InsideQC Forums

Rock2ober 2014: The Movie (Part 2) is (finally!!) here:

https://www.youtube.com/watch?v=OYMj3RIf-EI

Behold :

Rock2ober 2014: The Movie!!!

Calling ALL Quakeworld TF fans,

The LEGEND returns, the MONSTER is back, the BIGGEST TF game of the year is here! This is the seventh annual Rock2ober event, the annual get together where we attempt to fill a quakeworld server w 32 players like days of yore.The big question is, will we make 32 ...

Pretty sure in that screenshot, there should be a jump or rocketjump waypoint over the obstacles between a and the player,
In this example, there isn't. The current logic is: if(the bot gets stuck) UseWaypoints() - and current ai returns A as the 'final waypoint'.

there should also be a ...

Just to clarify (in case I'm not explaining myself clearly): currently the bot successfully navigates the nodes and ends up at the final waypoint. The final waypoint is always one that is the closest and has 'line of sight'. This would be fine if the bot needed to attack, but in this case I want the ...

It's all about correctly determining the cost to find B, actually. In the example you showed, a blocked waypoint should represent a HIGHER cost than the not so close but visible alternative.

Hi frag.machine
Yes, the thing is, in this situation .. how can WaypointA realise that it's NOT a valid wp ...

You might consider switching to A*. Dijkstra explores all possible waypoints and can be expensive. A* is a bit smarter, so it can be a big performance gain if you have a lot of points.

I have some code here (that I heavily referenced ACEBot source for): http://paintball2.cvs.sourceforge.net/viewvc ...

Is c0burn's source avaiable?
I made an custom bot back in the days with waypoint navigation, and if the bot sees the player he will start moving and attacking the player.
Also i added the "tracker system" from aicafe, so if the player walks around a corner / thru a teleporter the bot / enemy will ...

Update:

I've put aside my 'newbie guesswork' code and focused on studying c0burn's 'actually working' implementation. One thing I'm confused about is how to make it so the player can call his bot from any position in the map, and have the bot find him, and end up standing right next to him, not the ...

Thanks Zop, I was just responsing to Cobalt when your post came through.

Update: Thanks Cobalt for your suggestions. Spike has confirmed that issue was an..um.. er.. how shall I put it.. 'undocumented feature', of the FTE dedicted server. I had ruled out a server bug because two people confirmed ...

Hey people,
Here's a video of me demonstrating the issue, which might clarify the situation better.

ps If anyone can think of a better/more concise way to describe the 'subject' of this topic, please let me know.

[edit: I changed the 'subject' from 'Baffling SV_RunClientCommand bug.. ' because this might not necessarily be a SV_RunClientCommand specific issue]

Hi all,

IIUC every map change is a 'fresh beginning', the engine re-loads ALL variables and functions from the dat, including built-ins, and 'entry ...

... dynamically sized arrays are not supported either (although I could theoretically implement that, there's no support right now). have fun.

Well .. is it accurate to say that declaring an entity makes space for an entity pointer, and until it's spawned it doesn't take up much memory? eg ...

1) Take a look at the last release of CustomTF coop that gizmo maintained. I sent him some code changes that would address some issues with grunties getting stuck and I know he had some things he did to the gruntie bot code to improve it.

Hey Ratbert! Yes, thegizmo added some finesse to Grunt ...

find finds strings. don't use it for floats or ents. that way lies madness (or instability).

Ah.. ok, thanks.

findchain finds a chain of entities. if you call it multiple times within a loop, or recursively, then you lose.

Ahhh.. so my 'super function' with multiple nested loops wont work? Hm ...