Transparent*
Still exist in 1.5.1
Fixed for solid blocks, still getting stuck in non-opaque blocks sometimes.
Update Affects Versions to 10b and pre-release.
I thought the BUD nerf will do the job, but it fixed nothing. Why was the BUD ability removed then?
In 13w10b they react a bit faster but not as good as skeletons and still walk in the holes
I think the problem came from dispensers that are too sensitive when powered. The droppers are so fast that they react at Redstone updates not like an instant update but like frequent block changes. Maybe adding 1 tick delay before item drop will fix the problem.
Please add a picture from https://mojang.atlassian.net/browse/MC-10667 It is easier to reproduce and a bit different.
Or a picture from setup like here, that should be more realible http://www.youtube.com/watch?v=E4FuwPPcpds
In the 13w09c still exist.
Since 13w07a they're acting more like in 1.4.7, but the lag is still there.
They will walk 3 blocks before the AI update tell that they are not in the stream.
You can reproduce it easily using Snow Golem like on the second screenshot. Put Creeper right after a Zombie, and it will start fighting the stream.
Skeletons doesn't have that lag and they react instantly, bypassing the hole.
It's more noticeable with high levels of effect.
At level 4 effect (80% reduction) wood plank breaks in 15 seconds and stone in 7 seconds using hand.
At level 5 effect (100% reduction) wood plank is unbreakable and stone breaks in the same 7 seconds as without potion effect. The stone is unbreakable only for pickaxe.
If you use any other tool or bare hand, they ignore the debuff.
Seems like it came from Efficiency tools nerf. We can't use Efficiency shovels for mining anymore, but as the downside effect we now have Mining Fatigue (counter-effect) broken.
p.s. ну и ник у тебя
From your explanation all Entities that use this class have the same bug, so I updated the title and added example with Villagers and Zombies.
Joss Eden, that's right. It is AI update lag.
Markku, the environment doesn't matter. If there will be a wall instead of a hole to the void, the would just bump into the wall.
I suggest this formula:
children parameter = av + rand(-0.6;1)*(6/(av+5) - 1)
av - average parents' parameter;
parameters here are float of range 0 - 1 (use as factor)