Actually...

>>Thanks for the help, but I'm going to try finding these laws of 3D myself

Actually, I was trying to get you to find them yourself. I can give a much more detailed explanation, if you like, but I think it's better just to give it some time and let you find things for yourself.

>>Now I see that if I have a square at (50,50,0) and another square at (50,50,20) they should be exactly on the same spot on screen.

The diagram program I gave you DOES show things in a "flattened" view. This makes it easy to find the relations you need.

But the final starfield program should NOT be "flattened"; it should show things in perspective view.

Note that two stars that differ only in their Z co-ordinates WILL appear at different locations on the screen (at the two purple circles):

SCREEN 12
LOCATE 14, 3: PRINT "Eye";
LOCATE 12, 9: PRINT "Screen";
LOCATE 10, 24: PRINT "Star 1";
LOCATE 10, 49: PRINT "Star 2";
LINE (120, 170)-(430, 170), 8
LINE (30, 240)-STEP(400, 0), 4
CIRCLE (30, 240), 3
LINE (90, 200)-STEP(0, 80)
LINE (90, 229)-(400, 170), 2
LINE (400, 240)-(400, 170), 2
LINE (30, 240)-(90, 229), 9
LINE (90, 240)-(90, 229), 9
PSET (400, 169)
PSET (401, 170)
PSET (399, 170)
PSET (400, 171)
LINE (90, 215)-(200, 170), 2
LINE (200, 240)-(200, 170), 2
LINE (30, 240)-(90, 215), 9
LINE (90, 240)-(90, 215), 9
PSET (200, 169)
PSET (201, 170)
PSET (199, 170)
PSET (200, 171)
CIRCLE (90, 229), 3, 5
CIRCLE (90, 215), 3, 5