Answer 1

Hmm, I guess I can throw out a few words. I'll *highlight* some of them.

Usually, every object is represented as a set of *vertices* (just three numbers, the x, y, and z coordinates). Then you play connect the dots and form *polygons*, and lay a *texture* (an image) over each of the resulting faces. In really old games you can even see the individual vertices (corners): http://netbookgaming.de/wp-content/uploads/2012/02/systems21.jpg

Some special logic (*Z-Buffer*) finds out which objects are in front of the others and then you use the graphics card (which applies some *shaders* so stuff looks cool) to draw objects onto the screen in the right order.

Usually though, you deal with pre-made *frameworks* and *graphics engines*. Your designer will make *models* in his favourite modeling program and you will use some pre-made code to import that model and put it onto the screen without you knowing what really happens.

That code will then make sure everything is interacting correctly. For example, it'll need to make sure that when you run at a wall you don't go right through - *collision detection* using *hitboxes* - or that when you throw something, it comes back down - *physics engines*.

There's a lot to game development, but there's some stuff to look up!