CheatDetector 0.5.2.1

[ImportBot]

Originally Posted by russel5*:

 

i will add an option to choose that in the next update

Thanks!

[ImportBot]

Originally Posted by russel5*:

 

Hi, I have a question. Suppose the situation. Player has bought a new account and bought cheat and went to the server. Set up a one shot kill and started killing people on the server. The player has the first level, his statistics are not yet determined, but it kills a lot of people. Your plugin will ban it?

[ImportBot]

Originally Posted by HexaCanon*:

 

Hi, I have a question. Suppose the situation. Player has bought a new account and bought cheat and went to the server. Set up a one shot kill and started killing people on the server. The player has the first level, his statistics are not yet determined, but it kills a lot of people. Your plugin will ban it?

Not this plugin but there are others that u can use to set rate of kills.

[ImportBot]

Originally Posted by grizzlybeer*:

 

Hi, I have a question. Suppose the situation. Player has bought a new account and bought cheat and went to the server. Set up a one shot kill and started killing people on the server. The player has the first level, his statistics are not yet determined, but it kills a lot of people. Your plugin will ban it?

no it will not (as of now, version 0.1.2)

the scenario you describe is excatly what i have planned "for later" along with detecting players that had their stats reset.

 

the limitations for these 2 planned features are:

 

1. cheater on the server with clean (zero) stats

-only hs/k nad k/m will be checked because that is the only thing the server reports

-some weapons and all (_) vehicles are reported as "death" by the server

-for hs/k and k/m the check will need at least ~ 25 kills (for example with 19 headshots which is 75% or 2 minutes which is 12,5 k/m) to be accurate

 

2. cheater that had their stats reset (zero stats)

-i see no way to determine that using battlelog alone, because playtime and all scores are zero

-would need to check through another stats website (since i dont want the plugin to save all the stats of the players). for example bf3stats. problem is that your procon layer server provider might block ALL traffic except for official battlelog, which is needed anyway

 

i will add info about these 2 planned features in the first post

 

-grizzly

[ImportBot]

Originally Posted by IAF-SDS*:

 

Also, I have an idea for a great plugin and it might exist. It's somewhat related, but I was thinking if there was a plugin that would automatically run the players' names thru Cheat-O-Meter and automatically issue a ban (with optional temp ban or kick) if the palyer's results from Cheat-O-Meter came back with X% confirmed cheater (configurable). http://www.team-des-fra.fr/CoM/

 

Do you know of a plugin that does this?

A couple months ago, I was having a discussion with TimSad about a plugin that would check Cheat-O-Meter and automatically ban a player when cheater confirmation exceeded X percent (could be set to whatever you want). I figured this would be a nice integration into his VoteBan plugin.

 

Grizzly, in my opinion, what you have created here is one of the top five "must have" important plugins, right next to Panther's TrueBalancer and other great plugins like those from Falcon and TimSad (I use more than a dozen wonderful plugins and there are too many to mention).

 

Absolutely FANTASTIC work Grizzly! Without your work and the work of the other plugin developers, I could not see this game being sustained. The whole community is lucky to have such effort from you guys.

 

There are more hackers and cheaters than most realize. Recently, I came across this youtube interview (published November 2012) with a BF3 hacker where he discusses exactly what he does to cheat, how he hacks, and other details including intentionally spraying bullets into the air to throw off his stats to avoid cheat detection. Im not sure if you guys have seen this, but its interesting and thought everyone here would like to see it.

 

Part 1 - BF3 Hacker Interview

http://youtu.be/bTWZzEQ_nfw_t=1m10s

 

Part 2 - BF3 Hacker Interview

http://www.youtube.com/watch_v=JOE_tJzGD88

 

Part 3 - BF3 Hacker Interview

http://www.youtube.com/watch_v=FD0Eql4Ft3g

 

Part 4 - BF3 Hacker Interview

http://www.youtube.com/watch_v=juxQPfUWoho

 

 

 

It sounds like you've addressed the limits imposed by EA/DICE from too many requests by plugins with automatically detecting when the BF3 Battlelog Request Limit is reached and adjusting the queue timeout accordingly, but I hope you can work with Morpheus and the other guys to confirm resolution.

 

I will be installing this plugin on our servers.

 

Thank you again Grizzly.

[ImportBot]

Originally Posted by PapaCharlie9*:

 

Hey Grizzly, congrats on your plugin. You clearly put a lot of work into it.

 

Request: I'd like to extract the values from your checkKillsperHit and checkKillsperMin functions. I believe Insane Limits users would be interested in using those averages for their custom limits. Would that be okay with you? My plan is to publish the values in a post in Plugin Enhancements, not to incorporate the code into a plugin.

 

BTW, I don't mind that you derived your Battlelog stats fetching code from micovery's Insane Limits -- after all, he published the outline for the method for everyone to use. That said, it would be a good gesture to credit micovery and Insane Limits in your plugin code and documentation, particularly as you are asking for donations.

 

In reading your code, I found some areas that are a bit concerning with respect to layer resource usage (CPU, memory, net bandwidth) and Battlelog loading. I suggest you make the following improvements.

 

1) For now, it is good that this.batchsize remains 1 and new BackgroundWorker threads are created only once per Max(time since last OnListPlayers, this.batchtimeout). However, in the future when you will need to change this code as Morpheus warned, you'll be able to shorten or even eliminate the this.batchtimeout delay. That could lead to one thread being created per player if you are not careful. For future proofing, you should put an upper limit on the number of BackgroundWorker threads that can be created, regardless of this.batchsize.

 

2) Even though you are adding a delay between the stats fetch for each player, two of your three fetchWebPage calls to Battlelog are back-to-back with no delays or local processing time in between, and the first only has a little local processing. Multiply that across multiple game server connections on the same Procon layer instance and you can easily exceed the 15 in 20 seconds rate restriction if the delays happen to line up -- indeed, any layer with 6 or more active game connections is likely to generate Battlelog errors, even less if TrueBalancer and/or Insane Limits are also active. (Exercise for the math inclined: assuming the best case scenario where only CD is active and the start of the 30 second delay for each game server connection is offset by 20 seconds, what's the minimum number of game server connections that are guaranteed to saturate the Battlelog rate? Hint: it's less than 20).

 

3) The fetchWebPage for weapon stats can easily exceed 10 seconds -- I see a range of 12 to 20 seconds on my server -- it's a gigantic JSON blob. It would be a good idea to check for plugin disabled before and after doing the fetch for weapon stats to save doing unnecessary fetches from Battlelog for the most expensive case.

 

Cheers!

[ImportBot]

Originally Posted by grizzlybeer*:

 

PapaCharlie and IAF SDS: thank you very much :-)

 

thx for all the tips papacharlie. i will certainly try to keep all that in mind. really a lot of good infos thank you :-)

 

this plugin lacks some credit for sure. i will fix that asap. sorry for not putting it in the apropriate places in the first place. you can use any code or values you like from this plugin just credit better than i do ;-)

 

 

edit: just noticed there was another post 1minute before this :-)

[ImportBot]

Originally Posted by HexaCanon*:

 

is the average headshot/kill is something that is already set ? like M16 = 25% , M98B = 70% and so on..

 

or does it get calculated with your programing magic ?

[ImportBot]

Originally Posted by ty_ger07*:

 

is the average headshot/kill is something that is already set ? like M16 = 25% , M98B = 70% and so on..

 

or does it get calculated with your programing magic ?

If you look at the source code, you will see that the average values are already set.

 

Code:

switch (weaponname)
            {
                case "MK11":
                    avgkh = 0.314;
                    break;
                case "SVD":
                    avgkh = 0.3206;
                    break;
                case "SKS":
                    avgkh = 0.2555;
                    break;
                case "M39":
                    avgkh = 0.3187;
                    break;
                case "Pecheneg":
                    avgkh = 0.2093;
                    break;
                case "G36C":
                    avgkh = 0.1638;
                    break;
                case "M412 Rex":
                    avgkh = 0.3513;
                    break;
...

He said this already:

 

the idea to have an average hs/k (and k/m, k/h, accuracy) came from istats (i-stats.net).

they have a huge database of over 400000 bf3 players and calculated the averages for every weapon.

i "adjusted" these average values to be a little more forgiving. think about lmg with 200 rounds per mag having a very (too) low accuracy. and things like that.

[ImportBot]

Originally Posted by dyn*:

 

I do like the sound of this plugin.. however.. I am a bit hesitant to install it because we're already using TrueBalancer and Insane limits on the box (same IP).

 

Others who are also in the same boat as me seem to be giving it a try. What's your take if you've tried it ?

[ImportBot]

Originally Posted by HexaCanon*:

 

If you look at the source code, you will see that the average values are already set.

 

Code:

switch (weaponname)
            {
                case "MK11":
                    avgkh = 0.314;
                    break;
                case "SVD":
                    avgkh = 0.3206;
                    break;
                case "SKS":
                    avgkh = 0.2555;
                    break;
                case "M39":
                    avgkh = 0.3187;
                    break;
                case "Pecheneg":
                    avgkh = 0.2093;
                    break;
                case "G36C":
                    avgkh = 0.1638;
                    break;
                case "M412 Rex":
                    avgkh = 0.3513;
                    break;
...

He said this already:

cool, time to work on my IL limit ^^

[ImportBot]

Originally Posted by Tomgun*:

 

EA_C7F378A53BFDE71E53CF2C662870C73C Guid Permanent SUSPECTED CHEATING: Letintin AN94 hs/k: 129.25%

 

http://battlelog.battlefield.com/bf3.172212534/pc/

 

EA_7A99C9E45D54F0AF58E65BD11BCB1878 Guid Permanent SUSPECTED CHEATING: Kylarn Taurus 44 scoped hs/k: 99.54%

 

http://battlelog.battlefield.com/bf3.363522063/pc/

I would agree with both of the above but cant work out its 88.47% on the below

 

EA_D2A3A8591399C616EDBB4E59DFDC1677 Guid Permanent SUSPECTED CHEATING: JuanSnowJumKun M249 k/h: 88.47%

 

http://battlelog.battlefield.com/bf3.568325642/pc/

[ImportBot]

Originally Posted by russel5*:

 

but cant work out its 88.47% on the below

that means killing 865 times he was hit in the victim 977 times. Percentage of k / h 88% is impossible. Perhaps he used no recoil / no spread, ie all the bullets were flying just a sacrifice

IMHO

[ImportBot]

Originally Posted by russel5*:

 

Hi, grizzlybeer. See this http://battlelog.battlefield.com/bf3.387332257/pc/

This is not impossible, but your plugin not check it

[ImportBot]

Originally Posted by grizzlybeer*:

 

Hi, grizzlybeer. See this http://battlelog.battlefield.com/bf3.387332257/pc/

This is not impossible, but your plugin not check it

55 hs with 100 kills is a lot, i agree. but not anything i want the plugin to act upon. the average is ~23% on the rex.

but given 55% hs with >130 kills still wouldnt trigger anything. its 239% above average. suspicious would be considered 270%, impossible 320%

 

i would consider this player not even suspicious.

keep in mind that there will always be cheaters "below the radar" of any detection engine and i want this plugin to only act if there really is proof.

 

still i will keep your idea in mind, but for now i need 130 kills :-)

[ImportBot]

Originally Posted by russel5*:

 

ok, and what about Psay?

[ImportBot]

Originally Posted by grizzlybeer*:

 

coming :-) but first i would love to see this plugin in the procon gui packages tab :-)

phil is already on this. it should not take long. that will make things much easier for most users i think.

 

expect the requested features then

[ImportBot]

Originally Posted by PapaCharlie9*:

 

cool, time to work on my IL limit ^^

Hang on, Hex, I'm going to combine the data from CD and Chuck Norris into a single table that everyone can use for limits. I'm going to use Singh400 reciprocal form (shots / kills instead of kills / shots).

[ImportBot]

Originally Posted by LumpyNutZ*:

 

Hang on, Hex, I'm going to combine the data from CD and Chuck Norris into a single table that everyone can use for limits. I'm going to use Singh400 reciprocal form (shots / kills instead of kills / shots).

+1 after opening server again, we really need something like this.

We are not so far for getting hackfree servers anymore

[ImportBot]

Originally Posted by ty_ger07*:

 

cool, time to work on my IL limit ^^

Hang on, Hex, I'm going to combine the data from CD and Chuck Norris into a single table that everyone can use for limits. I'm going to use Singh400 reciprocal form (shots / kills instead of kills / shots).

+1 after opening server again, we really need something like this.

We are not so far for getting hackfree servers anymore

Why are you guys trashing other people's threads with your off topic conversations? Both this thread and the chuck norris thread have off topic comments for a future competing plugin code. Move that talk to your own thread!

 

Can an admin please move all this off topic stuff to this thread:

www.phogue.net/forumvb/showth...K-IN-PROGRESS)*

 

Thank you!

[ImportBot]

Originally Posted by HexaCanon*:

 

Why are you guys trashing other people's threads with your off topic conversations? Both this thread and the chuck norris thread have off topic comments for a future competing plugin code. Move that talk to your own thread!

 

Can an admin please move all this off topic stuff to this thread:

www.phogue.net/forumvb/showth...K-IN-PROGRESS)*

 

Thank you!

i am only here to see how he did the headshot/kill percentages.

 

i gave him direct questions.

[ImportBot]

Originally Posted by pharbehind*:

 

This plugin just kicked this guy with a 99% rating: quiksilverUK

[ImportBot]

Originally Posted by grizzlybeer*:

 

This plugin just kicked this guy with a 99% rating: quiksilverUK

look at the hs/k on the asval and ak74m:

CD - IMPOSSIBLE STATS: quiksilverUK AS-VAL hs/k: 122,33% (4,73)

CD - IMPOSSIBLE STATS: quiksilverUK AK74M hs/k: 76,34% (3,24)

 

all other stats are fine, but still these 2 remain.

[ImportBot]

Originally Posted by pharbehind*:

 

Is it possible to cheat with specific weapons, and not across the board?

[ImportBot]

Originally Posted by PapaCharlie9*:

 

Is it possible to cheat with specific weapons, and not across the board?

Certainly. The Medpack one-shot kill is just one example.

[ImportBot]

Originally Posted by IAF-SDS*:

 

Hi

again I got banned from a server for "cheating" with the following reason:

Game disconnected: you have been banned from this server. Stated reason: SUSPECTED CHEATING: BattleRushFPS XP1 QBU88 hs/k: 125%

 

1. I didn't use the QBU88 in this round I used only the M40A5.

2. How to fix this problem (because it isn't the first time)? by the way my PunktBuster is updated and my gamefiles are not edited (no colortweaker or something like that)

and I NEVER used OR currently using a cheat/hack in Battlefield 3

 

Server: http://battlelog.battlefield.com/bf3...-1000-Tickets/

Battlereport: http://battlelog.battlefield.com/bf3...4/1/325033265/

 

Regards BattleRushFPS

grizzlybeer,

 

The above quoted thread from the BF3 PC forum at http://battlelog.battlefield.com/bf3...4348370076676/ looks like a CheatDetector ban to me, but I may be wrong.

 

How is it possible for a player, cheat or otherwise, to get hs/kr higher than 100%?

 

Is it a glitch in BF3, or can an aimbot/hack get you more than 100% in headshots as compared to total kills?

 

If one bullet is killing more than one person like shown in this video: http://www.youtube.com/watch_v=7GCXriHgIRk [youtube.com] would that result in more headshots than kills?

 

 

His profile shows http://battlelog.battlefield.com/bf3.325033265/pc/

 

Weapon: QBU-88

Kills 292

Headshots 365 (how is it possible to have more headshots than kills_)

 

 

If this is a bug in the game, do you want to have CheatDetector ignore a particular weapon's headshot number when it exceeds total kills since it should not be possible to get more headshots than kills in quantity with that particular weapon?

[ImportBot]

Originally Posted by dyn*:

 

grizzlybeer,

 

If this is a bug in the game, do you want to have CheatDetector ignore any stats that exceed 100% since it should not be possible to get more headshots than kills with a particular weapon, even if one bullet kills more than one person (assuming that in itself isn't done with cheating)?

Not a bug.

 

To my knowledge there are 2 ways to get a head shot kill percentage above 100:

 

1: Fire 1 bullet and kill multiple people. Say 2 people are in a line and the bullet goes through 2 heads. Boom. 200%.

2: There have been hacks which produce kills (head shots) without actually requiring bullets to be fired. I'm sure you can find some youtube videos of individuals just standing around and killing everyone.... the whole time the hacker isn't even firing bullets. Not sure if this method is still possible - but this is the reason why some cheaters had 2,000 % headshot.

[ImportBot]

Originally Posted by Ike*:

 

Isn't the headshot statistic kills that were headshots. If so, killing 2 people with 1 bullet should not let this go over 100 like it would with accuracy

 

Sent from my Nexus 7

[ImportBot]

Originally Posted by MorpheusX(AUT)*:

 

A sniper should maybe be able to get a HS ratio of over 100%?

Imagine 1 soldier standing right behind another one. The sniper headshots the first one, bullet goes through and headshots the second one. That would make 200% headshots with 1 bullet, wouldn't it?

[ImportBot]

Originally Posted by Ike*:

 

A sniper should maybe be able to get a HS ratio of over 100%?

Imagine 1 soldier standing right behind another one. The sniper headshots the first one, bullet goes through and headshots the second one. That would make 200% headshots with 1 bullet, wouldn't it?

Comparing it to kills though, 2 kills with 2 headshots

 

Sent from my Nexus 7