Zed Ontargs

Another confirmation that this happens when loading old (1.12.x) Void Superflat worlds, similar crashdump.

Ah, searched specifically for torch placement. Apologies and thanks.

Possibly useful debug info: OTHER players breaking their tools makes a sound in LAN games, but my own tools do not. The server is telling my client about their sounds, but neither the server nor the client are telling me about my own sound.

Yes, any block update nearby can cause the ladder to break.

Done, see MC-117605

In Pre-Release 3, ladders can still be placed on improper blocks if a solid block is beside the ladder. Removing the solid blocks does not cause the ladder to update and fall off.

In Pre-Release 3, melons and pumpkins are fixed, but fences and walls still connect to jack-o-lanterns. Is this intended?

Still no narrator on Windows 8.1 and 17w18a because of SAPIWrapper_x64.dll

Can confirm the same issue on 17w15a:

java.lang.UnsatisfiedLinkError: com.mojang.text2speech.NarratorWindows$SAPIWrapperSolutionDLL.say(Lcom/sun/jna/WString;)J

Issue still exists in 17w06

Appears to be fixed in 1.11 / 16w50.

While this doesn't happen on slabs with average horse speeds, horses with high speeds or affected by speed potions can glitch out when climbing slopes made of slabs just as they would on stairs.

Confirmed for 16w43a. I was testing the fixed unbreaking calculation, and noticed that my pickaxes silently disappeared when they broke.

Still an issue in 16w40a.

Not able to confirm that big trees are completely absent in 16w40a here. They are doing the goofy "only one size" thing though.

Still an issue in 16w39c, and the Nitwits make things much, much worse than in pre-Nitwit versions. A villager breeder is almost mandatory in order to get some useful villagers.

Still an issue in 16w39c.

A reply to the Reddit suggestion that Mojang make the observer weakly power blocks instead of strongly powering them:

"Hmm. After experimenting with the assumption that it'd weakly power blocks, we run into a new problem: you can't use it in a conventional BUD-triggered piston setup anymore. The single point of redstone power can't reach any useful distance without a delay-causing repeater, which leads to the update loop.

"I had to use a chain of observers and dust in order to instantly pass the signal, which makes these messes (strongly powered on the left, weakly powered on the right): http://sli.mg/a/ZtIbp3

"If Mojang goes that route, I hope they increase the power of the output signal (if that's possible for a weak source), otherwise it's less useful than the makeshift ones we have now."

Simply adding a delay to the observer might cause more problems than it fixes. Re-posting a comment of mine from Reddit:

"Right now, if an observer triggers a piston with no delay, it doesn't react to the piston head passing in front of it. If you add in a delay, you get an update-loop caused by the piston head. Try it with a repeater and see.

"In order to solve that, unless Mojang builds a minimum cycle length into the observer block, you'd need an entire control circuit to make the piston ignore the piston-caused updates. It would also mean that we'd need to time everything else around the lockout period, making observers even less useful."

I hope, if this does get "fixed", it does so in a way that won't make it harder to use than any of the existing powered-block or quasi-connectivity BUD systems. It'd be a shame to turn it into a "furnace minecart" situation right after adding it.

Yikes, this got worse in the 16w39 snapshots. I'm getting the stitched-together mess seen in B.png