all.
There don't seem to be many front ends for it, except there is one that is
extremely innovative.  It is called symmax.  Highly recommended.  It
presents a whole new way of looking at putting together complex
mathmatical models.  Maxima is written in CommonLisp, BTW, and Symmax is
written in Perl and uses some sorta perl-gtk bindings.

Anyway, I love mathematica, but can't afford it.  Yacas is almost as good,
and if anything easier.  Maxima is so much more powerfull, but much
harder.
 
> > Ouch.  My local school has one of those machines which would be cool to
> > revive (the machine has some cool imaging boards in it).  
> 
> Imaging boards? Do tell.

Some sorta VME board with video inputs and outputs.  It can do different
types of real time video processing in hardware like LUTs and other
transforms.  I don't think that it is really a video capture board, but I
think that it allows frame grabbing, and I've seen some printouts (made by
hooking a video printer to the video outputs of the board) that showed it
tracing a moving object, but maybe that is just a trick of the hardware
image processing.

Anyway, the most spectacular use of this board was to build a robot that
could putt.  There was an arm with a putter attached, and 1 (or 2, don't
remeber, think that board could take two inputs at once, either that or
machine had two boards installed) cameras attached, and the robot watch
the ball move and build a model of the terrain to use for judging the next
shot.

This of course is a field of work that hasn't been touched in something
like 5-6 years.  Current research is related to haptics, which isn't quite
as appealing to me.  Mainly because the way they are going about it is to
just guess, then get real surgeons to test to see if it feels right.
Doesn't feel as much like research as the previous projects did.

I'm trying to do backround on MRI and CT scans to see if I can use such
data alone to build a better haptics model.  What I need is to be put in
contact with an expert in medical imaging, but so far no one has returned
my emails.

For the haptics research, see http://cs.millersv.edu/haptics/

--
Joshua Boyd