Accurate collision detection ??

Hi,

I am working on a breakout clone and I am a bit stuck on how to efficiently
code up collision detection.

I have a simple algorithm that detects collisions between rectangles so I
can tell if the ball has hit a brick. The problem is I am not sure how to
do the calculations to determine where on the brick the ball hit, so I can
then do the appropriate collision response. For example if the ball hits
the horizontal edge of a brick then its y-axis velocity should be changed,
and if the ball hits the vertical edge of a brick then its x-axis velocity
should be changed.

The only thing I can think of is to create a line that represents the
trajectory of the ball and then check against each surface of the brick to
see if it intersects. I can reduce the number of surfaces to check on the
brick by looking at the direction the ball of travelling eg. if the ball is
travelling toward the top-left of the screen then I only need to check the
bottom horizontal surface and the right vertical surface of the brick. Is
there a simpler/smarter way to do this?

Anyone know of a good reference site that talks about this?

At a later stage I would like to add rotation to the ball, but for now I
just want to get something simple running.


Thanks
Ty

Ty - You probably already have a solution, but I found this article.
http://www.mvps.org/directx/articles...ng_spheres.htm
on collision detection for spheres

To Newsgroup Games.programming
I am a curler ( a Canadian game ) which has similar collisions to pool table
ball to ball collisions.
Whenever I watch or participate in curling, I think that judgment skills
could be enhanced by a better understanding of the collision physics
involved. (both planned offset and sweeping judgment skills - plan b shots,
etc)
Being an engineer, I start thinking about programming a collision simulator
but am rusty on my circular collision equations.
I was hoping that someone from this newsgroup could point me in the right
direction to get started on .circular collision equations.
I need
a) equations for
- simple circular collisions
- more complex rotation collisions to account for direction
changes due to spin
- very advanced - to take ice interaction into account.
Objective:
Set up on overhead view of house - the area where the collisions occur
Populate with rocks ( maybe in future marry a TV frame grabber and verify
simulation with actual shots)
Show the possible outcomes of initial rock collision for
a) plus/minus angle windows for next collision
b) translate that into the horizontal offset view
c) duplicate for multiple collisions
Regards
Peter



===========
"T" <csaweb@ozemail.com.au> wrote in message
news:va2Jb.73413$aT.66309@news-server.bigpond.net.au...
> Hi,
>
> I am working on a breakout clone and I am a bit stuck on how to
efficiently
> code up collision detection.
>
> I have a simple algorithm that detects collisions between rectangles so I
> can tell if the ball has hit a brick. The problem is I am not sure how to
> do the calculations to determine where on the brick the ball hit, so I can
> then do the appropriate collision response. For example if the ball hits
> the horizontal edge of a brick then its y-axis velocity should be changed,
> and if the ball hits the vertical edge of a brick then its x-axis velocity
> should be changed.
>
> The only thing I can think of is to create a line that represents the
> trajectory of the ball and then check against each surface of the brick to
> see if it intersects. I can reduce the number of surfaces to check on the
> brick by looking at the direction the ball of travelling eg. if the ball
is
> travelling toward the top-left of the screen then I only need to check the
> bottom horizontal surface and the right vertical surface of the brick. Is
> there a simpler/smarter way to do this?
>
> Anyone know of a good reference site that talks about this?
>
> At a later stage I would like to add rotation to the ball, but for now I
> just want to get something simple running.
>
>
> Thanks
> Ty
>
>