[Date Index] [Thread Index] [Author Index]

Re: Intensive numerical calculations

• To: mathgroup at smc.vnet.net
• Subject: [mg80373] Re: Intensive numerical calculations
• From: Cristian <LumisrobTogliquesto at yahoo.com>
• Date: Mon, 20 Aug 2007 03:39:53 -0400 (EDT)
• References: <f9s445$all$1@smc.vnet.net>

On Tue, 14 Aug 2007 11:36:37 +0000 (UTC), Bill Rowe
<readnewsciv at sbcglobal.net> wrote:

>On 8/13/07 at 4:33 AM, LumisrobTogliquesto at yahoo.com (Cristian)
>wrote:
>
>>How much is it distant Mathematica from the speed of the c++
>>language compiled ?
>
>A well written program in a compiled language like C++ will
>always execute much faster than a similar program written in
>Mathematica code. But I strongly believe this type of comparison
>isn't useful. 

I also thought so but if I have to make intensive calculations I have
to turn to c++, matlink etc.............then boh....

>The execution speed advantage of C++ is more than
>offset by the development time for the program.

true on light calculations
not true for greater projects

>=46or any non-trivial program written in Mathematica code, you
>will be making use of Mathematica function that would have to be
>re-written in C++ code.And even if this is not the case, things
>that take several lines of C++ code tend to take much fewer
>lines of code in Mathematica. 

But if these objects I cannot use them to perhaps make intensive
calculations it was better that I immediately wrote them in c++ ..imho

>This is an inherent advantage of
>using a higher level functions available in Mathematica and not
>available in C++.

scientific interpreters exist  also for the c++ with numerical libraries....
then at the end.... boh

cristian