Problems with Modulo

I am getting an error message in this script. The script is supposed to make a prim
orbit a center point.


float i = 0;
float diameter = .10;

default
{
state_entry()
{
llTargetOmega(<0,0,1>, 0.1, 1);
llSetTimerEvent(.5);
}

timer()
{
i = i + .025;
i = (float)i % (float)TWO_PI;
llSetPos(llGetPos() + <llCos(i) * diameter, llSin(i) * diameter, 0>;
}
}




The problem is I get this error:

(14, 30) : Error : Type mismatch

If I do another kind of operator like + or * it works fine. It just doesn't like that I am
trying to mod a number by TWO_PI.

Any Thoughts?

The modulo is, by definition, only an integer operation. It finds the remainder of two integers... i.e., 11 / 3 is 3 with a remainder of 2. So 11 % 3 = 2. It's often used to find even numbers, because if X is even, then X % 2 = 0, but if it's odd, X % 2 = 1. My point here is that such an operation is fundamentally incompatible with floating points. Thus the type mismatch error.

So the question becomes, what exactly are you trying to accomplish on the line where you've used the modulo?

Well as with cos once you get to 2*PI you are basically back were you started. If you
where looking at a graph of cos(x). So I just didn't want i to get infanantly large. I guess
I could just do

if(i > TWO_PI)
i = 0;

Instead of doing

i %= TWO_PI:

But thanks for telling me what you did. That way I can move on to other things.

Hi Hiro,

I don't know if you've thought of this approach and discarded it, but here's something else that might work:

- rez a very small prim (say, a sphere or a cylinder flattened to a disk).

- set your orbiting object in the same plane as the new prim, and at a distance from it equal to your desired orbital radius. Then, link it to the new prim, selecting the new prim last of all. Since the new prim is now the root prim of the linkset, its coordinates will become the coordinates for the entire set.

- center the tiny prim inside your base object, and then apply llTargetOmega() to a script in the root prim. The root prim will act as the center of rotation.

== danny d.

I'd do :

if ( i > TWO_PI )
    i = i - TWO_PI;

You'll get smoother movement.

- Ace

That is an awesome idea Danny. I did not think of that. Thanks for the tip it is perfect for
what I want to do.

For what it's worth, fmod() (modulus for floating point numbers) is part of xycalc

/15/03/9686/1.html