Live data from Hacker News

We outsmarted CSGO cheaters with IdentityLogger

mobeigi.com

101–110 of 372 posts

Re: We outsmarted CSGO cheaters with IdentityLogger

#101
post #53

I hope they asked permissions for storing those cookies. Otherwise they're violating various EU laws.

I'm not a lawyer, but I think this actually has some interesting things to think about. Not all cookies require consent under the ePrivacy directive, there is an exception for cookies that are "strictly necessary for the delivery of a service requested by the user". I think that'd fit in this case, since providing a cheater free experience is part of the "service" the players are looking for. At the same time, the eP…

No, that doesn't count. Companies have tried arguing that their ads' tracking cookies are strictly necessary otherwise they wouldn't be able to offer their services (ads pay the bills). And yet, they require consent.

Preventing cheaters is similar. And this is blatantly a tracking cookie.

Re: We outsmarted CSGO cheaters with IdentityLogger

#102
post #25

For UT2004, you can ban by player GUID (a hash of the CD key) or IP. With the game abandoned by Epic, a number of key generators have cropped up, which makes GUID bans useless. IP bans only go so far with VPNs costing $2 these days. The main solutions we have today are IP ban + VPN blocking using a database of known VPN subnets and adding them all to the firewall, and a similar fingerprinting technique which scans th…

How about just a whitelist? I can't imagine there are a ton of legit ut2k4 players left?

Re: We outsmarted CSGO cheaters with IdentityLogger

#103
post #61

Earlier quoted context omitted.

My idea: 1. Determine minimum human reaction times and limit movement to within those parameters on the client side. (For example a human can't swing their view around [in a fps] in a microsecond so make that impossible on the client) this will require a lot of user testing to get right, get pro players and push their limits. 2. Build a 'unified field theory' for your game world that is aware of the client side const…

This is a slippery slope which we can view in real-time looking at the speedrunning community. Many current real person runs are using strategies once thought to be computer-only. A Mario run from 2024 would be viewed as totally impossible in 2004.

This isn’t really a relevant concern for online games since speed running is mostly rehearsed play with predictable game mechanics, not inhuman response to novel stimulus.

Re: We outsmarted CSGO cheaters with IdentityLogger

#104
post #25

For UT2004, you can ban by player GUID (a hash of the CD key) or IP. With the game abandoned by Epic, a number of key generators have cropped up, which makes GUID bans useless. IP bans only go so far with VPNs costing $2 these days. The main solutions we have today are IP ban + VPN blocking using a database of known VPN subnets and adding them all to the firewall, and a similar fingerprinting technique which scans th…

What about banning VPNs?

Re: We outsmarted CSGO cheaters with IdentityLogger

#105
post #70
post #60

Earlier quoted context omitted.

This still leaves you wide open to cheaters using mobile data tethering and proxies. Have you considered more advanced network analysis? It's one of the areas I have an interest in (professionally and personally) so if you want any suggestions let me know.

> This still leaves you wide open to cheaters using mobile data tethering and proxies Is latency going to be good enough on mobile data (especially if they're also using proxies) for a FPS, though? Sure, they're using cheating software, but I wouldn't be surprised if the software gets the information it needs to cheat too late often enough for it to be useful.

I regularly played CSGO in Europe because the North American ranking system were screwed up.

I got to Supreme (2nd highest rank) with 150 ms ping. The people I queued with hit Global.

It's possible to play legitimately with very high ping. The higher ping put us at a disadvantage, but the skill gap between regions made it worth it to arbitrage.

Re: We outsmarted CSGO cheaters with IdentityLogger

#106
post #25

For UT2004, you can ban by player GUID (a hash of the CD key) or IP. With the game abandoned by Epic, a number of key generators have cropped up, which makes GUID bans useless. IP bans only go so far with VPNs costing $2 these days. The main solutions we have today are IP ban + VPN blocking using a database of known VPN subnets and adding them all to the firewall, and a similar fingerprinting technique which scans th…

sorry for the not-so-smart question.

the cheats are software, software has certain quirks, like the way it aims or the way it tracks. And I'm willing to bet it has enough distinctiveness from human aiming to be classified. Couldn't a classifier work on the behavior of the cheating software itself, rather than use IP bans?

Re: We outsmarted CSGO cheaters with IdentityLogger

#107
post #61

Earlier quoted context omitted.

My idea: 1. Determine minimum human reaction times and limit movement to within those parameters on the client side. (For example a human can't swing their view around [in a fps] in a microsecond so make that impossible on the client) this will require a lot of user testing to get right, get pro players and push their limits. 2. Build a 'unified field theory' for your game world that is aware of the client side const…

This is a slippery slope which we can view in real-time looking at the speedrunning community. Many current real person runs are using strategies once thought to be computer-only. A Mario run from 2024 would be viewed as totally impossible in 2004.

No one does multiplpayer speedruns.

Re: We outsmarted CSGO cheaters with IdentityLogger

#108
post #25

For UT2004, you can ban by player GUID (a hash of the CD key) or IP. With the game abandoned by Epic, a number of key generators have cropped up, which makes GUID bans useless. IP bans only go so far with VPNs costing $2 these days. The main solutions we have today are IP ban + VPN blocking using a database of known VPN subnets and adding them all to the firewall, and a similar fingerprinting technique which scans th…

How about just a whitelist? I can't imagine there are a ton of legit ut2k4 players left?

Yes, we have a whitelist ability also, but it is definitely a last resort. The game is mostly dead and difficult to discover for new players. We don't want that roadblock if we can avoid it.

Re: We outsmarted CSGO cheaters with IdentityLogger

#109
Couldn’t you stop cheaters by just looking at how their telemetry metrics are different from the baseline? If you get to a point where the cheater has to cheat to only be as good as a median player in the lobby in order to evade detection, you’ve effectively neutered it.

Re: We outsmarted CSGO cheaters with IdentityLogger

#110

Server side only anti-cheat is one of the problem domains that I'd really love to work on at some point in my career. This is the type of adversarial arms race that just seems really fun to think long and hard about.

The state of the art is pretty boring and you can learn about user command payloads in an afternoon.

The world is much more complex now that YOLO-based aimbots exist, and I think the real answer is that anti-cheats are now defeatable, period.

You can craft a private binary that has no hash registered to any major anti-cheat service on the client-side, and on the server-side you’re limited to what is allowed by game rules.

Since there’s no mechanisms for preventing super human reflexes, and there probably shouldn’t be, it’s an issue that cannot be solved anymore.

So you need community judgement, and that too is boring. Good players being accused of cheating in Counter Strike is a years old and entertaining problem.

Post reply on HN