MC-125400Bubble columns spawn inconsistently
The first time I tested bubble columns, I made a 2x2x2 cube of water source blocks over magma blocks. One out of the 4 columns produced a bubble column. But repeating this, none seem to, even with the exact same setup. It seems bubble columns spawn only inconsistently.
EDIT After some more testing, this seems to be related to block updates. If you have a 1x1x3 column of water sources (the 3 is 3-tall in the Y axis) over a magma block, contained by glass, you'll properly get a bubble column. If, however, you break one block of glass at the bottom of the column, letting the water flow out, half the time you'll get a bubble column with no flowing water, the other half you'll get just a regular column of water with the water flowing out. And you can change between these simply by updating the magma block underneath. The expected behavior would be that, regardless of block updates, you'd have a bubble column and then flowing water out the side.
EDIT 2 After some more testing, I was able to reproduce the issue in a consistent manner. Screenshots of the process are attached. It's a 1x1x3 column of all water sources, a magma block below, and a contained 1-block outlet on the side of the bottom water source.
"Starting" shows how the column looks normally – notice it's water and not bubbles.
"First Placement" shows what happens when I click once with a water bucket on the magma block, thus placing a fresh water source at the bottom of the column. Notice the water now flows out properly, but it's still not bubbles.
"Second Placement" shows what happens when I click again in the same place with a water bucket. The bottom source turns into bubbles and it stops flowing out (you can see the outlet water starting to diminish).
"AutoReset" shows that, once the outlet water disappears completely, suddenly the column turns back into water sources and the bubbles disappear, resetting everything to the initial starting state.
I'm starting to think this has less to do with block updates, and more to do with the way the water source at the bottom interacts with water around it. Either way, now that I've found a consistent reproduction, you can build it yourself and see the odd and surely unintended behavior.
Linked Issues
![[Helper] Michał](../../../data/avatars/users/43368.png){kind=small}
Created Issue:
Bubble columns spawn inconsistently
The first time I tested bubble columns, I made a 2x2x2 cube of water source blocks over magma blocks. One out of the 4 columns produced a bubble column. But repeating this, none seem to, even with the exact same setup. It seems bubble columns spawn only inconsistently.
The first time I tested bubble columns, I made a 2x2x2 cube of water source blocks over magma blocks. One out of the 4 columns produced a bubble column. But repeating this, none seem to, even with the exact same setup. It seems bubble columns spawn only inconsistently.
EDIT After some more testing, this seems to be related to block updates. If you have a 1x1x3 column of water sources (the 3 is 3-tall in the Y axis) over a magma block, contained by glass, you'll properly get a bubble column. If, however, you break one block of glass at the bottom of the column, letting the water flow out, half the time you'll get a bubble column with no flowing water, the other half you'll get just a regular column of water with the water flowing out. And you can change between these simply by updating the magma block underneath. The expected behavior would be that, regardless of block updates, you'd have a bubble column and then flowing water out the side.
The first time I tested bubble columns, I made a 2x2x2 cube of water source blocks over magma blocks. One out of the 4 columns produced a bubble column. But repeating this, none seem to, even with the exact same setup. It seems bubble columns spawn only inconsistently.
EDIT After some more testing, this seems to be related to block updates. If you have a 1x1x3 column of water sources (the 3 is 3-tall in the Y axis) over a magma block, contained by glass, you'll properly get a bubble column. If, however, you break one block of glass at the bottom of the column, letting the water flow out, half the time you'll get a bubble column with no flowing water, the other half you'll get just a regular column of water with the water flowing out. And you can change between these simply by updating the magma block underneath. The expected behavior would be that, regardless of block updates, you'd have a bubble column and then flowing water out the side.
EDIT 2 After some more testing, I was able to reproduce the issue in a consistent manner. Screenshots of the process are attached. It's a 1x1x3 column of all water sources, a magma block below, and a contained 1-block outlet on the side of the bottom water source.
"Starting" shows how the column looks normally – notice it's water and not bubbles.
"First Placement" shows what happens when I click once with a water bucket on the magma block, thus placing a fresh water source at the bottom of the column. Notice the water now flows out properly, but it's still not bubbles.
"Second Placement" shows what happens when I click again in the same place with a water bucket. The bottom source turns into bubbles and it stops flowing out (you can see the outlet water starting to diminish).
"AutoReset" shows that, once the outlet water disappears completely, suddenly the column turns back into water sources and the bubbles disappear, resetting everything to the initial starting state.
I'm starting to think this has less to do with block updates, and more to do with the way the water source at the bottom interacts with water around it. Either way, now that I've found a consistent reproduction, you can build it yourself and see the odd and surely unintended behavior.
![[Mojang] Agnes Larsson](../../../data/avatars/users/10122.svg){kind=small}
[Mojang] Agnes Larsson![[Mod] violine1101](../../../data/avatars/users/92858.png){kind=small}
relates to
I thought I'd add these images to the bunch to show that this issue still might exist? Water columns from the image only seem to be forming in the North East and South West corners of 2x2 water streams. I'm not certain if this pertains to this particular ticket but I thought I'd put it here.
https://imgur.com/a/evXXJ
EDIT this is most likely caused because the bubble columns won't form unless there's other water blocks next to the water streams?
I could produce bubble columns in naturally generated lakes above y=63, so my guess would be you just encountered the inconsistent behavior described in MC-125400.
Correct, see MC-125400
I made a fix that should be consistent. However, this will probably change when we have the real water fixes in. Current bubble column rules (might change!):
- No horizontal neighbor is allowed to be air
- At least one horizontal neighbor must be a pure WATER block
This can give ugly looking things i.e. when placing bubble columns in water falls. This is planned to be fixed with the other water fixes.
The comment tells that it should be consistent, but it isn't - in 2x2 column only 2 diagonal streams will occur. In L shaped of 3 columns 1 or 2 streams of bubbles will be created. As long as I can't be sure - it is not consistent. So I would open this, or MC-125400
Since the default randomtickspeed value is 3, this issue is virtually unnoticeable in normal gameplay, but setting the value to 1000 will cause source water blocks above magma blocks to evaporate so quickly that the bubble columns form and deform at random and do not retain a stable state. The issue is a direct cause of the fix to MC-125400, in which bubble columns require at least 1 adjacent source block of water, and no adjacent air blocks. A proposed fix to this issue is that bubble columns should require at least one adjacent source water block or at least 1 adjacent bubble column to be able to form and maintain itself.
The current behavior of needing at least 1 pure water block next to the column causes severe issues if you have a magma block under water surrounded on all four sides by other magma blocks; i.e. a plus shape. A bubble column is not allowed to form in the center, because it is surrounded by four bubble columns and no pure water blocks. Then, since magma blocks randomly evaporate water above them, the center magma block will at some point do that (a very high frequency with a high randomtickspeed value). Now, in that brief moment that the water evaporates and becomes air, the four surrounding bubble columns become invalid, do to no horizontal block being allowed as air. This causes the four surrounding magma blocks to now evaporate the water above them, and given the fact that water source blocks will be reformed indefinitely in the right circumstances (bottom of the ocean), this behavior continues to loop until one of the magma blocks is removed. Adding more magma blocks outward makes the effect more problematic.
Same as I said in the ticket I made, it is a direct result of a fix to MC-125400. The water evaporates, becoming air, which invalidates adjacent bubble columns. A bubble column surrounded on all sides by other bubble columns will become invalid (no source water blocks adjacent to it), then the water block above the middle magma block will be able to evaporate, become air, and invalidate the 4 surrounding columns. This pattern continues and causes a chain reaction. The magma blocks do not evaporate the water above them when there is a bubble column directly above them. If the water is evaporating, it means a bubble column can't form in that spot already. Please update the ticket with these details for clarity on the issue. The behavior you described is intentional and is simply a side effect of the actual bug. Bubble columns require at least 2 consecutive blocks of water above the magma block to be able to form. When the water above evaporates, a bubble column can no longer form until the water fills back in.
This is a direct result of the fix to MC-125400, so it may be invalid.
Working as intended (for now), see [Mojang] Agnes Larsson's comment in MC-125400:
I made a fix that should be consistent. However, this will probably change when we have the real water fixes in. Current bubble column rules (might change!):
- No horizontal neighbor is allowed to be air
- At least one horizontal neighbor must be a pure WATER block
This can give ugly looking things i.e. when placing bubble columns in water falls. This is planned to be fixed with the other water fixes.
For now it seems intentional, because of the "fix" to MC-125400 requiring at least one "pure" water block next to the column. However, the comment by Agnes Larsson also states that this behavior is subject to change in the future.
Can confirm this, found this issue too when trying to build a simple item elevator with soul sand producing an upward bubble column. It works in some cases, sometimes not. Placing/removing water source or solid blocks diagonal above the soul sand block sometimes activates or deactivates the bubble column while above the soul sand there's still the required minimum of 2 blocks of water. This feature is undoubtedly one of the most useful additions of this update and allows people to create very efficient and compact upward/downward transportation systems for entities, although I would really expect this to work more consistent in all cases.
Update: See the comment of Agnes Larsson in
MC-125409. Right now, bubble column generation depends on neighbour blocks (including the corner blocks), and they don't form if air blocks are present amongst them. This should explain most of the observed issues with bubble column generation.@[Mojang] Agnes Larsson, on
MC-125409:@redstonehelper - you mentioned about different person, I guess... @FruBasilicum is not @Excited
Oops, had too many profiles open at the same time. Fixed, thanks.
I made a fix that should be consistent. However, this will probably change when we have the real water fixes in. Current bubble column rules (might change!):
This can give ugly looking things i.e. when placing bubble columns in water falls. This is planned to be fixed with the other water fixes.
@[Mojang] Agnes LarssonSo we can't make 3x3 waterbubble column?
@Michał Tried a 3x3 column, the center column (which has no water source blocks as neighbours) still creates a bubble column, seems to work fine to me. I guess having other bubble columns as neighbours instead of water sources works too. You just need to have one column with water source blocks as neighbours to begin with.
UPDATE:
Tried 2x1 and 3x1 column configurations, can confirm this. With the first one, only one bubble column activates, the other one remains as water source blocks. Expading it to 3x1 activates all three of them.
I still think the dependency on neighbouring blocks should be removed completely, it would make bubble columns even more flexible and easier to build. At least the current mechanics allow players to build some "relatively" compact 2x1 column item elevators, already tested some which work totally fine. Single column elevators would be preferable for making actual pipe systems though.
@_zombiehunterIt's "fixed for future version, not 18w07b.
@Michał I know, I can read. I answered your question and dropped a short opinion on how the columns would preferably work in the future compared to their current state. It wasn't a complaint about anything benig broken in 18w07b, I've even mentioned that I got some item elevators to function properly
The current behavior of needing at least 1 pure water block next to the column causes severe issues if you have a magma block under water surrounded on all four sides by other magma blocks; i.e. a plus shape. A bubble column is not allowed to form in the center, because it is surrounded by four bubble columns and no pure water blocks. Then, since magma blocks randomly evaporate water above them, the center magma block will at some point do that. Now, in that brief moment that the water evaporates and becomes air, the four surrounding bubble columns become invalid, do to no horizontal block being allowed as air. This causes the four surrounding magma blocks to now evaporate the water above them, and given the fact that water source blocks will be reformed indefinitely in the right circumstances (bottom of the ocean), this behavior continues to loop until one of the magma blocks is removed. Adding more magma blocks outward makes the effect more problematic. Adding screenshots that show the behavior.
@Aaron Rhodes: Please create a separate ticket for this issue.
This appears to possibly still be an issue? Bubble columns will still form somewhat inconsistently and sometimes form and deform at will. This might be partially due to this bug: https://bugs.mojang.com/browse/MC-125723
Scroll up in the comments on this ticket and read the fix that agnes larsson added. The fix to this issue is that bubble columns require an adjacent block of water to form, and this behavior is subject to change. Technically, the ticket you made is the intended behavior at the moment, but it will likely be changed for reasons similar to the one I gave.
Still this is highly inconsistent: I can't guess if in Lshaped columns of water with soulsand/magma on bottom it will create one or two bubblecolumns.
I made a test. Created L shape columns with soulsand. After adding water - center was bubblecolumn. After destroying and placing back soulsand - edges were bubblecolums. Inconsistant. Screenshots of this attached.
It isn't inconsistent depending on the order that you set this up in. If you place soulsand in one of the 2 corners first, then place soulsand in the center, then soulsand in the remaining corner, you will have 1 bubble column in the center. The same happens if you place soulsand in the center to begin with. If you place soulsand in the 2 corners first, then place soulsand in the center, you will have two bubble columns. The order in which you placed the water has the same effect. You will get consistent results if the same bubble column gets activated first each time.
@Aaron Rhodes How about /fill or moving all soulsand with pistons at once? Will it remain unchanged or change or random?
I would assume it would update starting from the north-west corner all the way to the south-east corner. Given a big enough area, the chain reactions that start can get complicated and cause a seemingly random occurrence. If you repeat the same setup and the same method for changing the blocks a few times and get inconsistent end results, then it could still be considered an issue.
It would be very handy to get solid confirmation on "intended behavior". It appears the current behavior is to require 1x2 minimum with one column having bubbles and the other still, preferably for player elevators on the side away from the entrance. I'd prefer a single column of water, but I found I could do a "dead-fall" diagonally with a "still-water" buffer directly adjacent to both the bubble column and the "dead-fall" column. This would allow a way to break going up or down to use more than just a "bottom" and "top" floor. It's a solution I can live with. If anyone finds a screenshot necessary, I'll put something together on Minecraft Forum. Otherwise, clarification on "intended behavior" would be quite welcome once this is firmed up for the 1.13 release.