Engine

[FrostyCoolSlug]

If you plan on keeping it open, there will have to be LARGE numbers of content checks, because we arnt just talking about 'holes' here, we will get people who will happily re-build the entire game with hacks and cracks inside it. Keeping it open source, it wouldnt be a case of 'plugging holes' because the holes may not be there to plug i hope you understabd where i'm coming from here

[EDIT]
I've become very interested in this project, so i am gonna hang around
[/EDIT]

[botmeister]

If you plan on keeping it open, there will have to be LARGE numbers of content checks, because we arnt just talking about 'holes' here, we will get people who will happily re-build the entire game with hacks and cracks inside it. Keeping it open source, it wouldnt be a case of 'plugging holes' because the holes may not be there to plug i hope you understabd where i'm coming from here

Well, people could do the same to Linux or any other open source software but I'm not aware of it happening. Usually there's an official "approved" version for people to download, just like we see with Linux and many other OSS packages.

Someone with a hacked version of the game would have a very hard time convincing cautious server operators that they should use it over the official version. On the other hand, I agree with you that preventing cheats with an open source client will be a challange, but there's usually a solution to every problem.

Anyone with anticheat experience in here that can comment on this?

[Pierre-Marie Baty]

I believe the key to making successful online games is a strong policy of game clients validation before allowing any client to connect to a game server. CRC checks, config checks, passwords, unique IDs, private/public keys, etc, etc, etc. so as to ensure every client connecting has valid, official, approved software.

[FrostyCoolSlug]

the problem would be that all these can be faked.. So long as they know what the server expects, it can in some way be faked.

Also, how will banning be implemented? Because it wont be possible to ban by anything but nickname / IP, which in most cases can be easily changed, and there would be no way to ban cheaters from all servers etc..

Maybe we should concide this as well

[botmeister]

Originally Posted by FrostyCoolSlug

the problem would be that all these can be faked.. So long as they know what the server expects, it can in some way be faked.

Also, how will banning be implemented? Because it wont be possible to ban by anything but nickname / IP, which in most cases can be easily changed, and there would be no way to ban cheaters from all servers etc..

Maybe we should concide this as well

Authentication can be done with a separate system, similar to WON or STEAM. The identification system may not have to be closed source because the connected server would do the authentication through a trusted authentication server. The trusted servers would be published by the official "owners" of the game system.

Verification of the client software is likely to be a complex topic, but I know there are ways to prevent cheats even when the client is open source. One simple method (for example) is to allow only pre-compiled "official" versions of the client to be authenticated by connected servers. The source may be open, but if you compile the code yourself, it will exclude "something" critical that only the official precompiled version contains (which is kept secret), without the correct version at hand, connections could be refused by remote servers. If need be, the server binaries can also be precompiled as official releases. Hopefully someone with cryptography and authentication expereince will join the project and lend a hand to sort this one out.

thats where Keys come in, each key lends to a GUID each GUID is, matched to the server record of install,

on install guid is sent to server the server sends a signal to pc with a new key that new key is the one that is specific to pc, and thats what is checked by the verf server, if 2 of the same key are online both keys are flagged and neither key can be used

[SoUlFaThEr]

its nice to know that my name for this company is still a hit enough to make a forum on it.

me = Chief Level designer/Waypointer

i cant wait........are you guys still serious about making a game with our amazing bots? or whats the deal here?

whats the scenario base?

[FrostyCoolSlug]

Originally Posted by Ava3ar

thats where Keys come in, each key lends to a GUID each GUID is, matched to the server record of install,

on install guid is sent to server the server sends a signal to pc with a new key that new key is the one that is specific to pc, and thats what is checked by the verf server, if 2 of the same key are online both keys are flagged and neither key can be used

interesting concept, maybe we could use the same sort of authentication for connecting, so before connections are made to the server, the game will scan hardware, then send that info, where the Auth server will then generate a key based on the info provided, and compare them, I'm sure theres something hardware based that gives each component a 'unique' id, then maybe from there, we could use Mobo, BIOS and Processor Id to generate it (The rest can be easily changed).

The 'missing code' bit is a clever idea, and in theory would work a treat, maybe with a pre-compiled 'Authentication' DLL (or .so based on platform) that comes with the server version (the only part that wouldnt be open source), and somehow make it so it cant be reverse engineerd. This will add the ability for server to add certain features and compile it themselves without having to worry about authentication. (Rather than a seperate 'Official' server. Opinions?

[mirv]

What about level editors?
I've seen quite a few open source engines around, but not many editors, which has stopped me from even considering most of them.
I just think that while a game is nice and all, people won't do much with it if there's nothing to make levels, etc, with.

mirv

[FrostyCoolSlug]

its not worth concidering level editors untill we have decided things like:

1) What engine we are gonna use (i just looked at the Ogre3d Demos, and.. well.. its great :p)
2) what type of map we are gonna use.. This one is a little more complex, WE havnt completly decided how the game is going to work, if its going to be very open (ala bf1942), which would require a lot more work (but could be more rewarding gameplay wise), or if its going to be 'closed' areas (ala Half-Life), in which case we could use BSP (which has loads of editors, but may limit gameplay). Its a careful mix, but at the end of the day, its up to the developers to decide :p