Summary:
A moving block is now considered a full block. This will cause, some machines with block movement don't work properly.
Possible Examples:
Players or entities glitch into to wierd places when pushed
It breaks many shulker loaders/unloaders, as the items glitch all over the place (ignoring the existing issues with pushing items with pistons)
It stops minecarts sticking to rails being moved, as there is a solid block in the way of the minecart when the game attempts to pull it along, breaking any flying machines with minecarts on
It stops the player being pulled along from the side by honey blocks, as the player glitches away from the block, which they need to be in to be pulled along
It can stop slime blocks bouncing the player, as the moving block is placed and the player glitches away from that block. The player needs to be in the block to be bounced. The player will often glitch to the side, which will result in them bouncing, although this is not ideal for slime block launchers. This breaks elevators that bounce players or entities upwards using slime blocks
Steps to Reproduce:
Place a sticky piston facing upside on the under
Place a slime block on the piston
Stand on the slime block
Activate the piston
Expected Results:
The slime block must launch you upwards
Observed Results:
The slime block not launch sometimes you upwards
Original Description:
UPDATE:
Moving blocks now have a collision box identical to the block being moved, however client-side this hasn't been implemented. The way pistons move entities has also been changed, making it (in my opinion) better than before, although not quite perfect.
All that's left to do is extend the collision box changes to the client. It causes wierd client-server desyncs and buggy behaviour. After that this bug is pretty much fixed.
When moving a block such as a rail or a glass pane, the moving block created is a full block no matter what. This breaks many things involving pushing entites. The main role of moving blocks used to be just to prevent two pistons moving blocks into the same space at the same time. The player needs to be inside the moving block for physics to work correctly.
It is most likely a result of fixing MCPE-53815. Moving blocks were turned into full, solid blocks to prevent players "bouncing" downwards when being moved. While this is a good change to some extent, fixing many issues with players falling off of flying machines, moving blocks being fully solid causes the wierd behaviour listed above.
Entities glitching out of moving blocks and being prevented from entering from the side breaks many interactions between entities and pistons.
It would make more sense for them to be paritially solid, like scaffolding, so you can stand on it but not glitch out when in it. Maybe even like cobwebs.
Linked issues
is duplicated by 100
relates to 2
Attachments
Comments 50
This issue is also affecting slime blocks, when bouncing using the blocks it will cause the player to snap to another block, in some instances this will prevent the player from bouncing on slime blocks at all. If several are side by side it is as if the player was snapped into a different place and pushed up and down
With regard to shulker loaders and unloaders, what I have seen is that designs that previously worked with less that 1% loss of shulker boxes (and in many cases, designs which I had NEVER seen lose a shulker box) are now experiencing loss rates in the neighborhood of 30%.
As an example, there is a 1-wide loader design by navynexus that breaks the shulker with a horizontal sticky piston and trap door that I tested in earlier 1.14 versions, put 100 shulker boxes through them and every single one was collected. In 1.14.20, in a 50 shulker test it lost around 20 boxes.
Breaking shulkers with a sticky piston and trap door had been one of the most reliable ways to do it. Shulker boxes would have a bit of random velocity to them, but they were pretty much guaranteed to stay in the same space and drop down. Now the shulker box is likely to glitch through the piston head and despawn.
As another example, I have a manual shulker unloader (a "crafting station") which breaks the shulker box from above with a vertical sticky piston and either a trap door, end rod, slab, etc. - again, a design I had tested with hundreds of shulker boxes in prior versions and not observed any being lost in the machine. All of these options now result in the shulker box glitching upward through the piston head a fair percentage of the time, and remaining there once the next shulker is dispensed.
Shulker loaders and unloaders on Bedrock have always been a bit unreliable on Bedrock in my experience because, almost without exception, any method you use to break the box will result in it having some random velocity that can send it flying, glitching through corners, popping out the back of a piston, etc. But with some careful design a loader could still be quite reliable. But a lot of designs that were quite reliable before 1.14.20 are now very unreliable. Given the relatively high cost of acquiring more shulker boxes in-game (i.e. by end raiding) I see this as a pretty serious problem.
I can confirm all of the problems listed above to be an issue as well.
I imagine this is a consequence of fixing MCPE-53815, which most likely meant changing properties of moving blocks in order to guarantee the stability of being pulled along properly when standing on top of moving honey blocks. However, just as it was mentioned in the various comments and initial bug report, this is a very detrimental change to not only player-created contraptions designed to fulfill certain tasks, but also basic functionality such as honey blocks dragging you along when standing on the side of one, which also is shown to be one of the core features of honey blocks in various official MC media. Slime blocks too seem to be having issues launching you up when standing on them - sometimes they will launch you up but offset you to the side in a weird position, sometimes you will just get stuck inside of that block. This new property of moving blocks also affects entities that are not just the player, just like the aforementioned shulker box items getting into bizarre places due to this behavior.
This issue seems very harmless for normal gameplay, but it actually hurts a variety of gameplay aspects that are expected to work as intended, yet results in this unexpected behavior. Shulker boxes are something that are not often acquired by the average MC player, but with the major focus on this update being on honey blocks and everything surrounding it, the fix to MCPE-53815 is understandable, yet it breaks other core functions of the honey block and other well-established mechanics.
Confirming what navynexus describes for slime block, I have uploaded two videos of a broken slime block on sticky piston setup. In the "slime bouncing broken" video you can see that the slime block behaves no differently than stone when you stand on the center. In the "slime bouncing glitch to side" you can see that landing near the edge of the block preserves the bouncing functionality but moves the player off the block.
@Ella Rose: after several attempted fixes, Mojang chose to resolve this ticket and track continuing issues with moving blocks on a new ticket. The new ticket is MCPE-96035. Please add your vote and any information that is not already covered to that ticket.
Things got better in 1.19.50.20 with the fix for MCPE-96035, but I can still observe the moving rail example occurring.
[^Moving blocks [1.16.0.63].mcworld]
[media][media][media][media][media][media][media][^Moving block testing [1.16].mcworld]