Recent content by valeskas

Seems to be fixed now. At a glance, the client RemoveBuff now sends instigatorId to server and it no longer echos it back.

Actually, not exactly true. dnSpy supports debugging unity games, with conditional breakpoints and decompiled source and the ability to change source of specific methods and compile it back. All the good stuff. Any good programmer can debug it, given the easy repro. (waiting 150 hours is not...

In this case in the server log: 2025-10-06T15:13:44 4744.641 EXC Attempted to read past the end of the stream. at PooledBinaryReader.ReadByte () [0x00008] in <511861985a8443b48f3c22e037ca2d7e>:0 at PooledBinaryReader.ReadBoolean () [0x00000] in <511861985a8443b48f3c22e037ca2d7e>:0 at...

The server log is from a different date: 2025-10-06T19:05:58 18678.715 INF World.Unload

Investigated the issue under the debugger. My guess is this: The client on spawning does EntityBuffs.AddBuff in Equipment.Apply. After that, in EntityAlive.switchModelView it reequips stuff, calling EntityBuffs.RemoveBuff, then EntityBuffs.AddBuff again. EntityBuffs.RemoveBuff does not pass...

On a second thought, the idea is fine. Expecting a onSelfEquipStop first, onSelfEquipStart second is reasonable. Buffs obtained in these events should be just fixed to exist as expected.

This bug is not only about Water Purifier. It affects anything relying on buffs applied at onSelfEquipStart on dedicated server. For example Nerd Outfit without intellect mastery 3. For example Impact Bracing. For example Nerd Boots, Rogue Boots, Miner Boots, Raider Boots Probably Bandolier, a...

Windows can be configured to write a memory dump on blue screen, and ppl can take a look at it. You can ask chatgpt for details "how to configure windows to write a full memory dump on bluescreen" "how to check memory dump to determine what cause the blue screen?"

Fun. Looking at GCUtils.PostUnload, public static void PostUnload() { GC.WaitForPendingFinalizers(); GCUtils.FullCollect(); } it can be some finalizer, not garbage collecting itself. If you feel like it, you can download process explorer, configure MS public...

Horde nights struggle not because CPU is weak. They struggle because alghorithms are not tested and optimized for big player counts. Algorithims with quadratic or worse scaling are not something that can be compensated with hardware. If you want it improved, best approach is to gift devs...

Seems to be this error from log: 2025-09-15T23:02:44 14.633 INF [EOS] Connection test failed: Invalid URI: The hostname could not be parsed. I think it tries to open https://api.epicgames.dev/sdk/v1/default in this particular case. Idk why it can fail, maybe try opening this url in the browser...

You did not change anything, but the devs changed the launcher. And its unlikely that they have tested it on computers with non-english locales. My guess is not particularly important though, since you are playing anyway.

My guess is that you have non english locale, that renders date a bit differently. And launcher has a bug about interacting with localized dates when it should not. Possible workaround - bypass the launcher by launching the game directly, it should be something like "c:\Program Files...

Seems to be fixed in 2.3b9. Nice. For those curious - now SetOversizedStabilityDirty uses spatial index and checks the stuff in the block chunk, not entire map.

I assumed the change is about driving physics. Vehicles steer differently in storms, and now 4x4 no longer do that.