Verified that I receive the same result patterns as you on my machine.
One other possibility outside of what you have already mentioned would be
something JIT-related. Perhaps there is an optimization that can be
performed if the locking sections of the code are in another function but
otherwi
On Jul 29, 4:38 am, Ken Wesson wrote:
> Yes. In fact this should work with records, too -- all the "native"
> fields of the record need to be copied for each version, but most will
> typically be pointers and most of those will typically be to existing
> objects. Only the ones that were changed wi
On Jul 29, 8:58 am, Michael Gardner wrote:
>
> An atom would be more appropriate, unless you're coordinating updates to the
> tree with updates to some other ref.
Ah, for some reason that didn't occur to me, thanks. Since the tree
in this case acts as a supporting index on spatial data, it pret
Hi all, this is my first post.
I'm currently building a tree data structure (quadtree to be exact)
while doing some exploratory game server programming using Clojure.
This tree will end up with a lot of "entities" stored in the nodes,
whose positions determine which node they end up in. As game s
