A Solution for all wall hacks

P

pencilhead

Guest
Hi, my first post here. :)

I play a lot of counter strike and the use of cheats is by far the worst thing about the game.

I was wondering why there are so many wallhackers, and realised that it is a flaw with the multiplayer engine, that the server tells the client where entities are when they are not in view.

I was wondering if hl2 will have this same flaw, or will Valve have improved the net code? If they fixed this it would be great, and would cut out one of the worst cheats around.

What do you people think?
 
As far as i understand about how ppl do the wallhack, Im pretty sure they find a way to use the gl_wireframe console command in multiplayer. Multiplayer is supposed to disable the use of cirtain console commands (this being one of them) but ppl figure out how to enable it.
 
They take advantage of the opengl driver... hence alot of hacks are called openglhacks
 
Generally they use a modified OpenGL DLL file which modifies the OpenGL API calls the engine makes to cause the walls to become semi-transparent, or to have player models rendered on top of everything else.

Thankfully, HL2 uses D3D only, so you won't get OpenGL-style wallhacks. You might still see the ESP-type which use the positions of players being sent to you even when they're not in view, or locks onto the sound being emanated from them, but Valve have had a while to build the system from the ground up to avoid this kind of thing.
 
Aww, I thought u had the solution! :(

can u please change the title of your thread?
 
I think you might be missing my point. If the server didn't give the client the data concerning player locations until they were in view, any form of transparent wall hack/wireframe isn't going to work. The client simply wouldn't know where the players are hence making wall hacks impossible.

I have heard that some multiplayer games use this system. I earnestly hope that Valve have implemented this fairly obvious change and at a stroke remove the possibilty of wallhacks.
 
pencil, you are right and your theory is definately a possibility.

The way the half-life1 multiplayer netcode works, is that it detects when you are going to be approaching a certain player, and then it starts sending data packets telling you where that person is shortly before you see them. That explains why wallhacks don't enable you to see everyone on the map, just people around a corner and not that far away.

If they were to change their netcode so that the data packets only start getting sent WHILE the 2 players are able to see each other, then you won't see people appearing until shortly after they are in sight (for example you are running under the tunnel in dust as a T, and there are CTs trying to snipe you. you won't see them, and they won't see you, until maybe a second (depends on ping) after you're both in each other's view, which isn't acceptable).

If there is some type of way to predict exactly when players are going to be in each others view, your method would be correct and wallhacks using that method are all negated.

I just woke up I need some cereal...

Oh yeah, and welcome! :)
 
but pencilhead, the server doesn't have time to calculate whether or not entities are in the view of the player. That would require it to do some sort of ray-tracing or something that it really doesn't have the time/resources to do.

And we can all thanks ASUS for the openGL hack, they had it as a "feature" of one of their drivers a couple years back; just check a box, no hacking required. They pulled it after the gaming industry screamed foul, but the damage was already done.
 
"I think you might be missing my point. If the server didn't give the client the data concerning player locations until they were in view, any form of transparent wall hack/wireframe isn't going to work. The client simply wouldn't know where the players are hence making wall hacks impossible."

As far as I know halflife already does this, to an extent. Player entitys outside a client's potential visable set's data doesn't get sent to that client. I dont think it checks their aim vector, but that would be a reasonable option providing the connection is stable, which on 56k modems it really isn't.
If this were the case then your enemy would appear out of nowhere due to latency. ideally one would want to start sending data when the enemy is at a certain offset from the player's field of vision. the offset would have to be pretty reliable or the above would still happen on bad connections.

HL2 as we know does not use opengl, so the driver hacks simply won't work. That's not to say there isn't a d3d alternative, if there is i'm unaware of it.

More rigorous checks for client hooks would be good too, though I haven't done much research into it and so can't say if it can be improved.

Rest assured though, I'm sure valve have thought of this ages ago and will already have a decent anti-cheat system in place, and i'm willing to bet they'll keep up to date with the latest hacks and patch the game up to compensate :)
 
Back
Top