In particular, this makes swamp and jungle nitwits almost, but not quite, impossible to get: you can only get them by curing Zombie villagers that happen to have the nitwit skin in those biomes, or in the extremely rare circumstance that a village from another biome extends into a swamp/jungle and a nitwit happens to generate in that part of the village.
(I'm actually working on a huge mob farm specifically to get swamp nitwits on the server I play on because I wanted those green coats with green robes to be monks in a build I'm making)
This issue makes Swamp and Jungle skinned Nitwits the rarest mob in the game!
Seems to occur reliably if you're looking at kelp when you press F3
You're welcome, but maybe calling the mods incompetent isn't the best way to get them to want to be helpful? Regardless I hope they'll revisit both this issue and MC-91621. Considering that Hermitcraft is so high profile I'm surprised that this isn't getting more priority from Mojang, honestly.
Resolving this as duplicate seems wrong to me - this and MC-91621 are two separate problems that just happen to interact in an unfortunate way to lead to the particularly nasty lag that Hermitcraft and others are seeing in 1.13.1. This bug is that the fish-specific AI routine is excessively slow, which would still be a problem even if MC-91621 were fixed. MC-91621 is about ALL mobs, not just fish, constantly spawning and despawning, which always causes lag even when there's nothing wrong with the AI of the mobs in question.
Thank you - it really seems fixed now! One remaining glitch - if you put a map of the End in an item frame in the Overworld, the green marker will show up on the map even though it's in a different dimension. Should I file another report for that?
Hermitcraft's cubfan135 just got bitten by this issue in his latest video https://www.youtube.com/watch?v=yS9Nj4dZH1Q - as I understand it the Hermitcraft world was created in one of the 1.13 prereleases so it's never been a 1.12 world. Probably still a pretty big file though...
I did.
The behavior I'm getting is slightly different - re-clicking the existing banners does work to add the markers back, but they disappear again the next time I log out. Unfortunately my world is even bigger - over 1Gb - so I can't upload it either. It's a singleplayer world that's been played in every major release since at least 1.7 (prior to that I wasn't paying attention to version numbers), in case that makes a difference.
This is "sort of" fixed in 17w30b - the markers aren't actually black any more, but I'm pretty sure they're darker than they should be. Also some of them are still disappearing when I log out and log back in again. See https://imgur.com/a/ugewLwj
Confirmed that this still happens in 18w30a
Not sure if this is the same issue or a related one, but I also noticed that banner markers sometimes don't get saved properly and disappear after logging out and back in again. The pattern doesn't seem to be consistent though.
In my testing it definitely seems like every banner added after the first named banner appears black. If you remove the first named banner then the problem moves to the second named banner that was added. This means it's impossible to have more than one correctly-colored named banner on the same map.
The player's marker is affected the same way - if their marker is added to the map after a named banner, then it appears black as well. Removing any named banners added before the player was added fixes the player's marker as well.
Is it 100 chunks from where the cartographer spawned, or 100 chunks from where it is when you try to unlock the trade? What would happen if you took one that didn't give you a mansion map, put him in a minecart, transported him a few thousand blocks in some random direction, and then tried again at unlocking the trade?
Still happens in 1.13-pre2. The new banner markers persist but the old map-in-item-frame ones don't, so you get weird effects where sometimes there's a green dot overlaying your banner marker and sometimes there's not.
Still happens in 17w50a
Still happening in 1.11.2. Can someone with sufficient permissions update the affected versions?
Duplicate of MC-1528?
Still happening for me too in 1.11.2. I'd really love to see it fixed - makes my big map wall almost useless.
Can confirm fixed in 17w18b - created a new world with seed "17w13a", found parrots in less than 5 minutes of searching.
Yes, can still reproduce in 1.18pre1. Took a map of the end, placed it in an item frame in the overworld within the coordinates of the map, and the green marker shows up.