A way to deal with aimbotting

[Null]

Currently (and it comes to no surprise) actual aimbots are leagues better than even the best players. I'm talking every shot a headshot killing 10 deuces consecutively aimbots. For whatever reason people don't enjoy kicking aimbots when their team is steamrolling because of one, something to do with not enjoying a challenge or proper game. The current detection scripts, though effective, are now outdated since the people designing the scripts are now less greedy and don't try to spawn more bullets than the server normally allows.

There are two simple tweaks which can be implemented to pretty easily get rid of the problem. First of all, change the votekick script to only require a number of votes equal to the number of players on the team initiating the kick plus one or two (so that it won't be abused by simply kicking all members of the opposing team). This one's a gimme, there's no extra code to implement just some numbers to swap around in whatever votekick script is already in place.

The above tweak of course assumes that people are adequate judges of aimbotting and actual packet manipulation (an actual "hack") which is often not the case. It is however also possible to detect aimbots based solely on statistics, keeping a headshot:body-shot ratio for each player and checking against a pre-calculated (and possibly periodically tweaked) average range for the player's respective weapon. Another useful statistic might be hits:misses, again, taking into account weapon and server averages. It wouldn't really add all that much overhead at all, maintaining the calculations would, in the worst case, cost only 2 logical comparison(hit or miss, if hit: head or body) and 4 primitive math calculations (iterate the hit:miss/head:body counter by 1 and divide the two for the ratio) and (if the averages are to be tweaked) a O(n) operation to add the data from all players to the current server average. I of course don't know if this would cause more network overhead (I don't know all details of your implementation) but these are checks which would be relatively painless to implement in a perfect world and would force current aimbots out of circulation and new aimbots to operate within the parameters of only an above average player.

EDIT: Clarity in some of the technical aspects

[Sniper101]

Yep, nice idea (read the text 5 times to get what you mean), but I think it would increase the lag in game.

[Null]

Yep, nice idea (read the text 5 times to get what you mean), but I think it would increase the lag in game.

I don't think it necessarily would add all that much lag. Really all that would be added with the checks is about 64 bytes of RAM use per player and an extra few calculations per player (7 or 8 at maximum, it depends how everything is set up) every time we want to check for abnormal accuracy. Even that can be dependent on whoever maintains the server, it could be as sporadic as every time x kills are made or every y minutes.

The place where I could see some independently maintained servers getting lag from such a script would probably be during heavy SMG spam, but assuming that the server is already resolving bullet collisions (which it has to be) and non-collisions (which it very well could be) it only adds about 2-3 calculations per bullet fired.

Oh, and sorry about the lack of clarity. This is something I thought up in aloha pinpoint after seeing a teammate hit every headshot he aimed for for 3 minutes pretty much straight, I scribbled it down over the next few minutes. Even if my suggestion with statistics doesn't "make the cut" as it were a tweak to the votekick would make a huge difference.

[Moonkey]

No... You can't solely determine an aimbot from statistics. Nor do all aimbots use head-shots. Some players who use wall hacks (Seeing players through walls) can be called Best players, or professional. The votekicking would be semi-inaccurate but probably OK.

[bobbotheclown]

Doesn't the votekicking system already require like 50 percent of the players to vote yes?

[kmaj]

Doesn't the votekicking system already require like 50 percent of the players to vote yes?

Yes, and it's too much imho.

[bobbotheclown]

but aren't most teams autobalanced so it's 50-50?

what null is suggesting is already in the game, or so i think.

[topologist]

Working on a pyspades plugin that'll do exactly this. Just need to do some testing (which I can't do for a while). Nice suggestion!

[Paratrooper]

I heard somewhere that they could send a fake player spawn packed behind a player. If he turns, it must mean that he/she's an aimpotter.

[runHarder]

Realistically, there is no way to eliminate the scourge of the aim bot.

[BlueToonYoshi]

There should be something that if the client knows you are using a program to hack, it kicks you automatically!

[runHarder]

There should be something that if the client knows you are using a program to hack, it kicks you automatically!

Nice in theory, but would only work against the casual botter.

[Venator]

Better idea:
1.Track their IP
2.Search for their house
3.Get a bag and wrap it over their heads
4.Beat them up with a baseballbat
5.??????????
6.Profit

[Sniper101]

Better idea:
1.Track their IP
2.Search for their house
3.Get a bag and wrap it over their heads
4.Beat them up with a baseballbat
5.??????????
6.Profit

GENIUS. LETS DO IT (but I don't have a baseball bat......).

[Venator]

i can borrow you one