Sign in to follow this  

CSG unpredictability

This topic is 4521 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I've recently been writing a Laidlaw-esque (the details are different to make my life easier) CSG implementation with AABB trees, but the results from the program can vary widely when they shouldn't - I suspected it wasn't working properly so during the phase where you determine which polygons are inside and which outside I cast a ray to the nearest face in a random direction instead of from the polygon normal. I'd expect that to cause some problems but in fact depending on whether I enable of disable bounding box tests, make the AABB tree all in one node and random chance caused by the vector direction I get variations from around 3000-4500 polygons output from my test level. I'm pretty sure I've covered numerical inaccuracy in the code and I can't figure out where the problem is coming from (and I don't think it's bad input data). The trouble is I can only get the problems to show up on large datasets. Does anybody have any really bright ideas of what might be wrong or how I can track this one down?

Share this post


Link to post
Share on other sites
Sign in to follow this