Very Computer

I've just installed a new version of Linux (kernel 3.2.0) and I
am having some strange problems with the math.h library.
When I write a program like this:

#include <math.h>
#include <stdlib.h>
void main () {
  printf("%f",atan(1));

Quote:}

Henson Daytona

btw man atan gives : return value of atan is in RADIAN

--
 Universite Paris VI             |       Jean-Francois Boisvieux         |
 C.H.U Pitie-Salpetriere         |                                       |
 Departement de Biomathematiques | Tel: 33 1 45861998 Fax: 33 1 45865685 |

 75634 Paris cedex 13 France     |                                       |

   I've just installed a new version of Linux (kernel 3.2.0) and I
   am having some strange problems with the math.h library.
   When I write a program like this:

   #include <math.h>
   #include <stdlib.h>
   void main () {
     printf("%f",atan(1));
   }

   When I write that, I ALWAYS get 0.785398 for the atan of 1, which
   is supposed to really be 45.

Oh, dear!

Didn't you ever wonder what that DEG/RAD switch was doing on your
calculator?

Try multiplying 0.785398 by 4, I think that might clue you in...

It's as easy as...
--
   O_   ---- Peter Dalgaard
  c/ /'  --- Dept. of Biostatistics
 ( ) \( ) -- University of Copenhagen

| I've just installed a new version of Linux (kernel 3.2.0) and I
| am having some strange problems with the math.h library.

First of all, this is not a kernel "problem" (it's not a "problem at
all) -- its handled by libm.

| When I write a program like this:
|
| #include <math.h>
| #include <stdlib.h>
            ^^^^^^^^
printf() is defined in <stdio.h>

| void main () {
|   printf("%f",atan(1));
| }
|
| When I write that, I ALWAYS get 0.785398 for the atan of 1, which
| is supposed to really be 45.

Do you know the difference between radian and degree? Both values
are correct -- you have to supply the correct unit. From the man page:

RETURN VALUE
     The atan() function returns the arc tangent in  radians  and
     the  value is mathematically defined to be between -PI/2 and
     PI/2 (inclusive).

[...]
| I figured out that (tan(1))*(atan(1)) is very close to the value
| but I don't think it has anything to do with it.

oh, it has. Both, tan() and atan() return their values in units of
radian. If you want a degree return value, you have to
multiply it with 180/PI, i.e. you could use the following macro:

#define my_atan(x)      (atan(x)*180/M_PI)

| Please help me, I've included the -lm thing, and it used to work
| with my old kernel 1.0.8., but I can't figure out what's going

That would make me wonder ...

| on with it, it does a clean compile with no errors.
|
| Henson Daytona

Thomas
--

     <<o>> <<o>>                     Westfaelische Wilhelms-Universitaet
          ^                          Einsteinstr. 62, D-48149 Muenster, Germany
          -                          (+49) 251 - 83 3747