Fighting Online Game Cheating in Hardware

Monk writes "Multiplayer games these days have one problem. Cheating. Cheating is out of control because of failed attempts by software such as Punkbuster, and VALVe's Anti-cheat (VAC). Now it seems that could change change with Intel's own Anti-cheat Software/Hardware."

Wall hacking

[Anonymous Coward]

It appears to be yet more DRM designed to ensure that peripheral inputs match those received by the game.
This does not address the issue of cheats that allow the player to have information that he would otherwise not have, such as seeing through walls. Nor can it detect proxies.

Like all DRM, it sounds like it will cause legitimate users more problems than it will cause to cheats and crackers.

Wrong term.

[khasim]

How about just adding cheats as elements to the game? Players like radar? Add it.

The players don't like radar. The cheaters do.

Following your logic, the game would offer the ability to instantly kill any enemy, at any range, automatically. Regardless of intervening obstacles.

Yeah, that sounds like a fun game.

Cheaters want those because cheaters don't want to play by the same limits that everyone else does.

Re:Great..

[JNighthawk]

Maybe you've never worked on games before, but you seem very naive about it.

Sure, you can build an ultra-secure game that will be near-bulletproof, but you know what? That game wouldn't be fun. You'd have to wait for server auth before you could do anything, so this would only work for non-real time games.

And, finally, on top of what I said, the direct issue brought up (keyboard/mouse movement spoofing) cannot be fixed by games. Period.

Re:Mod me offtopic...

[Blakey Rat]

I think that was the origin of "Do not want," but it's in regular use on a ton of websites now, especially Fark.com where it's often used in comment threads and sometimes headlines. Also, a lot of "LOL Cat" images have incorporated it. It's just a popular meme at the moment.

Re:there is no technological fix

[EsbenMoseHansen]

In an online chess game, where the rules are fixed, I would challenge you to trick the server other than by hacking it. There is two things: Game data and interpretation of this data.

There are at least 2 possibities: Changing the rendering of the incoming data in a favorable manner (e.g. highlighting opponents, pickups or what-have-you) and having a custom client that plays or help you play. The classic example is the aimbot, that is a client that helps you aim your shots.

Re:The problem with anti-cheat software..

[Sigma 7]

This is why you won't find cheating in a good RTS or MMO, the server or peers can run everything in parallel.

Actually, there are easy ways to cheat in RTS/MMOs:

• Maphacks/wall hacks allw you to see through obstacles. This obstacle is FoW in RTS games, and due to their nature, most RTS clients tend to be syncronized and require having a copy of the game.
• MMOs can be botted. While not "cheating", it is an unfair advantage since it allows you to get powerful players more quickly than what is normally accepted.
• RTSs, as they generally need a copy of the game state on each client, can have critical information about other players displayed (e.g. how much resources they have/spent.)

Maybe the AI could run on a cheater's system, but I have never seen good AI in an RTS.

The AIs you see on most RTSs are lobotomized - they are written to be "good enough", whether it's through scripts or through randomness.

As an example, look at Starcraft - while the AI is relativly strong for new players, it is weak compared to the "build orders" that are posted on various websites, which are then memorized by master players. There's no reason why this can't be placed in an AI to make it stronger.

Another example is Galactic Civilizations (which isn't an RTS, but the same concept applies.) In most difficulty levels, the AI is crippled but is still a threat to most players - the only "cheat" is that it knows the location of good planets (which isn't much of an issue, since they were probing the universe before your race invented hyperspace). The threat is caused by the economic optimization - it picks the best tax rates and maximizes production efficiency. As a side note, there were reports of the AI somehow bypassing tech tree requirements - this complaint was eventually resolved, either through a patch or by identifying how they did it (e.g. tech trading with other races.)

The AI discussions were common with the game Total Annihilation, as it was the only game at the time that allowed AI patching. No matter how well you made your AI script, it was limited with implementation bugs - for example, the AI engine had a failsafe in case the script was faulty (or if it got nuked) where it would start building resource production on an economic shortage, but would never turn off the failsafe. Another bug would be the "5 peewee" rush, which could paralize the enemy AI commander and kill it.

It's not a lack of technology that limits RTS AIs - it's the lack of implementation. There hasn't been any serious attempts to make a strong AI.