is there any research done to address the cheating problem?

Started by
8 comments, last by GameDev.net 18 years ago
hello dear folks, I wonder if there has already been done some systematical research to address the cheating problems in onling gaming, if so could some one recommed an link to an article or research paper ,etc. .. thanks so much 824learner.
Advertisement
What problems specifically?
Yes, developers now build cheat codes in the games in order to support the problem cheaters have with hacking.

God bless them.

Maybe you want to be more specific?
AfroFire | Brin"The only thing that interferes with my learning is my education."-Albert Einstein
For first person shooters, you could always send the player only the data they can actually see on their screen, instead of the entire game state. Of course, this would probably mean very high server requirements, but of course, no wallhacks. Doesn't fix aimbots and stuff though.

Be more specific, there are many genres, game types, etc of online games.
You need to define what you mean by "cheating" also. There's the stuff pretty much everybody agrees is cheating like hacking your client to give you extra ability, etc.

However it's not at all unusual for developers to say something is cheating and for players to say it's just being creative. Here players are doing something that was not intended by the game designers but is allowed by the (unhacked) game client and may be perfectly acceptable in some other context.

Then there stuff like buying items with RL money. As far as the game is concerned it's just an item trade and is perfectly acceptable and intended. But the fact that RL money changed hands may be against the EULA.
-Mike
thanks, I think I am mainly interested in cheatings in MMORPG, there surely are many kinds of cheating problems in MMORPG, it will be great if there are some survey of various kind of cheats in MMORPG, or address one specific kind of cheating, ...

thanks,
824learner

Quote:Original post by jonahrowley
For first person shooters, you could always send the player only the data they can actually see on their screen, instead of the entire game state. Of course, this would probably mean very high server requirements, but of course, no wallhacks. Doesn't fix aimbots and stuff though.

Be more specific, there are many genres, game types, etc of online games.

Doesnt sending the entire games state is more costly then sending what you just see?


It's all about the wheel.Never blindly trust technoligy.I love my internal organs.Real men don't shower.Quote:Original post by Toolmaker Quote:Original post by The C modest godHow is my improoved signature?It sucks, just like you.
Quote:Original post by The C modest god
Doesnt sending the entire games state is more costly then sending what you just see?
Costly in terms of network bandwidth, yes. But if the server has to do visibility detection on everybody for a 128 person Tribes game, you're going to have some issues to work on.

[Edited by - Promit on April 1, 2006 12:00:05 PM]
SlimDX | Ventspace Blog | Twitter | Diverse teams make better games. I am currently hiring capable C++ engine developers in Baltimore, MD.
Quote:Original post by Promit
Quote:Original post by The C modest god
Doesnt sending the entire games state is more costly then sending what you just see?
Costly in terms of metwork bandwidth, yes. But if the server has to do visibility detection on everybody for a 128 person Tribes game, you're going to have some issues to work on.


This is a scenario where a physics processor (or a gpu doing physics calculations) could be very well utilized on the server.

Niko Suni

Quote:Original post by Nik02
Quote:Original post by Promit
Quote:Original post by The C modest god
Doesnt sending the entire games state is more costly then sending what you just see?
Costly in terms of metwork bandwidth, yes. But if the server has to do visibility detection on everybody for a 128 person Tribes game, you're going to have some issues to work on.


This is a scenario where a physics processor (or a gpu doing physics calculations) could be very well utilized on the server.


Some mmorpg games already do this. For example, the old neocron mmorpgfps ran the same code as the clients for collision and line of fire and sight calculations. This means every collision, movement and other physics were calculated from the 3d model of each scene. The result was correct line of fire calulations when someone tried to fire through a fence or some crowded room. I noticed this when I saw empty cartridges from a machine gun bounce on the frame of a window in synchron on both my and a friend's machine. The server side geometry was dynamically scalable, much like in the client. The trick to do this was to run each simultion frame on the server for each player and npc so the server knew what will be on the client's screen before the client even started to render it. Packet loss was compensated with dead reckogning. Setting it to zero on the client made the client run in lockstep with the server (with rendering fps as low as the physics fps). Hacking was impossible, and the server disconnected the client on the first synchronization error. The technology was dead simple, the graphics engine had the physics engine integrated. The only downside of this was that even with dx7 cards, most of the geometry calculations were done in software. (the original target hardware were vpu-s, but they had to downscale this to dx7) The npc movement mesh was accessible to players as a navigation overlay for finding important places faster. The typical maximal playable server load was around 64 players per node (aka. zone). This resulted in relatively little zones, especially in the crowded city. Zoning was done by writing out the client state and loading it back on the next zone server.

This topic is closed to new replies.

Advertisement