http://ourworld.compuserve.com/homepages/taftan
Comments are appreciated.
Taftan Data
mailto:Taf...@compuserve.com
There is no FORTRAN, no C, no C++, no Pascal, even no Matlab. If
you want to do heavy number crunching, use Excel!!! Do engineering, be
consultant in M$ way.:-)) Is this the future? Or the past? Or what?
I saw a lot of strange commercials in this group, but this one is
greatest in the past 1 year.
Good luck!
--Julian Stoev
You do need waking up. The world is passing you by. Many people (many ! ! !
!) are using Excel as a front-end for serious calculation; the calculation
engine can be a dll (dynamic link library) written in (yes, really) Fortran
(even Fortran 90 !), C++, Pascal, etc. etc.
I have a transient heterogeneous (vapor, solid at different temperatures)
reactor model that was written in Fortran, compiled as a DLL, and I use Excel
as a quick front-end for this model. Before you jump to any conclusions, I
also use a Silicon Graphics Indigo 2 High Impact with R10000, for CFD
calculations that run for several days on end, and use AVS for
post-processing.
So, you are incorrect in assuming that anything related to Excel is mickey
mouse, reconsider:
what is important is the end result, not the tools that are
used to get there.
Why write Fortran over & over, use hokey interfaces etc., when Excel
interface is easy to use, minimizes programming of result archiving, etc.,
when an Intel machine can do the the job just as well ? Lastly, leading CFD
vendors are starting to sell serious ($25 000 per year) CFD analysis software
written for (yes, it's true) Intel with Windows front end !
Windows has won the battle for the desktop - it's over. Now NT is taking
manufacturing on in leaps and bounds (check out platoforms that Honeywell is
starting to support for a fact-check). Take a walk through refineries and see
what the engineers & operators are using. Windows & Intel are being heavily
used in offline applications.
Paul
Paul wrote in message <353ADB71...@mediaone.net>...
>Julian,
>
>You do need waking up. The world is passing you by. Many people (many ! ! !
>!) are using Excel as a front-end for serious calculation; the calculation
>engine can be a dll (dynamic link library) written in (yes, really)
Fortran
>(even Fortran 90 !), C++, Pascal, etc. etc.
>
>I have a transient heterogeneous (vapour, solid at different temperatures)
On Mon, 27 Apr 1998 00:51:45 +0100, "Francis Lovering" <f...@mcmail.com>
wrote:
John Hughes
yaya...@deltanet.com
>I am considering dropping MathCad and concentrating my efforts
>on Excel. Any comments regarding MathCad?
For me they don't really fill the same niche, and in fact can be used
complimentarily. I did a process simulation using Lotus and MathCad
together where Lotus ran the material and energy balance and MathCad
did the linear algebra for the reaction matrix. I couldn't have seen
doing the whole thing with either alone. Plus I find spreadsheets
less than optimal where you have to be able to look back at the
mathemathics after the fact and do revisions - it's just too hard to
piece the total calculation together after the fact, whereas this is
trivially easy with something like MathCad.
regards,
j. getsoian
Eric Lucas
J Hughes wrote:
>
> Pete Psingpy wrote:
> > Based on the comments published on this topic, I am considering dropping MathCad and concentrating my efforts
> on Excel. Any comments regarding MathCad?
>
It's easy to set up and now with simulink, it's just a matter of joining the
boxes to for a control strucutre and model. Very easy to use.
Joe
In article <354731...@deltanet.com>,
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
John Popelish
-Peter
On Wed, 29 Apr 98 15:09:38 +0400, "John Getsoian"
<jget...@nojunk.csi.com> wrote:
>On Wed, 29 Apr 1998 06:55:44 -0700, J Hughes wrote:
>
>>I am considering dropping MathCad and concentrating my efforts
>>on Excel. Any comments regarding MathCad?
>
I haven't used Excel but I have used other spreadsheets and Fortran (I
to am that old) but I have found that APL2 gets me from problem to
solution fater than any calculation system (including DERIVE - a
competitor to MATHCAD). Anything that involves arrays is really easy in
APL (matrix inversion requires one character). Just to mention a few, I
have written functions for statically indeterminate beams, columns,
graph plotting, ...
Ted
I guess for my $0.01 on this subject...
As engineers I feel that we can and should treat our software as
just another set of tools we need to perform our job - rather than
a single solution for all of our computational difficulties.
I feel fortunate that I've had an opportunity to use several approaches,
such as C, FORTRAN, Speadsheets, Mathcad, Matlab... I am far from being
an "expert" in any of these but feel that it has been relatively easy
to learn what their relative strengths and weaknesses are - allowing
me to pick up the right "tool" out of the box when things get rough.
That said, my personal biases are:
Spreadsheets: Great for organizing knowledge and handling bulk data
Mathcad: I really like the interface, but wish it had the mathematical
power of Matlab. I have a hard time performing operations on
huge data sets in Mathcad ... but haven't tried the most recent
release yet.
Matlab: Painful to learn but very powerful. Some strange, frustrating
bugs.
In short, what I suggest is that you become at least somewhat familiar
with every tool you can get your hands on. In the short run, the extra
knowledge may allow you to be more effective. In the long run, it may
help you out if you switch jobs and land in, say, a "Matlab-only" or
"Mathematica-only" company...
Good luck!
Mike
--
"Si hoc legere scis nimium eruditionis habes."
(If you can read this, you're overeducated.)
regards
Francisco
If I remember correctly (someone pls correct me if I'm wrong), it
started out life in the 1980's as a CPM O/S package. It was
reincarnated somewhere along the way in the early 1990's, when it was
bought by a different company.
Although it has its adherents, I believe that this group is relatively
smal. TKSolver is not considered to be "mainstream" - i.e. in the same
class as Mathematica, Maple, Mathcad; and Matlab.
-- W. R. Smith, PhD, P. Eng., Senior Scientist, Mathtrek Systems --
EMail(replace "_at_" by "@", "_dot_" by "."):
support_at_mathtrek_dot_com
--------------------- http://www.mathtrek.com ---------------------
-Mathtrek Systems - Home of EQS4WIN Chemical Equilibrium Software -
If I remember correctly (someone pls correct me if I'm wrong), it
started out life in the 1980's as a CPM O/S package. It was
reincarnated somewhere along the way in the early 1990's, when it was
bought by a different company.
Although it has its adherents, I believe that this group is relatively
small. TKSolver is not considered to be "mainstream" - i.e. in the same
J Hughes wrote:
> Pete Psingpy wrote:
> > Based on the comments published on this topic, I am considering dropping MathCad and concentrating my efforts
> on Excel. Any comments regarding MathCad?
>
> John Hughes
> yaya...@deltanet.com
>
> > Indeed I agree as well. We use Excel for quite a lot including dynamic
> > modeling. It's quite incredible what it can do, yet it is also very
> > easy to use. Not exactly the same issue, but I bet you'll find that a
> > large majority of chem engineer's are using as their primary calc
> > software.
> > -pete
> >
> > On Mon, 27 Apr 1998 00:51:45 +0100, "Francis Lovering" <f...@mcmail.com>
> > wrote:
> >
> > >I think that I agree with the message below.
> > >I can do in minutes in Excel what took an age in Fortran (yes I am that old)
I won't debate that many people still use Quattro Pro, but in my opinion
much more can be accomplished using Excel.
Rick Palmer
charlie sloan wrote in message <354DD475...@sam.neosoft.com>...