Bot Dev server

[Ansley1]

OTHER THEN THE REALM WHATS ANOTHER BOT DEV SERVER? damn caps ...who runs the newest version of bots?

[Frumph]

If you see it up, the Frumph one runs the latest bot server.

[Secrets]

WildcardX and I have set up a non-legit bot server for people to play around on and break.

You can find it on the list under the name Official Bot Test (Non-Legit/Latest Bot Revision). This will always be running the latest stable bot code.

Currently, it is running revision 1099 W/ Bots. This server is mainly to test out new features amongst other things with bots that have been commited to SVN, fix crash issues/general oddities with bots, and for people to have a place to try out new bot features before adding any of them onto their servers.

The Hidden Forest runs bots, as well, and I am going to keep them up to date with the latest stable bot logic, however the non-legit server will be getting things first to try out. (as it has nothing to lose!)

[WildcardX]

You can think of it as a "community bots sandbox" server. A server you can stop at to evaluate how the bots behave. Right now, for example, we are evaluating the behavior differences between bots r1099 and r1110.

[Frumph]

yeow, there's an extra posupdate going on that's making the bots act as if they're hyped up on coffee

[Frumph]

I found it, there are two SendPosUpdates active at the same time every cycle when the bot is standing still, one is around line 2430 the other is around line 2403 both in the if(AImovement_timer->Check()) { area.

Those two posupdates are happening constantly causing the jittering, specifically the one at around line 2430 is the cause

I figured it out by setting a Say("line: 2430 Sending pos update."); at each pos update part in the code so the bot would say the refresh and it was spamming the screen constantly

So it's checking if(moved) so somewhere in the same cycle it's setting moved back to be true each time when it shouldn't ? Not sure - dunno why that particular one would keep triggering a pos update, I know it's pretty much required somewhere around there to do a posupdate to stop the ghosting when the guy stop's walking. But is there a better place for the posupdate is there a particular location that makes the stopping moving which the posupdate would be better placed at so it's not doing it every cycle? (which is happening at the sendposupdate around 2443, so n/m that). hrm.

[Frumph]

Try this out, it's looking great on my server with just a jitter every 6 seconds:


@@ -1583,11 +1583,10 @@

if (tic_timer.Check()) {
//6 seconds, or whatever the rule is set to has passed, send this position to everyone to avoid ghosting
- if(!IsMoving() && !IsEngaged())
- SendPosUpdate();
-
+// if(!IsMoving() && !IsEngaged())
+// SendPosUpdate();
+
SpellProcess();
-
BuffProcess();

if(curfp)
@@ -2068,7 +2067,7 @@

SetRunAnimSpeed(0);

- if(IsMoving()) {
+ if (IsMoving()) {
SetMoving(false);
moved = false;
SetHeading(0);
@@ -2395,9 +2394,7 @@
}

if(AImovement_timer->Check()) {
- if(!IsMoving())
- SendPosUpdate();
-
+ // if(!IsMoving()) SendPosUpdate();
// now the followID: that's what happening as the bots follow their leader.
if(GetFollowID()) {
Mob* follow = entity_list.GetMob(GetFollowID());
@@ -2419,9 +2416,11 @@
else {
SetHeading(follow->GetHeading());

- if(moved) {
+ if (moved) {
SetMoving(false);
- SendPosUpdate();
+ if (tic_timer.Check()) {
+ SendPosUpdate();
+ }
}
}

[WildcardX]

excellent i'll try this code out now. if there is still a jitter every 6 seconds, then i'ts probably caused by the pos update that is sent in the process() method after the tick timer fires.