Hey all,

The following code is meant to convert an int to a float in C, but its not working!!! I need to use ints because of the mod (%) operation, but i need the results as a float.

{
int tempx;
int tempz;

float fx;
float fz;

/* sizex, sizez and resolution are all predetermined int values
rand initialised using srand(time(NULL));
*/
tempx = ((rand())%(sizex * resolution)) - (sizex * resolution / 2);
tempz = ((rand())%(sizez * resolution)) - (sizez * resolution / 2);
fx = tempx;
fz = tempz;
}

I've tried using:

fx = (float)tempx;
fx = *((float*)&tempx);

But nothing seems to work, i get valid tempx and tempz values, but they go crazy when converted to float.

Ie. tempx = 9865, tempz = 5806, converts to fx = 0, fz = 1086538880

Anyone able to help me?

Have you tried this:
Typecasting an int to a float is surely not what you want. That tells the computer to interpret the bits in the int as a float; those two data types most assuredly do not use the same format. Though, I will say I'm surprised the original "fx = tempx;" didn't work. I could swear I've done something like that before. I'll need to reboot into Linux to double-check.

fx = (float)tempx;

This should work.

fx = *((float*)&tempx);

This will produce utter garbage. The binary representations of floats and ints are not interchangeable.

Besides, the code you posted (just assigning fx = tempx) works fine for me.

Must be my compiler...

By the way, how do you know what the values of fx and fz are? Through a debugger, or do you display them with a printf()? If printf(), make sure you're using the right conversion character. If you try to print a floating point as an int, you'll get crazy results.

Haha now thats embarrasing, yeah i was using fprintf("%d",fx);

Sorry guys, thanks for your help!