News for hackers. Stuff that matters.
Re:An Exact Algorithm for Finding Minimum Oriented Bounding ... by forrestthewoods on Sunday November 01, 2015 @05:37PM (#10488733)
I don't understand your comment. You say OBB isn't any better than GJK on arbitary convex polyhedra. Followed by saying GJK has issues with roundoff. And that testing it with random polyhedra is a bad test.

So, uh, what's your point? Because I don't know.

And none of these really has anything to do with the root post. OBBs are always going to be used. Always. Different methods for finding OBBs is useful. Fast ways to find "good" OBBs and slower ways to find "perfect" OBBs both have value.

Re:An Exact Algorithm for Finding Minimum Oriented Bounding ... by Animats on Sunday November 01, 2015 @06:25PM (#10488963)
The point with GJK is that it's a good algorithm, but hard to implement right. I struggled with this years ago, Steven Cameron at Oxford worked on it, and it's a solved problem now. I'm tempted to open-source the collision code from my old Falling Bodies product, but it's pre-STL C++ from the 1990s and I don't want to overhaul it now.

On collision detection generally, the performance advantage of OBBs over convex polyhedra is small and may be negative. If you use convex hulls, you get better looking collisions and it doesn't cost you much, if anything.

Re:An Exact Algorithm for Finding Minimum Oriented Bounding ... by sillysaurus3 on Sunday November 01, 2015 @09:19PM (#10489574)
I'm tempted to open-source the collision code from my old Falling Bodies product, but it's pre-STL C++ from the 1990s and I don't want to overhaul it now.

You should release it and put that in the description.

The lack of STL means it's more likely to be thoughtfully designed. It's a merit, not a demerit. Or to put it a bit bluntly: Don't worry about that! I'd love to see the code.

Re:An Exact Algorithm for Finding Minimum Oriented Bounding ... by sillysaurus3 on Monday November 02, 2015 @04:04AM (#10490785)
I try not to do this, but you might be interested to know that four people upvoted me. So there are a bunch of people who want to see your code. I assure you, no one minds whether it's STL or not. The solution is the interesting bit.
Re:An Exact Algorithm for Finding Minimum Oriented Bounding ... by AndrewKemendo on Sunday November 01, 2015 @08:16PM (#10489359)
If you use convex hulls, you get better looking collisions and it doesn't cost you much, if anything.

Very interesting.

Can you go into more detail here on what you mean by better looking?

Re:An Exact Algorithm for Finding Minimum Oriented Bounding ... by MrScruff on Monday November 02, 2015 @03:15AM (#10490656)
The convex hull will represent the geometry more accurately than an OBB because it can have >8 points and doesn't have to be cuboid. So the collisions will be more representative.
Re:An Exact Algorithm for Finding Minimum Oriented Bounding ... by Animats on Monday November 02, 2015 @03:47AM (#10490732)
Right. In particular, OBBs roll like square wheels.
Re:An Exact Algorithm for Finding Minimum Oriented Bounding ... by speps on Monday November 02, 2015 @04:06AM (#10490795)
I'm tempted to open-source the collision code from my old Falling Bodies product, but it's pre-STL C++ from the 1990s and I don't want to overhaul it now.

If you open source it, someone will port it, add tests and make it into a nice library... So don't hesitate.

Re:An Exact Algorithm for Finding Minimum Oriented Bounding ... by forrestthewoods on Monday November 02, 2015 @12:32AM (#10490199)
"you get better looking collisions and it doesn't cost you much, if anything."

Yeah I dunno about this. Generating good convex polyhedra isn't free. Their cache missing extra memory isn't free. Nor is the test free.

I suppose it all really depends on the use case. The right balance of early out vs precision depends on your data set. Run too many early outs and it can be a net loss. Run too many precise tests and it's a net loss.