Kubikx

With a .bat file: "%ProgramFiles%\Java\jre1.8.0_144\bin\java.exe" -server -Xincgc -Xmx2G -jar forge-1.11.2-13.20.1.2386-universal.jar nogui But I add the "nogui" param yesterday and it start running much better. I have no idea how the window can affect performance that much, but it does. The few lags that are still there from time to time are probably cased by mods.

I tried 1.12 with no mods and result is even worse. Average ticks are over 300 ms and game crash in few minutes. Crash log Any idea what can cause that high tick times?

I try 2 GB instead of 10 GB and it help a bit. (It uses only around 1100 MB) But it is still far from expected performance. Messages like Can't keep up! Did the system time change, or is the server overloaded? Running 6294ms behind, skipping 125 tick(s) still appearing multiple times per minute. CPU usage is low, memory usage is low and no sign what could cause this.

profile data: profile s profile u profile p profile e profile l I think profile s shows the problem, but I have no idea what exactly it means. My start params are "-Xincgc -Xmx10G". Total system RAM usage is under 50% so I think it is ok to offer 10 GB to the server. Removing GC switch have no effect on this. I tried Recommended forge version and Latest with almost same results.

Logs: fml-junk-earlystartup.log fml-server-latest.log latest.log I want to play with just few mods and the server start lagging after first player connect. I reproduce the same with clear installation of Forge (No other mods installed). My system have Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz (12 CPUs) and 32GB of RAM Steps I do: When I start server and connect I wait for world to generate. Then the average tick was over 200 ms. I fly around in creative to check if all the world is loaded, and I saw no missing chunks. I stop the server and start it again. I connect and wait for world to load. The result was the same even after 20 minutes waiting on spawn. The average tick was about 200 ms and spiking over 500 ms when I try to do something.