## Re:Newbie: Using Delphi 2.0 for Math

In article <32c8520b.10580...@netnews.worldnet.att.net>,

jchenwo...@postoffice.worldnet.att.net (Jeff Chenoweth) wrote:

##### Quote

> On Sun, 22 Dec 1996 23:14:09 -0500, "Pavel E. Guarisma"

> <pegua...@eos.ncsu.edu> wrote:

> >I've been thinking of "upgrading" to Delphi 2.0 because, as I understand

> >it, it is an Object Oriented Pascal Compiler plus a GUI development tool

> >(true?). What I'm working on right now is my Operations Research

> >Master's thesis and I would like to make it a full fledged application

> >with a snazzy GUI. My thesis is very mathematical in nature and I

> >require good precision and double-real variables (which I know Delphi

> >2.0 has). Most operations consist of matrix operations (matrices should

> >not exceed 20x20) and solving systems of differential equations (using

> >Runge-Kutta or some Predictor-Corrector method).

> >So, what do you think? Am I correct in my Delphi 2.0 facts? Can the

> >desktop version handle this kind of project or do you think maybe it's

> >overkill? Anyone out there have any experience with Delphi in a math

> >setting?

I missed the original post here. My advice there would be DO get

Delphi, but get the Developer version, not the Desktop version.

Mainly because there's a math.pas included in the Developer version.

Not that there's anything in the math unit that's going to help with

your matrices and differential equations, it just seems silly to get

the Desktop version under the circumstances (especially since you'll

qualify for the "academic" version, which is pretty close to free.)

##### Quote

> I use Delphi 1 and 2 to provice engineering programs to our structural

> group. In 1.0 I had to find a set of math functions to handle mundane

> tasks like raising a number to an exponent, etc.etc. All thats

> included in version 2.o (I have the developer version).

> However...I should warn you that there are NO built in functions or

> procedures for matrix manipulation. I wrote all mine to do solution

> of simultaneous equations . I too came from a Fortran background. My

> biggest thing was going from row -> column orientation to column->row

> for all my old subroutines I had to convert. Also, working with 2-d

> arrays of dynamic scope is a {*word*81}! I ended up working with some

> monster arrays and using some tricks with Pointers so I could exceed

> the 64K limit!

Have you upgraded to D2 yet? Things like this are suddenly

much easier with the flat memory model. (Hmm, you still need to

address the entries yourself if it's a dynamic 2-d array, but the

64K nastiness is gone.)

David Ullrich

-------------------==== Posted via Deja News ====-----------------------

http://www.dejanews.com/ Search, Read, Post to Usenet