I just found a Mushroom Island biome directly adjacent to a Savanna Island.
From what I've researched online, the chance of this happening is less than 0.001%.
Can someone confirm this for me and tell me in the comments?
Here's the world seed, but I'm playing on a server, so I don't know if it changes the terrain generation: 8503596189782575358
Oh, and the coordinates: x: -3777, y: 85, z: -2316
(I went into creative mode just to take this better screenshot, I'm not a cheater...)
Without knowing the version of the game, or more specifically what version of the game the terrain generated in (and if mods which impact terrain generation were used on the server generating it), seeds can be misleading.
Initially, I checked and found that this does not exist with the current (1.18+) terrain generation. Here is what is at those coordinates.
There is a small lovely island off shore at spawn though.
It`s version 1.7.10, the server don´t have mods, the terrain generation it´s peculiar. Originally, the server was for version 1.0, but I upgraded it to version 1.7.10, so the terrain generation was changed. And the server wasn´t generated in large biomes or amplified.
Does this awnser your question?
(I´m brazilian, so don´t pay attention to my ugly english.)
Yeah, I figured out that it was generation in a version between 1.7 and 1.16.
The my understanding, the reason mushroom islands are almost always surrounded by water is because the game uses a certain value that will usually only be reached in an ocean. I think it was something like a value of 1 = land, 0 = ocean, and -1 equals mushroom island? But it might be different for versions before 1.18 so maybe they are more common touching land in versions before that?
Rollback Post to RevisionRollBack
"'Tis foolishness! If all were so easy, why, none would suffer in this world!"
Yeah, I figured out that it was generation in a version between 1.7 and 1.16.
The my understanding, the reason mushroom islands are almost always surrounded by water is because the game uses a certain value that will usually only be reached in an ocean. I think it was something like a value of 1 = land, 0 = ocean, and -1 equals mushroom island? But it might be different for versions before 1.18 so maybe they are more common touching land in versions before that?
The game does check if a point is surrounded by ocean before assigning it to Mushroom Island but it only checks adjacent points (not along diagonals):
Also, these initial "points" go through multiple stages of zooming which randomly shift their boundaries to form the intricate fractal shapes in the old generator, which otherwise looks deceptively simple, and this can effectively merge adjacent land masses (or convert them to mostly water; the game always places land at the (0,0) cell but it may not result in a proper landmass there), and yes, allow "hot" and "cold" to rarely directly touch despite code that separates them (I have similar code in TMCW to separate hot/cold ocean from land extremes which has a total of three passes to help avoid any such intersections, I think 1.7 only has one, same for 1.13 and land/ocean extremes). On average it appears to grow land since about 25% of the world in 1.6.4 is land while there is only a 10% chance that a given point will be a continent, which itself is just a random 10% chance that a point will be land; likewise, Mushroom Island is about 2% of the world (twice its initial chance).
Of interest, if you remove the "fuzzing" effect from the zoom layers you get biomes which are perfect squares, which is how I achieved this result in a "debug" world type, and show how simple the underlying biome placement really is (it just starts out as a simple grid of square cells, in this case I place biomes ordered by their ID so it is more apparent, otherwise the code shown above just loops through x and z and randomly sets each point to 1 (land) or 0 (ocean), Later stages may decide the actual biomes, or if part of a biome should be turned into another biome ("hills") or its boundary ("edge"); rivers are generated by creating a "biome" map which has two "biomes" and turning the boundaries between them into rivers, then overlaying that onto the actual biome map):
I just found a Mushroom Island biome directly adjacent to a Savanna Island.
From what I've researched online, the chance of this happening is less than 0.001%.
Can someone confirm this for me and tell me in the comments?
Here's the world seed, but I'm playing on a server, so I don't know if it changes the terrain generation: 8503596189782575358
Oh, and the coordinates: x: -3777, y: 85, z: -2316
(I went into creative mode just to take this better screenshot, I'm not a cheater...)
Without knowing the version of the game, or more specifically what version of the game the terrain generated in (and if mods which impact terrain generation were used on the server generating it), seeds can be misleading.
Initially, I checked and found that this does not exist with the current (1.18+) terrain generation. Here is what is at those coordinates.
There is a small lovely island off shore at spawn though.
https://mcseedmap.net/1.21.5-Java/8503596189782575358
If I change the version to one between 1.7 and 1.17 (I used 1.16), then there does appear to be one at those coordidnates.
https://mcseedmap.net/1.16.5-Java/8503596189782575358
I'm not sure of the chances of it happening though.
"'Tis foolishness! If all were so easy, why, none would suffer in this world!"
If you're having performance concerns with Minecraft, I hope this may prove useful.
A retrospective of the most important game to me (or, a try to stay awake while I never stop talking about something challenge).
It`s version 1.7.10, the server don´t have mods, the terrain generation it´s peculiar. Originally, the server was for version 1.0, but I upgraded it to version 1.7.10, so the terrain generation was changed. And the server wasn´t generated in large biomes or amplified.
Does this awnser your question?
(I´m brazilian, so don´t pay attention to my ugly english.)
Yeah, I figured out that it was generation in a version between 1.7 and 1.16.
The my understanding, the reason mushroom islands are almost always surrounded by water is because the game uses a certain value that will usually only be reached in an ocean. I think it was something like a value of 1 = land, 0 = ocean, and -1 equals mushroom island? But it might be different for versions before 1.18 so maybe they are more common touching land in versions before that?
"'Tis foolishness! If all were so easy, why, none would suffer in this world!"
If you're having performance concerns with Minecraft, I hope this may prove useful.
A retrospective of the most important game to me (or, a try to stay awake while I never stop talking about something challenge).
The game does check if a point is surrounded by ocean before assigning it to Mushroom Island but it only checks adjacent points (not along diagonals):
if (var17 == 0 && var13 == 0 && var14 == 0 && var15 == 0 && var16 == 0 && this.nextInt(100) == 0) { var10[var12 + var113] = (byte)Biomes.mushroomIsland; }Also, these initial "points" go through multiple stages of zooming which randomly shift their boundaries to form the intricate fractal shapes in the old generator, which otherwise looks deceptively simple, and this can effectively merge adjacent land masses (or convert them to mostly water; the game always places land at the (0,0) cell but it may not result in a proper landmass there), and yes, allow "hot" and "cold" to rarely directly touch despite code that separates them (I have similar code in TMCW to separate hot/cold ocean from land extremes which has a total of three passes to help avoid any such intersections, I think 1.7 only has one, same for 1.13 and land/ocean extremes). On average it appears to grow land since about 25% of the world in 1.6.4 is land while there is only a 10% chance that a given point will be a continent, which itself is just a random 10% chance that a point will be land; likewise, Mushroom Island is about 2% of the world (twice its initial chance).
for (int var6 = 0; var6 < par4; ++var6) { for (int var7 = 0; var7 < par3; ++var7) { this.initChunkSeed((long)(par1 + var7), (long)(par2 + var6)); var10[var7 + var6 * par3] = this.nextInt(10) == 0 ? 1 : 0; } }Of interest, if you remove the "fuzzing" effect from the zoom layers you get biomes which are perfect squares, which is how I achieved this result in a "debug" world type, and show how simple the underlying biome placement really is (it just starts out as a simple grid of square cells, in this case I place biomes ordered by their ID so it is more apparent, otherwise the code shown above just loops through x and z and randomly sets each point to 1 (land) or 0 (ocean), Later stages may decide the actual biomes, or if part of a biome should be turned into another biome ("hills") or its boundary ("edge"); rivers are generated by creating a "biome" map which has two "biomes" and turning the boundaries between them into rivers, then overlaying that onto the actual biome map):
TheMasterCaver's First World - possibly the most caved-out world in Minecraft history - includes world download.
TheMasterCaver's World - my own version of Minecraft largely based on my views of how the game should have evolved since 1.6.4.
Why do I still play in 1.6.4?
Ok let´s just pretend that I understanded everything of that Minecraft code and continue with my live.
But thank you guys with the help!