this post was submitted on 08 Nov 2024
293 points (98.7% liked)
PC Gaming
8607 readers
634 users here now
For PC gaming news and discussion. PCGamingWiki
Rules:
- Be Respectful.
- No Spam or Porn.
- No Advertising.
- No Memes.
- No Tech Support.
- No questions about buying/building computers.
- No game suggestions, friend requests, surveys, or begging.
- No Let's Plays, streams, highlight reels/montages, random videos or shorts.
- No off-topic posts/comments, within reason.
- Use the original source, no clickbait titles, no duplicates. (Submissions should be from the original source if possible, unless from paywalled or non-english sources. If the title is clickbait or lacks context you may lightly edit the title.)
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Yes, but not through standard methods. Even AI aimbot can be filtered, but the amount of RND required is likely to much for a single studio to bear alone. I believe we are more likely to see neural network trained bots largly replacing real players using an off the shielf model. Just a guess, not an expert.
There is already a solution using relatively simple analytics and building a profile of the player. It becomes very easy to find cheaters because it is easy to analyze how fast and directionally they aim. It is obvious when someone is using macros for instance or a aimbot.
The problem is this does not require intrusive programs that are essentially spyware for your OS. This is what attracts the big studios to these solutions not their effectiveness.
There is a workable solution but let's be honest. Cheaters are often whales and spend a lot of time and money on the game. It is bad engagement to send them away.
Big studios already recognize this. So to be blunt they allow a certain amount of cheating because they don't want to really solve the problem.
I was speaking to the long term, 5-10 year in.the future. Analytics is a current solution and as far as I know works well. I was just talking vaguely about long term problems and solutions.
I think the best thing I’ve heard for long term solutions is to fix a lot of the cheating using server side solutions. In a game like CoD, that means the server doesn’t send you player positions unless you absolutely need to know them.
The other thing honestly is just increasing the investment required to cheat. That could mean that in order to play competitive game modes, you need to have signed in at least once for 4 weeks straight and played the game. Or you need to be a certain level. Issue hardware bans and IP bans to people. Require phone number verification.
What those things do as barriers is actually increase the potency of current detection methods. This should also carry over to accounts. I’m not sure why steams VAC ban system isn’t more popular. As in accounts need to be flagged as a whole when cheating in just one game is found.
There are many solutions but it’s just not a big deal for companies as the prior person said. Plenty could be done to at least make cheating harder and cost more time/money. But that won’t happen
Presumably because this opens players to significantly damaging abuse from server operators. Players aren't the only ones who fuck around.
I don’t mean individual servers. What I more meant was let’s say a game uses a standardized anti-cheat. Like EasyAntiCheat or Battleye or similar. And whoever runs your game service (Steam, PSN, Xbox) can vet these anti cheat programs and allow them to create a record on your account of cheating.
And obviously these things get false flags so you can account for that, give people strikes and allow appeals. And games would have the option of banning you for: having too many strikes total, violating only a specific anti-cheat X times, or ignoring this system except to place extra suspicion and resources on those already having strikes.
Also having an account tied to hardware is a no brainer and I’m surprised that this doesn’t get employed often. I know IDs can be spoofed but that’s another barrier potentially.
Cod is peer to peer. Clients host the game server.
They use a hybrid system now and only use peer to peer when dedicated servers aren’t enough, so they could just swap to purely dedicated servers.
However ignoring that, even a peer to peer system can do similar tricks if you don’t isolate the host peer to just one machine. That can even be done by spot checking with a company owned server. You use the server as a verification peer and have it as a backup host to the assigned peer. If your verification peer gets different ram values or what not, you shut the server down at the very least and place that peer on a suspicion list.
But even if they went the cheap route, just distribute the peer network. Let’s say that you have a game of 12 people. You could make it so that each peer is only assigned a certain part of the simulation and players (with overlap on assignments) and cannot track the entire simulation. It’s more complicated than a single server hiding info from you, but they could at least make it to where you’d need multiple infected peers to take over a lobby.
I think you were spot on about training a neural network with player data. It is already happening without a doubt.