Hi Andi,
On Wed, 2005-05-04 at 09:29 +0200, =?iso-8859-1?Q? Andreas=20Br=FCning ?
> Thanks Dirk,
>
> that's a nice idea. But is it faster? I don't have special viewports,
> the user can look at the mode! l from every site. I search for the
> poygons, which are kompletly invis ible. I only load a model and show
> it. Now I try to take less rays and find the neighbors of the
> neighbors of the founded visible polygon. (and so on...) It's a very
> tricky problem. May be there is a good combination.
Well, completely invisible is pretty hard to do, but you won't be able
to do that totally correct with rays either. Image an object enclosed by
a sphere with a small hole. One of your rays would have to hit the hole,
to decide that the object is visible, and the chances of that are
proportional to the size of the hole, which can be pretty much
arbitrarily small.
You can think of the rendering approach as sending a whole lot of rays
from a single point into the scene (or parallel rays if you use an
orthographic projection), it's just faster than sending individual rays,
and you don't have as much control over every single ray.
Given that it's hard to decide the area-area visibility beforehand,
current hardware supports a method to do it on the fly. Look for
'occlusion culling' as a keyword to find papers about it.
Hope it helps
Dirk
-------------------------------------------------------
SF.Net email is sponsored by: GoToMeeting - the easiest way to collaborate
online with coworkers and clients while avoiding the high cost of travel and
communications. There is no equipment to buy and you can meet as often as
you want. Try it free.http://ads.osdn.com/?ad_id=7402&alloc_id=16135&op=click
_______________________________________________
Opensg-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensg-users
