David J. Romrell
- Fireswamped
- JIRAUSER551450
- Europe/Stockholm
- Yes
- No
MC-200358
MCL-19238
Missing language javafml version [33,)
It says the current java is 31.2 even though its latest
modding bug that will save tons of work if fixed and make support cost lessfake forge is pain
not resolved you idiots
not resolved
you idiots
not resolved
all
fake forge
Think of it this way increasing that value allows more cpu usage and causes increased memory usage.
so you are using a boatload of pseudo/virtual memory (doesn't show as using mem but is)
main solutions:
a) (least optimal due to experimental nature)-XX:ActiveProcessorCount=4 ... is like using a horse instead of a car.
b) (optimal method) remove all usages of -XX:ActiveProcessorCount cause it won't help, and then try my arguments again. c) (optimal method) use config mod YOSBR to fix the situation. d) try regular server with solutions a, b, c
Note: all minecraft versions have more issues when low disc space on any drives, especially low disk space on temp folder drive and make sure main directory drive has a lot of disk space.
conditionals to check:
- drive space for all drives that are being used(including drives for temp folders Minecraft occasionally uses)
- make sure to allocate a decent amount of memory for game.
- something* I can't figure out what, is using more ticks than it should be.
His thread: "DestroyJavaVM" prio=5 Id=30 RUNNABLE
those are old versions unlikely to get fixed cause these versions have security vulnerabilities.
Java Arguments are most of the issue it seems, I cannot reproduce issue.
Isomehow broke all java and jdk's on my system.I broke stuff you can close this.
Caused by a game crash with nocrash log.I still should have gotten a crash log.
Although still likely valid, this issue needs updated files without them I cannot replicate.
Viradex that is a different issue.
I think this is still occurring in 1.21.1 or something similar.
2024-09-06_18.59.25.png
Due to java version 1.8 is the key cause here, I still am interested in seeing if this still happens with newer java versions.
I cannot reproduce the issue using the following in a datapack.
{ "depthNoiseScaleX": 1, "depthNoiseScaleY": 1, "depthNoiseScaleZ": 1, "mainNoiseScaleX": 1, "mainNoiseScaleY": 1, "mainNoiseScaleZ": 1 }
So with the datapack the original issue is almost non-existent, due to this only makes the collision code look worse though.
with around 1500 of any items/blocks/objects falling you generally get ~0-
10 FPS
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
Steps to reproduce: /fill ~10 ~10 ~10 ~ ~ ~ minecraft:acacia_sapling
Observed results:
7fps in gif shared
Expected results:
around ~30 FPS (should perform approximately the same as chests)
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
Steps to reproduce: /fill ~10 ~10 ~10 ~ ~ ~ minecraft:acacia_sapling
Observed results:
7fps in gif
sharedExpected results:
around ~30 FPS (should perform approximately the same as chests)
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
Steps to reproduce: /fill ~10 ~10 ~10 ~ ~ ~ minecraft:acacia_sapling
Observed results:
7fps in gif or mp3 of same test
Expected results:
around ~30 FPS (should perform approximately the same as chests)
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
Steps to reproduce: /fill ~10 ~10 ~10 ~ ~ ~ minecraft:acacia_sapling
Observed results:
7fps in gif or mp3 of
sametestExpected results:
around ~30 FPS (should perform approximately the same as chests)
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
Steps to reproduce: /fill ~10 ~10 ~10 ~ ~ ~ minecraft:acacia_sapling
Observed results:
7fps in gif or mp
3of testExpected results:
around ~30 FPS (should perform approximately the same as chests)
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
Steps to reproduce: /fill ~10 ~10 ~10 ~ ~ ~ minecraft:acacia_sapling
Observed results:
7fps in gif or mp4 of test
Expected results:
around ~30 FPS (should perform approximately the same as chests)
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
Steps to reproduce: /fill ~10 ~10 ~10 ~ ~ ~
minecraft:acacia_saplingObserved results:
7fps in gif or mp4 of test
Expected results:
around ~30 FPS (should perform approximately the same as chests)
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
Steps to reproduce: /fill ~10 ~10 ~10 ~ ~ ~ enter item/block/object
Observed results:
7fps in gif or mp4 of test
Expected results:
around ~30 FPS (should perform approximately the same as chests)
with around 1500 of any items/blocks/objects falling you generally get ~0-30 FPS
Steps to reproduce: /fill ~10 ~10 ~10 ~ ~ ~ enter item/block/object
Observed results:
7fps in gif or mp4 of test with saplings(is ~ same for any item/block/object)
Expected results:
around ~30 FPS (should perform approximately the same as chests)
nvm, it appears you spent a long time on this.
David J. Romrell If you believe this issue should be reopened, you can leave a reason why it should be reopened in the comments, and then post a request to reopen it in "java requests" channel in the Mojira discord server ( https://discord.gg/rpCyfKV ).
![[Mod] Jiingy](../../../data/avatars/users/91802.png){kind=small}
David J. Romrell that is incorrect, this issue is triaged and has already been resolved as 'Fixed'. Please do not post comments like these.
![[Mod] Asteraoth](../../../data/avatars/users/92886.png){kind=small}
David J. Romrell How are you trying to reproduce this? Did you open MC-26765's world? Can you try the new provided world?
fake forge site by
1
address http://files.minecraftforge.net/
although this seems to be correct, I think this is a java change not a Mojang change.
1."Player placing/destroying blocks" and "Player movement and actions" is usually due to built-in Minecraft shader-based code. 2. This still often occurs without large tick lag. Although is way more noticeable during large tick lag though. 3. I would go under the assumption this issue is true for all versions although worst in stated versions. 4. using java arg -DspriteAtlasTextures:false also would help
So this is mainly an issue with (minecraft's caches) and an (issue with mobs), best option atm is using modrinth "aesthetic and performant" page
also I noticed the JVM flag -XX:ActiveProcessorCount=8 is causing worse performance on server and host computer.
although I am unsure if this is something mojang wants to fix, this is a threading issue. If mojang was to fix this what I would recommend is making JVM threads take higher priority. This way it does not break optimizations to connect to some older versions, yet blocks connections when it is needed.
Simple explanation: -XX:ActiveProcessorCount=8 this java argument apparently decreases minor GC's and results in more major GC runs. So even though it is a valid java argument treat it as experimental.
Final main answer unless none of my solutions work, 1. use standard server 2. use config mod YOSBR to fix the situation
3. (least optimal due to experimental nature)-XX:ActiveProcessorCount=4 ... is like using a horse instead of a car.
4.(-XX:+UnlockExperimentalVMOptions -XX:+UnlockDiagnosticVMOptions -XX:+AlwaysPreTouch -XX:+OptimizeStringConcat -XX:+UseCriticalJavaThreadPriority) doesn't fix issue, just states I want performance. {note to minecraft moderators: (minecraft's caches) and an (issue with mobs) and (GC) bug increased by paper although still present in regular}{}
partial solution use:
-DspriteAtlasTextures:false
this issue is less common still happens.
-Xmx2G option causes this problem(setting max memory instead of minimum memory)
Found a partial fix, (-DspriteAtlasTextures:false) will cause fps to range from 40 to 60 instead of 10 to 20 using same Chest Test.zip , note that in a lot of cases this problem is due to low memory amount.
This seems to be a continuation of MC-5417 unless given more context...
I cant reproduce the issue when using the partial fix for 5417, so either my partial fix applies here or it has been fixed or I need another world example that I can load that shows the issue.
Could either help or fix -DspriteAtlasTextures:false
using -DspriteAtlasTextures:false will either help or fix.
I may be incorrect, I think they are burning just the visual for burning is not happening.
This seems to be a rendering issue, try placing the bookshelf.
First check if using vpn or proxy or similar, because those are known issues. If none of those are true try network reset.
is this a mod environment?
is very likely related to MC-3263
could be related to MC-3263
This seems to be related to MC-3263
That response does not make sense(Deserts are not dry biomes, and savannas are dry biomes?)
That was a beach biome, not a desert biome. You can easily tell by the fact that water is right next to it.
Your statement although correct does sound like what I said. The reason I state that it does sound that way is in game there were 3 biomes(desert, savanna, beach) and the ones most visible were desert and savanna.
@tryashter I do not think MC-98598 is same issue although related. Due to bodell23 trying my fix and it not working, suggests they are different issues. Anyone going to fix the resolution tag on this one?
Issue with "newAi", I would remove the first zip file.
Issue with "newAi"
This seems to be due to a modified server for the ?? part. Your server surprising is vanilla, and the issue occurs every 2 seconds, usually worse when crossing a chunk border.
Yes, this is a memory leak in jdk 21 G1GC, also present in MC-273262 ,
MC-271216, MC-273024, and many more. Two easy solutions, one more challenging 1. Implement along the lines of using java args such as (ParallelGC by default with 6 threads, a minimal GCTimeRatio such as 0, and minimal GCPauseMillis=1) 2. Remove G1GC from valid java arguments. 3. Other solutions such as fixing Minecraft's caches could work.They added which shows in multiple logs as "newAi" and it causes a variety of tick issues. Also a little bit of this is caused by lighting I just found out.
Is this a commands issue or a /locate issue?
I was wrong, it seems to be a combination of GC, Profiler, Lighting recalculations. Amplified by using non-vanilla launcher. Viradex your comment distracts from the actual causes.
@Greymagic27 this is not windows dependent ANY computer will eventually run into this issue if using G1GC. I would recommend reopening this and then merging most of the performance category issues that still use G1GC or JDK less then 22.
I can reproduce the issue, can be reproduced with torch or other light sources as well(FPS drop dependent on gamma and glass used).
still an issue just turning off Ambient Occlusion fixes though.
may be incorrect, seems to be lighting recalculating
Using: ( "GCTimeRatio": 0,
"ao": false,
"spriteAtlasTextures": false ) Patches the issue, would prefer a more optimal solution.
Mostly seems to do with "minecraft:noise" cause using custom noise dramatically decreases problem.
Altering noise and altering "foliage_color", "water_color", "water_fog_color". Completely removes so performance drop is equivalent to regular swamp.
J4PC the mod patches alike to my fix, does not completely remove.
Cannot be reproduced in vanilla with specified version.
@QUAM12 can you restate steps to cause, the command had no effect on my game.
I was wrong, it seems somehow it renders passengers even though they are not seen. This is a problem with command blocks being run from server(GPU usage goes to 100%).
@COMETC2021A1 your account is not strictly vanilla you are launching with a modlauncher or with an old java version such as 8.
confirmed although this seems to be a variation of MC-276284
still present just way better.
This is fixable I fixed most of it using
{ "ao": false, "disableShader": true, "spriteAtlasTextures": false }what version of java?
confirmed(although regardless of java version)
This is a known issue with macs, known possible fixes exist at:
MC-271105, and by forcing the game to launch using JDK22(MC-273262).Ok still try both fixes, let me know if either works or if neither does.
This is not reproducible with vanilla. On the f3 screen you can see this for yourself if you look at SH S:?? O:?? M:?? ML:?? and at local difficulty:??
Here is a valid server install: ([Minecraft Server Download | Minecraft)|https://www.minecraft.net/en-us/download/server]
I have been wondering what the cause of atlas memory usage was from, Connor although your png of windows is correct it still stores the whitespace as memory.
1st try reproducing with Geforce experience disabled/off . 2nd if issue still present try with game bar is disabled/off as well. 3rd note your computer is not listed as arm64 capable. If still not successful move over to
MC-273401as to not bug devs.@hethand Minecraft.exe is in a temp or invalid directory surprised it gives that error though.
Wither, this occurs only on non-vanilla servers usually due to spigot.
Mojang please do not default post a invalid on this.
not saying it should be reopened just stating that I want to have it actually thought about before deciding to set as invalid.
can others verify I know its happening I just want to know if its on just my pc?
incorrect version tag stating 1.21, it stops at 1.20.2 mostly.
ent and AgentM try my solutions, see if they work.
Connor Steppie, this is already being tracked under MC-5417 is what I think mojang is saying.
read
MC-273401andMC-271105and MC-273262Java version 8 is likely the issue, if not it is a device specific issue shown at
MC-151530could be something else I am not being easily able to reproduce it.
cause I am not sure if this is truly vanilla/non-modified can you open f3 screen and take a screenshot?
can you see if this occurs during day as well?
seems a duplicate of
MC-277418IDK why it was in vanilla logs, when it was mainly a fabric issue. Although the sculk sensor showing up as unknown is annoying.
reason I asked it because some people get unofficial Minecraft servers thinking its official.
think I can confirm, it seems related to: MC-250246. And if individual is willing, I would assign Les3awe.
can only consistently cause the GPU increase. Am I missing something?
I appreciate you giving a better test environment. Note I am using java 22.0.1 instead of 21.0.3:
Let me know if any more info would be helpful.
I also cannot reproduce on the latest.
I can somewhat reproduce: java version is making it worse.
the torch and other light sources seem fixed.