Rounding to next highest number?

[NoKetch]

I have a simple program to output log2(x) but the result is not whole numbers.
Here is the code:

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

int main()
{
double x;

printf("Enter a Number: ");
scanf( "%lf", &x);

if ( x >= 0 )
printf("\nResult: %lf\n", log2(x));

system("PAUSE");
return 0;
}
----------------------

I get results like this:

Result: 4.321928
Result: 5.459432

How do i round these numbers to the next highest whole number so that the
output would be:

Result: 5
Result: 6

I'm not a very experienced programmer, so the explanation should be fairly
simple.

Thanks for any tips


Dave

 

[Mark A. Odell]

(e-mail address removed) (NoKetch) wrote in

How do i round these numbers to the next highest whole number so that
the output would be:

Result: 5
Result: 6

I'm not a very experienced programmer, so the explanation should be
fairly simple.

Try ceil() for up, and floor() for down. See:

http://www.acm.uiuc.edu/webmonkeys/book/c_guide/2.7.html#ceil

 

[Christopher Benson-Manica]

How do i round these numbers to the next highest whole number so that the
output would be:

Use ceil(), from the math library. #include math.h and do whatever
may be necessary on your system to link to the math library. Check
your friendly documentation or man page for more details.

 

[Mark A. Odell]

(e-mail address removed) (NoKetch) wrote in



Try ceil() for up, and floor() for down. See:

http://www.acm.uiuc.edu/webmonkeys/book/c_guide/2.7.html#ceil

Or you could at 0.5 to the result and then cast to int. E.g.

(int) (4.732 + 0.5) is 5.

 

[Thomas Stegen]

Or you could at 0.5 to the result and then cast to int. E.g.

(int) (4.732 + 0.5) is 5.

That depends on if you want to round away from zero or in a positive
direction. What you have rounds in a positive direction

 

[Mark A. Odell]

Rather than stew about what happens here, I think you should use ceil().
After all that IS what you want to do. Why make some other poor soul
figure out what you were up to? Eschew obfuscation!

I agree. On a small 8-bit micro, I might do this instead of ceil() if I
hadn't used any FP library functions anywhere else. Ceil() is the better
choice in most cases I would think.

 

[osmium]

That depends on if you want to round away from zero or in a positive
direction. What you have rounds in a positive direction

Rather than stew about what happens here, I think you should use ceil().
After all that IS what you want to do. Why make some other poor soul figure
out what you were up to? Eschew obfuscation!

 

[Mark McIntyre]

I agree. On a small 8-bit micro, I might do this instead of ceil() if I
hadn't used any FP library functions anywhere else. Ceil() is the better
choice in most cases I would think.

ceil() can lead to unexpected results when reading in data from
external sources, due to FP inaccuracies. Well I reckon it always
surprises to see ceil(5.0f) == 6 anyway.