CrazyBreeze1074

I created the new ticket upon suggestion from a moderator, Golden Helmet.

It looks like leftover orphaned entity keys are not being removed in 1.19.20. I counted 5409912 keys in data taken from one of Prowl8413's servers a day or so before the update. About 16 hours after the update there are still 5454388 orphaned entity keys. Only 6 of the pre-update keys are not part of the post-update database.
 
 
 
 
 
 
 
Message #bugs
 

I confirm on a server with many users this unlisted entity data contributes to a large expansion in world size. On this server I found 10,370,925 unlisted entity keys. Only 86,906 were part of the digp key space. The consequence to the world is the database if 6.8 GB after counting the actor keys. And it goes down to 3.34gb after removing the 10,284,081 unlisted entity keys.

This bug still affects the 1.18.12 hotfix.

I performed an analysis of spawner abundance on a largish 1.18 world as requested by GoldenHelment. I used rbedrock for analysis; seed 26874156; and 251676 chunks of data. I found skeleton spawners and normal spider spawners generated about 1/2 as often as zombie spawners.

Spawner        Freq
------------  -----
cave_spider    2225
silverfish        1
skeleton       1146
spider         1057
zombie         2205

I repeated this analysis with a smaller 1.18 world using seed 1808432794 and 31757 chunks of data. For this plot I binned the data into 10 block wide y-intervals due to data sparsity. Again, skeleton spawners occur ~1/2 as often as zombie spawners.

Spawner        Freq
------------  -----
cave_spider     302
silverfish        1
skeleton        177
spider          171
zombie          357

Here's a skyblock world I worked on last on 8/22. Upon opening the world in the current version, I find floating water and leaves in what should be voided chunks.
https://drive.google.com/file/d/14C0TrIveWZqPbEA-KnByu8qDy_YRvado/view?usp=sharing

I analyzed a new bedrock world from the 1.18.0.24 beta When deepslate ores are considered, it appears Bedrock has approached parity with the Java snapshot as described by the original poster. Here is a table detailing the block locations at the minimum, maximum, and mean values and histograms of ore generation by y-value.

Here is some analysis of ore counts for several thousand chunks in Bedrock beta 1.18.0.22. As you can see, emerald ore is not distributed much below y = 50. Deepslate coal, copper, and emerald are completely missing in the world.