Capo gets banned for cheating

[Doomkid]

Wait, isn't all this just a second reason to ban Capo? For his brief tenure in the Doomworld multiplayer section, he managed to shit talk endlessly and start conflict with just about anyone over just about anything, while simultaneously having a massive unjustified ego. Looks like there's sufficient evidence he was at the very least looking for cheats/hacks, which honestly is a pretty bad offense on its own.

If he really is cheating now, he's simply hit an ever lower low..

[mifu]

Well since we are no longer sure where we stand on this anymore, we are getting the demos re-reviewed again. For now, capo is unbanned till proven that he has cheated. Also I am going to set up some sort of community watchdog for shit like this. This group/watchdog will hopefully consist of community members who have a keen eye for this. I will announce more details once I get a decent draft as to what I want for this.

[Mobius]

So there is justice after all.

[Doomkid]

*raises hand for Zandronum neighborhood watch*

Naturally I can only be on the night shift, but still.

[Mobius]

*raises hand for Zandronum neighborhood watch*

Naturally I can only be on the night shift, but still.

He said a watchdog not a circle jerk

[Catastrophe]

Well since we are no longer sure where we stand on this anymore, we are getting the demos re-reviewed again. For now, capo is unbanned till proven that he has cheated. Also I am going to set up some sort of community watchdog for shit like this. This group/watchdog will hopefully consist of community members who have a keen eye for this. I will announce more details once I get a decent draft as to what I want for this.

I'm willing to help you out with this. Pretty much everyone I've called out in the past ended up being confirmed cheaters.

[Frits]

More fishy to me are the following things:
Capo's demo has no puffs only from him, everybody else has puffs. My demos shows he has puffs. Map14 has some weird aiming going on, and there is a 4 kill streak 30 seconds into the map that looks exactly the same on both demos, where the crosshair hits the center of the targets via a lock-on. If you look at my demo, many kills it looks like he isn't actually aiming on the person except for a few strange ones (lag/ping difference, the "raziel effect"). Map15 shows he has snap-ons, but when the target dies it jerks off aim very quickly and looks very strange, not smooth at all.

For those unaware of how an aimbot can work, it can a) only work when you hold a bound button, this means it is toggle-able. The player does not have to fire while actually using the bot either. b) The field of view can be set so the player can lock onto a target if the target is within a particular range of the crosshair. This can be set from 1-180 degrees (you have to aim onto the person before locking on (1) or the target can be fully behind you (180)).

Can someone PM me decay's POV demo?

I don't know whats up with the puffs maybe a dev can give some feedback on this?

- A lot of people aim with the ssg by snapping on to the opponent in doom, you never see someone tracking the opposite player all the time with an ssg out. Can you say when these shots happens in terms of time left on the map?
- It's normal he is off in your POV demo, that's why you always need a demo of his POV.
- He has a two aim lock-ups (aiming at nothing at these time) but these happen exactly when a player with high velocity bumps into him.
- His CG tracking is not unusually strong on map14 or map15, those players didn't even dodge.

It's hard to tell if someone is cheating on a hectic dm demo. You can find suspicious things if you want to find suspicious things. If you look at a demo from any good player in DM you are bound to find something that's not right if you really want to. Capo just shit talked after he stomped a pub and people took this too seriously.

[Shane]

The demo lacks proper evidence that he is using an aimbot. Most of the questionable shots I saw were mainly in the beginning, but those were due to desyncs with the demo and his ping. Most aimbots operate as tracking the closest enemy within sight and capo definitely did not exhibit those behaviors. Don't just view the ssg as the go to for hackers, how one performs with other weapons besides is also just as important.

As for him simply admitting that he is hacking this is just a typical capo comment meant to dig for negative feedback, however should be an offense on its own if players are now looking over demos to fact check his incredibly inappropriate boasts. And asking for hacks may be bad, but asking is not the same as posting or using hacks, like you won't see someone getting banned asking for porn sites, still just capo being capo.

I wouldn't mind helping out with demos I have experience prior when I was working with cyber when he was master admin for about a year as well as demo reviewing when I was admin on funcrusher.

[Doomkid]

*raises hand for Zandronum neighborhood watch*

Naturally I can only be on the night shift, but still.

He said a watchdog not a circle jerk

huh

[Mobius]

The demo lacks proper evidence that he is using an aimbot. Most of the questionable shots I saw were mainly in the beginning, but those were due to desyncs with the demo and his ping. Most aimbots operate as tracking the closest enemy within sight and capo definitely did not exhibit those behaviors. Don't just view the ssg as the go to for hackers, how one performs with other weapons besides is also just as important.

I think the latest one just fires when the target (enemy player) is on the center of the screen or whatever the bot presumes to be the center of your weapon. Essentially it just fires at the most "optimal" time a player is in sight (it clicks fire for you). Now whether or not it's 100 percent is unknown cause I only tested it once in shoot.wad with a bot offline and at certain angular motions I'd miss.

As far as the watchdog group I'd nominate the following

• Jenova
• Bluewizard
• Cyber
• Razgriz

I wouldn't trust a group of players without history or keen eye-sight to detect cheaters and having just regular staff to me isn't enough.

[Jenova]

Most aimbots operate as tracking the closest enemy within sight and capo definitely did not exhibit those behaviors.

skullhack, the most popular aimbot/wallhack for skulltag and probably early versions of zandronum did not operate this way. It locked onto whoever was closest to your center of view. This is demonstration of poor knowledge at best.

Definitely a good point but I think it's useless to compare modern hacks to (insert port)hook because it's pretty likely that they're completely different developers. It was easy to catch cheaters before because there was only one cheat and it was publicly accessible. If there was ever any doubt that someone was wallhacking/aimbotting, it was just a simple task of loading the hack yourself and comparing playstyles.

Now, with no public cheats and apparently a bunch of private ones, this becomes much harder. There's people offering $$ for developers to make private cheats for them on Zandronum, although I'm not sure how legit that actually is. I'd suggest creating an irc channel dedicated to having a select group of people review demos of potential cheaters separate from the staff, since it appears being a staff member and being good at the game are mutually exclusive on Zandronum.

[Frits]

I think the latest one just fires when the target (enemy player) is on the center of the screen or whatever the bot presumes to be the center of your weapon. Essentially it just fires at the most "optimal" time a player is in sight (it clicks fire for you). Now whether or not it's 100 percent is unknown cause I only tested it once in shoot.wad with a bot offline and at certain angular motions I'd miss..

This is a triggerbot. It fires when you move your cross over an enemy (or when an enemy runs into your cross). There is no aim locking or dodgy tracking which means if it's used right it becomes really hard to detect by eye. Only AC software can stop these kinds of hacks. The signs of walling and too good of a institutional knowledge are still there though.

[Mobius]

I think the latest one just fires when the target (enemy player) is on the center of the screen or whatever the bot presumes to be the center of your weapon. Essentially it just fires at the most "optimal" time a player is in sight (it clicks fire for you). Now whether or not it's 100 percent is unknown cause I only tested it once in shoot.wad with a bot offline and at certain angular motions I'd miss..

This is a triggerbot. It fires when you move your cross over an enemy (or when an enemy runs into your cross). There is no aim locking or dodgy tracking which means if it's used right it becomes really hard to detect by eye. Only AC software can stop these kinds of hacks. The signs of walling and too good of a institutional knowledge are still there though.

You are literally reiterating what I posted.

[Ænima]

I'm still not sure exactly how these hacks work specifically in Zandronum.

Do they exploit clientside prediction in some way? How much are clients "allowed" to know about other players' positions? How much memory has to be scanned to find this information and is there a way for the developers to obfuscate it or make its location dynamic?

[Torr Samaho]

Do they exploit clientside prediction in some way? How much are clients "allowed" to know about other players' positions?

Clients know the positions of all players. Depriving the clients of this knowledge would break things, like spying in client side demos.

How much memory has to be scanned to find this information and is there a way for the developers to obfuscate it or make its location dynamic?

Since the source is open, you can't really do any obfuscation.

[ZZYZX]

Even ZDaemon with all it's closed source is easy to make a hack for. Just saying. Although, in case of ZDaemon, they really seem to be obsessed with obfuscation of master server protocol which looks like horrible mess in the disassembly, yet do nothing to prevent the hacks, even though they don't need to pull any code from ZDoom and as such can introduce the required drastic changes to their code...

Either way, apparently with the client becoming opensource there are LESS hackers now, most likely because now coding hacks isn't of any interest, as anyone with VS2005 can do it.

[Ænima]

Do they exploit clientside prediction in some way? How much are clients "allowed" to know about other players' positions?

Clients know the positions of all players. Depriving the clients of this knowledge would break things, like spying in client side demos.

How much memory has to be scanned to find this information and is there a way for the developers to obfuscate it or make its location dynamic?

Since the source is open, you can't really do any obfuscation.

Ah yes, thank you for the clarification. I forgot about demos. They also need to know where to render other players, right?


Anyways I don't think it would be entirely foolproof or even worth the developers' time to pursue anti-hack methods, in that case. I guess it's just a matter of catching those who do it. Eventually they'll screw up when it really counts.

[Torr Samaho]

Ah yes, thank you for the clarification. I forgot about demos. They also need to know where to render other players, right?

Exactly. Otherwise, coop spying or free spectate mode wouldn't work in demos. Also, even if you'd axe these demo features, you'd still have to have some leeway on visibility checks (allowing wallhacks to show players slightly before they get in sight) to prevent adverse side effects for honest player (like players seemingly appear out of nowhere when they get in sight).

[Mystical-Acer]

Now the ultimate question.. Who's the next cheater? :}

[Dusk]

How much memory has to be scanned to find this information and is there a way for the developers to obfuscate it or make its location dynamic?

Zandronum is open source. A cheat does not have to scan any memory or hex edit anything. You just edit the code and make a cheat build. And thus we also cannot obfuscate it, or prevent cheats from being made in any way. The only anti-cheat is the community. This was a drawback that was accepted when the project went open source in the first place, as hacks cannot be stopped anyway.

I don't have the time or skill to review the demos myself, but I'd say that if capo's demos lack puffs it sounds somewhat suspicious. The client is, by default, responsible for rendering puffs. Therefore if there aren't any puffs at all, it is indicative of a modified client.

Even ZDaemon with all it's closed source is easy to make a hack for. Just saying. Although, in case of ZDaemon, they really seem to be obsessed with obfuscation of master server protocol which looks like horrible mess in the disassembly, yet do nothing to prevent the hacks, even though they don't need to pull any code from ZDoom and as such can introduce the required drastic changes to their code...

Either way, apparently with the client becoming opensource there are LESS hackers now, most likely because now coding hacks isn't of any interest, as anyone with VS2005 can do it.

I think there are more cheaters than you might realize but just have not been caught. In any case, thanks for your... insightful and useful... post.

I think he has a point. From my understanding, most cheats for Skulltag were made just for the hell of it. With Zandronum being open source, there's no "hell" anymore in that. I guess that's why there aren't much public cheats around and this crap is being spread in private.