I've been playing with this for the past several hours and I keep reaching dead ends.

What I'd like is some sort of environment in which I can set up a series of "mirrors" off of which particles/photons reflect. The idea is it would allow you to try out designs for things like periscopes and prisms.

The hardest part seems to be finding a good algorithm for handling collisions between the photons and mirrors... They often work for a few scenarios, but perform erratically in others. Here's one such example:

Code:

  boolean collide(Mirror m) {
    if(m == lastCollision)
	return false;
    
    // Endpoints of mirror
    float x1 = m.xPos + 0.5 * m.len * cos(m.angle);
    float y1 = m.yPos + 0.5 * m.len * sin(m.angle);
    float x2 = m.xPos - 0.5 * m.len * cos(m.angle);
    float y2 = m.yPos - 0.5 * m.len * sin(m.angle);

    // Endpoints of photon
    float x3 = xPos;
    float y3 = yPos;
    float x4 = xPos + speed * cos(angle);
    float y4 = yPos + speed * sin(angle);
    
    // Calculate the intersection
    float ua = ((x4 - x3) * (y1 - y3) - (y4 - y3) * (x1 - x3)) / ((y4 - y3) * (x2 - x1) - (x4 - x3) * (y2 - y1));
    float ub = ((x2 - x1) * (y1 - y3) - (y2 - y1) * (x1 - x3)) / ((y4 - y3) * (x2 - x1) - (x4 - x3) * (y2 - y1));
    
    if(ua >= 0 && ua <= 1 && ub >= 0 && ub <= 1) {
	// Hack based on the Law of Cosines and some guesswork
	float a = m.len / 2;
	float b = dist(m.xPos, m.yPos, xPos, yPos);
	float c = dist(x1, y1, xPos, yPos);
	float ang = acos((sq(a) + sq(b) - sq(c)) / (2 * a * b));
	angle -= PI - 2 * ang;
	
	print("collision ");
	lastCollision = m;
	return true;
    }
    
    return false;
  } 

It successfully identifies all collisions, but the reflection is incorrect.

Any thoughts?

Just a thought but you might want to look up interference. Two light wavefronts (and these wavefronts are spherical mind) of the same wavelength will cancel one another out. This is how holograms are made. The laser passes through the holographic plate, hits the subject and bounces back to interfere with the laser light still coming forward. The spherical interference patterns create parabolic mirrors of silver halide. This is how holograms appear three-dimensional.

When taking the hologram the setup needs to be perfectly still. A vibration of greater than the wavelength of the light used will disturb the interference pattern.

Another tidbit of information is that first surface reflection mirrors are used in the setup, unlike normal mirrors where the reflective surface is behind a pane of glass.

This is probably irrelevant but I'm just sharing.