I mostly play online shooters. Been playing them for about 25 years and the amount of hackers in various games is unbelievable and it pisses me off how much its down played everywhere. I cant be the only one...
Now, there are various articles here and there that briefly mention it when its extreme for a particular game. Places like Reddit seems to have a complete ban on letting players show examples of it, and I believe for a few reasons.
First, they probably dont want to discourage new players from trying the game if they see people complaining about cheaters. Second reason being, I believe, they dont seem to want to spend much money resolving the issue.
I feel the largest drawback in this policy is that no one seems to really be discussing it and therefor we are not addressing it. Wouldnt it be best if there were thousands of people brainstorming a solution of some kind?
Two games I spend most of my gaming time on are Overwatch and Apex Legends. Overwatch doesnt have a large advantage to using Aimbot or wall hacks given the nature of the games so its pretty rare that I cross another player that is even suspect.
Apex Legends on the other hand is out of control. I am not a pro player but I have 1,000 hours on the game and play in diamond rank so I am not terrible player and have pretty decent mechanics. That said I feel like I have and extremely good sense of when I come across a cheater, especially when I view them afterward.
Living on the west coast puts me in a time zone where something happens at a specific time of day. My friends and I call it the Chinese hour. Around 1am Pacific very low level players start to shit on me. Every other match is lost to a cheater, and many times its a whole squad of cheaters. Their ping is so bad that I can see them skipping around. Their looking and movement is garbage yet they know where everyone is and have laser aim. Its stupidly obvious. I cant express how annoying this is and I cant understand why we cant resolve the majority of this issue.
First off why cant they just region lock Asia from the US? That would tamper down some percent of them. Second, why dont they just block users based on Ping? If you have 250ms ping then you cant connect to the server.
Another proposal is just autoban or issue warning based on accuracy? We have data from the pro players and we know what the best of the best can do. Anyone who have better aim than the pros should get a few warnings then autoban. Another way would be to autoban based on sight snapping to enemy targets through walls. The games know where everyone is and what they are looking at. So if another player is snapping to targets repeatedly through walls that they clearly wouldnt know the position of autoban. Its one thing if they look toward an area they hear footsets or gun fire, but its another thing when they are having a hard time shooting another player because they keep snapping to another target behind an enemy they are engaging.
I give and example. Cheater is inside a building, holding looking at a door, they can see through the walls and press a button to aimbot when firing so the first person through is downed. Then they shield and crawl back out side. Its at this time you can see them trying to aim at another player peaking the door, but keep snapping to the downed target. I have seen this, time and again. I have seen them snapping to targets far off in the distant that are directly behind and enemy that is near.
With little consideration I have just thrown out a few ideas. I am sure other people could develop or expand on these concepts to eliminate cheats with little human review. In general a cheating account tends to be new. You could probably expend the parameters of very old accounts, and tighten them on newer accounts, especially from problem regions.
I would love to hear from someone who is knowledgeable in this area or field. Why is dealing with cheating software so difficult? why isnt there a team automatically downloading the software from cheating websites and just reverse engineering it? The software sells, and eventually people get banned, and just create a new account immediately. Eventually there is a fix for that software and the cheaters just update to get around it and it starts all over again. Why isnt there someone just spending 20 bucks immediately the date the update is released and creating a fix?
My intuition is that they dont want to really invest the time and money into it and if they ignore its its mostly fine as long as they dont lose players and if they can keep people from talking about it then the majority of players wont realize how bad the issue really is.
Thoughts? Other than "git gud, noob" tolling?
This would limit the cheats "max skill" on the cost of the max being set in stone, also if poorly implemented it would lead to false positives which is a problem with all your solutions especially since there is no method to prove your innocence.
Also possible to implement to some extent and in your example it is clear cut, however same as previously this would only shrink obvious cheats. Since how do you determine if they heard the footsteps or not?
You do not have full control over the system which runs the game and the input, and there is a large variety of actions/variables which is hard to distinguish legitimate and non legitimate. And In an arms race the cost is to much to get to the fine grain required unless of course you want to play something with very limited set of variables.
Of course you could increase the servers control of the hardware aka stadia(which would increase the difficult of cheating since you no longer can get the clients data to parse from and now would need to parse from the image/sound itself, but this would only slow it down until the individual get the processing power/code to do so) But that would to give even more control to corporations which as you can see is not used in your best interest.
Learn to code?
My suggestion is a return to the old days where the servers could be in the control of the community rather than the corporations due to the fact they could then implement the granularity of control and checks they want without us requiring to give up our own control of hardware. So you could have the casual servers which has lenient cheat control/moderation and the more esport/hardcore which would perhaps mostly be run in lan matches. (of course this is a pipe dream since corporation will never give up their control due to microtransactions).
Thanks for the reply. Id like to keep the discussion going. Though I am not sure that going back and forth about specific rules or parameters themselves serves any purpose in this moment of the discussion. Rather are there descriptive ways to define human from inhuman input, in such a way that would make it useful for detecting cheaters? If the answer is definitively no, than the parameters are meaning less.
I believe there is a distinct and detectable difference between human input and software driven input. Software for Aimbots must follow a pattern, therefor its action is detectable like a fingerprint. One would think that some fairly basic algo should be able to detect a pattern fairly easily, especially once identified. Given the fact that most cheaters are customers of software they have purchases.
I have reviewed various cheat software and they offer quit a variety of features. You can choose how the aimbot actually behaves in some aspects. For one, you can select that the aim bot automatically targets the nearest limb first, with a dedicated amount of bullets first going to the limb before then locking onto the head.
With this protocol, you should easily see an inhuman input pattern emerge from those implementing the software versus those who do not. Why cant someone just code into the game, after (x) reports of cheating, review user mouse input for series of matching identical or near identical mouse input?
One could expand on such variable to develop a comprehensive anti cheating system. One could easily download the same software that I can simply google and then test myself to see how its operate without even looking at the code.
There are only so many pro players for any FPS based game, and it would take little effort for a dev team to identify those exceptionally talented players and white list them from the software if needed.
There are ways to appeal a ban, at least with Apex. My bothers account was banned after it was hacked while he was deployed for a few months. This was as simple as reviewing the login IP location. Not that the other example's solution would be as easy, but the path exist already to some extent.
Why not staff a dozen or so employees for immediate review? Give out a Twitter or discord handle for reporting live? The could focus on recent reports, as in a single player has been reported 20 times in the last hour or two. They could just review the game play of that player live.
I had one night, a while back, a few months ago. Some buddies and I are playing ranked matches and we can see the same team wiping the lobby for several hours. We can easily see the names in the kill feed, and ended up losing to the same team 4 times, winning once, over the course of 2.5 hours.
We could not have been the only people reporting them over and over. We ended up just quitting playing ranked for the night because it was to demoralizing. Similar circumstances arise over and over and we just end up not playing the ranked mode or switching games because cheaters dont restrict themselves to ranked.
I honestly think the issue is worst than most imagine it to be. Its just that most people dont have 4,000 hours on Apex Legends. I personally have only 1,000 hours but 3 of my frequent players have over 3,500 hours. I actually think that amount of time on the game is kinda nutty, but Im not here to judge.
I know what its like to come across some of the best players in the game. Everything about them is distinctive. When you see the way they move and the decisions they make, reveal their skill within moments, as an experience player. I know what its like to be shot at by the Best-of-the-best. Its very hard to compete with them but they are obviously good. This is strikingly different than been laser'd down in an instant by someone who moves and acts like a potato. This person loots at the speed of a senior citizen, the movement is terrible, but they are never in doubt of the enemy location, and have laser accuracy at a moment in need.
If software can drive a car down the road, detect faces down to a specific individual, I see no reason why gaming software that has millions of hours of human input cannot differentiate natural human input from that driven partly or wholly by software.
Not sure you can in entirety dismiss the query to a yes or no in regards to if there are methods to distinguish a human input from a computer input, since that is obviously possible, the doubt comes into how large margin of error you allow.
But how would you go ahead and create this algorithm which computes input and then classified them as valid or non valid,First you would need to gather a large database of input in the format of whatever the game uses, This input would then need to be classified into series of inputs(the data need to be over time due to a classification on each individual input would not catch anything except for input that is out of bounds and give a lot of false positives) (This step would make it probably more difficult to use data from other games due to even though fps are similar to each other there are sequences which are fair in others (e.g aimbot exist in overwatch as part of the soldiers gameplay)) thus the data itself need to be prepared for each game. You then need to also gather data for a subset of different hacks and how they affect players input and generate some data from this, then the algo needs to be trained on this data ( my knowledge is not up to date atm on how to build the algo itself with the latest techniques) and it's result tested until you at least get 90% accuracy, once this is done you then need to implement the system to run the trained algo and here comes a new problem, you do not want this algorithm to affect performance of the game (we are already screwed with communications over the internet degrading performance) so this would undoubtedly stop it from running at the same time as the match is going (might work if there are low amount of player input to consider to have it done in realtime e.g 5 vs 5 or 3 vs 3) which means you need to save all the match data and then run it after the match is over even then this solution will still give false positive especially at the start of its deployment when you have minimum of data thus getting to my point of having a process of appealing a ban (Most of these security software in gaming works by having their process hidden in how they function in attempt to stop people from learning how to bypass them too quickly which would limit appeal since it would be the corporation itself judging if they are in right or wrong and simple cases such as your brother would be fine (unless they would find that it is more profit to ban people and make them buy the game again and again) but the hard to define would most likely be defined in favor of the corporation, which I would be fine with if there were a option to run my own server of the game thus the ban would only limit my ability to play on the corporation server but not the game itself)
The above example would cost quite a bit to implement and since it would most likely only be good for one game and perhaps even only one version of the game and if the game has a low entry cost such as apex then it would only train cheat maker in making a better software thus we get into the idea of an arms race, which I think current corporation do not find to be profitable at the moment.
The current software for driving a car down the road is limited to some extent when it comes to handle things in realtime and with other human drivers and pedestrians thus can mostly be classified as assisted driving rather than self driving in that it need a human operator on standby to handle more difficult environments.
Now to my very limited understanding of current anti-cheat software (due to their own secrecy) to some extent they use machine learning and algorithms but they mainly uses them in order to try and determine if the extra process you are running is a legit process or not similar to how anti-virus software try to match signatures of possible viruses running on a computer the problem being is you can quite quickly build a virus with some intended effect and then check with all the anti-virus vendor if they catch it and if they do try to obscure the process so it no longer matches their signatures thus the defender always have to react to the attacker meaning that they will most likely be one step behind.
So once more the optimal solution is to minimize your liberty even more and give the corporation full control since it is the most cost efficient method for them in order to remove all cheating, while my solution of having more decentralized will only limit the extent of cheating by making people naturally group together with non cheaters over time but not remove it in its entirety.