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."

there is no technological fix

[circletimessquare]

for a social problem

anything designed by a man can also be broken by a man

the only remedy for human antisocial activity is human social activity. no technology will change that fact. and if you think it can augment those who intend good, then you're right but you must also bear in mind that it can also augment those who intend evil

this applies to security cameras, file trading on the internet, etc. as well as game cheating

Re:Add the cheats as features to the game

[boaworm]

Because many of these games aim to be realistic, that's why people play them. Adding an "aimbot" as a powerup is not something that would have happened the 101:rd airborne when they dropped down over normandy, so when you play that scenario, neither do you want it or should have it.

Re:Add the cheats as features to the game

[Anonymous Coward]

Yeah, make an FPS game where everyone automatically has immortality, omnipotence, omnipresence & every conceivable weapon.
Sounds a lot of fun.

It seems rather futile though..

[boaworm]

The whole concept of anti-cheating is based on making a chip comparing input on mouse/keyboard to input into the program.

So how about:

1: Software that wraps this chip, and returns "true" all the time ?
2: Cheats that does not emulate keyboard or mouse input ? (like radars, spike skins, you name it)
3: Software that generate keyboard/mouse interrupts ?
4: The fact that someone would not buy a CPU/MB with anticheat stuff in it if you intend to cheat. You'd just have a dummy driver emulating this hardware or something.

This only seems to be able to solve a very small portion of cheats.

Wow!

[smiltee]

Exactly like DRM, I am sure this restrictive method will work flawlessly! I think Intel is making the right choice by using something you can't update against an entire army of hackers!

Great..

[Khyber]

I'm looking forward to the time when I can't play a game online because some POS hardware/software thinks that my MP3 or video encoder is a cheat mechanism.

Lame, very lame. And you KNOW this will eventually happen. Some harmless software program running at the same time as a game will screw your online play without lube.

Why can't the game devs shift focus away from DRM & etc. and try building a solid product that doesn't NEED a third party anti-cheat software running? It's called internal testing, FFS. You made the software yet you can't find the holes, meanwhile some smartass 15 year old Russian just reads your code and goes "Oh! Look at what we have here!"

The problem with anti-cheat software..

[Animaether]

..is that the server, at some point, has to trust the data the client is sending. Now there's client-side anti-cheat software that will do things like try and make sure that external applications (not entirely unlike the old TSR cheats of lore) aren't altering the data in RAM before it sends the info back to the server. But that client-side anti-cheat software can-and-will be defeated. Eventually there might be an anti-cheat relying on TCPM sort of things, but eventually somebody will just make a TCPM-less version indistinguishable from the TCPM type by the server.

So the only proper anti-cheat lays with the server. But there you hit a problem. You can, for example, prevent some cheats that way. Somebody lobs 2 nades while the server knows he only has 1? Cheating. Somebody moves all over the screen, faster than the player can actually run? Cheating. Wait - or a laggy connection.. or a bug. Tread with caution there. Caution means a margin. A margin means a margin for cheating. Okay, so you don't have your cheat make your player run at 200% - you just make him run at 105%. Still an advantage, and the anti-cheat won't catch it because of the margin. And even when you can detect all the -technical- cheats (more ammo, faster reloads, increased speed, greater jetpack fuel (if there's any), that leaves you with the cheats that cheat the User Input. Aimbots and the like - which can be extremely difficult to detect.

In the end, you can't 100% prevent cheating. But you can make the landscape unattractive enough to cheat in by at least trying to prevent it and having an actual human being look at suspicious behavior from time to time.

( I admin at one of the more popular Soldat servers - we're virtually cheater-free because the cheaters know they'll be busted in no time and their cheating fun ruined by us /kill'ing them (rather than banning - as they'll just be back) and ousting them in public. )

Just one problem?

[Quarters]

Multiplayer games these days have one problem. Cheating.

Really? Just one? What about:

Bad design

High prices

Poor performance

Steep system requirements

Bugs

Anti-cheat systems are flawed.

[Angelwrath]

All software anti-cheat systems are flawed because they include things other than cheating. I get kicked by Punkbuster for high ping on gaming servers.

The trouble with anti-cheat systems is that the developers have no ethical standard. They tolerate inconveniencing legitimate players to ensure that the cheaters are stopped as well. The law would see things differently. The law believes in letting some criminals go to ensure that it never punishes an innocent man. Flawed though it may be, it works far more often than it fails. Punkbuster is the complete opposite, and what's worse is that Punkbuster is full of bugs. I get kicked from servers several times a day and the only message I get is:

"Punkbuster

[Ok]"

All complaints to the company fall on deaf ears. And because EA chooses PB, I am stuck with a company granted an artifician monopoly by another company, and have no choice but to have a greatly diminished experience. Nothing is worse than screwing a gamer over in the heat of a competitive match, and that's what PB does too often.

Re:Not in the game anymore

[illegalcortex]

Not really. The real problem is that there's always a small minority that wants to cheat. They drive off the large majority that just want to play a good game.

Re:there is no technological fix

[Kjella]

To go back to classic crime term "Motive, means and opportunity". You can either try to take away their motivation, take away the means or take away the opportunity. Time and time again we've shown that to change human nature is very very difficult. To take away the means is usually to take away the tools, which are usually overbroad and takes away legitimate uses. Taking away the opportunity is usually the most appropriate and effective.

I have a lock on my door. It's to take away the opportunity. It's a lot better than trying to outlaw lockpicks and crowbars and everything else that might be used for breaking and entering, and it's a lot easier than to remove, tag or secure all my belongings so there's no point or to make sure burglars are tracked, arrested and punished with such efficiency that it doesn't pay off, even if the door was open.

Any sort of security, locks, alarms, encryption can probably be broken if not directly, then indirectly. Would it be a challenge for a pro team to break in here, install a keylogger and capture my encryption password? Hell no. But it's a pretty good defense against anyone casual, it's mostly about keeping honest people honest. Which is really a nice way of saying most people are crooks, they just haven't gotten the right opportunity yet.

Re:there is no technological fix

[AnonymousDivinity]

there is no technological fix for a social problem. anything designed by a man can also be broken by a man.

I don't know about you, but I for one like having locks on my doors. Are they 100% perfect at keeping determined individuals out? Of course not. But that's not their purpose. These kinds of measures merely need to make an activity "not worth it" to those who have some motivation (the aforementioned societal problem). Economic deterrants do work well, at least on a statistical basis.

As for cheating devices, if one were to construct an anti-cheating system that would require a hundred million dollars worth of high tech, rare equipment to break - do you think some gamer is just going to have that kind of money lying around? I'm not saying Intel's solution is of this nature, but this absurd notion on slashdot that technology cannot help/solve societal problems is total bullshit. A lot of social problems are highly context/environment dependent (mostly as a result of human psychological quirks, and evolutionary behavior), and technology can do a lot to alter the environments where people interact to the point where many harmful behaviors are discouraged or stopped altogether.

It's a reputation problem.

[Colin Smith]

You can't trust the person, you can't trust the hardware or the software you can't trust anything which comes back from the client machine.

Da fix? A cross game registry of gamers with identities linked to real addresses and bank details. Something which all the online games can query, though I'd go with hashed values for bank details/address etc rather than real ones. You get caught cheating, you get marked as such. To get rid of the marking you need a new identity.

Will it stop it? Mmm look at the athletes who take drugs, I doubt it. What getting caught would do though is ruin the gaming life in all the games which use the registry. Gaming environments could be split into two areas. One for trustworthy gamers, one for cheating scum.

 

Re:The Scarlet Letter

[Charcharodon]

I'd rather have banning. A key ban is the way to go. Sure let them sign up for a new account, that'll be $50 please. If a person makes enough of an nussance of themselves follow it up with a credit card number ban. Sure most people have more than one card, but the truly cronic bastards would be face pretty quickly with a long time ban if they didn't straighten up.

Personally I've been leaning back towards LAN parties. Cheaters are much easier to deal with, you just chuck an empty beer bottle at them after the first offense. The second offense involves dragging them out back for a little wall to wall counselling session.

Re:there is no technological fix

[The Clockwork Troll]

The social problem has an obvious solution: accountability.

If banning of an anonymous ID is the worst any cheater might endure, and they know it, they're going to operate as you would expect someone with impunity to operate.

The obvious solution has obvious problems. The social solution leaves a worse taste in our mouth than cheating. That's why we're chasing it technically.

Re:Add the cheats as features to the game

[kasperd]

The objective of the game would be to develop the best program to play the game.

That kind of games are fun for those of us who know how to code. Most of the gamers out there wouldn't stand a chance in a game that involved coding in order to play it. But there still remains a few questions, do you run the program on your own machine talking to a server? If so, is the program supposed to play by itself, or is each player going to be a person and a program cooperating? Are people with a beafy machine and a fast Internet connection supposed to have an advantage? You probably cannot design it such that they will not have an advantage. Alternatively, you submit your program, and everything is then run in a controlled environment.

STOP MODDING UP MEANINGLESS SHIBBOLETHS, PEOPLE.

[mumblestheclown]

For fark's sake people. A statement like "there is no technological fix for a social problem" is just important-sounding nonsense. Really? We seem, after all, to have prevented the problem of people physically reaching out across the internet and strangling people... I have yet to see anybody do this (as much as I'd like to sometimes). Parent poster completely ignores the obvious problem with his arguments: that ALL defense mechanisms are not about absolute defense, but about reducing the rate of successful attacks and/or increasing the barriers to entry (such as technical sophistication, equipment, time, etc) that an attacker must invest in to be successful. Security guards and alarm systems do not prevent all bank robberies - but it is safe to say that there would be more robberies if those things didn't exist. Same here. You may have technological issues as to exactly how much such a hardware defense would decrease the amount of cheating, but it seems fairly obvious that, if implemented, this figure would be greater than zero.

Re:The problem with anti-cheat software..

[bjorniac]

But then good players would be lumped together with cheats, which is no fun whatsoever. I play CSS somewhat competitively, but love playing on a few pub servers too. It's no fun at all, though, when someone who obviously hacks comes in. Sure, there are noob servers out there and anyone who's good should leave them alone so people can have fun there, but all your scheme would do is shift the problem to servers for good players.

Re:The problem with anti-cheat software..

[brkello]

The only problem with that is a lot of the time the admins just aren't very good players. They are unable to tell a good player from someone who is cheating. I know my brother and I have been banned from many CS servers by admins who can't believe we can play like that. The sad thing is, I am really not that good. Just the admins have no idea how bad they are.

Re:STOP MODDING UP MEANINGLESS SHIBBOLETHS, PEOPLE

[yada21]

We seem, after all, to have prevented the problem of people physically reaching out across the internet and strangling people...

We've prevented it the same way we've prevented trespass by teleportation and time-travel assisted stock fraud. The method revolves around not building the technology to enable it.

Incredibly poor logic and a crappy analogy.

Re:there is no technological fix

[irc.goatse.cx troll]

latency of two moving people around some obstacle means you either let them both know where eachother is before they should be able to render, or you'll be able to induce lag to allow yourself to teleport around the game which is just as bad.

Then theres issues of "can it be seen through?" for example when I replaced all fences (which in a real engine blurs to solid after some distance). Is it cheating to tweak your drivers with rivatuner to change how it blurs them so you can see through them? What about replacing the texture with an empty texture?
Replacing the enemy models with sold colors?

Even defining cheating with 100% accuracy is impossible, saying you can stop cheating is laughable.

Re:The problem with anti-cheat software..

[PachmanP]

Conversely, servers that better players frequent would be more likely to recognize and bust cheaters. Whereas the n00b servers, people would be more likely to just think the guy was really good, die alot, and give up on the game.

Re:*sigh*

[localman]

Ah, ESR. I like this quote:

"If Quake had been designed to be open-source from the beginning, the performance hack that makes see-around-corners possible could never have been considered -- and either the design wouldn't have depended on millisecond packet timing at all, or aim-bot recognition would have been built in to the server from the beginning."

Which is really just another way of saying that it wouldn't have been developed at all. Great solution.

I hate cheating too, but I'm afraid it'll always be there. I just assume on public servers that there is some cheating. When I get sick of it I set up a private game with people that I trust.

Cheers.

Re:The problem with anti-cheat software..

[brkello]

This seems like a horrible idea. Maybe it appeals to those who suck at the game...but then what is the point? You already have a filter that does this. Go on different servers and find one the suits your skill level. You don't need statistics to kick you if you have a lucky game. Besides, people would just find out the rules and play around them...making sure they empty their clip in to the wall before they get out of their spawn or purposely miss that next head shot.

Fine, use statistics to detect cheating...but to detect and kick good players? Man, why the heck are you playing online if you don't want to improve or don't want competition?

Re:The Scarlet Letter

[Anonymous Coward]

Camping is not cheating. It may be lame to you, but they're not cracking any executable or intercepting packets to make them shoot someone in the head as soon as they come around a corner. Huge difference, I'd say.

Re:The problem with anti-cheat software..

[Anonymous Coward]

And what happens if you *are* really that much better than everyone else? Are you no longer allowed to play because you are so much more skilled than other players? In other games the very best players are rewarded - do you suggest we ban them from playing in the online gaming world?

Re:The problem with anti-cheat software..

[lena_10326]

Several CS aimbots require fullbright coloured models and actually do screenscrape.

And the models would be server-side where you can't paint them a special color for the aimbot to target, which was the point of the previous post regarding John Carmack's statement.

With just a video stream, there's no way the aimbot could calculate the vertical angle to the opponent because it can't assume he's at the same elevation as you. The aimbot also doesn't know whether you're looking up or down so it doesn't know your vertical viewing angle.

For it to work, you'd have to get yourself level with the enemy and look straight 90 degrees to the ground, otherwise you're guaranteed to miss. Missing several times in a row by consistently shooting over or under someone by a few feet will arouse suspicion and you'd be labeled a cheater very quickly. Being forced to get level eliminates a major strategy of aimbot cheaters: camp in a distant high spot that's difficult to get to and provides you cover.

By the way, some games let you view left and right without turning so in that case the horizontal angle toward the opponent would also be unknown.

Aimbots really only work when it has access to game state, which is position and velocity of you and the opponent as well as your viewing angle. Those require a client-side game.

Re:The problem with anti-cheat software..

[fractoid]

This would punish cheaters, sure; but it would also punish those who just happen to be good (on that map in my case).

To be honest, I don't care whether they guy that headshots me 5 seconds before I see him, every single time, is cheating or is just really good. Either way it makes the game suck for me, and it presents no challenge for him. Users *should* be grouped by ability level (whether natural or assisted) so that everyone can actually enjoy the game while they're getting good at it. Surely you didn't have as much fun 'owning noobs with your leet sniper skillz' as you would have playing against equally skilled players?