CODEWITHSUNDEEP
cfloating-pointbinary
as3rdaccount
as3rdaccount

Reputation: 3941

What is this bitwise shift left doing?

I am going over someone else's code written a few years ago which is pretty badly documented. At one point there is the following code snippet: 

#define BINARYPOINT 16
....
float x;
int y;
int z;
....
....
//x has new values now
y = (int) (x*(1<<BINARYPOINT));
z = arctan2(y);
x = z;

In the code y is being explicitly used to compute the arctan of it, function arctan takes an int and gives the arctan of it which is another int. Now my trouble is here, I don't see how computing the arctan of x*(1<<BINARYPOINT) will give me the arctan of x?

Can anybody please explain what is going on here?

Thanks a lot in advance.

Upvotes: 2

Views: 169

Answers (1)

K. Brafford
K. Brafford

Reputation: 3849

That code is doing Binary Scaling. The floating point number is converted to an integer representation with a B16 binary scale, and then the function arctan2 is called. This is a routine that takes an integer argument and knows that it is using a B16 scale factor. It's not the standard atan call that takes a double argument.

Note that the name of the function, arctan2, makes it look like it's intended to be the two argument version of the arctan, so you should check into that and make sure you're making the right call.

Upvotes: 4

Related Questions