Re: Is intrinsic MIN inefficient?
From: Jeff Krimmel (madscientistabc03_at_hotmail.com)
Date: 05/12/04
- Next message: Jeff Krimmel: "Re: Is intrinsic MIN inefficient?"
- Previous message: Gib Bogle: "Re: Is FORTRAN a dying language? (not a troll)"
- In reply to: robin: "Re: Is intrinsic MIN inefficient?"
- Next in thread: Clive Page: "Re: Is intrinsic MIN inefficient?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Tue, 11 May 2004 23:50:56 -0700
On Thu, 06 May 2004 15:02:05 +0000, robin wrote:
> Jeff Krimmel <madscientistabc03@hotmail.com> writes: > Greetings,
>>
>> I have a program that takes five days to run to completion, and I used
>> gprof to profile the execution on the most recent run. One subroutine of
>> the 28 takes 28.59% of the execution time, and this subroutine is
>> basically of the form:
>>
>> DO i=start,end
>> result(i)=MIN(bunch of algebraic computations, 1d6)
>> END DO
>
> I would investigate the "bunch of algebraic computations".
> Is there an unnecessary conversion or somesuch there?
No, not that I can tell, though I am a novice at much of this. It is a
series of summations of terms like
(array(a,b) - array(a,b-1))**2
except with "b" and "b-1" being slightly different in the other terms.
Around twelve of these terms are added, and MIN is called to determine if
this sum is less than 1d6.
Jeff
-- Add an underscore between 'd' and 's' and remove the first three letters of the alphabet for email.
- Next message: Jeff Krimmel: "Re: Is intrinsic MIN inefficient?"
- Previous message: Gib Bogle: "Re: Is FORTRAN a dying language? (not a troll)"
- In reply to: robin: "Re: Is intrinsic MIN inefficient?"
- Next in thread: Clive Page: "Re: Is intrinsic MIN inefficient?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|