Follow-up Comment #3, bug #21992 (project freeciv):
> (I suspect this isn't amenable to autogame testing, since the
> AI is expected to improve. Probably the correct test is against
> unmodified code and nreqs-based ruleset.)
Tested unmodified code with patch #4411 reverted (i.e. with nreqs) against
patched code with current classic ruleset (with present=FALSE). Diverged at
turn 36 when AI decided to build Great Library instead of Great Wall :(
Don't think this should happen, so holding patch back for now.
Fixed code with nreqs ruleset behaved the same as unfixed code with nreqs,
Might run a couple of repeat autogames to rule out unrelated nondeterminism.
(I double-checked patch #4411 itself, and didn't spot any semantic changes.)
Reply to this item at:
Message sent via/by Gna!
Freeciv-dev mailing list