Thread: Rounding float to nearest integer

Hi Guys,
I have a situation where a float calculation is made,
and the result is converted back to an integer.


Is there a neat solution to convert to the nearest integer,
rather than just copying the float to the integer to always round down?


So instead of this:

Code:

float afloat = 5.9;
int aninteger = 0;


aninteger = afloat; // aninteger = 5

Something neater, and hopefully faster than this:

Code:

float afloat = 5.9;
int aninteger = 0;


aninteger = afloat; // aninteger = 5
afloat = afloat - aninteger; // afloat = 0.9
if (afloat > 0.5) {aninteger++;} // aninteger = 6

Or is this the best I can hope for?
Cheers, Brek.

Well it depends on what exactly you want from your rounding.

value = floor(value+0.5); would round 5.9 to 6 like your example.

You can trick floor into working for other decimal places by sliding the point.

value = floor(value * 100 + 0.5) / 100; would round something like 4.668 to 4.67 or 1.234 down to 1.23.

Yes ok thanks Really just anything above 0.5 to the next integer will do for now.
I’m using an old C30 compiler that doesn’t have round() functions in math library.

You didn’t catch that I forgot negative numbers exist though!

I just did a check if the number is below 0.0, and used a slightly different
version of the above in that case.