Visibility calculation (Ex. Quake 3 Clusters)
Author
I've tried to google search this for hours, to no avail. I am looking for a good algorithm to work out the visibility clusters in a Quake-3 style BSP (although I am happy with this method I would also take suggestions on a better method for visibility).
Thanks for all suggestions / links, ~SPH
Are you talking about actually compiling the PVS? I don't have a good answer for you - I have a general idea of the process but have never coded it. Are any of the Quake-related tools open-source? That might be a good place to start.
The second Watts/Policarpo 'Fly 3D' book has a description of the process. However, IIRC it bails at the end and uses point sampling for computing portal visibility, rather than the exact method using anti-penumbras. But it does describe at least part of the algorithm.
Several years ago I had a detailed tutorial on the process that I got somewhere online. Unfortunately, it's long gone and I have no idea where it came from :(
Sorry I can't help you more. From what I understand PVS compiling is actually kind of a tricky process due to floating point issues, as well as the overall code complexity.
The second Watts/Policarpo 'Fly 3D' book has a description of the process. However, IIRC it bails at the end and uses point sampling for computing portal visibility, rather than the exact method using anti-penumbras. But it does describe at least part of the algorithm.
Several years ago I had a detailed tutorial on the process that I got somewhere online. Unfortunately, it's long gone and I have no idea where it came from :(
Sorry I can't help you more. From what I understand PVS compiling is actually kind of a tricky process due to floating point issues, as well as the overall code complexity.
Author
Well all that I have in this is a simple BSP and I need to calculate the visibility from all of the leaves to all of the other leaves (in Quake 3 this info. is stored in the form of 'clusters'). I need this to be able to draw only the visible scene and such. The problem does seem to be quite complex, but I would expect there to be a tutorial or at least a paper available on it?
Thanks, ~SPH
Thanks, ~SPH
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
