Ping Spike with Mine and Slash GUI

[glossy crescent]

Whenever I open MnS GUIs for more than a moment my ping starts to spike, then very slowly go back down. Depending on the GUI it spikes more (like talents spikes me a ton). This seems to occasionally happen for other mod GUIs too. This is a fresh instance on a fresh private server, and seems to only be happening to me. Any other info available on request as well.

[woven bridge]

that's because everytime you open talent tree, fancy menu will dump tons of error message to your latest.log. As for now I still can't find a way to fix it.

but if you really mean "ping spike", this is not the answer. Log spam will not affect your ping

[glossy crescent]

It is an actual ping spike, player list shows me go from 60ms to over 4k if left long enough, which then slowly drops down over time

[woven bridge]

prob just your network problem

[glossy crescent]

A network problem that ONLY happens when I open a specific set of menus, and absolutely nothing else?

WHich the network would have 0 way of knowing what menus I was opening?

[frozen warren]

same stuff started happens to me and my friends. We are sitting in discord watching yt, doing basic stuff but in game we have 3k sometimes 9k ping. It started happening after switching to newest version.

only when opening MnS gui

[glossy crescent]

If I had to guess, might be related to getting MnS info from the server since it is pretty specific to that set of guis

[frozen warren]

rn it's gotten worse and we can't play the pack anymore. Like whenever we try to alloce point our ping skyrocket and needs few min to go down

[glossy crescent]

Yeah it is especially rough on the talent tree, skill trees a bit less for me usually though

[woven bridge]

@quiet ivy have you done any changes from 0.9.5 to 1.0.5? I've tested the talent tree in dev but it doesn't spam log.

this only happen on cte2

[steel shoal]

The ping spike indeed is here and it's very strange. Maybe for me it's something different but sometimes it's just impossible to play due to skyrocketed ping
But nice catch with the log, gonna take a look why it's doing this

[quiet ivy]

Likely culprit is modern UI or fancymenu. Lots of changes tbh between those version, I don't remember off the top of my head

[glossy crescent]

Either of those able to be disabled safely without dunking a ton of stuff or are they HEAVILY integrated?

[steel shoal]

Our log is full of errors because FancyMenu is unable to work with widgets with negative coordinates.
I asked them if this is intended or just a bug, let's wait for an answer
https://github.com/Keksuccino/FancyMenu/pull/1225

[woven bridge]

ok now i know why cuz if Y is negative then the result string is not a number...

but if we have the situation like this, it still will clash the id

and this

uh the talent tree is too conplicated that edge case could be normal...

[quiet ivy]

You can remove them, and if youre able to determine which of them causes the ping spike it would be really helpful for me!

[steel shoal]

FancyMenu code is well made, so you will never have id clash, don't worry about that. This makes sure they all are unique https://github.com/Keksuccino/FancyMenu/blob/v3-1.20.1-3.6.0/common%2Fsrc%2Fmain%2Fjava%2Fde%2Fkeksuccino%2Ffancymenu%2Fcustomization%2Fwidget%2FScreenWidgetDiscoverer.java#L116

[steel shoal]

This is the same version but with support of negative widget IDs. Can you try and let me know if it any better for you?
At least log spam is gone

To install move out of mods folder existing FancyMenu mod and put this one instead

[glossy crescent]

Will give it a try in a bit when I can hop on

[glossy crescent]

So it made the talent tree load faster which is nice, but ping still spiked

[steel shoal]

Are you 100% positive it happens only with GUI?

[glossy crescent]

Before, just doing stuff, opening refined storage, backpacks etc

After doing some gui stuff spiked a little but currently isnt spiking as much as it did when I had a new point to spend, interesting

Very interesting actually, used respec pots to get some points back to respend and so far has been fine, however when I had a completely new point to spend for both it spiked like mad. Lemme kill a couple mobs real quick to level and see if it happens again

Hmm there was a noticable lag for a moment before it would assign the new point in the talent tree, but didnt spike this time. That is strange af

Ooh there is a new one, started spiking this time when I tried to open the skill hotbar

isnt as large of a spike but still a very noticable one, also the hotbar was delayed a few seconds before actually opening

It kept increasing even while typing that last message

So as the ping decreases, there is still a noticable lag time to open the skill hotbar, but ping hasnt increased again (yet at least). Maybe it has to do with the initial load for a tab each session? Gonna re-open the game and test each one again real quick

So restarted game then opened talents, made a change with 1 of my respec points (took awhile), and it spiked. So yeah I think it may have to do with the initial loading for a tab now, seems once it has loaded once it doesnt spike again but initial load (potentially with changes during that load) spikes it

Also happened if I clicked a skill in the spells panel without points to spend, the error log message for no points took a few seconds to show up then ping spiked

Subsequent attempts to do the same thing the error came up near immediately rather than delayed

[glossy crescent]

So after those various tests, yeah pretty positive it is only with the GUI, but it seems like it is related to each tabs initial loading, and potentially logging for them as well

Ah nope spiked again after trying to click a skill with no skill points after it eventually lowered again

[glossy crescent]

It being ping related I am expecting it is likely an issue with syncing the gui and the player state from the server since the gui wouldnt be handled serverside, but the trigger for the issue is a client side action that should be changing server side state

[steel shoal]

Theoretically you can also install spark, start profiler, make ping skyrocket, stop profiler and send it here. Possibly it would contain something useful
https://www.curseforge.com/minecraft/mc-mods/spark

[glossy crescent]

Sure, can try that in a bit once I am back at home

Does spark work clientside only or need to get it up on the server too? Also I remember seeing prism has a few profiler options so may check those out too

[steel shoal]

It works client side, yeah

[glossy crescent]

Ok cool

[glossy crescent]

https://spark.lucko.me/kbQnKBaGlW

clicked a skill twice to get the delayed error, and unspecced + respecced a single talent tree point and ping spiked. Skill error was still delayed

Hmm spark says spiked at 4.8 gb of memory, have a full 8gb allocated for the jvm at the moment

[steel shoal]

Compare yours with mine https://spark.lucko.me/ufQIeyRiuz
From the first glance your hardware and installed libraries just take more time to render all gui
Example 1: drawing curious slots dlovin.inventoryhud.gui.renderers.CuriosRenderer.render()takes my game 69ms while yours is taking 600ms
Example 2: opengl library com.mojang.blaze3d.vertex.VertexBuffer.drawWithShader() takes my game 36 ms while yours is taking 1180ms

I have no recommendations here, just interesting observations

[glossy crescent]

Yeah did notice the curios menu taking a bit longer to render, which while odd wasnt causing a 2-5k ping spike xD

mesa might have a funky opengl implementation going on for the second one, but not 100% sure there. Im on linux so when it comes to stuff like drivers it can be a much weirder thing, but yeah rendering issues havent been causing ping spikes

[steel shoal]

I just noticed our thread have different duration, that's probably why render timing are different too. Question is should be there x10 render difference to x5 difference in thread duration

[glossy crescent]

linux vs windows scheduling can be wonky

[steel shoal]

Actually, let's pray and summon someone who really knows, if you even go in the right direction
@low totem , hi, do you know if it even possible for GUI to cause ping spikes?

[glossy crescent]

IF it was somehow related to the rendering, my guess would be that somewhere in the render call it is updating the stats for some reason, instead of updating the stats THEN rendering with a cached value of it

to not suddenly make rendering tied to server-client interaction

[steel shoal]

You can go through all these nodes in that tree manually, maybe something will catch you eye

[glossy crescent]

Havent fully used spark before, where is the call tree? At the moment mostly just seeing stats based on time rather than order

Hmm even without finding that though, this looks like it was ONLY tracking the renderer when it is purely clientside, guessing would need serverside to also get other info?

[steel shoal]

Press "All" at the most top row of buttons.

But yeah, first time using spark too, either client-server is not there or it's very light and hidden in depths

[glossy crescent]

Ok found that some, I am not seeing the numbers you are for my render times

like this you said was showing as 1180ms for me?

Ah wait diff call to it, lemme look for that

[steel shoal]

Pretty sure we are looking at different parts of the tree, but I really didn't noticed self: 4 ms there, so it probably fine

[glossy crescent]

Did find the curios one, but looks like it is lazy loading something, interesting

Yeah the self sections are showing render time, the total times I am pretty sure is just how long the window was rendered

Like I had the skill screen up for a few seconds total during the minute or so this profiled, which makes sense for the numbers shown

But yeah since spark is showing only renderer stuff it looks like, and this all looks more or less as Id expect, I am pretty condfident it wouldnt be related to the rendering side of the gui but the data updating/access side

[steel shoal]

Yeah, interesting to see if there is already something to profile such inner logic

[glossy crescent]

prism has profilers as well so may try one of those and see if it gets more than the renderer

[steel shoal]

Yea, to export all threads you can try /sparkc profiler start --thread * --not-combined https://spark.lucko.me/ALj73OThtM

[glossy crescent]

Cool, will give that a shot before the prism ones as well then

[glossy crescent]

https://spark.lucko.me/0GsuRd41m1

took a few tries for it to trigger this time, thats odd but did still happen

Highest time per tick label is prob more useful than % it looks like

[steel shoal]

Yup

It says 9ms to render everything, lol

[glossy crescent]

@frozen warren @jolly flower are both of you on linux by chance?

[steel shoal]

Is your game still on? Can you add --only-ticks-over 100 to spark command?
Maybe okay-ish ticks ease the delay of those ticks where you ping goes up

[glossy crescent]

sure can reboot it and do another with that

https://spark.lucko.me/ZJCZpQGxKc

hmm I did 100 with --not-combined but seems like it is still combining some of them for some reason

[steel shoal]

Was it lagging this time? Just to be sure

[glossy crescent]

Yeah, around the 1 minute to the end moment

I stop the profiler once the lag shows up

something like that should be around when the ping visually increased significantly, had a couple spikes during the whole thing just not the big one yet

this one stands out a little

Which if it is that, could be a race condition T_T

[jolly flower]

Servery Are on Linux based conteiner system on our host

I can set up it as well so we have more data maybe from more servers.

Btw what I experience is a bit different with more variables bcs it happen only to ppl that already have a higher ping (but constant if afk) recorded a session of 3,5h when afk with no ping spikes

But to players with ok ping the issue doesn't exist but to players that have 50-100 ping there are spikes to the 10 thousands

And .. I have servers in Amsterdam in the main data center that has best continental connection. Probably same datacenter for other known games so the state should be stable

[glossy crescent]

Huh, that is strange. Doesnt sound like it is quite the same issue though o_O unless it is happening to people that are opening the gui specifically

[jolly flower]

I can maybe pop up firewall on stick to the VPS so it goes thru something like opensense so I can log packets so we get direct traffic information

[glossy crescent]

Situation is also flipped for my setup at least, server is on a friends extra windows pc, im running linux though. I am the only one with this issue so far though

Could def be a possibility!

[jolly flower]

I tested it with him and we saw connection with the menus opening mainly when mine and slash was happening and player is directly loading the (show me what you doing menu)

[jolly flower]

Maybe try to run VM of windows with mc and see if the issue is persistent?

And if possible install Wireshark and log the MC APP

I think the direct packets would be helpful

[steel shoal]

Maybe I am a bad reader, but did you find a solution to these spikes? Have same issue and connection is stable only in the night, lol

[glossy crescent]

Currently dont have a windows vm set up or I would. Part of why I was asking you and Deka if you were on linux or not. That said, it should functionally be about the same since that is the entire point of Java, running on the JVM to enable cross-compatibility

[jolly flower]

No I'm looking into the issues bcs it bothers me. Sry for bad readability English isn't my first language 😅.

[steel shoal]

Yeah, haha, this issue is really pain in the ass

[glossy crescent]

I may just dive into the repo directly to see if my theory is right xD do you know which mod (or mods at this point) handle the MnS gui? I know Advanced Teams at least expands it, dunno if there is something else managing it nowadays

[jolly flower]

Ok that is true but! Big But! I thing java VMs use difrent drivers to connect to the main hardware on Linux and windows not sure but from different virtualization it is usually the case

And Linux used to be rly bad with some relatek drivers like real bad

[steel shoal]

It's MnS who create the menu and FancyMenu that provides api for that

[glossy crescent]

Fair, definitely could be possible. I gotta track down a windows iso if I am gonna do that though xD hopefully Deka is on windows so I know it isnt a linux issue. It sounded like only some on your server are having issues though so it b eing a linux based server isnt causing it for all

[jolly flower]

I have friend that I met on canceled server and I hate it when stuff doesn't work I work in IT at local data center but I'm not like expert but I do related stuff and this is making my head hurt ...

[glossy crescent]

Ah ok, so sounds like I gotta dig into both then, heck

Still Mahj's fork or is it using a diff fork now?

[steel shoal]

Does Mahj has fork?

[glossy crescent]

Had one as of a few years ago, looks like a diff one then since that fork is VERY inactive lol

[steel shoal]

Pretty sure they use official one

[glossy crescent]

Looks like it based on relations for the pack

[jolly flower]

Tbh the 1.0 mod updates broke kidna a lot huh 😅

[glossy crescent]

Are you sure it is FancyMenu for that? This is looking like it is primarily for the main menu on game launch and isnt listed as a dependent for MnS

[jolly flower]

Btw did you play before update 1.0? So we know if it happened before 1.0 might be able to track it down closely to updated mods etc..

[steel shoal]

Just curious, is it a lot of work to understand if you have a packet loss?

[glossy crescent]

Yeah I have played a bunch in general, it is definitely something that is post 1.0

[jolly flower]

What you mean, the thing is packet loss is kinda hard to identify you have to have release and catch programs that check if this packet is send and received

Im not sure now if it can be caught only on client I usually don't dig into stuff that is heavily effected by packet loss, doesn't matter in cashed style apps

[jolly flower]

Ok so we might wanna check what mods where updated to newer versions then

So we can thin down the mods we have to check then check dependencies and we might get closer

[glossy crescent]

It seemingly only happens for the Mine and Slash GUI for me which is why I am pretty positive it is that

[steel shoal]

I run some sessions of mtr command from Hypixel, lol, and it says I have 40% packet loss during daytime

[steel shoal]

Actually, you can debug your game with IDEA and see if there are any delays

[glossy crescent]

Could attempt that, gotta grab Idea first though xD I am a neovim guy, I tend to manually debug stuff still 😛 lol

[steel shoal]

And now, during night, there is 0% packet loss, lol

[glossy crescent]

So skimming stuff at the moment and the skill gem panel being one of the ones that tends to lag a bit had me looking here https://github.com/RobertSkalko/Mine-And-Slash-Rework/blob/1.20-Forge/src/main/java/com/robertx22/mine_and_slash/gui/screens/OpenSkillGems.java

[steel shoal]

idea is very convinient in terms of doing things on the fly

[glossy crescent]

and it does look like this is one of the few places it sends a packet to the server

[glossy crescent]

For sure, I just like having a single editor set up for everything especially one that functions via terminal 😛 so I tend to not grab many full blown ides anymore

https://github.com/RobertSkalko/Library-of-Exile-Rework/blob/75da71850dd34924a9d12b64fa5894eff68b9e63/src/main/java/com/robertx22/library_of_exile/main/Packets.java#L42 I believe this right here may be our problem child function xD

All 3 things that I have done that seemed to cause that ping spike (trying to add a skill, opening the skill gems menu, or modifying talents) all call Packets.sendToServer with specific new packets for each of those actions

https://github.com/RobertSkalko/Mine-And-Slash-Rework/blob/cff80c4960061aebcf40e81a7b86b8db84b2344f/src/main/java/com/robertx22/mine_and_slash/gui/screens/spell/LearnClassPointButton.java#L54 for spell learn, https://github.com/RobertSkalko/Mine-And-Slash-Rework/blob/cff80c4960061aebcf40e81a7b86b8db84b2344f/src/main/java/com/robertx22/mine_and_slash/gui/screens/OpenSkillGems.java#L18 for gem menu, and https://github.com/RobertSkalko/Mine-And-Slash-Rework/blob/cff80c4960061aebcf40e81a7b86b8db84b2344f/src/main/java/com/robertx22/mine_and_slash/gui/screens/skill_tree/buttons/PerkButton.java#L131 for the talent tree selection

[steel shoal]

Yeah, imagine how cool it would be to be able to build whole flamegraph for the game

[glossy crescent]

Should in theory be doable in SP I think since it starts a local server, so if monitoring the entire process and its children would be possible there

[steel shoal]

Curious if you would have same issue on a local server

[glossy crescent]

Can boot up an sp world and see in a few

Would be kind of surprised if so though since if it IS a packet issue getting packet loss on a local server would be... worrying

[steel shoal]

But it would be approximately 999 times better if you will have issue on a local server than not

[glossy crescent]

Works fine in SP, gonna join a rando public real quick and just see if it lags out there too

well, not really rando but one of the official pubs xD

Yup, happens on the official servers too

So we do know that the client 100% IS sending packets for these specific requests, and I think spark tracked that

Yeah, looked like it was 0% lost, but I am wondering if they were delayed by something happening serverside maybe

and here is the correct interface xD so there is definitely more incoming (spark does specify it is system level logging though not just MC)

So this is the function that handles allocating or removing points from talents, which is one of the things that seems to be able to cause the spike (not always though, I think it might be due to how fast you can edit it which is related to my thought), which does end with it trying to sync the new player state as I originally expected.

HOWEVER this one is what happens when opening the skill gem menu which MUCH more consistently causes the issue, so I am kind of thinking it might have to do with the player loading manybe since that seems to be the main thing in common between these

Which is also shared by the trying to learn a skill packet, where trying to use points you dont have could potentially cause it to happen if you tried a few times, which again is wanting to load the player

So if I am right and it is ctx.getPlayer() causing issues since it is that specific call cropping up repeatedly, that function is causing issues particularly if it is called frequently, and it is called repeatedly in each of these functions. Potential easy fix would be to just call it once at the start of the received function and reference the player that way, which could at least alleviate some of the calls but not sure if that alone is enough to fix it

[jolly flower]

Are all the functions synchronous and do they use like asynchronous task burgers etc I'm not that into coding but might be related to some queues and waittimes before sending data

[glossy crescent]

Not entirely sure, thinking we probably need to get Robert in on this tbh since the problem seems to be in his library mod that sets up the rest of his stuff

[jolly flower]

Out of my scope of work, take me with a shaker of salt just throwing stuff out

[glossy crescent]

Nah you are on the right track I think, you also were with the packet idea

I just dunno much about how minecraft handles packets in general, networking is out of my wheel house a bit which is part of why I was saying we may need Robert lol but best as I can tell it is all coming down to the ExilePacketContext he has here

Which then does stuff related to forge and mc networking that I dont quite know

[jolly flower]

Bcs usually I remember that some issues where caused by synchronous tasks if they where deleted and the main queue should be asynchronous so it can be independent but some stuff might not work bcs the order of stuff send and received has to align by order

[glossy crescent]

Ah shit. If I am getting this right, it might be a damn forge issue >_>

[glossy crescent]

Some stuff that matters for sure yeah, just not so sure about in this exact case

[jolly flower]

Yep

[jolly flower]

Uh and we got forge update in 1.0

Can we check if there are reports on forge with others experiencing this ?

[glossy crescent]

So looks like this is the most recent forge version, cant really test against a server if I change forge versions to test though

Also slightly curious as to why actual forge instead of neoforge at this point but that is a diff thing all together lol

[jolly flower]

Ye neoforge would be nice ...

[glossy crescent]

Not seeing anything in forge's issues that would seem related. Yeah think we need Robert at this point, and/or consideration for neoforge potentially (havent been following closely on how much they have diverged at this point though)

[jolly flower]

Well a lot I think XD it would take a lot to try to make it work with neoforge now

[glossy crescent]

Ah there arent any 1.20.1 neoforge versions, rip

So no options then. Not sure if it is a forge issue or a diff issue within Robert's library mod I am not finding so far

[jolly flower]

I think it might be more of parsing issues maybe that sometimes there is an error or something (would be classic) do they have try catch or order error prevention method on there ?

Bc I think if it was pure packet loss etc you would disconnect or experience ruberbanding or other related issues, but this seems to be just delayed expected packets that are stuck somewhere in the void 😤

[glossy crescent]

Yeah, like i was saying I think it is just delayed not actually lost, and yeah there are try catches for most of these at least at some level

It has just gotten to the point where it is now using forge and mc networking stuff which I am not familiar with at all xD

[jolly flower]

Well time to do the scathi shit tody toda

https://jabelarminecraft.blogspot.com/p/minecraft-forge.html?m=1

Some old but golden info 😅

[steel shoal]

@quiet ivy about Fancy Menu, I was talking about this where it dumps tons of error messages when you open Skill Tree.
We created an issue, so hopefully we will get a fix one day https://github.com/Keksuccino/FancyMenu/issues/1226

[quiet ivy]

ah thanks

yeah i saw this

you just replied to my red string thing so i was a bit confused

but yeah ideally same update

[glossy crescent]

Is Robert in this disc or think Ill need to make an issue?

[steel shoal]

Yea, he is here as well

[glossy crescent]

Wanna ping him? 😛 I dunno which is the right one, and would like to get his input ideally

[steel shoal]

@silk lagoon hi, we've been searching for a reason to ping spike when, much likely, opening MnS GUI and would appreciate any your feedback on this

[glossy crescent]

Tyty

Yeah if you end up reviewing this post, basically everything I could find pointed to something happening with MnS packets so it is either at that point of something within forge from there

[steel shoal]

But what happening with packets? Did you find something out?

[glossy crescent]

Wasn't seeing what was specifically happening to them, wasnt seeing any packet loss, but like I was saying it seemed like they were getting significantly delayed

[silk lagoon]

no clue, my only idea was that im doing something bad with packets, like not using the correct queue method

hm

i might not be clicking release

try this jar on both server and client

[glossy crescent]

Will do

[glossy crescent]

Still spiked, but not as high at the moment at least

Seems to be behaving a little bit better after that initial spike though so far, will keep futzing with stuff to see if it spikes further

Yeah still spikes in general it seems. Gonna let ping move back down and see if it spikes hard again same session though, since it is seeming like that is less common now

Scratch that, yeah still happens same session once ping has gone back down, while it is still displaying higher it seems like it works kinda right for a bit, sometimes x_x This is strange as hell

[silk lagoon]

is this only happening to you specifically

[glossy crescent]

On the server I am playing on yes, but @jolly flower seems to be having a similar issue if not the exact same, DeKa I am not sure on since they havent mentioned much else yet

It happens when connecting to the official servers as well, and has only become an issue since 1.0 launched, previous versions have all worked fine

[silk lagoon]

well, does the entire server lag when you open guis, or just your client?

[glossy crescent]

Just my client, no one else is affected

Mostly just makes assigning skills and talents particularly annoying, and the skill gems. I am not quite sure why that specific menu needs a packet to open it though when the others didn't seem to need to, but that's a different matter. I did also see you have your own custom logging, where does that usually output to since it wasn't showing up as errors in the console for the server at least

[silk lagoon]

ah, then its less of an issue, and more likely related to what that guy above said about your MS being 1000x higher on some stuff

checked your log for spam?

[glossy crescent]

Yup, and that was a misunderstanding on his part when reading the profiler output

Everything I have found that has managed to trigger the spikes has been specifically once the client sends a packet to the server, #1397097265310470304 message outlined in this message

[silk lagoon]

gonna leave this to mahj

[glossy crescent]

Furthest I got down the stack was to where you were starting to use forge and mc networking stuff, hopefully it isnt a forge issue x_x

[silk lagoon]

idk why would it be a forge issue if it happens to only a select few people. it'd be more likely to be a mac or linux thing lol

[glossy crescent]

That was why I was asking Eclipse and DeKa if they were on Linux, I know I am and connecting to a server hosted on Windows, @jolly flower I think was doing the inverse and got it

but like I said, in theory that should be mostly irrelevant since the whole point of the jvm is to try to keep things consistent between platforms

[frozen warren]

I wasnt paying attention bcs me and friends didnt play modpack bcs of lags. We were running windows and for us it was only one friend at the start then after .5 update it was everyone i moved serwer to NA and back to EU just worse spot and it fixed itself. Atleast for everyone beside that one friend who had problems at the start

Only when opening mns gui

[keen stump]

Im playing on widnows on oficial server ping spikes to like 1.5k but i wanna play on my firends server and pings goes to 25k and beyond xd

[silk lagoon]

@quiet ivy do you think this is mns fault or some mod you updated?

[keen stump]

i read whole post here but i dont see any way to repair it now?

or can I just downgrade mod version or sth?

[steel shoal]

no ideas yet

[glossy crescent]

Only thing I can think of at the moment is basically limit how often you need to go to the MnS menus (which is rough early game especially), but otherwise no solutions yet

[steel shoal]

@glossy crescent did you install a debugger after all? Can you inject into sending packets line and see how much does it take to receive the response?

[glossy crescent]

Didn't yet, hadn't tried the prism profilers yet either to see if they have that or not

[keen stump]

was it an issue before? when i come back to mod pack i started to use the newest version is here, maybe going back 1 or 2 small updates can repair that?

[steel shoal]

I remind again about the IDEA, you just download it and it can profile minecraft without any other software

[glossy crescent]

I didnt try on 1.0.0 - 1.0.4, I do know it wasnt a problem for the 0.9.x releases though

[steel shoal]

Btw, do you have the same issue in SP? If I read correctly, it's still a server

[keen stump]

got issues only on friend server (on his router - HUGE issues), official serverf

on singleplayer its fine

[steel shoal]

Any chance you can try running server locally and check if it find there too?

[keen stump]

my firend one is like few houses away but still its on rounter

ill try with friend versions 1.0.0 -1.0.4 rn

on private server its unplayable rn xd

We were wondering maybe its the mod that shows what you do in your inverntory on muliplayer? Do you guys know how its called?

[wanton sonnet]

inventory hud

[quiet ivy]

Idk, maybe fancymenu

[keen stump]

I think we removed huge portion of lags

by removing

ping goes to 100-300 not 25k

[silk lagoon]

whats watut

[keen stump]

this thing

[glossy crescent]

https://www.curseforge.com/minecraft/mc-mods/what-are-they-up-to

[silk lagoon]

oooooh

[keen stump]

well this is destoryed as well (quest book)

but who care that it looks shit if you can play mod xd

[silk lagoon]

makes sense. that shit syncs your client side ui to other players..

however the fk it does that

[glossy crescent]

Hmm that does sound like it could be a likely source of it, interesting

Lmao yup that was it

/coro config client set watut-client.toml dontSendDetailedGUIInfo true stopped the lag it looks like

didn't even need to fully remove the mod, just disabled the client side sending (also sweet jesus they chose a verbose format for their config commands)

@keen stump very good catch with Watut!

[keen stump]

Acually it was my friend but hes too shy to type

[glossy crescent]

Then good catch friend! 😛

[quiet ivy]

ohh shit

if its watut

thats interesintg

ill take note and maybe remove if confirmed

[quiet ivy]

nvm see this

thank you

[glossy crescent]

no problem! Yeah it seems REALLY weird that it is that, but not gonna complain since that was an easy fix lol

[quiet ivy]

thats blessed