Hacker News new | past | comments | ask | show | jobs | submit login

If steam spends some time working with the larger "anti-cheat" vendors, I think they could really bump up the list. Many of the 'borked' games use anti-cheat, like battle-eye.



I'd much rather those games just become unplayable.

I've had so much trouble with anti-cheat systems that I will now refuse to buy or play any game that uses easyanticheat or battleye. I don't care how much it costs, how many of my friends are playing it, I'm not giving another penny to any company that uses those programs.

I get that it's not an easy problem to solve, but i'm so goddamn tired of having to shut down services dedicated to controlling fan speeds in my machine, having to disable network adapters, uninstall vpn software (not just turn it off, uninstall it), and wasting hours trying to figure out why it would crash on startup only to find that it was finding a pcap.dll in a random folder in my downloads from when I was messing with some traffic sniffing stuff on this system like a year before that, and it decided to crash without any warning or error...

Cheaters suck in games, and they can ruin it for many others, but there has to be another way other than forcing your paying users to jump through this many hoops. And if there isn't, then i'm happy to be doomed to playing against/with hackers and cheaters rather than not being able to play at all.

And I know it's kind of co opting the purpose of Proton, but if they refused to accommodate these anti-cheat systems, then maybe either the game developer or the anti-cheat vendor themselves will see some of the potential paying customers they are excluding and will significantly retool their systems to work.


> Cheaters suck in games, and they can ruin it for many others, but there has to be another way other than forcing your paying users to jump through this many hoops. And if there isn't, then i'm happy to be doomed to playing against/with hackers and cheaters rather than not being able to play at all.

Back in my day, we just played on community servers with moderators who'd just ban cheaters!

I know that's not really a solution with the scale of today's online games, but cheating isn't a new problem. It's only become a problem of today's world of matchmaking systems, global public servers, and the removal of community hosted servers. Still, I feel like just hiring a few more people to manually check suspected cheaters would reduce the need for such heavy handed anti-cheat.

On the other hand, Valve run 2 of the most popular online games (CS:GO and Dota 2), which both have Linux clients, and don't have the same level of intrusive anti-cheat compared to EAC and BattlEye.


> I know that's not really a solution with the scale of today's online games

Most of these games have not scaled up. "Scale" is not the issue here.

The issue is that companies want to provide their own server hosting, and charge users for the privilege of hosting games. This leaves no room for community moderation.


I've gotta wonder about your system config. I've got Wireshark, Viscosity, probably a TUN driver or two from another VPN install, fan control, system monitoring and I've had no problem at all.

Not saying it doesn't happen - I'm pretty sure there was one that didn't like VirtualBox and I haven't tried again since switching to Hyper V. Maybe I got lucky?


Yeah i'm sure i'm one of the outliers, but if I'm not voting with my wallet then why would they ever listen.

easyanticheat is by far the worst i've ever experienced. That was the one that would crash if it saw pcap.dll, and I use virtualbox as my VM software of choice so that could be part of it as well.

The fan control system is a bit weird since it's a pretty unknown little program, but it took me a while to find because at least easyanticheat wouldn't tell me which program was causing the error, so I had to manually close them until it worked. This is what actually put me over the edge and made it so I'll never buy another game with easyanticheat again unless they significantly rework how much access and power they need.

There are so many games out there now, and so many different developers and styles, that I don't need to put up with these kind of problems to play a good game. Especially when that software is almost always beatable by how often I'd still see cheaters in the games, which meant that the only ones being punished by this are the lazy hackers, or paying customers.


> Yeah i'm sure i'm one of the outliers, but if I'm not voting with my wallet then why would they ever listen.

I mean, if you are an outlier, voting with your wallet is never going to work. That is the problem of being an outlier.


It probably won't work, sure, but I'm not going to continue to support games that I can't play without downloading cracks. And I feel they are crossing a line requiring that much control and integration with my system to run.

And being an outlier doesn't mean you are insignificant. Sure, the latest Call Of Duty won't miss my income, but a smaller dev that was thinking about using easyanticheat because it was cheap and easy to integrate might think twice when reading comments from me, or if their demographic overlaps with people like me that won't put up with it.

I can't in good conscience talk about how much I dislike invasive anti-cheat programs while still buying every game I want that has them.


Depending what is his objective. If he would want to change their bahaviour, then it will not work. If he want to spare himself headaches and wasted time making their product work, it will work perfectly.


There is another way. Statistical analysis on game performance. No one is 3 standard deviations above the mean in terms of their skill. So ban anyone who hits that. Cutting that off leaves cheaters that are cheating just to be as good as pretty good players, but it keeps them defeatable.


> There is another way. Statistical analysis on game performance. No one is 3 standard deviations above the mean in terms of their skill.

Assuming normal distribution, 0.1% should be; with top games reaching 1+ million active players, that'd be on the order of 1,000+ players.

And if you start banning people for being too good with no other evidence of cheating, you'll drive of the people who would likely be your most avid non-cheating players.


And I'm not sure the assumption of a normal distribution is even correct. Players will naturally get better at a game the more they play it. Most measures of skill available to a game will end up picking this up more strongly than anything else. The end result would be automatically banning anyone who has been playing the game for too long -- this would be terrible for retention...


> No one is 3 standard deviations above the mean in terms of their skill. //

Maybe, but then surely the cheaters will just set their cheats to be 2.99 sd above the mean. Probably best to use it as a signal rather than a fixed limit.


I set up a separate user account for Steam precisely because I don't want anti-cheats sniffing my main account's memory through /proc. Anti-cheat can go fuck itself, doubly hard if it tries to install a kernel module just to play a damn game.


Unfortunately it probably will want to have itself loaded as a kernel module, if they ever were to decide to even begin supporting it.

Anti-cheat has been one of the biggest obstacles standing in the way of gaming on Linux for a long time.


Considering the lengths these anti-cheat vendors go to to detect bad behavior (including but not limited to kernel-mode drivers and kernel-hooks), I doubt they have any incentive to go along with this until a massive amount of momentum for gaming on Linux exists.


Valve and steam are large enough to provide incentive themselves if they so choose.


I remember not being able to play anything BattlEye because it breaks on a Windows Insider version late 2017, and stuck for half a year until the next stable update cycle. It’ll be a loooong time until they even consider to support anything remotely interesting at all.


Is it possible to do those anti-cheat checks on wine/linux? Valve could work with the vendors to have them work on it.


From what I understand, the anti-cheat checks work, it's just that the permissions model on Linux is so much more coarse than on Windows that there isn't really a good way to tell the difference between a cheat and benign behavior. They end up with so many false positives that it's better to just not let the game run at all. Even something as simple as running "ps" on Linux looks through procfs and can show where in memory different things are loaded, probably getting flagged by the anti-cheat.


Linux being open source and gpl licensed is a problem for anti-cheats.


Why does the operating system's license matter for software running on top of it? The software can still be proprietary.


An anti-cheat relying on a kernel module would have to provide the source code of that module (the glue part, at least). One could then modify and compile that module as he want. Also, you could easily write a sandbox environment at the kernel level dedicated to a specific anti-cheat.


You can recompile every component of your system to lie to the s*y anti-cheat.


This smacks of "security through obscurity" to me. Windows has been thoroughly reverse-engineered by black-, gray- and white-hats: is there a single windows .dll, .sys or other component whose purpose is remains a mystery or cannot be subverted by a person with physical access?


It is. Any anti-cheat system that relies on having explicit control of the client is inherently flawed, but it is considered "good enough" by most people.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: