Message-ID: <19990505115507.28110.qmail@www0d.netaddress.usa.net>
Date:  5 May 99 04:55:07 PDT
From: Oscar Almer <firstcheesemaster AT netscape DOT net>
To: djgpp AT delorie DOT com
Subject: Re: [optimizing a C program]
X-Mailer: USANET web-mailer (M3.0.0.105)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by delorie.com id IAA06667
Reply-To: djgpp AT delorie DOT com



The optimivation flags you are looking for are -On (where n is a integer
between 0 and 9), -expensive-optimizations and -fast-math(i think it is
called). Compile with these flags set and you should get around 50% more
performance. At least i did. 

>"Bonifati" <abonifati AT telsa DOT it> wrote:
>how can i optimize this little program which calculates N!  ?
>i don't now what optimization flags to set into RHIDE


>/* this program works; i have tested it */

>#define MAXPIECES 36068  /* 36068 are those necessary for 42950! */
>#define N 1000           /* from 0 a 42950 */

>/* a "long int" must be  32 bit (at least) */
>/* "short" must be at least 16 bit, or use int instead */

>unsigned long Fact[MAXPIECES], Product;
>unsigned short i, i2, Carry, Index, EndIndex;

>main()
>{
>   /* calculates N FACTORIAL */
>   Fact[0]=1;
>   Index=EndIndex=0;
>
>   for (i=N; i>1; i--)
>   {
>      Carry=0;
>      for (i2=Index; i2<=EndIndex; i2++)
>      {
>         Product=Fact[i2]*i + Carry;
>         Fact[i2]=Product%100000;
>         Carry=Product/100000;   //This could be written with binary
//shifting instead. See any tutorial for documentation.
>      }
>      /* manage last carry */
>      if (Carry)
>         Fact[++EndIndex]=Carry;
>      /* adjust Index to optimize */
>      while (Fact[Index]==0)
>         Index++;
>   }

>   /* print */
>   printf("%ld", Fact[EndIndex]);
>   for (i2=EndIndex; i2>0;)
>      printf("%05ld", Fact[--i2]);
>   putchar('\n');
>}

Personally, I woudn't do it like this. I would have used a recursive function
to  calculate n, and than make a function to call itself with n-1 as
parameter, like this:

int nfact(int n)
{
int n1,fact;
n1=n-1            
if ((n-1) == 0)
  return 1
fact=n*nfact(n1)  //recursive call to same function
  return fact
}

But i am not sure this (specified) function does work. it is only something i
got from the top of my mind. the tecnique should work, though.

Oscar Almer

____________________________________________________________________
Get your own FREE, personal Netscape WebMail account today at http://webmail.netscape.com.