- Joined
- Nov 17, 2010
- Sythe Gold
- 0
- Messages
- 58
Well with the latest revision client, fagex has implemented their latest technology for bot detection. (Apparently this has been in the client for ages, but this and new systems are in full-effect as of now) they can fully detect when methods and variables within the client are altered. The client goes through a list of checks, and if an exception is caught, a packet is sent to the server and the account is flagged. This particular detection system is called ClassChecking.
With this being discovered, it's safe to say that censored is 100% detectable. censored (to my knowledge) edits methods within the client using bcel. Everything about censored at this point in time is fully detectable. I don't know much about Nexus, perhaps you guys can tell me more.
Nexus uses reflection, which means it takes the client and interprets the data and methods and uses it to our advantage(in simple terms, of course). Meaning nothing is really altered within the client. But someone told me that Nexus still does change methods within the client, however I'm not sure about this. Anybody have any information regarding this?
This new detection system also includes detecting when the mouse jumps from (x1,y1) to (x2,y2), as opposed to moving, as if the mouse were actually being moved by a human. This has been in previous revisions but has been fully implemented in the latest one, 634.
With all the new ways of detecting when someone has deob'd the client or manipulated it, fagex finally has the technology to reimplement free-trade and the wilderness and fully detect botters and RWT'ers. This is the only reason they are even considering it.
EDIT: Just for clarification; rsbots.net, censored, and any website that you purchase an auth code from for the same botclient, that's all Nexus.
I'm learning this as I go, so it may not be 100% accurate. But this is from looking at the packet:
Quote:
The client starts a system called ClassChecking and checks methods for errors and altercations in the client. queryID is sent first as an integer for each check, then errorID is set to 0 and if any exceptions are caught, it sets to a negative number. When errorID is at a negative number, the client sends the byte to the server and stops writing data. If the errorID is 0, the client will continue writing the data as normal.
Apparently this has been in the client for ages, but this and new systems are in full-effect as of now. It can be fully avoided by injecting the original classes at runtime. But I'm not sure what bot clients do that, if any at all.
what they gonna do ban like 20% of their members?




























