Quote:
Originally Posted by cbk1994
If you're going to do that, you should be sure to turn off the wall check in the move function. That way you aren't doing the work twice.
|
Yes, of course.
Quote:
Originally Posted by -Ramirez-
Which wastes more server CPU time as a result of being done via script, rather than native code. Sure, it works, but nobody should have to bother.
Have fun with sending multiple move() calls without previous ones finishing. It looks awesome, unless the pathetic skippiness has actually been fixed.
|
Yeah, and about a year after I do it the hard way, despite having said nothing the whole time I was working, Stefan will release move3, which does exactly what move said it would do.
In fact, the Multiple Move Calls seems to be what kills the Wallcheck: It sends a "Go from 10,10 to 30,30, and stop if you hit a wall." It hits a wall at 20,20, and runs a new move call: "Go from 30,30 to 40,10, and stop if you hit a wall", so it warps to 30,30 (never figuring it maybe stopped when it hit the wall?) and continues its route unimpeded...
It all sounds quite familiar...I think there was a mailman NPC on one of the servers that was always getting stuck behind players; it seems as though Move was rewritten so that NPC could go about its business regardless of impedence...though frankly they should have just turned off the wallcheck for that one NPC rather than destroying the move command...but maybe I'm wrong.
In any case, it's easy enough for me to use timeout for move-monitoring. The only downside is, of course, I've used the only timeout the NPC has.