Thread: issue with math functions

i have the following

Code:

#include <stdio.h>
#include <stdlib.h>
#include <math.h>

double power( double, int );

int main()
{
    //declare variables
    double iSum = 0, iTotalNum = 0, iMod = 10, iTmp = 0;
    int iPower = 10;

    /* on the basis that 2^15 == (2^3)^5 ie 3 * 5 = 15  */

    //calculate 2^10
    iTotalNum = power( 2 , iPower );

    //calculate (2^10)^10 ie 2^100
    iTotalNum = power( iTotalNum, iPower );

    //calculate ((2^10)^10)^10 ie 2^1000
    iTotalNum = power( iTotalNum, iPower );

    iTmp = fmod( 10, 3);
   
    printf("%lf\n\n\n\n%e\n", iTmp, iTotalNum);

    return 0;
}

double power( double iBase, int iExponent )
{
    //break out clause
    if ( iExponent == 1 )
    {
        //end of line so return 1
        return iBase;
    }

    return iBase * power( iBase, --iExponent );
}

when i try and build/compile it i get |29|undefined reference to `fmod'|
||error: ld returned 1 exit status|
||=== Build failed: 2 error(s), 2 warning(s) (0 minute(s), 0 second(s)) ===|

this happens with pow and sqrt but with them if i write in the numbers manualy rather than using assigned variables they work ie

Code:

pow( x, y) //this fails
pow( 4, 2 )//this works

You most probably need to link against the math library - libm.

For GCC you need the '-lm' compile option.

The GCC compiler will optimize / inline some simple functions to improve performance rather than calling the library function, and pow(x,y) is one of them.

how do i do that in codeblocks

Originally Posted by cooper1200

how do i do that in codeblocks

The first google hit I got was:

gcc - C-Prog Code::Blocks -- math.h library not linking within IDE ... works fine in command line. Where is the file/folder to link in IDE? - Stack Overflow

thanks 2 am and im getting square eyes never mind thinking for myself

thanks that worked

ok here is the final program. code blocks now compiles it with out any errors or warnings.

Code:

/*
2^15 = 32768 and the sum of its digits is 3 + 2 + 7 + 6 + 8 = 26.
What is the sum of the digits of the number 2^1000?
*/

#include <stdio.h>
#include <stdlib.h>
#include <math.h>

double power( double, int );

int main()
{
    //declare variables
    double iTotalNum = 0, iMod = 10, iSum = 0;
    int iPower = 10;

    /* on the basis that 2^15 == (2^3)^5 ie 3 * 5 = 15  */

    //calculate 2^10
    iTotalNum = power( 2 , iPower );

    //calculate (2^10)^10 ie 2^100
    iTotalNum = power( iTotalNum, iPower );

    //calculate ((2^10)^10)^10 ie 2^1000
    iTotalNum = power( iTotalNum, iPower );

    while ( iTotalNum > 1 )
    {
        iSum += fmod( iTotalNum, iMod );
        iTotalNum /= iMod;
    }

    //iTmp = iTotalNum - iTmp;

    printf("%lf\n\n%lf\n", iSum, iTotalNum);

    return 0;
}

double power( double iBase, int iExponent )
{
    //break out clause
    if ( iExponent == 1 )
    {
        //end of line so return 1
        return iBase;
    }

    return iBase * power( iBase, --iExponent );
}

however i don't get the result i expected (surprise surprise). i get 1195.955762 i understand that doubles are large floats but soon as it has got rid of the e notation (looped 301 times) it starts adding decimals

sorry meant to add that 2^1000 is 1.07150860719e+301 at least thats the number i am working with

Originally Posted by cooper1200

sorry meant to add that 2^1000 is 1.07150860719e+301 at least thats the number i am working with

2^1000 will be about a 300 digit number, and is far more than any standard data type can hold.

You should investigate using the "double dabble" algorithm.

Double dabble - Wikipedia

Originally Posted by hamster_nz

2^1000 will be about a 300 digit number, and is far more than any standard data type can hold.

double can hold up to 2^1023.

I would try using pow() again instead of your power() function. It could have been the missing -lm that caused it not to work with variables for you (the compiler probably replaced pow(constant, constant) with a constant value without a call to pow())

Originally Posted by christop

I would try using pow() again instead of your power() function. It could have been the missing -lm that caused it not to work with variables for you (the compiler probably replaced pow(constant, constant) with a constant value without a call to pow())

that was the issue the missing -lm was causing all sorts of problems i think