Seriously why is it so impossible to conquer hackers. I really don't understand why it is so difficult.!!!!!!
Because the game makers don't control the computers that run the clients. Think about an FPS with a wall hack. The server will need to send the position of all of the players to the client. A hacked client can hook into that information, parse that information, create overlays and then hook in with the display info to create the wall hack with the overlays. The server is never going to realise that any of that is happening. From the client software's perspective it is just making function calls as per usual, and doesn't really see what the hack is doing. The hack is running outside of the game itself and just grabbing and manipulation information on the fly most likely through memory. The game can't really sense or feel or intuitively know that is happening. There are probably more sophisticated tools, that can monitor the executable and memory as it runs and see if anyone is manipulating the information, but that becomes a cat and mouse game really, with one patch comes the next hack. The easy answer you might think is not to send all of the positional data but only what the player sees, but there is interpolation going on on the client side, so that a players position can be sort of guessed between server ticks. This avoids lag and overhead that would be present without some form of interpolation. Basically your question is kind of like asking why is it impossible not to get viruses. Its basically the same idea. So this is me half guessing and half remembering so don't take it as gospel or anything.
Reactive hacker banning is much more profitable than hacker prevention. Which is basically allowing hackers to play the game for some period of time until they are banned. When there is a will there is a way. The hackers have the will, the companies making the games do not.
Some games have more problems with it than others. I think a lot of the problem has to do with what functions are handled on the server and which are handled by the client. the more of the work done on the client side the greater the opportunity for hackers.
You could sign the executable, run it in a signed environment with a TPM, and encrypt all communication between the client and server. That would dramatically reduce the attack surface for hacking a multiplayer game. But then people go nuts over UWP apps that even vaguely approach these requirements, so nobody wants this sort of security actually implemented because of the downsides in the lack of user customisation.
Mod heavy single player games like Sykrim, don't need to run as a UWP, but a game like Battlefield 1 where there is no customisation and is almost exclusively online-multiplayer oriented probably would benefit from running as a UWP I don't think there is anything necessarily wrong running signed executables in a sandbox, the problem is the implementation. You need to leave a hook for performance monitoring tools. You need to be able to side-load applications into that environment from third party vendors like Steam, GOG, etc. There was at least a lack an ability to run a Gsync/Freesync monitors in UWP (that may have changed) The list of negatives goes on and on and they tend to remove the advantages of owning a PC. Running games on UWP is potentially a good thing, but there needs to be way more done to bring it to where it needs to be first.
Applications can provide an API for this, but hooking into active programs will need to be prohibited. This is already the case by default (as of November 2015 update 1115, prior to that it was a user option). Fixed as of May.
Like I said it is potentially a good thing but they need to allow the same functionality as we have now for it be worth it. I imagine that most problems can be solved but the mods is always going to be a big that will never make it across.
