compute sine using very simple floating arithmetic : Sin ASin « Math « C / ANSI-C

Home
C / ANSI-C
1.assert.h
2.Console
3.ctype.h
4.Data Structure Algorithm
5.Data Type
6.Development
7.File
8.Function
9.Language Basics
10.Macro Preprocessor
11.Math
12.math.h
13.Memory
14.Pointer
15.setjmp.h
16.signal.h
17.Small Application
18.stdio.h
19.stdlib.h
20.String
21.string.h
22.Structure
23.time.h
24.wctype.h
C Tutorial
C++
C++ Tutorial
Visual C++ .NET
C / ANSI-C » Math » Sin ASinScreenshots 
compute sine using very simple floating arithmetic

/*
Practical C Programming, Third Edition
By Steve Oualline
Third Edition August 1997 

ISBN: 1-56592-306-5
Publisher: O'Reilly
*/

/* Usage:                                               *
 *      sine <value>                                    *
 *                                                      *
 *      <value> is an angle in radians                  *
 *                                                      *
 * Format used in f.fffe+X                              *
 *                                                      *
 * f.fff is a 4 digit fraction                          *
 *      + is a sign (+ or -)                            *
 *      X is a single digit exponent                    *
 *                                                      *
 * sine(x) = x  - x**3 + x**5 - x**7                    *
 *               -----   ----   ---- . . . .            *
 *                 3!     5!     7!                     *
 *                                                      *
 * Warning: This program is intended to show some of    *
 *      problems with floating point.  It not intended  *
 *      to be used to produce exact values for the      *
 *      sin function.                                   *
 *                                                      *
 * Note: Even though we specify only one-digit for the  *
 *       exponent, two are used for some calculations.  *
 *       This is due to the fact that printf has no     *
 *       format for a single digit exponent.            *
 */
#include <stdlib.h>  
#include <math.h>
#include <stdio.h>

/*
 * float_2_ascii -- turn a floating-point string        *
 *      into ascii.                                     *
 *                                                      *
 * Parameters                                           *
 *      number -- number to turn into ascii             *
 *                                                      *
 * Returns                                              *
 *      Pointer to the string containing the number     *
 *                                                      *
 * Warning: Uses static storage, so later calls         *
 *              overwrite earlier entries               *
 */
static char *float_2_ascii(float number)
{
    static char result[10]/*place to put the number */

    sprintf(result,"%8.3E", number);
    return (result);
}
/*
 * fix_float -- turn high precision numbers into        *
 *              low precision numbers to simulate a     *
 *              very dumb floating-point structure.     *
 *                                                      *
 * Parameters                                           *
 *      number -- number to take care of                *
 *                                                      *
 * Returns                                              *
 *      number accurate to 5 places only                *
 *                                                      *
 * Note: This works by changing a number into ascii and *
 *       back.  Very slow, but it works.                *
 */
float fix_float(float number)
{
    float   result; /* result of the conversion */
    char    ascii[10];      /* ascii version of number */

    sprintf(ascii,"%8.4e", number);
    sscanf(ascii, "%e", &result);
    return (result);
}
/*
 * factorial -- compute the factorial of a number.      *
 *                                                      *
 * Parameters                                           *
 *      number -- number to use for factorial           *
 *                                                      *
 * Returns                                              *
 *      factorial(number) or number!                    *
 *                                                      *
 * Note: Even though this is a floating-point routine,  *
 *       using numbers that are not whole numbers       *
 *       does not make sense.                           *
 */
float factorial(float number)
{
    if (number <= 1.0)
  return (number);
    else
  return (number *factorial(number - 1.0));
}

int main(int argc, char *argv[])
{
    float   total;  /* total of series so far */
    float   new_total;/* newer version of total */
    float   term_top;/* top part of term */
    float   term_bottom;/* bottom of current term */
    float   term;   /* current term */
    float   exp;    /* exponent of current term */
    float   sign;   /* +1 or -1 (changes on each term) */
    float   value;  /* value of the argument to sin */
    int     index;  /* index for counting terms */

    if (argc != 2) {
  fprintf(stderr,"Usage is:\n");
  fprintf(stderr,"  sine <value>\n");
  exit (8);
    }

    value = fix_float(atof(&argv[1][0]));

    total = 0.0;
    exp = 1.0;
    sign = 1.0;

    for (index = 0/* take care of below */ ; ++index) {
  term_top = fix_float(pow(value, exp));
  term_bottom = fix_float(factorial(exp));
  term = fix_float(term_top / term_bottom);
  printf("x**%d     %s\n"(int)exp, 
      float_2_ascii(term_top));
  printf("%d!       %s\n"(int)exp, 
      float_2_ascii(term_bottom));
  printf("x**%d/%d! %s\n"(int)exp, (int)exp,
    float_2_ascii(term));
  printf("\n");
  new_total = fix_float(total + sign * term);
  if (new_total == total)
      break;
  total = new_total;
  sign = -sign;
  exp = exp + 2.0;
  printf("  total   %s\n", float_2_ascii(total));
  printf("\n");
    }
    printf("%d term computed\n", index+1);
    printf("sin(%s)=\n", float_2_ascii(value));
    printf("  %s\n", float_2_ascii(total));
    printf("Actual sin(%G)=%G\n",
      atof(&argv[1][0]), sin(atof(&argv[1][0])));
    return (0);
}

           
       
Related examples in the same category
1.How to use asinHow to use asin
2.How to use sinHow to use sin
3.Hyperbolic sine: how to use sinhHyperbolic sine: how to use sinh
4.Calculate arcsine: how to use asinCalculate arcsine: how to use asin
5. Calculate sine: sin
6.Calculate hyperbolic sine: how to use sinh
7.copy one file to another
www.java2java.com | Contact Us
Copyright 2009 - 12 Demo Source and Support. All rights reserved.
All other trademarks are property of their respective owners.