Test Server: test.grim.ac
Please note that GrimAC is incompatible with ProtocolLib 4.x. If using ProtocolLib, you must update to ProtocolLib 5.x development builds: https://ci.dmulloy2.net/job/ProtocolLib/
GrimAC is an experimental open source predictive anticheat. It is free on spigot while in beta testing. It will eventually become paid and/or will include offer additionally subscription based paid checks.
The anticheat is stable for both 1.8 and 1.19 servers with both versions being tested by over a hundred players concurrently. It is written for 1.19 and has been backported.
There is a 1:1 simulation to the player's movements to prevent illegitimate actions. It detects 99.99% antikb, 0.0001% speed increases, and nearly any movement cheat tested on it. It can even detect Optifine FastMath in both reach and movement.
The entire prediction engine is ran on netty to avoid lagging the main thread. Performance on a 40 player server was taken to be 0.4% extra CPU load per player on the netty threads, with 0.0005% main thread CPU load per player (negligible). This is magnitudes faster than any other public anticheat.
Geyser players are fully exempt.
For 3.01 reach please enable both pre and post transactions in the reach config, with 1.8 players on 1.8 servers. By default, pre and post transactions are disabled to save bandwidth. 1.9 players are checked to about 3.04 reach, with an extended fight.
For any issues, please report them on the github or on discord: https://discord.gg/FNRrcGAybJ
license:
Installation notes:
If you use Geyser, place Floodgate on the backend server so grim can exempt bedrock players. Grim cannot access the Floodgate API if it is on the proxy.
If you use ViaVersion, it should be on the backend server as movement is highly dependent on client version.
Please note that GrimAC is incompatible with ProtocolLib 4.x. If using ProtocolLib, you must update to ProtocolLib 5.x development builds: https://ci.dmulloy2.net/job/ProtocolLib/
GrimAC is an experimental open source predictive anticheat. It is free on spigot while in beta testing. It will eventually become paid and/or will include offer additionally subscription based paid checks.
The anticheat is stable for both 1.8 and 1.19 servers with both versions being tested by over a hundred players concurrently. It is written for 1.19 and has been backported.
There is a 1:1 simulation to the player's movements to prevent illegitimate actions. It detects 99.99% antikb, 0.0001% speed increases, and nearly any movement cheat tested on it. It can even detect Optifine FastMath in both reach and movement.
The entire prediction engine is ran on netty to avoid lagging the main thread. Performance on a 40 player server was taken to be 0.4% extra CPU load per player on the netty threads, with 0.0005% main thread CPU load per player (negligible). This is magnitudes faster than any other public anticheat.
Geyser players are fully exempt.
For 3.01 reach please enable both pre and post transactions in the reach config, with 1.8 players on 1.8 servers. By default, pre and post transactions are disabled to save bandwidth. 1.9 players are checked to about 3.04 reach, with an extended fight.
For any issues, please report them on the github or on discord: https://discord.gg/FNRrcGAybJ
license:
- You are allowed to copy, distribute, modify, fork, sell, and take code from this plugin as your own, provided you follow GPL
- You are encouraged to look and modify the source code on github
- You may use this plugin and derivative works infinitely and indefinitely for as many servers as you wish.
Installation notes:
If you use Geyser, place Floodgate on the backend server so grim can exempt bedrock players. Grim cannot access the Floodgate API if it is on the proxy.
If you use ViaVersion, it should be on the backend server as movement is highly dependent on client version.
