% cat  rootsub.C

/* rootsub - Find the roots of a function using subroutines and functions */

/*  Written by: C. Severance  - Tue Dec  7 22:44:26 EST 1993 */

#include <stdio.h>
#include <iostream.h>

void calcroot(int &,float &,float, float, float) ;

main() {
  float root;
  int found;

/* Do the printout's differently showing iostream and stdio too */

  calcroot(found,root,-3.0,0.0,0.1);
  cout << "found = " << found << " root = " << root << "\n";

  calcroot(found,root,-3.0,0.0,0.01);
  cout << "found = " << found << " root = " << root << "\n";

  calcroot(found,root,0.0,5.0,0.1);
  printf("found = %d root=%f\n",found,root);

  calcroot(found,root,0.0,5.0,0.01);
  printf("found = %d root=%f\n",found,root);

  calcroot(found,root,5.0,10.0,0.01);
  printf("found = %d root=%f\n",found,root);

}

/* Routine calcroot */

void calcroot(int &good,float &rt,float start,float end,float delta)

{
  float tmp,x;
  float func(float);

  printf("Scanning from %f to %f by %f\n",start,end,delta);

  good = 0;
  rt = 0.0;

/* Check the input parameters */

  if ( start>end || delta <= 0.0 ) {
    printf("error in parameters - no root calculated\n");
    return;
  }

/* Calculate the function value at the start and then loop until */
/* The function changes sign.  This point is returned as the root. */

  tmp = func(start);

  for( x=start; x<=end; x+=delta) {
    if ( tmp*func(x) < 0 ) {
      rt = x;
      good = 1;
      return;
    }
  }

/* No root found */

  return;
}

/* The function to be evaluated */

float func(float val)

{
  float retval;

  retval = (val*val*val) - 0.15 *(val*val) - 2.8449 * val + 0.704975;

  return(retval);
}
% g++  rootsub.C
% a.out 
Scanning from -3.000000 to 0.000000 by 0.100000
found = 1 root = -1.7
Scanning from -3.000000 to 0.000000 by 0.010000
found = 1 root = -1.72
Scanning from 0.000000 to 5.000000 by 0.100000
found = 1 root=0.300000
Scanning from 0.000000 to 5.000000 by 0.010000
found = 1 root=0.250000
Scanning from 5.000000 to 10.000000 by 0.010000
found = 0 root=0.000000
%