I am running Project Ozone 3 on MultiMC.
Live in Singapore, server in US. Ping latency around 250ms. Traceroute getting maybe 5-20% packetloss.
The early game is fine when there isn't much built. But then when I have lots of machines up and running, I sometimes get prolonged lag spikes. These can last for hours, hammercore registers the lag incrementing up to over 120000ms and then I get disconnected. Before that actions still go through, like when lag is at 30000ms, I open a chest and the GUI pops up after that 30seconds.
The problem only happens like half the time though. Sometimes, the ping rises after initial loading during dimension travel, and may get some rubberbanding to the initial spot, but the ping will decrease back to around 250ms, even in the base.
I have tried troubleshooting.
- Memory usage remains at around 4-5 out of 6gb allocated.
- Checked the profiler graph and most of the memory use is either in blockentities or rendering.
- Check my cpu usage, seems okay. My secondary disk, HDD, sometimes peaks at 100% for a bit. I assume maybe swap memory? But minecraft doesn't even use up the 6gb I allocate to it.
- Check my internet connection. Speedtest 40mbps wifi, ping 5. Upgraded it via powerline adapter, now 160mbps ethernet, ping 5.
I am guessing it is something to do with the packetloss. I am thinking the client-server connection builds up a backlog of stuff to send causing the increasing ping in-game. However, I connected to VPN and found a connection with little to no packetloss on the traceroute and the issue still remains.
Another theory is bandwidth issue on the trans-oceanic connections due to the increasing bandwidth usage globally with COVID-19.
Not sure how to check/confirm any of these theories and whether there is any solution to this issue.
Any help from anyone would be greatly appreciated!