#26.1 Snapshots
1 messages · Page 7 of 1
Depends on the contribution. If more than one person contributes, ports go faster
ports are private
um so while yes registerForHolder uses the type directly, uses were not changed to use ? extends T so it requires a cast every time
so the double generic is still needed 💔
Yes, but there is a team iirc?
Idk
Obviously not part of said team
there is a team yes
did u guys start
or are planning a big update like Create 6
porting has started yes
based
Did you create something to port it? /Jk
boooo
fluid model support is merged and building
I predict
today
or tomorrow
maybe tmrw cuz they probably wanna get some more bugs in td
think we borked removing chars from text fields at some point
when i open chat start typing out some long string then press hold backspace to remove stuff
it only removes the first char, have to keep tapping backspace or select everything to remove it all
this does not happen in vanilla
noticed this in my pre2 mod dev env while trying to run commands
couldnt easily backspace from incorrect tab complete
uhm why do we keep the pre-1/2 branch around and not just merge and continue on the port branch?
Presumably the port to the pre was not considered "complete"?
imo it's complete enough to continue on the main port branch
Can't reproduce, tried in creative inventory search, chat and a custom UI's EditBox. Are you sure you're not just being slightly impatient? The key-repeat takes a moment to trigger
oh its something in my mod env
which is weird cause i dont touch anything that could cause this
welp time to grab a shovel and find out why my mods which are simple furniture/dice/food infusion mods are causing key repeats to not work
(yeah also noticed its more than just backspace, holding any key does not repeat)
furniture mods messing with keybinds as a free QOL feature is standard modding
thats even more confusing, not including a single one of my mods it still occours
include(
// "Registree",
// "ApexCore",
// "ItemResistance",
// "InfusedFoods",
// "FantasyDice",
// "TestMod",
// "FantasyFurniture",
// "ApexCompatibilities",
)
//includeFurnitureSets(
// "Nordic",
// "Venthyr",
// "Bone",
// "Dunmer",
// "Necrolord",
// "Royal",
// "Decorations"
//)
i now have a bare bones mdk no mod code yet key repeats still broke
does it happen in prod or just dev
havnt tried compiling my mods but i know it doesnt happen in neo dev or vanilla prod
unsure how i would test neo pre2 prod
not entierly sure how i would install it to prisim
happening in a fresh mdk download too
but not vanilla or neo dev
weird
the only changes i made to the mdk and key repeats still dont work for me
that envvar is to fix ELS crashing on startup which has been a thing since ~snap8 iirc
-# #1445764870564745228 message
key repeats work fine up until 26.1.0.0-alpha.13+snapshot-10
but moment i try 26.1.0.0-alpha.14+snapshot-11 or newer key repeats stop working
so its related to the ELS seg fault?
i have ELS disabled in neo dev (fml.toml>earlyWindowControl=false)
if i reenable it and add the below env var to client runs key repeats now fail in neo dev
environment("__GL_THREADED_OPTIMIZATIONS", "0") // fix ELS crashing on 26.1-snapshot-8+
yep disable ELS and key repeats work fine (can even keep the env var)
tldr
for some reason when i have EarlyLoadingScreen enabled key repeats dont work
disable ELS key repeats work fine
-# fml.toml>earlyWindowControl=false
ive had ELS disabled since ~snap8 cause it started seg faulting
-# #1445764870564745228 message
i also need to specify the following env var when ELS is enabled which works around the seg fault
tested and key repeats work with and without this env var
(not entierly sure why or what it does, just that it allows me to launch with ELS)
-# environment("__GL_THREADED_OPTIMIZATIONS", "0")
Which GPU?
Is this a bug in the NVIDIA driver?
nvidia geforce rtx 3060 and this is what nvidia-smi spits out for versions
NVIDIA-SMI 580.126.18 Driver Version: 580.126.18 CUDA Version: 13.0
I see
Then yeah, that function is what I expect
Essentially the way the nVidia OpenGL driver works is that when you perform an OpenGL call, what the driver actually does is take your command and enqueue it in a worker thread
Then the function returns immediately
Essentially allowing you to keep working while the driver does all it has to do to complete your OpenGL call
Disabling threaded optimization disables this worker thread IIRC
So every command is executed without copying
That doesn't mean that it is executed synchronously, as stuff like uploading to the GPU might still happen off-thread, but most commands should be executed on-thread now
This is my extremely high-level knowledge as someone that only dabbles in GP, so I might have oversimplified or got some details wrong, but this should be the general gist
Why that causes the crash, no idea
disabling __GL_THREADED_OPTIMIZATIONS (setting to 0) is not whats causing ELS to crash
it crashes if that env var is not defined (have not tried setting it to anything other than 0)
-# or maybe i misunderstood what you were saying?
but my key repeat issue only happens if ELS is enabled
if i disable ELS key repeats dont work, with and without that env var defined
By default, it is enabled; the environment variable disables it
iirc
Yeah that’s how I understood it and without the env var ELS crashes really early on into the start up
but this NVIDIA optimization breaks a lot of things even Sodium disables it by default
Does this also affect Windows with an NVIDIA GPU?
You can also force this off by enabling synchronous GL debug callbacks IIRC
since logically the callback can only be synchronous if the GL method call is synchronous (to the driver, not the GPU)
I do, how am I supposed to test this?
uh just launch the game i guess, i cant even get a window without ELS seg faulting
like it crashes basiclly instantly
if you get any further than this with ELS enabled (and without the env var/optimizations enabled)
the issue is not occuring for you
oh im just going by what embeded said it was
trust them to know what it is more than myself
The lines above the exception do look suspiciously like a JVM crash header
this is the error i just got (from the video i posted)
Failed to create gist due to an exception: Server returned HTTP response code: -1, message: 'null' for URL: https://api.github.com/gists
and the hs_err if thats helpful too
Ah THATS a segfault 😄
oh lord
New version detected: 26.1-pre-3.
Yes
26.1 - Pre Release 3
- Primer: https://github.com/ChampionAsh5357/neoforged-github/blob/update/26.1/primers/26.1/index.md
- Article: https://www.minecraft.net/en-us/article/minecraft-26-1-pre-release-3
- Changelog: https://misode.github.io/versions/?id=26.1-pre-3
- Notion: https://apexmodder.notion.site/2524f070f88880648482e3cbdcba566a?v=2524f070f88880b29f60000c52785ecb&p=3264f070f88880afbadce050e5dc7c0f&pm=c
- SnowMan: https://github.com/neoforged/Snowman/commit/935700989716f4da0ebce7b187224ae3e38acf11
SlicedLimes Videos:
- Main: https://www.youtube.com/watch?v=e8k3g3MO2kU
- Pack: N/A
||<@&1067092163520909374>||
A third pre-release is here for the Tiny Takeover drop with more fixes for dolphins and baby mobs! Here's a quick showcase! #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. This is an unofficial update video that aims to...
small one
we are yet to see if that actually fixes them....
they do like to drown....
rc1 on friday?
4096...
quick port
pre3 pr version: 26.1.0.0-alpha.0+pre-3.20260317.152434
use main release #1445764870564745228 message
they ysually do rc1 on Thursdays
so long as they dont do anything tomorrow, im switching isp tomorrow so prob wont be around
Considering how trivial pre3 is, I think we're good to merge and tag another release
what format would the tag name be? release/26.1.0.0-alpha.15+pre3 or release/26.1.0.0-alpha.15+pre-3
-# or doesnt it rly matter
sticking to mojangs naming scheme sounds good, went with that
26.1.0.0-alpha.15+pre-3 should be available once CI finishes
-# ~~ https://github.com/neoforged/NeoForge/actions/runs/23208455445?pr=2879 ~~
if it starts that is... runner taking little bit to pickup the job
oh ofc as i press enter github picks it up
we need to poke those custom actions we have and update them
Awesome @plucky hollow! I taught you once how to publish releases and now I can just chill and let you do all the work 😄
or... get rid of them and go back to good ol' human labour
that sounds like ez way to accidently publish mc sources or something we shouldnt be doing
https://i.imgur.com/SDVT5Pj.png only one class messed up I call that a win
all those missing textures are "triggering" me
A third pre-release is here for the Tiny Takeover drop with more fixes for dolphins and baby mobs! Here's a quick showcase! #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. This is an unofficial update video that aims to...
Sewing Kit and Tool Belt didn't need any code changes (vs pre-2)
Packing Tape was on snapshot-8 and did need one little code change
I just had the screen change from the GuiGraphics change
yeah most who have already done alot of the snap/pre porting shouldnt need to change much for pre3
now yall who are still on 1.21.1 or some shit, port now or else your gonna cause some physical damage upon yourself later
Yeah 1.21.1 to 26.1 is yuuuuuuge
1.21.11 -> 26.1 is not too bad
1.21.1 -> 26.1 has all of 1.21.3, 1.21.4, 1.21.5, 1.21.8, 1.21.10, 1.21.11, and the changes to 26.1 itself
God forbid the people still on 1.20 lol
yeahhhhhhh
21.4 for anyone with model gen is oof, im sorry
specially when adding the neoforge transfer system stuff to the mix
Please provide input on this implementation of cauldron interaction registration events https://github.com/neoforged/NeoForge/pull/2999
If anyone has a better idea just use that branch, this is literally the most primitive implementation I could come up with.
whats wrong with just creating a new dispatcher and calling the put methods myself? like vanilla does
thats has been working just fine in mod dev
only thing i would like is better way to obtain/modify already existing interactions
like im doing for the water bottle
to allow my potion bottle -> potion cauldron
vanilla does water bottle -> water cauldron
before my PR the put methods are package-private and registering to non vanilla and not your own dispatchers would be impossible.
also the id mapper is private
ahh yeah i do have the following AT too which makes everything public since its all nice useful helpers/default impls
# Cauldron Interactions
public net.minecraft.core.cauldron.CauldronInteraction$Dispatcher *()
public net.minecraft.core.cauldron.CauldronInteractions *()
also yeah i guess the event is nice to register to another mods interactions
but personally id just have a mod specific entry point which does similar to what im doing
and grab that mods CauldronInteraction.Dispatcher and call the put methods on it
but the event is nice for that i suppose
i also dont think these events should be fired on the neoforge bus
they feel like they should be fired on the mod bus
Just be careful with threading if you're registering directly
anywhos left all these as comments on github so that they dont get lost here in discord
What does the .0.0 mean?
26.1.0 is the mc version, including the .0 to mean "no hotfixes yet"
the last .0 is the neoforge build, which is 0 because it's alpha so not yet part of the standard development cycle
26.1.0.0-beta, followed by 26.1.0.1-beta, ...
Interesting
same as current naming but one version part more to ensure no conflicts with mc versions
Ah! Makes sense!
dang it the bootstrap method runs before mods are loaded
Attempt number three of "fixing" block model submits: https://github.com/neoforged/NeoForge/pull/3001
With that said, needing the dispatcher to construct the block brings up another issue which I didn't consider before: dispatchers need to be registered before blocks but items (and therefore also blocks) need to be registered to fill the dispatcher, so the event split is unfortunately needed and needs to be split further by firing them at different times.
So the dispatcher registration needs to be before registration and the interaction event needs to run after registration.
We should also have a look at flammability (I can't find a datamap or an event for it)
Isn’t that patched into blocks via a method or two?
flammability in vanilla is two maps on FireBlock that are only used on the Blocks.FIRE instance with a method that is private
#modder-1․21․1-support message
Yeah. Didn’t NeoForge patch flammability into being methods of the Block class?
flammability is 2 values not just one
https://github.com/neoforged/NeoForge/blob/3f86c8c82dd7993ce1aaaf7dee0802031a99f740/src/main/java/net/neoforged/neoforge/common/extensions/IBlockExtension.java#L676
https://github.com/neoforged/NeoForge/blob/3f86c8c82dd7993ce1aaaf7dee0802031a99f740/src/main/java/net/neoforged/neoforge/common/extensions/IBlockExtension.java#L720
Yes, and aren't multiple methods patched in to handle that?
I mean, that might have changed for 26.1; but I didn't think it had.
but having to override that for normal blocks like a normal log seems ass
I agree, 100%.
I've commented on it before, I think.
I'd vastly prefer a datamap or something similar to that. Ideally, vanilla would patch it to be part of the block properties (or something similar that would work for multiloader more cleanly) but I don't think that's going to happen.
I believe the explanation was something to the effect that the flammability behavior is a part of the block's behavior, and thus should be handled via a method.
those override are older than the code I'm porting and that used the vanilla method so this is probably not intended to be the only way
I'm just saying what I've heard in the past.
I mean, iirc we have (or used to have) systems where datamaps are used by default yet still a method exists for e.g. per-state/per-stack behavior
let's ask datamaps mc gee
I think we do that for furnace fuels?
yeah that sounds about right
@bold forge flammability datamap yay or nay
[Reference to](#modder-1․21․1-support message) #modder-1․21․1-support [➤ ](#modder-1․21․1-support message)(side note: Looking at that code I think there's a slight issue there in how neo patches in overrides. For one, there's no way to change the flammability of other blocks that use neo's callback, e.g. one cannot make mod A make mod B's block flammable/fireproof. But the major issue I see is that the vanilla system allows to have different kinds of fires that burn different things, but neo's patches work against that. Subclassing can restore it, but at the cost of copying the complete code of the FireBlock, as those patches are in private methods and in the middle of the logic instead of in the getters vanilla has to isolate flammability determination from the logic.
I think the better implementation would have been to use this.burnOdds.getOrDefault(p_221165_.getBlock(), p_221165_.getFmamability(...)); in getBurnOdds.
)
[Reference to](#modder-1․21․1-support message) #modder-1․21․1-support [➤ ](#modder-1․21․1-support message)yes the part that vanilla flammability is per fire block is the reason I haven't yet added any data map for it
it's not per fire block tho... it always uses Blocks.FIRE even for soul fire
wheeee rc1 time
New version detected: 26.1-rc-1.
<@&1067092163520909374>
Why today my internet has been not working all day
Any link?
I'm not ready for 26.1 porting, oof
I expected rc after live
rc, live, and then release day soon after live
rc now and in live they tell us the release date
yeah, sure, but I didn't expect it
I can maybe get my usual message out but someone else pick up the port
I read that as "this will be the version that we ship for the eventual fall release"
plot twist, tiny takeover is 26.3
rc with no date is interesting
I can
they are waiting for the Minecraft Live on saturday
the 24th release is still likely
they "always" do rc-1 on thursday and release tuesday
on the bright side, the diff is tiny
just version updates?
no bugs fixed, at least in the changelog
26.1 - Release Candidate 1
- Primer: https://github.com/ChampionAsh5357/neoforged-github/blob/update/26.1/primers/26.1/index.md
- Article: https://www.minecraft.net/en-us/article/minecraft-26-1-release-candidate-1
- Changelog: https://misode.github.io/versions/?id=26.1-rc-1
- Notion: https://apexmodder.notion.site/2524f070f88880648482e3cbdcba566a?v=2524f070f88880b29f60000c52785ecb&p=3284f070f88880c08698c15f90ae5f40&pm=c
- SnowMan: https://github.com/neoforged/Snowman/commit/4c9105d034a3bf9fdab7a8fd02b97aae3b553c56
SlicedLimes Videos:
- Main: N/A
- Pack: N/A
i remember seeing that in cg output, i think they have something stripped in there
oh, wait what, that seems different
I think some inlined constant messes with that. Mojang change their constants for releases and rcs to make the game crash less on weird edge cases
who's gonna read the JVM bytecode
this happens all the time when going to and from snapshots iirc
so these are the actual meaningful code changes
"meaningful" is a stretch
dafuq
the second sounds like decompiler artifact
isn't that first change in the second diff the same thing
ah no
extra try
yeah
but the inner try is just a try-with-resources with no catch
so it is the same in practice
that smells decompiler artifacty
i'll have cg stats in a few, i have recomp bytecode diffs :)
it's still -Pupdating=true right?
- version bump
- a couple decompiler artifacts
- a super call that is no longer there
the super method in pre-3 looks like this
I think this is also beta code that has been stripped
rc-1
1.21.11
CG went from https://ss.ln-k.net/9b14b.png to https://ss.ln-k.net/f3189.png
so, in summary, there are no code changes
version bump + built with development flag toggled off resulting in slightly more stripping
RIGHT, this method has a folded constant in it
only change here in CG is the isNonRecoverable if. Otherwise bytecode RT requires that the try-catch and try-with-resources are not merged, it changes exits slightly
every release someone makes the same discovery...
There is a code change and there is a static final that changed value from true to false, what do you think happened there?
hmm, actually we dont attempt twr merge post java11 yet, so maybe thats fine
I mean we didn't think anything specific we just wanted to make sure the decompiler didn't fail in a way that could cause issues after rebuild :P
PR is ready https://github.com/neoforged/NeoForge/pull/3002
no changes only patch shifts
do we need this comment? 
Its probably if (CRASH_EAGERLY && (alwaysThrow || !ioException))
mm, not quite actually, unsure
If no critical issues are found, this will be the version that we ship for the eventual full release.
Good luck everyone that waited to port, here it is
No primer changes this time around
can we merge https://github.com/neoforged/NeoForge/pull/3002 since it is no functional changes?
Done
I would appreciate some feedback on this PR
oh no. now i have to port from 1.21.11 to 26.1. how ever will i cope? /s
Is it possible to add a tag for RC1?
Possible? Yes. Completely pointless? Also yes. RC1 is completely identical to Pre3, except for the two things that always change when release candidates start
oh
With appendHoverText being deprecated, what is the intended way to add tooltips now? I see that most of it is handled on a per-data component basis in ItemStack#addDetailsToTooltip, but we don't have an event for that yet? (Unless that's in another place)
Keep using appendHoverText(). Making the mess that is vanilla's component tooltip handling work for mods is rather involved
so ignore the deprecation for now, got it
If you're curious, here's a convo about that whole topic including a potential but complicated solution: #neoforge-github message
Everything looks sound to me
rc?

Has anyone spot checked all the datapack hooks?
Caused by: java.lang.NullPointerException: Cannot invoke "net.minecraft.server.packs.repository.Pack.streamSelfAndChildren()" because "pack" is null
at TRANSFORMER/[email protected]/net.minecraft.util.thread.BlockableEventLoop.pollTask(BlockableEventLoop.java:137) ~[minecraft-patched-26.1.0.0-alpha.0+rc-1.20260319.191853.jar:?] {neoforge:access_transformer}
at TRANSFORMER/[email protected]/net.minecraft.util.thread.BlockableEventLoop.managedBlock(BlockableEventLoop.java:147) ~[minecraft-patched-26.1.0.0-alpha.0+rc-1.20260319.191853.jar:?] {neoforge:access_transformer}
at TRANSFORMER/[email protected]/net.minecraft.client.gui.screens.worldselection.CreateWorldScreen.openCreateWorldScreen(CreateWorldScreen.java:191) ~[minecraft-patched-26.1.0.0-alpha.0+rc-1.20260319.191853.jar:?] {}
at TRANSFORMER/[email protected]/net.minecraft.client.gui.screens.worldselection.CreateWorldScreen.openFresh(CreateWorldScreen.java:132) ~[minecraft-patched-26.1.0.0-alpha.0+rc-1.20260319.191853.jar:?] {}
at TRANSFORMER/[email protected]/net.minecraft.client.gui.screens.worldselection.CreateWorldScreen.openFresh(CreateWorldScreen.java:116) ~[minecraft-patched-26.1.0.0-alpha.0+rc-1.20260319.191853.jar:?] {}
at TRANSFORMER/[email protected]/net.minecraft.client.gui.screens.worldselection.WorldSelectionList.handleNewLevels(WorldSelectionList.java:177) ~[minecraft-patched-26.1.0.0-alpha.0+rc-1.20260319.191853.jar:?] {neoforge:access_transformer}
Not sure if our bug or not, happened when my datagen failed and tried to create a new test world
@lethal onyx i see you requested my review on cauldron interactions pr
i do want to pull it into mod dev first but as mentioned yesterday my internet is not happy rn
engineer should be coming around today to take a look at it tho so hopefully it gets fixed
New version detected: 26.1-rc-2.
oh no
<@&1067092163520909374> RC2 incoming with a GUI bugfix
https://www.minecraft.net/en-us/article/minecraft-26-1-release-candidate-2
I’m making food
why no
I'm not home 
But my internet is fixed so il get the usual message out as soon as I can
too bad, rc time
Because I just finished my rc1 port ~7 hours ago lol
singular bugfix
what in the name of Hylia--
ah, a smol bugfix
(and a teensy refactor alongside it)
I don't know why we have this bot but I'm gonna blame maty just because
26.1 - Release Candidate 2
- Primer: https://github.com/ChampionAsh5357/neoforged-github/blob/update/26.1/primers/26.1/index.md
- Article: https://www.minecraft.net/en-us/article/minecraft-26-1-release-candidate-2
- Changelog: https://misode.github.io/versions/?id=26.1-rc-2
- Notion: https://apexmodder.notion.site/2524f070f88880648482e3cbdcba566a?v=2524f070f88880b29f60000c52785ecb&p=3294f070f88880eebc16e128599a2dd3&pm=c
- SnowMan: https://github.com/neoforged/Snowman/commit/8f42f359a7c9ce8778cf17f3794848dc699a97d7
SlicedLimes Videos:
- Main: N/A
- Pack: N/A
||<@&1067092163520909374>||
oh whoops i forgor to remove thge ping
It's fine, RC2 so two pings
iirc, its the same workflow but it checks two sources: the version manifest, and the launcher changelog manifest
evidently, one of em is updated faster than the other
three thousand four
I mean it's 5 major versions, of course it's a difficult port /j
25 major versions if you just port Minecraft and ignore NeoForge
That isn't how the version history works, 26 is a year not a major
Would be 22 major versions
Your one line primer: https://github.com/neoforged/.github/pull/39/changes/77d0ab9c0e18de4c573019398749d609b8fb348c
(well, its actually adding three and removing one, but same thing)
If nobody has any further comments on this, then I'll merge the PR tomorrow
And another one: https://github.com/neoforged/NeoForge/pull/3005
Hmm, I have two comments:
- This API feels very low-level compared to our other model-related APIs; in particular, it's not clear just from the docs if I should be setting entries in the list, appending to the list, etc
- If the extension is always called with an initially empty list, do we even need the
booleanreturn value or can we just check if the list is non-empty after the call?
- The parameter doc does say that values need to be appended
- That's a fair point, I'll change that
I've clarified the doc a bit. Does that make more sense or is it even more confusing now? 😅
Yes, much better
I subconsciously skimmed the param doc the first time
New game drops! Developers! More!
It’s time for Minecraft LIVE again! Kick things off with a fun-packed community pre-show, get the scoop on upcoming game drops from our developers, go behind the scenes, meet awesome guests, and discover exciting stuff to come. After the broadcast, stay tuned in for the Deep Dig After Show where our developer...
so whats everyones guesses for if they will announce the drop date and if so when?
"we're releasing it this hour!"

March 24
someone make a recap after it is done 
and what farm Agnes has gone to this time 
<@&1067092163520909374> we have a date March 24th
if you have not started porting to 26.1 now this is your final chance to start porting and maybe get ahead of the curve
theres quite a few changes that can be a pain to update to depending on what your mod does
can find the WIP primer here
-# which unless anything major happens will be the final version of the primer
is this actually where Agnus went to?
I'll resume work on the blog post either tonight or tomorrow 🙂
No, it's the sound designers
You see I solved that problem by not updating my mod since 1.20.2, (Actually I am just lazy and every time I actually want to work on my mod, I end up going back to doing other stuff before releasing an update, that 1.20.2 update was released in March 2025 as well... well after that version released)
hahahah
we knew this was in dev for a year now
with how they introduced this drop, i assumed they were going to add something great mechanically
Was your assumption fulfilled?
have you seen the live?
Yes. I was very underwhelmed.
yes, as was i
the sulfur cube is not a bad addition
it's not mechanically shallow
but it's just
nothing
technical minecraft will benefit from it, sure
Exactly.
fun minigames too
I don't really see myself using it for much, though.
but SPS, which is the game's worst facet, will not see much of an improvement
I mean... maybe my family server could set up some sort of game? There's only three of us, though... so, not really too useful.
(single-player survival)
Exactly. I'd love more singleplayer survival features.
i'll be brainstorming EoMC features that use cinnabar and sulfur, because this is so ass
Mercury as a liquid.
at least the sulfur gas particles are a nice base for my cauldron puff particle, i'll just be stealing those
(Cinnabar is an ore of mercury)
that'd be cool but it needs a proper use
This probably isn't the channel to brainstorm for that, though, unfortunately.
don't wanna do another sulfur cube 
yes

good god
Vulkan x Minecraft in 2 weeks
it would be funny if they implemented vulkan with the april 1st shapshot
would be a good time to test it tbh
AFAIk, not all of the game will be fully vulkan
they will do it in parts
yeah afaik they are focusing on getting it functional
before optimizing it, so i wouldnt even expect any performance improvements once vulkan initially drops
I wonder which extensions will be used...
.jar /s
That's not how transitioning from one graphics API to another works. The entire game will use Vulkan, it just won't take advantage of all of the API's capabilities
i could piecemeal the api to us so to speak
but if that was to happen the game would still run on lwjgl
just the vulkan backend would exist and be unused
that would at least allow shader mods to migrate over with as the api changes
but i dont see them doing that
LWJGL or not has nothing to do with switching to Vulkan
more than likely they will merge in the vulkan branch once its ready
ah yeah, but what i was trying to say is correct tho right
they could drop unused parts but i doubt that would happen
There will, as far as I understand it, be both backends for a while and some way for the player to switch between them. The game will at all times run fully on a single backend though, not a weird mix
i could imagine having the game run on some crazy hybrid opengl-vulkan backend would throw so many if nto all the rendering/shader mods out the window
I wonder if the Minecraft Java videos on Minecraft Live use the Vulkan backend, because there are rendering bugs in the video
@zenith dove Mind rebasing your PR on the 26.1 update branch? Thanks!
will do!
I don’t know if it’s gonna get common use, but I get a feeling that there’s something to be done with a mob that you can split like a slime, that then regrows into its bigger size.
?
i think i'd have to open a new PR for that - maintainers should be able to switch the target branch, i don't think there'd be any conflicts at this point
GitHub just hides the dropdown
Figured it out
It's only available in one specific page of the pr view
Jank ai generated code and all
i think it's just on the PR's homepage? i don't have any PRs targeting me at the moment to check it on
Got some broken links
will fix
this is correct
what else would it use?
I presume it will still use glsl
directly for opengl, and via a spir-v compiler for vulkan
yeah glsl isnt bound to opengl its just the shader language it uses
oh that's interesting
For at least while both GL and VK are supported, the B3D api will take in GLSL
The Vulkan backend will handle the differences
Once Vulkan only, this may change, it may not
How do I lobby for HLSL?
You don't
However, once VK only, the API could take SPIR-V instead, which you can compile HLSL to
yea technically opengl can take SPIR-V too, I'm ignoring that
I mean, you can also compile GLSL to SPIR-V
which I'm gonna assume is what's gonna happen
an interesting question will be once the switch to Vulkan-only happens, whether the shaders will continue to be shipped in GLSL or if there will be a change to precompiled SPIR-V
[Render thread/WARN] [mojang/GlProgram]: neoforge:pipeline/item_translucent_unlit shader program does not use sampler Sampler2 defined in the pipeline. This might be a bug.``` 
You can ignore that
there was no release for rc2, right?
there's no release for rc1/rc2 of neo itself yet
rc1 didn't matter because it was identical to pre3
you can use the PR build though
nah I'll just make a release
it's unfortunate that we have so few people with bandwidth to help; I see many rejects and issues 😄
The only rejects that should still be relevant are the fluid-related ones. Everything else should be done, it's just not cleaned up yet
it's a shame that we will have to move back to the main branch where doing anything requires approvals 😄
I hope that it won't create too much friction
As for making an RC2 release: I would ideally like to get https://github.com/neoforged/NeoForge/pull/3005 in but I'm waiting for pup to have time to respond, I want to make sure I understood her question correctly and not just blindly merge it without
ah ok; I almost made it heh 😄
In terms of issues with the port, the worst known one at the moment is this: https://github.com/neoforged/NeoForge/issues/3008. Seems to be exclusive to Linux and Nvidia and potentially Wayland specifically. The segfault (which is not visible in the log linked in the issue) is completely unhelpful and the stacktrace on the Java side is useless as well
As far as I can tell it doesn't happen on AMD, though the only one who I know tried reproducing it on AMD is embeddedt
Here's where they tried: #1445764870564745228 message
let's try
uh the log in that issue report says
Kernel driver in use: amdgpu
It lists both
Amd iGPU
wait yes nevermind
Yea, the crash from apex is in libegl_nvidia, so, Nvidia specific
oh, the launcher layout has changed again
I'll see if I can repro it tomorrow (Nvidia/Wayland)
I have a laptop with nvidia gpu (Bazzite KDE, so wayland), if anything needs testing there
yeah, works fine on amd/arch/wayland
Wouldn't hurt to have another datapoint, and maybe we can get a useful stacktrace out of one test subject 
Testing the same scenario with the RC2 PR build would also be interesting as that includes the FML update with the LWJGL bump, though I doubt that would change anything
linking to 26.1 PR for easy finding it from my laptop: https://github.com/neoforged/NeoForge/pull/2879
hm
pr2879.net.neoforged.neoforge 26.1.0.0-alpha.0 rc-2.20260321.154221 is unavailable.
The package version you're trying to access has either been deleted or never been published.
The generated artifacts link is wrong (it doesn't url-encode the +), you need to go through "View all package versions"
I don't see an installer jar listed in the assets
There's a link to the installer at the bottom of the bot comment
might not matter, though. the launcher isn't loading
oh!
okay the flatpak worked
(I have never run mc on the linux install in my laptop so it's downloading assets ... slowly...)
do we not produce a debug.log file?
pre-3 worked also
so I was trying to figure out if it launched with x11 or wayland
The debug log still exists. You can also see X11/Wayland in the crash report
Yes, what I mean is that you can force a crash to find out whether X11 or Wayland is in use
what was the key for that
F3+C for ten seconds, even works from main menu
Created Gist at the request of @sleek perch: https://gist.github.com/NeoCamelot/3127b38eafb2430f740976d66b77a859
It's using the wrong one: GFLW Platform: x11
x11
yeah
no idea how to make it run with wayland
the flatpak for the vanilla launcher is set up to run on x11
Uninstall X
I wasn't able to run the launcher outside the flatpak
oh well gotta be heading to bed
theres a debug thing for it
or, well, that applies to vanilla, idk how that applies to ELS
It doesn't apply to ELS but ELS also doesn't set the window hint for preferring X11 when both are supported
Oh huh. You can add me as a datapoint for the linux bug, I've been running fine for quite a while (Ryzen, iGPU)
another nvidia/wayland crash for the pile, though it's entirely inconsistent
https://mclo.gs/2a21BBh
84 lines | 2 errors
pssst I don’t think this is the right channel for mod support for 26.1
- please don't backseat mod
- this was part of a greater discussion around Neo and the graphics issues, not a specific mod support question
My b
there is a debug flag named PREFER_WAYLAND
rc-3 today?
possibly, since there was a fix #mojira message
im not expecting a RC this close to release
like it drops tomorrow, just merge that into the release
Thursday: 26.1.1 drops
i did repro the ELS crash w/ wayland, but interestingly i also got a crash on shutdown when using xwayland, does anyone else repro the shutdown crash?
note: the crash is a sigbus in libnvidia-glcore.so
it didn't crash on shutdown for me
i've personally only repro'd a crash on startup
is your session x11 or xwayland?
I can try again when I get home
wayland, using idea installed from their website
wack
wayland, here, with the launcher flatpak
i wonder if its the specific drivers i have installed causing this one then
I have whatever the Bazzite variant with KDE desktop + nvidia has
New version detected: 26.1-rc-3.
i also have driver 570
theres yourr answer
the pattern holds
it has been 0 days since a bug in the ender dragon fight needed to be hotfixed
damn it, guess i was wrong
oh! that sounds like what I've been experiencing in my dev end, though I'm not sure if MC is running on XWayland
26.1 - Release Candidate 3
- Primer: https://github.com/ChampionAsh5357/neoforged-github/blob/update/26.1/primers/26.1/index.md
- Article: https://www.minecraft.net/en-us/article/minecraft-26-1-release-candidate-3
- Changelog: https://misode.github.io/versions/?id=26.1-rc-3
- Notion: https://apexmodder.notion.site/2524f070f88880648482e3cbdcba566a?v=2524f070f88880b29f60000c52785ecb&p=32c4f070f88880dd9af7c4dc949c5501&pm=c
- SnowMan: https://github.com/neoforged/Snowman/commit/fd82ed7d3efdc1190d14b1785a901ca7c1af4f48
SlicedLimes Videos:
- Main: https://www.youtube.com/watch?v=amh7jYPWGFw
- Pack: N/A
||<@&1067092163520909374>||
A third release candidate is here for the Tiny Takeover drop with a... dragon fix! Here's a quick showcase! #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. This is an unofficial update video that aims to be the most com...
I am running Wayland though

I'm running wayland kde + nvidia id you need another datapoint
called it
snowman being slow to pickup the job: https://github.com/neoforged/Snowman/actions/runs/23441183982/job/68193986415
even tried retriggering it
Xwayland is the best x11 emulator since x11
If you are using Wayland, and the app is using x11, that's xwayland
currently fixing it, the runner isn't running at the moment
https://github.com/neoforged/NeoForge/pull/2879 updated to RC3
I tried running Minecraft using a nvidia GPU through WSL (although it doesn't allow for using a linux nvidia GPU driver instead seemingly sending operations to the windows driver) and didn't have issues with starting it up through the vanilla launcher even when trying to get the early display window to use wayland
runner's fixed, job is now running
I really should look into offloading that into a free runner + S3 cache, as suggested by multiple people before
but the time
(also wondering where I'd put that cache
)
I doubt WSL would use the offending NVidia GL linux driver, i don't know at which level it forwards
Use the minio on the backup box perhaps, otherwise, can host
It's possible we're doing something out of spec that usually causes no issues but causes crashes in NV, or it's just a quirk we'll have to workaround
oh right, that reminds me, I also need to look into replacing MinIO on the backup box
or looking at that OSS fork
11/10 code fix
They snuck in a tiny update to window render state size as well. Tiny code update for tiny takeover.
No primer stuff today
for reference, the Neo window has the Wayland logo on it, and when I go to quit,
[22:52:52] [Render thread/INFO] [minecraft/Minecraft]: Stopping!
#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00007f7254be160e, pid=552461, tid=552518
#
# JRE version: OpenJDK Runtime Environment Temurin-25.0.1+8 (25.0.1+8) (build 25.0.1+8-LTS)
# Java VM: OpenJDK 64-Bit Server VM Temurin-25.0.1+8 (25.0.1+8-LTS, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc, linux-amd64)
# Problematic frame:
# C [libEGL_nvidia.so.0+0xb660e]
#
# Core dump will be written. Default location: Core dumps may be processed with "/usr/lib/systemd/systemd-coredump %P %u %g %s %t %c %h %d %F" (or dumping to /home/sciwhiz12/dev/personal/VoxelTools/run/client/core.552461)
#
# An error report file with more information is saved as:
# /home/sciwhiz12/dev/personal/VoxelTools/run/client/hs_err_pid552461.log
#
# If you would like to submit a bug report, please visit:
# https://github.com/adoptium/adoptium-support/issues
#
if that helps in some way
different DLL though, I get libEGL_nvidia.so
yes, i reproed this
hmmm. I haven't tested vanilla yet
i also got a crash on shutdown using xwayland/x11
i've so far only seen it on neo - fabric testing has been fine despite me launching it more often
hmmmmm
if this is related to the ELS crash ive been having
yeah vanilla is fine its purely ELS crashing (for me)
disable ELS neo loads fine
I'm home now, is there some way to force wayland with neo?
different from what I'm experiencing then; I can load Neo fine up to the main menu, but quitting from there causes the crash
yeah my crash is instant, when ELS is enabled
without ELS everything is fine
its also a segfault too so
this launcher is ancient I just noticed
Does the segfault have a stacktrace? Is there a Java stacktrace as well and if yes, what does it look like?
I posted above in reply to Rogue what shows up in my debug console after I quit
is it consistent for you? i've only seen it like once
The segfault should produce a hs_err_pid log with a bit more info, including a stacktrace if we're lucky.
yep, every time i try launch with ELS enabled
i get the same crash and segfault error
Not in dev, no
No stacktrace either, unfortunate
Given I vaguely recall things being manually forwarded from 1.21.11 to the 26.1 branch, has anyone done a look to make sure that all features/fixes that have been added to 1.21.11 since the 26.1 branch has started have been forwarded? Or is there some number of commits that we still need to move?
I managed to run mc under wayland (via an Ubuntu distrobox), and it crashed on quit
with the iGPU
I'm confused, though. no debug.log files
wait, it was definitely running on the igpu... why did it crash with libegl_nvidia...
🤷♂️ yeh it crashes on quit with the nvidia lib even if it runs on igpu
@round spear do you mind taking another look at https://github.com/neoforged/NeoForge/pull/3005?
yeah that answers my question
wasn't sure how long the cache persisted for, but seems like it isn't long so isn't a problem
I don't think it matters which GPU ot runs on this is more likely a bug in the OpenGL clientside dll no?
"dll" used in lieu of .so here 
A third release candidate is here for the Tiny Takeover drop with a ! Here's a quick showcase! #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. This is an unofficial update video that aims to be the most comprehensive gu...
FYI, the version IDs have changed now 26.1-rc3 -> 26.1-rc-3, so you should update the changelog links to my site
wait wsl can do that ?
is this for all versions or just 26.1+? and would this be for rc only?
Yes for all versions since 26.1. For -snapshot-, -pre- and -rc-
ah ok good to know, il have to update the pinned messages for the affected versions at some point then
which looks to be all the pre releases and rcs
https://github.com/ApexModder/Snapshot-Doc-Generator/pull/2/changes
been meaning to make those urls smarter in my app at some point
cause they can easily be generated on the fly from the full long version id
yes, gui apps work through WSL
https://learn.microsoft.com/en-us/windows/wsl/tutorials/gui-apps
@graceful laurel https://github.com/neoforged/NeoForge/commit/b93282b532b4093a538b7156a52679e4e340d570 is already correct in 26.1, right?
yeah i have that
just the basic stuff
Yes
is it over?
looks to be
bot clearly wanted to re-read every version rdy for tiny take over tomorrow
did the bot glitch, or did mojang push a new version manifest?
jeez
also that's not based on the version manifest
it's based on the launcher update notes endpoint
i dont think you got them all i still see 1.13-1.20 pre1
anywhos do we know what the articles url format will be tomorrow
i believe main releases used to have -java-edition- in the url
unsure if that changed with the version format changes
if i have it right it should be https://www.minecraft.net/en-us/article/minecraft-java-edition-26-1
prepping tomorrows pin message
https://github.com/ApexModder/Snapshot-Doc-Generator/pull/3
Question, can we add a toTemplate() method on DeferredItem because it's annoying me and I think I may not be alone with that sentiment
i mean the ctor overloads should allow you to pass the deferred item to item stack template
new ItemStackTemplate(deferredItem)
whats annoying about that?
it's just that ModItems.MY_ITEM.toTemplate() is a bit more convenient than new ItemStackTemplate(ModItems.MY_ITEM)
same reason we have DeferredItem#toStack(), for example
ehh i dont rly see it being convenient
and feels like "feature creep" type thing but for helper/utils
and its no less readable or convenient than if you were to be using vanilla items
new ItemStackTemplate(Items.DIAMOND)
new ItemStackTemplate(MyModItems.RUBY)
I would say prefer the constructor. Otherwise we have to add one both to the block and item variant
the only thing i will admit is inconvenient about the ctors is they want Holder<Item> so wont accept holders of different kinds (DeferredBlock)
and theres not rly much we can do about this nor should we rly
track your registered BlockItem holders and pass those
We can patch the Holder<Item> ctor to Holder<T extends ItemLike
it gets erased to Holder regardless
we shouldnt be relying on ItemLike
mojang has stated they dont like it since it exposes the raw registry value
and will at some point remove it
What does that mean
what im referring to: #1445764870564745228 message
basicly they dont like that it exposes the raw Item value
you should be handling Holders not raw values
fair enough
I don't think so, none of my stuff triggered, which uses the hash of the version list manifest as a marker.
@graceful laurel needs porting to 26.1, right? https://github.com/neoforged/NeoForge/commit/656a45bac15677767102426863e371f12717a9dd
hmm doesn't look trivial
hehe I think I got it
vanilla 26.1 isn't even out
sometime today, i dont think we got a release time only date
I see
think vanilla + 2-3 hours
yeah im expecting that myself, ~normal snapshot time
Did someone say vanilla+
missed a space
Quark is a mod by Vazkii
but unless something major comes with the release i couldnt imagine neo taking too long update
its already running on RC3 which dropped yesterday
https://github.com/neoforged/NeoForge/pull/2879
What about parchment?
not needed anymore due to unobf sources
mojang ships their param and local var names now
which is mostly what parchmenet was used for
Ooh right, neato
Will my sources retroactively update?
Like
had me
for a moment there thinking 26.1 dropped just now
same thing that happens whenever you update the mc version in your workspace, the toolchain genie decompiles mc again and gives you a new set of sources
you just update the neo version, (and gradle wrapper and gradle version if coming from 1.21.11 or older), and reload the gradle project
mdg is already updated if you are on latest
2.0.140
Yeah I'm on 21.10
Huh, I have 141
131+ i think is what should be running but yeah just run latest mdg
wait, do you have a time machine 
also dont forget to update your gradle wrapper to 9.4+ (iirc)
since that supports java 25 which 26.1 minecraft requires now
https://pr-99.neoforged-website-previews.pages.dev/news/26.1release/
preview for the 26.1 blog post
which should contain all this info iirc
I use gradle 8.14.4
had some issues with gradle 9 so I just went with latest gradle 8.x
weird ive seen so many people (including myself) fail to import anything unless they update gradle
due to java 25 sdks
ugh I hate that discord ignores autoplay choices when you send a message and loops gifs forever without any ability to stop them
mean while theres me with a plugin that ensures everything continues to animate even when window is not focused
like seeing the animated server icons, why that is only on hover confuses me
you like never see them animate cause of it
release now or in 1h?
If people want to approve and merge the primer now, feel free
likely 1h, maybe 2
oh god no
please don't release this monstrosity 😭
but it's also the perfect time to do an IP scan and find all streamer's unwhitelisted servers 🤣
Better use MDG 141, that will also remove the unknown version capabilities warning
I just use version '2.0.+' so I'm always on latest
hm no release yet...
release machine broke
im guess release ~45mins or so
T-20min
Please release, I'm bored 🙂
i have things to upload please release
Nope. Too busy refactoring rendering
Mojang needs to make sure everyone’s work is borked

https://www.playstation.com/en-us/games/minecraft/
Ready or not, here comes the Tiny Takeover drop! New versions of baby mobs have stormed Minecraft, causing cuteness overload across biomes and builds. Listen out for heart-melting sounds, craft name tags for your pets, and keep babies forever young with the golden dandelion. The babies are offi...

26.1 - Tiny Takeover
- Primer: https://github.com/ChampionAsh5357/neoforged-github/blob/update/26.1/primers/26.1/index.md
- Article: https://www.minecraft.net/en-us/article/minecraft-java-edition-26-1
- Changelog: https://misode.github.io/versions/?id=26.1
- Notion: https://apexmodder.notion.site/2524f070f88880648482e3cbdcba566a?v=2524f070f88880b29f60000c52785ecb&p=32c4f070f888807fa951d2171663cec5&pm=c
- SnowMan: https://github.com/neoforged/Snowman/commit/a60d071c7ddc89b3787a5a261f53aa02f12974c4
SlicedLimes Videos:
- Main: https://www.youtube.com/watch?v=8wqIYgk-uD0
- Data Pack: https://www.youtube.com/watch?v=UCZxN8Snfig
- Resource Pack: https://www.youtube.com/watch?v=XM-o2E5DhvA
||<@&1067092163520909374>||
The Tiny Takeover is happening - Minecraft 26.1 updates baby mobs, but also includes craftable name tags, new stone cutter recipes and more! Here's the ultimate guide to all the news! #minecraftemployee
This guide was made for Minecraft Java Edition. Many new features and some smaller changes also apply to Bedrock Edition.
slicedlime works as ...
The Data Pack version of Minecraft 26.1 is 101.1 and with that we get World Clocks, data-driven Villager Trades and much more. Here's an in-depth look at all the news! #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. Thi...
The Resource Pack version of Minecraft 26.1 is 84 and with that we get new Item Models, Block State Rendering improvements and more! Here's an in-depth look at all the news! #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare tim...

I have something you could do...
Now to wait for snapshots again 🤣
New version detected: 26.1.
holy shit
ITS TIME!
at least I can start ignoring snapshots until next year now
cricket noooooooo

bolb revenge
Modders, brace yourselves for the "26.1 update" spam in Curseforge comments
turned those off years ago
26.2 update when
guys why isnt neoforge updated to 26.1 yet 🥺
27.1 when
Dude, the correct title of the drop is in the video above your message

tiny take over
Drop
Tiny took over
tiny turnovers
I had one on discord 16h ago
i premade the mssage last night and jsu tcopy pasted it here but thanks 
:D Thank you for correcting it <3
Apple turnovers??
Inb4 it's Patchouli
@hasty breach Mojang dropped
where #26.2 Snapshots
you're correct (in world rendering should work now)
me when botania is going to take 10x longer to port
has neo decided on a versioning format?
@plucky hollow can i dm you
Time to see which config lib for Fabric will be out earlier than Cloth Config
Configs on fabric hold me back 
YACL was on snapshots
why?
also i just use kaleido config which isn't minecraft-specific and lets me port wherever
i need the snowman output zip
I swear... I searched "config" and marked 26.1-pre-1 and NOTHINg showed up config related 
Whoops, sorry for the ping
the whole jar ?
that site does the same job as snowman
it's only fabric
it doesn't give me an output zip
just click download on github 🙏 😭
Well I only need it on fabric. It's for my multiloader project and the other side is Neo for which I just use it's toml config system
any changes vs rc3?
Well what timing just had a blackout on my street
nope
Not ones that really matter
https://i.imgur.com/1X4HgD7.png
https://i.imgur.com/PSkYSWn.png
we have neoform, let's get this released
alright then, I'll finish doing some work things while waiting for 26.1.0.0-beta
And I'm waiting for 26.1.0.0-beta
is on a numpad
its right under 1
who uses numpad???
me
I do
some people do, but not me
so much easier to type numbers on a numpad
I just thought the wrong number
then numbar above letters i use mostly for gaming
I'm left-handed and the numpad is on the right 
22.0.0-beta
rn i'm waiting for neoforge to update and for cf to add 26.1
then i'll publish an update for my mod
yeah I have Sewing Kit, Tool Belt, Packing Tape, and Json Things ported to pre-3 and ready to get bumped to the final release
the rest I'll do as I get the chance
i have moderner beta and picture mode already ported to 26.1 prereleases
I specifically bought a laptop with a numpad 
my old laptop was full keyboard
my current one is tenkeyless
(weird name to refer to a grid of 4x5 on which 17 keys are placed, 3 of them double-sized)
my pc keyboard has a numpad too but i never actually used it 🙃
must be new here, the article and other related info is pinned 
XD
Hi
soo does anyone have the snowblower output
26.1 final == 26.1-rc-3
What do you even need that for?
(Not that anyone can legally send it to you)
no
ok
you will want to keep track of #builds
what's going on with the build being cancelled lol
currently we are trying to fix the version number generaation
it was the wrong version!
ah
ahhh alright
it's meant to be 26.1.0.0-beta
and it's trying 26.1.0-beta
26.1 + .0 because no hotfixes + .0 neoforge build
all good
You can do that on the Neo-patched JAR as well once the release is done
You should change the version numbering to be neo.[mc ver].forge.[mc hotfix].[ver]
So neo.26.forge.1.1
Horrible, yes
Or just run MDG in vanilla mode
i mean, you can use the pre release as well, wouldn't imagine a lot changed
But bad? Also
neo.forge.26.1.1
n.26.e.1.o.0.forge.0-beta
0.0.0-beta+mc.26.1
2e0F1rG3
for all those wondering 26.1.0.X-beta will be the version number
we are updated and running but the backend tooling is complaining about various things
so please do be patient while we work on getting the release out
the minecraft version is irrelevant build metadata afterall
it should be notbeta-26.1-neo-0.0-ohwait-yeahitsbeta
Double negation my beloved
it's not since any little change in mc source can break neo so we just reuse their version
It also clearly shows which mc version the mod loader is for
think I forgot a /j there
<mc_major>.<mc_minor>.<mc_patch>.<neo_build>[-beta]
is the format for the version number afaik
26.1 neo when? /J
notbetanotrelease-26.n.e.o.f.o.r.g.e.1.beta.1
Yes I was joking
Twenty.six.point.one.point.zerl
notbetanotrelease-26.n.e.o.f.o.r.g.e.1.beta.1.Twenty.six.point.one.point.zer.0.zer.0
this is horrible versioning ^
btw any signs of vulkan or is that the next update
yes
vulkan will come when its rdy, theres nothing set in stone on when that will be
true
vulkan is 26.2+ yes
The Tiny Takeover is happening - Minecraft 26.1 updates baby mobs, but also includes craftable name tags, new stone cutter recipes and more! Here's the ultimate guide to all the news! #minecraftemployee
This guide was made for Minecraft Java Edition. Many new features and some smaller changes also apply to Bedrock Edition.
slicedlime works as ...
XD
So that's why it's called chaos cubed
All the chaos is gonna be the porting
XD
hopefully they peice meal vulkan api to us via the snapshots
which will make updating "easier" rather than dump all in 1 and yolo it
I can basically guarantee you that the initial Vulkan implementation will come in a single snapshot, anything else just doesn't make sense since it wouldn't run
afaik the initial release with vulkan (i assume 26.2) will have both gl and vk backends available
my understanding was that the snapshots would have both, not necessarily the final release
but I find it likely that 26.2 will have both in order to get as wide testing as possible
we will stay in the testing period until we’re confident that the Vulkan implementation is stable, performant and ready
So potentially even 26.3
who knows
ah yeah
My guess is 27.1 will have VB and vulkan stable
26.1.0.1-beta
Soon™
ITS LIVE!
the new borderless fullscreen option uses a window hint not documented anywhere
our early loading system is probably not updated to whatever new stuff mc does
and only in the LWJGL fork of GLFW
seems like borderless fullscreen is completely broken in the Neo build, then
and is just always exclusive
GLFW.glfwWindowHint(131088, glfwBool(!options.exclusiveFullScreen()));
this is the hidden window hint
(called GLFW_SOFT_FULLSCREEN in LWJGL)
Why is it .1 and not .0?
some tech issues which required an additional push
yes
there was a whole thing about it
26.1 = the first major release of 2026
if there's hotfixes, they'll be 26.1.1, 26.1.2, etc
second major release will be 26.2
first major release of 2027 will be 27.1
estimate ~4 major releases per year
modpacks will probably stick to one major release per year and ignore the others

Any ETA on neo 26.2 ? /j
sometime this year 
-# id hope
What about 26.0
0 doesnt exist
This is the second time you have missed my humor
April foool
I guess it can't be helped
fools april
can someone help me in #user-support my dev world is frozen on world loading
26.1, neoforge 26.1.0.1
The Data Pack version of Minecraft 26.1 is 101.1 and with that we get World Clocks, data-driven Villager Trades and much more. Here's an in-depth look at all the news! #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. Thi...
The Resource Pack version of Minecraft 26.1 is 84 and with that we get new Item Models, Block State Rendering improvements and more! Here's an in-depth look at all the news! #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare tim...

amazing 1 line change

26.1.1 - Release Candidate 1
- Primer: https://github.com/ChampionAsh5357/neoforged-github/blob/update/26.1/primers/26.1/index.md
- Article: https://www.minecraft.net/en-us/article/minecraft-26-1-1-release-candidate-1
- Changelog: https://misode.github.io/versions/?id=26.1.1-rc-1
- Notion: https://apexmodder.notion.site/2524f070f88880648482e3cbdcba566a?v=2524f070f88880b29f60000c52785ecb&p=3344f070f8888073bc5ada9ba8ffc25c&pm=c
- SnowMan: https://github.com/neoforged/Snowman/commit/729991d8096a4201c127ae4ae1a5ee0b6ad4b257
SlicedLimes Videos:
- Main: https://www.youtube.com/watch?v=mjFo9nMD5Vo
- Pack: N/A
There's a new Hotfix in town with a new Release Candidate - 26.1.1-rc-1. You can skip this video. Thank me later. But if you're really interested, here's the news. #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. This is...
literally a one-line change 
(substantive change)
A single ! was added to a single line
lol
it seems to have been broken since 26.1-snapshot-7, based on the blame log
The release candidate was released a half hour into april fools in New Zealand
the april fools joke is that the RC is the april fools version and they've hidden all the features from the release notes as part of the prank
y'all don't count for april fools
Anti chat reporting people would love that version now
Mojang destroyed No Chat Reports with this simple trick, discorver how
smh, NZ erasure /j
Only CET/CEST exists for mojang
I'm pretty sure this doesn't need a primer
microslop
Agree. Mojang already handled it by having the bug fix in the notes
Is that the joke?
article about 26.2 since testing is here for bedrock beta/preview, posting here since 26.2 Snapshots is locked since no snapshots yet
https://www.minecraft.net/en-us/article/test-the-first-features-from-our-next-drop
Does nausea affect mobs? I thought it was one of those visual things that wasn’t all that impactful on things that don’t need to see via a screen.
not in current versions
maybe they will change this in 26.2
like, the effect does apply, but it does nothing
Would be funny to have the mobs walking in random directions when affected by nausea
That’s what I thought but was unsure if it changed recently and I missed it.
It definitely would be nice to have more negative effects work on mobs. Would certainly make my life easier lol.
That it would!
-# My mod has darts which can be tipped with items to inflict potion effects... they're almost completely useless against mobs, since almost all monsters are immune to most harmful effects!
There's a new Hotfix in town with a new Release Candidate - 26.1.1-rc-1. You can skip this video. Thank me later. But if you're really interested, here's the news. #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. This is...
[Jump to referenced message](#minecraft-updates message) in #minecraft-updates
<@&1067092163520909374>
New version detected: 26.1.1.
their april fools is to make us believe there's an april fools
why did it not let me ping 
#minecraft-updates message what
[Jump to referenced message](#minecraft-updates message) in #minecraft-updates
it did ping
and I just gave out Snapshot Guesser
there is no snapshot in ba sing se
Discord just doesn't want to add the ~300 people who have the role to the thread
oooh
The jar links have the same hash, so was it the metadata that changed?
26.1.1 - Hotfix
- Primer: https://github.com/ChampionAsh5357/neoforged-github/blob/update/26.1/primers/26.1/index.md
- Article: https://www.minecraft.net/en-us/article/minecraft-java-edition-26-1-1
- Changelog: https://misode.github.io/versions/?id=26.1.1
- Notion: https://apexmodder.notion.site/2524f070f88880648482e3cbdcba566a?v=2524f070f88880b29f60000c52785ecb&p=3344f070f8888019b47ff9ea8f18d857&pm=c
- SnowMan: https://github.com/neoforged/Snowman/commit/d6ab32ecb2efdd1b8522aa3949ce94605e7951df
SlicedLimes Videos:
- Main: https://www.youtube.com/watch?v=HqwNepcqAd4
- Pack: N/A
There's a new Hotfix for Tiny Takeover - 26.1.1. You can skip this video. Especially if you didn't skip the last one. But if you're really interested, here's the news. #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. Thi...
Conspiracy theory time /s
The bug was included intentionally, so that they can test everyone's automations for the full new version number scheme (year.drop.hotfix)
wait... did they push the publish button for the hotfix twice and the second release just replaced the first
We didn't do anything different from the way we usually do things. Maybe some caching issue on that bot's end?
no clue
There's a new Hotfix for Tiny Takeover - 26.1.1. You can skip this video. Especially if you didn't skip the last one. But if you're really interested, here's the news. #minecraftemployee
slicedlime works as a Technical Director for Minecraft at Mojang, but the YouTube and Twitch channels are personal projects run entirely in his spare time. Thi...
interesting april fools joke /s
My bot didnt have any issues like this. Although my scrape interval for the manifest is probably a lot longer (~5mins)
I wonder how many requests per second they get specifically from people polling for version updates
yes ps
It seems to be azure cached? or a blob or something? I doubt we'd even show up on metrics compared to launchers
I felt bad even setting my bot to 5 mins, as i couldn't etag cache with k8s at the time
Might have hit the thread member cap. Discord limits the amount of users that can be added to threads through pings. Pretty sure you can still manually join the thread past the cap
yeah was pointd out by someone else
Sus response given the day. 
What do you think is sus about it?
It's April 1st
Just because it's April 1st doesn't mean everything that exists on that day is sus
idk, this message is reading a bit sus to me....
/j
Security Unintended Service
New version detected: 26w14a.
<@&1067092163520909374>
LOL
ono
IT BEGINS
I guess old habits die hard
Ping?
Ah, there it is.

is there an article?
please don't be obfuscated

Back to old versioning, 26.1 was a mistake
Damn they went back to their own versioning system 😲
back to major versions as well
Please god, obfuscate it!
Article?
it is not obfuscated
Soon, I bet
Obfuscate Mod incoming 
no obfuscated, sadly
Who needs an inventory when you can have the world? Herd blocks, command items, and live in the moment with the Herdcraft update. Say goodbye to dull inventory management and embrace a better, mindful way to build and craft. It just makes sense!


