Chris Jacobson raised an interesting performance question
the other day - about using binary trees instead of linear
searches. His example was for the command list, and it is
probably agree'd that this would be bad because it would
ruin fine control over command precedence.
However, I wonder how much time in the mud is spent in
converting vnum's to rnums, which is done by the
real_[room|obj|mobile]() functions, also using a linear search.
I haven't profiled my mud (am doing so right now...) but was
wondering if anyone has profile data that would indicate this
would or would not be a useful effort?
Any other linear searches you can think of in the code that might
(or might not) benefit from a binary tree?
+------------------------------------------------------------+
| Ensure that you have read the CircleMUD Mailing List FAQ: |
| http://democracy.queensu.ca/~fletcher/Circle/list-faq.html |
+------------------------------------------------------------+
This archive was generated by hypermail 2b30 : 12/15/00 PST