I have a java project who makes the "windows' maze" and use the ray casting algorithm. Here's a screenshot :
Like you can see all the walls have the same height size. I would like to do the same but with different height size
private void castRay(int xOnScreen,double angle,double direction) {
R rx = castRayInX(angle,direction);
R ry = castRayInY(angle,direction);
// In case of out-of-space rays
if (rx.getDistance()==Double.MAX_VALUE && ry.getDistance()==Double.MAX_VALUE) {
graphics.setColor(BACKGROUND);
graphics.drawLine(xOnScreen,0,xOnScreen,this.image.getHeight());
return;
}
double distance = rx.getDistance();
double normal = rx.getNormal();
Color c = rx.getColor();
double coef = Math.cos((angle+direction+Math.PI)-normal);
Plot collision = rx.getPlot();
if (ry.getDistance()<rx.getDistance()) {
distance = ry.getDistance();
normal = ry.getNormal();
c = ry.getColor();
coef = Math.cos((angle+direction+Math.PI)-normal);
collision = ry.getPlot();
}
coef = Math.abs(coef);
int factor = map.length*SQUARE_SIZE;
double d = (double)(distance+factor)/factor;
coef *= 1/(d*d);
Color c2 = new Color((int)(c.getRed()*coef),(int)(c.getGreen()*coef),(int)(c.getBlue()*coef));
graphics.setColor(c2);
// graphics.setColor(c); // no illumination
distance *= Math.cos(angle); // lens correction
int h = (int)(this.screenDistance/distance*WALL_HEIGHT); // perspective height
int vh = this.image.getHeight();
graphics.drawLine(xOnScreen,(vh-h)/2,xOnScreen,(vh+h)/2);
drawEye(direction,collision);
}
private R castRayInX(double angleRay,double direction) {
double angle = angleRay+direction;
double x1 = eye.getX()+SQUARE_SIZE*Math.cos(angle);
double y1 = eye.getY()+SQUARE_SIZE*Math.sin(angle);
double slope = (y1-eye.getY())/(x1-eye.getX());
if (Math.cos(angle)==0) {
if (Math.sin(angle)>0)
return new R(Double.MAX_VALUE,3*Math.PI/2,BACKGROUND,null);
else
return new R(Double.MAX_VALUE,Math.PI/2,BACKGROUND,null);
}
if (Math.cos(angle)>0) {
int firstX = ((eye.getX()/SQUARE_SIZE)+1)*SQUARE_SIZE;
R r = new R(Double.MAX_VALUE,angle+Math.PI,BACKGROUND,null);
for (int x = firstX; x<map[0].length*SQUARE_SIZE; x += SQUARE_SIZE) {
int y = (int)(slope*(x-eye.getX())+eye.getY());
if (isOutside(x,y,Color.MAGENTA,this.showRayCastingX)) break;
Color c = colorAt(x,y);
if (c==null) c = colorAt(x,y-1);
if (c==null) c = colorAt(x-1,y);
if (c==null) c = colorAt(x-1,y-1);
if (c!=null) {
int DX = x-eye.getX();
double DY = y-eye.getY();
return new R(Math.sqrt(DX*DX+DY*DY),Math.PI,c,new Plot((int)x,(int)y, WALL_HEIGHT));
}
}
return r;
} else {
int firstX = ((eye.getX()/SQUARE_SIZE))*SQUARE_SIZE;
R r = new R(Double.MAX_VALUE,angle+Math.PI,BACKGROUND,null);
for (int x = firstX; x>=0; x -= SQUARE_SIZE) {
int y = (int)(slope*(x-eye.getX())+eye.getY());
if (isOutside(x,y,Color.MAGENTA,this.showRayCastingX)) break;
Color c = colorAt(x,y);
if (c==null) c = colorAt(x,y-1);
if (c==null) c = colorAt(x-1,y);
if (c==null) c = colorAt(x-1,y-1);
if (c!=null) {
int DX = x-eye.getX();
double DY = y-eye.getY();
return new R(Math.sqrt(DX*DX+DY*DY),0,c,new Plot((int)x,(int)y, WALL_HEIGHT));
}
}
return r;
}
}
private R castRayInY(double angleRay,double direction) {
// System.out.println("cast ray 2 Y "+angleRay+" "+direction);
double angle = angleRay+direction;
double x1 = eye.getX()+SQUARE_SIZE*Math.cos(angle);
double y1 = eye.getY()+SQUARE_SIZE*Math.sin(angle);
// System.out.println(eye+" "+x1+" "+y1);
double slope = (y1-eye.getY())/(x1-eye.getX());
if (Math.sin(angle)==0) {
if (Math.cos(angle)>0)
return new R(Double.MAX_VALUE,Math.PI,BACKGROUND,null);
else
return new R(Double.MAX_VALUE,0,BACKGROUND,null);
}
if (Math.sin(angle)>0) {
int firstY = ((eye.getY()/SQUARE_SIZE)+1)*SQUARE_SIZE;
R r = new R(Double.MAX_VALUE,angle+Math.PI,BACKGROUND,null);
for (int y = firstY; y<map.length*SQUARE_SIZE; y += SQUARE_SIZE) {
int x = (int)((y-eye.getY())/slope)+eye.getX();
if (isOutside(x,y,Color.CYAN,this.showRayCastingY)) break;
Color c = colorAt(x,y);
if (c==null) c = colorAt(x,y-1);
if (c==null) c = colorAt(x-1,y);
if (c==null) c = colorAt(x-1,y-1);
if (c!=null) {
double DX = x-eye.getX();
int DY = y-eye.getY();
return new R(Math.sqrt(DX*DX+DY*DY),3*Math.PI/2,c,new Plot((int)x,(int)y, WALL_HEIGHT));
}
}
return r;
} else {
int firstY = ((eye.getY()/SQUARE_SIZE))*SQUARE_SIZE;
R r = new R(Double.MAX_VALUE,angle+Math.PI,BACKGROUND,null);
for (int y = firstY; y>=0; y -= SQUARE_SIZE) {
int x = (int)((y-eye.getY())/slope)+eye.getX();
if (isOutside(x,y,Color.CYAN,this.showRayCastingY)) break;
Color c = colorAt(x,y);
if (c==null) c = colorAt(x,y-1);
if (c==null) c = colorAt(x-1,y);
if (c==null) c = colorAt(x-1,y-1);
if (c!=null) {
double DX = x-eye.getX();
int DY = y-eye.getY();
return new R(Math.sqrt(DX*DX+DY*DY),Math.PI/2,c,new Plot((int)x,(int)y, WALL_HEIGHT));
}
}
return r;
}
}
My Rclass has a Plot (x, y, z) for now I use WALL_HEIGHT a color, a distance and a normal for the light. For now this works but I would like to add a new function like castRayInZ but I don't have all the mathematics theory behind. My maze is made from a map like that :
private String [][]map = { // each: SQUARE_SIZE x SQUARE_SIZE
{ "Y300", "Z500", "X230", "Y112", "Z321", "X120", "X354" },
{ "X89", " ", " ", " ", "Y120", " ", "X232" },
{ "Z124", " ", "X276", " ", "X123", " ", "X" },
{ "Y290", " ", " ", " ", " ", " ", "X100" },
{ "X32", "Z430", " ", "Y500", "X120", " ", "X123" },
{ "X222", " ", " ", " ", " ", " ", "X210" },
{ "X12", "Y98", "Y763", "X146", "Y111", "Y333", "X321" }
where X Y Z is for the color (X for Red, Y for Green and Z for Blue just testing my light function) and I add a height for each square of my map. I set all of length to SQUARE_LENGTH for now maybe later I will reduce the size of each square to a pixel and enlarge my map by generating it. But I really want to know how can I change the height of each square. I'm working on it since 4 days now and I don't have any clues...
EDIT
I have some news, I changed the size of my walls but I have some strange stuff, here's a screenshot :
Like you can see I there's some strange things appear here. Here's my code :
private void castRay(int xOnScreen,double angle,double direction) {
R rx = castRayInX(angle,direction);
R ry = castRayInY(angle,direction);
// In case of out-of-space rays
if (rx.getDistance()==Double.MAX_VALUE && ry.getDistance()==Double.MAX_VALUE) {
graphics.setColor(BACKGROUND);
graphics.drawLine(xOnScreen,0,xOnScreen,this.image.getHeight());
return;
}
double distance = rx.getDistance();
double normal = rx.getNormal();
Color c = rx.getColor();
double coef = Math.cos((angle+direction+Math.PI)-normal);
Plot collision = rx.getPlot();
if (ry.getDistance()<rx.getDistance()) {
distance = ry.getDistance();
normal = ry.getNormal();
c = ry.getColor();
coef = Math.cos((angle+direction+Math.PI)-normal);
collision = ry.getPlot();
}
coef = Math.abs(coef);
int factor = map.length*SQUARE_SIZE;
double d = (double)(distance+factor)/factor;
coef *= 1/(d*d);
Color c2 = new Color((int)(c.getRed()*coef),(int)(c.getGreen()*coef),(int)(c.getBlue()*coef));
graphics.setColor(c);
distance *= Math.cos(angle); // lens correction
int h;
int hw = (int)(this.screenDistance/distance*WALL_HEIGHT); //WALL_HEIGHT value is 300px at default
if(rx.getPlot() != null)
h = (int)(this.screenDistance/distance*rx.getPlot().getZ()); // perspective height
else
h = (int)(this.screenDistance/distance*WALL_HEIGHT);
int vh = this.image.getHeight();
int y0 = (hw+vh)/2;
int y1 = (vh-h)/2;
graphics.drawLine(xOnScreen,y0,xOnScreen,y1);
drawEye(direction,collision);
My problem should be from castRayInXfunction :
private R castRayInX(double angleRay,double direction) {
double angle = angleRay+direction;
double x1 = eye.getX()+SQUARE_SIZE*Math.cos(angle);
double y1 = eye.getY()+SQUARE_SIZE*Math.sin(angle);
double slope = (y1-eye.getY())/(x1-eye.getX());
if (Math.cos(angle)==0) {
if (Math.sin(angle)>0)
return new R(Double.MAX_VALUE,3*Math.PI/2,BACKGROUND,null);
else
return new R(Double.MAX_VALUE,Math.PI/2,BACKGROUND,null);
}
if (Math.cos(angle)>0) {
int firstX = ((eye.getX()/SQUARE_SIZE)+1)*SQUARE_SIZE;
R r = new R(Double.MAX_VALUE,angle+Math.PI,BACKGROUND,null);
for (int x = firstX; x<map[0].length*SQUARE_SIZE; x += SQUARE_SIZE) {
int y = (int)(slope*(x-eye.getX())+eye.getY());
if (isOutside(x,y,Color.MAGENTA,this.showRayCastingX)) break;
Color c = colorAt(x,y);
int z = heightAt(x,y);
if (c==null) c = colorAt(x,y-1);
if (c==null) c = colorAt(x-1,y);
if (c==null) c = colorAt(x-1,y-1);
if (z == 0) z = heightAt(x,y-1);
if (z == 0) z = heightAt(x-1,y);
if (z == 0) z = heightAt(x-1,y-1);
if (c!=null) {
int DX = x-eye.getX();
double DY = y-eye.getY();
return new R(Math.sqrt(DX*DX+DY*DY),Math.PI,c,new Plot((int)x,(int)y,(int)z));
}
}
return r;
} else {
int firstX = ((eye.getX()/SQUARE_SIZE))*SQUARE_SIZE;
R r = new R(Double.MAX_VALUE,angle+Math.PI,BACKGROUND,null);
for (int x = firstX; x>=0; x -= SQUARE_SIZE) {
int y = (int)(slope*(x-eye.getX())+eye.getY());
if (isOutside(x,y,Color.MAGENTA,this.showRayCastingX)) break;
Color c = colorAt(x,y);
int z = heightAt(x,y);
if (c==null) c = colorAt(x,y-1);
if (c==null) c = colorAt(x-1,y);
if (c==null) c = colorAt(x-1,y-1);
if (z == 0) z = heightAt(x,y-1);
if (z == 0) z = heightAt(x-1,y);
if (z == 0) z = heightAt(x-1,y-1);
if (c!=null) {
int DX = x-eye.getX();
double DY = y-eye.getY();
return new R(Math.sqrt(DX*DX+DY*DY),0,c,new Plot((int)x,(int)y,(int)z));
}
}
return r;
}
}
Should I make a castRayInZfunction ? Or should I get my z value somewhere else ?


so you obviously know the basics of Wolfenstein raycasting techniques. To add variable height you need to do this:
so simply add another value to your cell info in your map table
map[][]. You code the stuff as strings that is odd ...Somewhere in the code (after hit was detected) you render vertical line per each ray. There you should compute the scan line size something like (assuming y=0 is top of the screen):
And should change to:
Where
projected_half_sizeis the line size computed for constant cell height as you got now,wall_size=<0,1>is scale andcenter_of_view_yis the y coordinate of the horizon line in your view. This will put your wall on the ground.now when you hit the first wall you stop. With variable wall height you can stop only when you hit wall of full size (
wall_size=1) or run out of map. You got 2 option to implement this.The first option is easy to implement but requires more memory and it is slower. The second one is fast and does not need any lists or stacks. But it involves a little bit more math for the scan line render (
O(1)if coded right)I played a bit with my demo from the link at the top. Now the result should look like this:
As you can see the highlighted cells on the map are pass through for rays above their height (so you can see bigger cells behind them).
Beware once you add movement altitude direction (jumps, stairs etc) then the end condition must be different (the rendered scan line hit the top of the view). Also projection part of y coordinate will be different and need to include the actual player altitude.
You need to add the rendering of top side. It is similar to rendering ceiling and floor. IIRC original Wolfenstein did not have this capability but the latter pseudo 3D games like DOOM did.
There are more possible approaches like Perspective Vision on Canvas but I think the easiest to implement (as we already got enough info) is to compute the top side part of vertical scanline coordinates in texture and just copy pixels. As we already know where the ray hit the cell and angle of player/camera is also known. For more info see: PCGPE 1.0 Doom techniques
So as a first step add hits for back faces too. That should look like this:
That is done by checking hits of the last hit cell first. Now if you remember the last rendered
ycoordinate from previous hit (of the same scanline) then if back face is hit instead of rendering the face render top side color from lastyto actualy(or copy pixels from floor/ceiling texture instead). Here the green color used for this:If it helps here is mine C++ (GDI/VCL based) code for this:
Just ignore the
performance.htime measurementtbeg,tend,tstr,OpenGLrep4d_double.hkeyboard and mouse handlerkeytaband port VCL related stuff (Canvas,AnsiString,File access, JPEG...).If you need help with understand the gfx stuff see
The usage of this class is simple declare an object of this class, and add the event to your window (mouse,keyboard,repaint ...). My VCL Window (single form with one timer on it) code looks like this:
And here the main iteration variables explanation:
And here the texture file:
Here is how it looks like after some more code tweaking and perspective correct texture mapping:
Here related QA to this one:
And more recent demo version (with map editor):