Coal_ore is nearly 0.52 times less dense and iron_ore is about 0.64 times less dense in 18w versions compared to 1.12.2 and earlier versions. Density is measured relative to stone plus andesite plus granite plus diorite plus dirt in both versions. If this is intentional, please ignore. Otherwise, it makes survival significantly more challenging, especially early in the game. The plots show density as a function of height for both versions.
For reference, the ratio of the densities in 18w08b and 1.12.2 for r.0.0.mca seed 12345 are:
0.52 coal_ore
0.64 iron_ore
0.66 diamond_ore
0.72 lapis_ore
0.64 redstone_ore
There is statistical noise in all of the measurements, most for diamond and lapis.
For what it is worth, I also analyzed the average patch or deposit size for coal_ore. The average size (number of coal_ore blocks) of a contiguous deposit of coal_ore in 18w08b is about 60% of the size of contiguous deposits in 1.12.2. There are also about 90% as many deposits in 18w08b as in 1.12.2. Basically coal_ore deposits in 18w08b are similar in size to iron_ore deposits in 1.12.2.
—
As of 18w20a/b nether quartz ore and lapis ore are still affected.
Linked issues
is duplicated by 2
Attachments
Comments 43
The volume analyzed was one region up to height 64 (512x512x64) for both versions. I believe the world generation algorithm changed for the 18w series, so the same seed doesn't generate the same world.
The difference really is quite big... if this isn't a coincidence depending on the seed, I can only imagine it being an intentional change.
A seed in 1.13 generates mostly the same world as in 1.12.2, even though world generation changed. Can you check it with the same seed as well? As @unknown said, it might be a coincidence depending on the seed and I want to make sure that it isn't.
I generated two worlds with seed 12345 and traveled to 0,0 in both 18w08b and 1.12.2. The attached screen shots show the terrain and biome are the same but trees, grass, flowers are in different places. The devil is in the details. It will take a while to force a save of a full region in both versions for analysis, unless you have a trick.
[media][media]
@WizzBangWAI:
think you test too subjective. Use the fill command or other objective testing tools. My small tests with the fill command show that the bug is fixed.
It would be good if Andrew Jankevics with his tool or somebody else could test a complete world with thousands of chunks.
@Xitomati
You're assuming I know anything about minecraft commands. How about sharing the process you use and the commands you use? Are you on a local world, or a sever world? I'm using a server world. Local world generation might be fine, server world generation might not be.
I restarted the world and cleared levels 1-20 of everything but ore in the same chunks I mined in. It's amazing, there's actually ore generated. I did the same on a backup of the previous world, pretty baron. Used the same seed on 1.12.2, ore generation looks to be identical to the current world. I may or may not have hit a fluke. For those wondering, I used the fill command to replace blocks with air.
I use the fill command and replace in a range of 170x 170 blocks (about 113 chunks) the chosen ore with for example air. I compare that with a world generated in 1.12.2 (singleplayer).
According to testing I've done today with 1.13.1 vs 1.12.2, this bug does not appear to be fixed after all. Ore generation is still significantly different.
Using similar methods to the above, and looking at a 181x181x64 sample from the same coordinates in the same seed, I counted every block of every overworld ore type in both versions, and here is what I found:
Ore Type | 1.13.1 | 1.12.2 | Diff% |
Coal | 25,375 | 23,308 | 8.9% |
Iron | 13,485 | 13,414 | 0.5% |
Gold | 2,320 | 1,193 | 94.5% |
Diamond | 290 | 495 | -41.4% |
Lapis | 435 | 507 | -14.2% |
Redstone | 4,785 | 3,598 | 33.0% |
This is a test of about half a million blocks. Obviously a larger sample size would be required to get an accurate Diff% for all cases, but this test should be enough to show that concentrations still vary very significantly between 1.12.2 and 1.13.1.
How many chunks did you analyze? Also, was it the same seed in both versions or a different one?