How punkbuster work(ed)

[=Death Hunter=]

I was just wondering. Punkbuster only worked if it was streamed? Or like it had to be running from a main source to work?

 

I guess i dont understand how the whole system worked. I guess the question im trying to ask is why cant the punkbuster system be like downloaded and run from our servers? Or systems like it? Is this possible?

 

Is it possible to create a program like punkbuster and run it on our servers?

[Ghostwind]

it probably cannot be downloaded because it is owned by a company that makes money off of it. According to thier website, we do not see direct costs, because they make contracts with the game companies, rather than charging the end user for the service. This is why they shut down so abruptly. The contract ended and they turned the lights off.

[stabak]

pb (i.e. parts from pb not the full system) is working, if you would you can run it on your server.

 

The problem in the whole story is that

-every player have to update his pb files manuel to the latest version, otherwise = kick

-every player who got a new install / lost his ETKey = kick

 

Even if you have not this problem pb will not be updated. Also you still have the same old known problems.

[SunLight]

There are still servers with pb, the problem is that it only detects old cheats, and not even all of them. There were cheaters around even when punkbuster wasn't dead.

And beginners who don't know how to install it properly wouldn't be able to connect, so many players whould have problem joining. In other words, the problems are more than the advantages.

 

It is certainly possible to create a custom anticheat for FA, but it would require some work and time to test it/debug it.

[=Death Hunter=]

 

It is certainly possible to create a custom anticheat for FA, but it would require some work and time to test it/debug it.

 

Thats pretty much what I figured

 

 

Thanks guys

[asp]

@SunLight, creating a program like punkbuster is more complex than you think. You could make a simple program which checks the ET folder for files which don't belong there (like .pk3) maps, but such a thing wouldn't be efficient. Punkbuster rather works like an antivirus, it scans the ET folder for strings which could be hacks (well, that's what I've figured out).

[SunLight]

@SunLight, creating a program like punkbuster is more complex than you think. You could make a simple program which checks the ET folder for files which don't belong there (like .pk3) maps, but such a thing wouldn't be efficient. Punkbuster rather works like an antivirus, it scans the ET folder for strings which could be hacks (well, that's what I've figured out).

In 2009 I was coding a server side anti-wallhack, after facing a lot of wallhackers in my clan's server I wanted to get rid of those bast*rds once and for all. I even managed to create a working prototype, but I still had to fix a couple of issues, and the author of antipro anticheat told me that it wasn't a good idea, and that it was too much cpu-intensive, so I gave up.

 

I still have somewhere a short demo made with it, if you watch it with r_shownormals you can see that when the player goes behind a box he totally disappears from view even with r_shownormals 1. So even a player with wallhack wouldn't be able to see him.

And keep in mind that back then full et source code wasn't even available, I had to do some test stuff in assembly, and check quake3 source to have an idea about some things. I never finished it, but I could have.

 

So I might be noob, and certainly I'm an amateur when it comes to this kind of things, but not to the point that I don't know how hard making something could be :, in other words if one of us had to do it, then that one would be probably me and not you but I don't wanna make the same mistake you did, and make assumptions on your knowledge or lack of it.

 

I have a couple of ideas on how I would code an anticheat myself, but It would require some time and especially a lot of debugging and testing. And of course cheaters would eventually bypass it, if they try hard.

 

A server side anti-wh cannot be bypassed, but it uses more cpu on the server, and it will inevitably create a slightly different game (some players could be invisible for you in some places, when in the original game you could see them, because traces are not always accurate, and you also need to remove 3d sounds for invisible players otherwise cheaters would do a soundhack, etc. etc.).

 

I didn't have such a complex thing (anti-wh) in mind when I talked about an anticheat, but something more simple yet effective (which of course I'm not going to write here). (More effective than SuxBuster for sure).

And remember that a simple anticheat is better than no protection at all.

 

But again I say it would require some work and testing by people who have some knowledge of c and a bit of assembly, and a fairly good knowledge of the game engine. If someone in FA has this kind of knowledge I would be of course glad to help, but it would take some time to create a working prototype, let alone all the testing.

 

Still doable, but it's much easier to review some demos and ban people that way.

[asp]

So I might be noob, and certainly I'm an amateur when it comes to this kind of things, but not to the point that I don't know how hard making something could be :, in other words if one of us had to do it, then that one would be probably me and not you but I don't wanna make the same mistake you did, and make assumptions on your knowledge or lack of it.

 

I understood from your previous post that you ment a program like Punkbuster as a "custom anti cheat". And from that assumption arised the question if you even knew how hard it'd be to create a program like Punkbuster. I just had the thought you underestimated it.

 

And, I don't have lack of knowledge. You made the same mistake as me actually.

 

 

For the rest, you can write some server side programs to protect the server more against cheaters, but it would indeed always change the game experience, even if it's written in C. If you got 50 players on a server which runs an anti-wallhack, I wouldn't want to play there cause of the lag.

 

Think all you want about the doable, I really doubt it's doable by 1 person.

 

But then again, you write like you know everything that's out there about programming, so go ahead and try.

[SunLight]

If you got 50 players on a server which runs an anti-wallhack, I wouldn't want to play there cause of the lag.

You would antiwh only enemies, not team mates. That already cuts the number in half. No antiwh for spectators as well. The game already doesn't show some of the players (the ones who are not in your pvs) so only the others remain.

Check a demo with shownormals and count how many invisible enemies you can see at any given moment, even in a crowded server.

Why I say invisible, because for the visible ones you can just check them every 2seconds or so, and assume they are still visible in between. After all, if someone is visible then runs behind a corner you know he's there already, you don't need to antiwh that.

 

For the invisible ones you obviously need to do multiple traces (stopping of course on the first positive one), and take into account prediction for player ping, leaning, jumping (just to mention the things I had already implemented back then when I made it). It would obviously use more cpu, but not necessarily cause lag. (And save a bit of bandwidth and client work for the players the server doesn't 'send' to you).

 

Of course it would be a slightly different playing experience, as you say, and as I have said as well. That is one of the main reasons for not doing it.

(The other being that, still wallhackers would have a slight advantage, in fact if you want to prevent players from appearing out of the blue, you have to do some kind of prediction, so they would see you behind the corner a bit before you do, making it even more useless. And also the problem of tracing going through bushes etc.etc.).

 

Think all you want about the doable, I really doubt it's doable by 1 person.

But then again, you write like you know everything that's out there about programming, so go ahead and try.

TZAC is made by 1 person only if I'm not mistaken. Someone more skilled than me of course, no need to say, but still 1 person.

In my post I never meant something full fledged like tzac or pb, scanning files on players hd was never my idea (and I don't really think pb does just that, by the way).

 

I actually never said you have lack of knowledge, it's an assumption of yours (well, that vague thing about pb only scanning files tricked me into doing a bit of assuming too everyone makes mistakes).

And it's actually not necessary to know 'everything about programming' as you say, if I may add.

[asp]

Don't know much about TZAC, but is it such a "wide" program as Punkbuster, or it is more limited? In the case it's almost the same, very much respect for the creator.

 

And when I said that PB only scanned files, I was a bit too quick. It's a bit wrong to say it that way. I even read somewhere that some anticheats perform heuristic searches, don't know if PB is among them.

 

And it's not necessary to know everything about programmer, but you'll have to be a seasoned programmer to write a good anti cheat, which is efficient and effective. I just don't see an amateur (with amateur I mean people who can work with arrays, a bit with classes and stuff like that but don't know any advanced programming techniques) creating such a program, that's all.

 

And I can see you have dug deeper into the case than I did, I never did actually. The only thing I know, again, is that it ain't easy to create such a program

 

PS: I'm not saying I could do it, not my cup of tea and I'm not that much of a programmer, I'm more into the analytic site of IT

Edited April 27, 2012 by asp

[SunLight]

Don't know much about TZAC, but is it such a "wide" program as Punkbuster, or it is more limited? In the case it's almost the same, very much respect for the creator.

I think it only supports et (etpro only) and cod, it started only for et. I don't like much the fact that the author is an ex cheat writer, but I think he did a nice work with tzac.

[asp]

Maybe the fact that he's an ex cheat writer explains it how he could create an anti cheat on his one?

 

Oh, well, as long as it works...