Nathan Ryan
- scudobuio
- scudobuio
- Europe/Stockholm
- Yes
- No
MC-88730![[Mojang] Grum (Erik Broes)](../../../data/assets/gravatar/grum.png){kind=small}
![[Mojang] Nathan Adams](../../../data/assets/gravatar/dinnerbone.png){kind=small}
![[Mojang] Felix Jones](../../../data/avatars/users/40484.png){kind=small}
Low
Some biomes appear to be missing from the world generation. I've been comparing between the same world generated in 15w51b and 16w02a, and areas of the world generated in the latter snapshot have different (more common) biomes in place of those of the world generated in the former snapshot. For example, forest seems to be generating in place of flower forest. Also, I haven't (yet) found these less common biomes in 16w02a where they weren't being generat
ingin 15w51b.The affected biomes include flower forest and ice plains spikes, though there may be others.
Some biomes appear to be missing from the world generation. I've been comparing between the same world generated in 15w51b and 16w02a, and areas of the world generated in the latter snapshot have different (more common) biomes in place of those of the world generated in the former snapshot. For example, forest seems to be generating in place of flower forest. Also, I haven't (yet) found these less common biomes in 16w02a where they weren't being generated in 15w51b.
The affected biomes include flower forest and ice plains spikes, though there may be others.
Prior to 1.13 "aquatic" snapshots, a non-source water block would convert to a source water block, generally, under two conditions:
(1) The non-source block was horizontally adjacent to two source blocks, and
(2) The non-source block was above an obstructing block, which included both solid blocks and source blocks. (For the purpose here, an obstructing block is a block that water would flow "around" when determining spread.)As of 18w07a, the second condition has been relaxed, such that the block beneath the non-source water block can be either an obstructing block or another non-source water block. (Note that a non-source water block still doesn't seem to act as an obstructing block.)
This makes certain kinds of cascading water effects (such as horseshoe water falls) impossible.
Demo: https://www.reddit.com/r/Minecraft/comments/7z3zzx/for_those_who_havent_seen_it_the_new_water/
[Edit]
As mentioned in the linked reddit thread, this carries with it potential for large-scale griefing and attacks to client/server stability.
In 18w10c, this appears to be exacerbated in that the block underneath also becomes a source block.
A boat with two non-player passengers will have only one passenger when the world is saved and re-loaded, if the x/z coordinates of either passenger are in a different chunk than the x/z coordinates of the boat. (The saved passenger is the one whose x/z coordinates are in the same chunk as the boat.)
How to reproduce: In a single-player world, summon a boat along a chunk boundary with two passengers, then save the world and re-load. For example:
{id:wolf}
{{/summon boat 8.0 63.0 0.0 {Passengers:[,
{id:chicken}]}
}}Upon re-loading the world, only the wolf will remain.
A boat with two non-player passengers will have only one passenger when the world is saved and re-loaded, if the x/z coordinates of either passenger are in a different chunk than the x/z coordinates of the boat. (The saved passenger is the one whose x/z coordinates are in the same chunk as the boat.)
How to reproduce: In a single-player world, summon a boat along a chunk boundary with two passengers, then save the world and re-load. For example:
/summon boat 8.0 63.0 0.0 {Passengers:[
{id:wolf},
{id:chicken}]}
Upon re-loading the world, only the wolf will remain.
[Edit: Fixed formatting of command, since the parser didn't seem to like the monospace. Sorry for the double notification.]
Unclear whether this is intentional or desired, as it is inconsistent with the fact that drowned otherwise always spawn without armor.
Unclear whether this is intentional or desired, as it is inconsistent with the fact that drowned otherwise always spawn without armor.
Dlawso the Really Lucky Rabbit skeleton horses are supported now, only zombie horses aren't supported
Nathan Ryan indeed
Does the collection villager have food? If so, the farming villager will stop sharing, regardless of how much food is in the inventory of the farming villager.
It is possible to get the horse out of the boat by placing blocks underneath and around the boat, removing the water from the enclosure, and then carefully destroying the boat with lava. Not really a practical workaround, but it will do if your favorite horse is stuck.
It appears that farmers are not "close enough" to the block of farmland they are targeting. If nudged forward a little bit, they will harvest/plant the block.
Have you killed the elder guardians? Not just three of the green ones, but the three large grayish ones.
Confirmed that there is some flakiness with mob spawning. Several times I witnessed all mob spawning cease (animals, hostile mobs, even squid) while experimenting with a passive mob farm. I've never been able to intentionally reproduce the behavior, however.
For me, I was experimenting in a survival world that had been opened to LAN with cheats enabled. Alternating between creative and spectator. When I used "/say @e", it reported a few item-holding zombies and some minecarts, I don't remember anything else, except maybe some items on the ground. I hadn't command-killed anything. I had just sword-killed some passive-mob spawns, and was flying around near the build limit and back down. After a while with no more passive-mob spawns, I realized that nothing was spawning, then started checking for entities with commands. This happened three or four times while I was experimenting. I non-methodically tried several things to restore the spawning, and eventually it would start again. I could never identify the criteria to reproduce the spawn-ceasing behavior or the spawn-restoring behavior, though.
In 1.8.8, placing ice/packed ice used the same sound as placing glass (which makes more sense, to me anyway). I don't know in which snapshot that changed.
@Michał Oh, yes, I understand that mobs don't spawn in spectator mode. I was mostly using spectator mode to phase through the walls of the farm, in order to check redstone and such. But I most definitely was in creative mode (even toggling between creative and survival) during the times when I noticed mobs weren't spawning.
I just had this happen to me in a new world using 15w44b. Same thing: Can hear the baby zombie, chicken agro as if controlled by a baby zombie, but can see no baby zombie. Killing the chicken also seems to eliminate the baby zombie.
Confirmed fixed for me in 15w45a.
I haven't experienced it since earlier than 15w44a (don't remember the precise snapshot), but I was never able to determine the criteria to reproduce it anyway. Very intermittent, only saw it a few times during long periods of experimentation. If I ever encounter this issue again, I'll be sure to cause a debug crash.
Duplicate of
MC-86733.Confirmed that the Minecraft textual interfaces do not support the direct entry of special characters using the typical Mac OS X input methods (which include Option+<key> sequences and holding down the related unaccented key until a popup appears).
Holding cmd (or shift) works in part, in that the accent is entered as a non-combining character.
Non-combining special characters work as expected. For example, holding down Option+p enters the Greek letter pi. The problem is only with combining characters, such as accents, umlauts, etc., that are expected to combine with a subsequent key press.
The largest area that can be operated on at once is 32768 (2^15) blocks, if I'm not mistaken. The area you were attempting to fill is significantly larger than that. The game should have reported this to you, rather than crashing.
Confirmed. Reproduce by setting down a singular block. Place a line of blocks one block higher, with a two-block gap. Sprint over the line of blocks, toward the singular block. The player falls across the two-block gap onto the singular block, but the landing is a bit...rubber-bandy. For a moment, it seems as though the player actually lands in front of the singular block, rather than on top.
I should add, what I think is probably happening is that, while falling, the player is hitting the top half of the side of the singular block. Then, the player can "walk" up the less-than-half-block height, in the same way a player can walk up a sequence of half slabs. In other words, the horizontal velocity seems to be taking precedence over the vertical velocity.
Related to
MC-75471.Possible duplicate of
MC-91388.I suppose the "bug" is that the villager AI guides villagers into the generated houses of a village in a purported attempt at staying safe from zombies, which makes no sense if the zombies can simply appear inside the houses. Really, it's rather silly that the pre-made village houses aren't lit well enough to thwart mob spawning.
Hah, that's going to make my Zombie-Pigman XP farm much more interesting. With a bunch of damaged pigman packed into one spot, an attack will 1-hit kill only one of them, but damage the rest. I expect that the new damaging sweep attack will cause other pigmen in the area to agro.
Duplicate of
MC-89963.Other ways (besides cactus) to get the horse out of the boat:
(1) Destroy the boat with lava (that is quickly removed).
(2) Block the boat in place and pour water over the horse and boat.
So I also thought, though this is apparently the intended behavior, as per the closing of this ticket. Guess I'll have an army of uber-powerful skeleton horses in my new 1.9 world...
Seed (and coordinates) identified in the screenshots. In 15w51b, this is a flower forest. In 16w02a, this is a forest.
The conditions for monument spawning have changed in 1.9, they are now more restrictive. If I remember correctly, a condition used to be that the center block of the monument had to be in a deep ocean biome. In 1.9, I think it's now a 16-block square must be in a deep ocean biome. So, many ocean monuments that would generate in a 1.8 world will not generate in a 1.9 world.
Probably related to
MC-95539. My guess is that the chicken suffers knockback up into the lava, where both it and its drops are consumed.Duplicate of
MC-75347Confirmed not fixed for 16w03a, world created in 16w03a.
Duplicate of
MC-90124.Possible duplicate of
MC-68565.Rabbit spawning is now biome dependent. Try using the spawn egg in a desert or ice plains, for example, and you will get a different result than on a beach.
Confirmed that maps don't always update. The behavior had been intermittent for me, so I had been unable to determine the criteria to reproduce consistently. Updating continued after quitting the game and loading back up, but I hadn't experimented further.
@Mustek I have experienced this bug when switching to a map that contained my coordinates. I have not been able to reproduce intentionally, however, so I have no idea the criteria that present the bug.
@redstonehelper This was in a new 16w03a survival world without having visited the Nether. I expect this report to stay resolved, as I can't provide criteria to reproduce, but some additional info anyway: First, the non-updating map had recently been copied (via the standard crafting table recipe). Second, while holding the non-updating map, the player indicator was present but frozen in its orientation.
There have been some tweaks to world generation since 15w44b. This has affected the specific trees that generate as large oaks, and it has also affected the configurations of large oaks that generate in the same spot across snapshots. (I've even seen some variability for the way in which a specific large oak generates in 16w04a; sometimes the same tree of the same seed will generate smaller or bigger, with different amounts of wood.)
In any case, I cannot reproduce a lack of large oaks. In each of my test worlds for 16w04a, large oaks are part of the world generation.
Just for completeness, I checked out the forest at spawn for seed "1", and found large oaks in abundance (though not one at x=175, z=246, interestingly). In any case, sounds like it's all fixed for you, good to hear.
Perhaps this is an issue of the controls having changed for snapshot 16w04a. Now use "w" for forward, "s" for back (slowly), and the paddle keys "a" and "d" to turn the boat left and right, respectively. The mouse controls the player's view only, and the player can turn its head only 180 degrees, centered on the direction in which the boat is facing.
I can confirm that this is still happening in snapshot 16w04a, in a world created in said snapshot.
Type "/gamerule mobGriefing". If the value is false, then villagers are prohibited from picking up items.
It seems as though villagers are trying to path diagonally. Surround a villager on all four sides by a cobblestone wall, but without the corners. The villager will keep trying to path to a corner, even though it can't fit between the cobblestone walls. Add the corners, and the villager won't try to path anywhere. The same behavior can be observed with other blocks that villagers consider movement inhibiting (full blocks, rails, etc.).
So, if a door is blocked such that the barrier is extended by one block to either side, it seems as though the villager will not repeatedly open/close the door, since the diagonals are also blocked.
Confirmed still happens in snapshot 16w06a, world created in same.
Villagers are still trying to path to unreachable corners in 16w07a. May relate to the issues partially addressed by
MC-94438.In those situations, you can switch to an axe for combat.
Possible duplicate of MC-4.
I have experienced this issue as well, albeit rarely (very rarely). Essentially, the player continues to move even after the physical key is released. Other movement keys are unresponsive. In some cases, the game recovers after a few seconds. In other cases, exiting the game and reloading seems to be the solution. Either way, on the rare occasions that this does happen, it's incredibly frustrating to see your player happily walk off a cliff as though told to do so by the Purple Man.
Duplicate of MC-92165.
Possible duplicate of
MC-96998.I would suspect that the underlying issue is that end structures don't fully generate when only some of the containing chunks are loaded. This would affect single player, as well, if the player died or simply moved away from a partially generated structure.
Duplicate of
MC-96219.Duplicate of
MC-97491.The height of a villager plus the height of a carpet (or even a lily pad) is now more than 2 blocks. That is why they do not fit. I doubt this is any thing other than WAI. However, villagers still try to path through a space through which they do not fit. That is probably a more appropriate framing for this issue.
Confirmed. In my testing of pre-release 2, I have taken fall damage from a high fall when passing through only one block of water. It's rare, though.
To oversimplify, a spider has a chance to drop a spider eye only if killed by a player. Death by falling will yield only string.
Please do not respond here to this comment, the mods have enough work to do. If you'd like to respond, please do so on the Minecraft reddit (https://www.reddit.com/r/Minecraft/comments/47inh7/open_issues_after_19_is_released/).
But please also cut the devs some slack. I understand the general frustration, given how eagerly people have been anticipating 1.9. I am also dissatisfied with some of the prioritization. But code is hard. Code that has existed and been modified over a long period of time is harder. We have incomplete information as to the difficulties of addressing performance issues in general, to say nothing of this one in particular. I'm sure the developers are aware of the negative implications for the upcoming release.
Cannot reproduce. In my testing of 1.9 pre-release 3, passive mobs are spawning, although I think the spawning may be nerfed a bit.
Mine was a single-player world created in 1.9 pre-release 3, using the normal world generation. I had regular spawning in customized terrain in the snapshots, however. I've not attempted in multiplayer.
"Transparent" is a loose term. It encompasses water and solid blocks like glass, as well as blocks that are not "full" blocks, like slabs. The conditions for iron golem spawning thus allow for part (or all) of an iron golem to spawn within glass.
Possible duplicate of
MC-90602. This seems like it's due to some cumulative pollution of a data structure. The effect increases over the duration of gameplay, and is fixed only by reloading the game (not just the world).I've been participating in some discussion of this on the side. How do we know that a minecart's position is governed client side? Myren has demonstrated that a passenger minecart—regardless of whether it contains a passenger—can be pushed by (client-side only) ghost blocks, but its position is periodically restored by the server. This seems to indicate that, while both the client and the server track the position of the minecart, the actual position is governed server side.
Myren has a nice video demonstration of the reproducibility of this issue: https://www.youtube.com/watch?v=ox3lAl2rXpM
Duplicate of
MC-91732.Duplicate of
MC-91732.Duplicate of
MC-93631.Duplicate of
MC-97102.Due to the activation sequence of observers, I would suspect that a stack overflow is the culprit. One experiment would be test the crash using various stack sizes for the JVM.
I should point out that removing random ticks from flowing liquids would also affect liquid dissipation. More specifically, it would slightly slow the already-relatively-slow dissipation, though it would make the rate more predictable (in which case it might make for some interesting medium-period clocks).
This is probably due to the mechanics of the CareerLevel tag. If I'm not mistaken, this value begins at 0 for vanilla-generated villagers. The first time the trading GUI is opened, the Career is reset, trades are re-generated, and the CareerLevel is bumped to 1. You can verify this by examining the entity data of a newly generated villager before and after the trading GUI is opened for the first time.
Relates to
MC-110390&MC-101424.You can force a crash by pressing F3+C for 10 seconds. I've never tried it from the death screen, though.
@Harvey Are your pumpkins at the world height (y=255)? If so, then this is a duplicate of
MC-68565.My understanding is that this is normal behavior for mobs (with a few exceptions, such as farming villagers).
That said, swimming turtles continue to move while no (non-spectating) player is near. I would consider that to be the bug.
In 18w10c, this appears to be exacerbated in that the block underneath also becomes a source block.
I noticed the same behavior for redstone dust found in some jungle temples in a newly generated world, but I couldn't determine the criteria for reproducibility.
More likely caused by
MC-130530.Crash report attached. kernel_task was just starting to exhibit the behavior when I initiated the crash.
Understood about the purpose of the reddit post. Personally, I think it would be a good idea to up the requirements. Although the game is playable, it requires a good deal of configuration that effectively trims the Minecraft experience (fast graphics, minimal particles, etc.). Running pretty much any version of Minecraft at more than 30 FPS on my pre-2014 laptop will turn the keyboard into lava.
Possible duplicate of
MC-91621.I'd hate to see this behavior changed. It's currently the only way to remove enchantments.
Typically, you'd combine tools that were obtained without crafting (bows from fishing, gear purchased from villagers, etc.) in order to add desired enchantments. Requiring at least one point of durability before combining items could work, of course, but it feels unnecessary.
Maybe the problem is just one of terminology, using the crafting table to "repair" the item. It doesn't make any sense to repair something that is pristine. I tend to think of it more as combining durability, although the difference is a minor semantic one.
Possible duplicate of
MC-132739The same also occurs with beds, but it does not occur with (extended) pistons.
While the 1.13 behavior of fire spread from lava is not entirely consistent with 1.12, lava does not start fires in blocks at the same y level as the lava itself, regardless of whether the adjacent blocks are flammable. (Also, note that the wiki description of fire spread from lava is incomplete.)
Still an issue in 1.13.1.
The performance is indeed better in 1.13.1. However, it's still not really playable to any extent. This may have more to do with some of the new chunk loading and entity culling. For example, depending on the render distance, I can see the entity count whip back and forth between what I assume is a normal number, and something in the high hundreds (even thousands or tens of thousands). It's during the abnormally high entity counts that I notice the most significant lag. Extreme numbers cause the game to halt on my machine.
I just realized that I had commented more in regard to performance in general. As to the subject of this ticket, I have no longer experienced a runaway kernel task correlating with Minecraft, so I suppose you could consider this issue closed.
Can confirm, up to and including 1.14.3 pre-release 1.
Given the importance of soul sand as a technical block, this change to world generation (unintended though it may be) is unwelcome.
In previous versions, the block above the crop needed to be light level 9 or greater, which means the crop block itself could be light level 8. Im not sure when this changed exactly, probably in a 1.14 snapshot.
Can confirm that something seem...amiss...with generation of eroded badlands. They look much less colorful and inviting than before. I've checked several generations in various worlds, and have found both red and yellow terracotta, but yellow only generating at y=105,106.
Duplicate of
MC-156586Checked some more worlds. Found more eroded badlands with all colors of terracotta, but the layering is more separated, seems less representative of strata.
Probable duplicate of
MC-136865Relates to
MC-127025.I can confirm. I have found several of these mis-generated fortresses.
Can confirm similar behavior. On my machine (older MacBook Pro), 1.17.1 runs in full-screen mode for a few minutes before kernel_task goes bananas and the game comes to a crawl. Exiting full-screen mode corrects the behavior. Returning to full-screen mode causes the cycle to repeat (few minutes decent performance, then overly high CPU usage).
Fiddling with video/Java settings has had no effect.
Do you have the world seed and coordinates?
I agree that it's likely the light level. You can always use the new light block by filling the layer directly above the wheat with light[level=9] or greater. Note that the wheat will not grow unless the level is 10 or greater.
Keep in mind that oak leaves generate in clusters around a single jungle log in jungle biomes. These leaves will of course drop oak saplings rather than jungle saplings.
Confirmed for 1.18 Pre-release 6, though the cause may now be different due to the new world generation.
I've added a screenshot of a shipwreck that generated completely on land, entirely within plains/forest biomes (for all y levels).
After exploring several mushroom islands in multiple worlds (1.18 Pre-release 8), it seems as though red mushrooms are very rare in general. I count dozens of patches of brown mushrooms for each patch of red mushrooms.
Can confirm. A bit more information:
Duplicate of MC-237145.
Possible duplicate of
MC-136282orMC-67150.Duplicate of
MC-243431.Did you accidentally press F1, which toggles the display of the HUD (on/off)?
Yes, the issue still occurs with the latest version of the launcher, version 1097 (1).
Older Mac OS, Big Sur (11.6).
I haven't measured the amount of time or experimented much with it. I've just noticed that if I start the launcher, start Minecraft, then immediately quit the launcher, it closes normally. It only seems to hang after leaving it open form some time.
For me, the steps to reproduce are pretty simple. Start the launcher, use it to start a Minecraft process, play Minecraft for a while, exit the game, then try to quit the launcher. Note that the hanging behavior is not specific to quitting; the launcher also stop responding in other respects, too. For example, the launcher will also hang when attempting to start another Minecraft process.
Can confirm on Mac (MacOS 11.6, Minecraft Java 1.18.2).
A consequence of this seems to be that, during the death animation of the mount, the rider has entity collision but not yet block collision. So the temporarily dismounted-but-not-dismounted rider, no longer moored to the dead mount, can be "pushed" by other entities into/through solid blocks until the dismount sequence has been completed.
Duplicate of
MC-9523