Talk:Diamond Ore

Natural generation
Under this specific section is following statement "Diamond ore only appears at Y level 16 and below, and can be found in veins of 3–8...". Are there not veins of 9? RobinEH (talk) 11:15, 1 August 2015 (UTC)

9 and 10! But the page now reads 1-12 which I doubt, does anybody have the knowledge to clarify if veins of more than 10 are possible? (without merging 2 veins.) Hexalobular 1 Jan 2016

Veins with 10 ore can be found in seed "-8246714680227619312" (without citation marks of course) at 377/10/1, 52/8/360, 1066/9/898, 593/12/1023, 312/9/307 and -22/8/-59 (and 9-ore veins at 361/11/-19, 77/12/-487, 1166/12/1374, 1096/9/1270, 1198/13/1266, 769/10/930, 170/7/586, 315/14/284, 331/11/1088, 138/9/944 and -475/12/372) All of these fit into a standard shape of a 2X2X2 cube with an extra ore block on top and another on the bottom (only 1 extra block for the 9 block veins) Screenshots at https://drive.google.com/folderview?id=0B5c6qRTAo-4nSjI2WUhzOXprQkE&usp=sharing Hexalobular (talk) 10:09, 4 January 2016 (UTC)


 * Could these be a result of viens combining? -PancakeIdentity (talk) 00:02, 23 August 2016 (UTC)

No, they would be more random and wouldn't all fit in the same standard shape. --Hexalobular (talk) 01:54, 28 August 2016 (UTC)

We need a better way to refer to specified vein size and actual vein size. I think that it is causing confusion. In the page it says 0-8 vein size but the vein size according to the mineral vein is fixed at 8. Furthermore, the size parameter does not correspond with the max amount of diamonds that should be spawned. Max amount is 10. So, if we want to put a range it should be 0-10. Correct? Guihori (talk) 20:36, 7 May 2020 (UTC)

Minecraft is still generating veins of diamond ore with 10 ores in Java 1.16.1 and people are still misguidedly changing the Wiki to read 1-8. Diamond veins with 10 ores: Java 1.16.1, Vanilla, Seed:-597086656834133 Large Biomes -10/9/-404, -454/11/119, -1/9/408, -67/13/434, -975/5/9, -867/6/86, -496/13/-133, -718/7/-135, -824/10/-70 all of these match the standard vein shape, 1 was straddling the chunk border, 4 had a 1 block gap, the rest had 2, 3, 5 and 6 block gaps to the closest border. I have also found 29 9 ore veins that also all match the standard shape. Flying around underground over a few thousand chunks with everything except diamond ore removed I have seen exactly no sign of any merged veins, I did find 2 places where 2 veins were touching but they were unmerged, taking up a total of 8 columns, 4 per vein. It beggars belief that all of those 9-10 ore veins could be the result of invisible vein mergers with none of them nor innumerable veins with fewer ores showing any signs of mergers, all of them fitting the standard shape. There must be some merged veins but they are obviously WAY too rare to explain all of the 9-10 ore veins! --Hexalobular (talk) 17:25, 30 June 2020 (UTC)

A diamond vein can't stick out of its chunk by more than 1 block, any merged veins must therefore be in contact with a chunk border. So any 9 or 10 ore veins that don't contact the chunk borders must be single veins! They can't be explained by mergers! --Hexalobular (talk) 08:44, 2 July 2020 (UTC)

Found 5 veins with 9 ores in Bedrock 1.16.1 (Win-10) Seed:16 328/9/2,  133/10/-78,  99/6/-106,  109/9/-150,  100/9/-157 --Hexalobular (talk) 16:11, 9 July 2020 (UTC) And 1 with 10 ores, 400/11/-602 --Hexalobular (talk) 18:14, 9 July 2020 (UTC)

I ran an external test with 1,000,000 cycles for this and was given an effective range of 2-8 ores for the given value of 8 as diamond has. Missing stone from carvers and the like would explain smaller veins of size 1, but doesn't explain > 8 size veins. Code breakdown here: https://pastebin.com/kgjt1Nye Sample rates here: https://i.imgur.com/F7QK18j.png Scimiguy (talk) 06:31, 2 January 2021 (UTC)

@Scimiguy What test? Did you look at actual veins or the value of a variable in the code? I don't understand the code but I have it from a good source, TheMasterCaver in minecraftforum.net, that the variable used doesn't set a max size but is used in some other way during the generation, which agrees with the comments in your code breakdown which talks of spheres and rings of varying sizes.

What does your graph show? 2 and 8 don't jump out at me when looking at it. Did you look at my 10 ore veins in game? Hexalobular (talk) 09:14, 2 January 2021 (UTC)

Could 2 attempts to add the same small ring account for 1 ore veins without overwriting? Hexalobular (talk) 09:19, 2 January 2021 (UTC)

Apologies - the C++ code linked was to generate bundles of spheres, so the 'sphere' in the code is actually an ore block. 1 sphere = 1 ore block. The code itself is an extraction of the java code of OreFeature's methods generate & func_207803_a, redone in C++ for easier visualization.

And I wasn't looking at a variable in the code, I was looking at the entire code for ore generation for diamond, using the variable as the input variable for the function. According to that, the output after a million tests gives a range of 2-8 ores per vein of diamond. Empirical evidence is better than anecdotal evidence, and until someone can actually PROVE that veins generate larger than 8 with actual evidence rather than anecdotal 'I saw a vein with 10 ores in it', I don't see why we shouldn't leave it as what the code outputs. If you have a source, provide it. I provided my source, and results.

I might also just re-create that exact code in java itself in a dev setup and run 100 million cycles on just value 8 to get the best possible output, I'll provide my results as soon as I can w/ attached code Scimiguy (talk) 09:24, 2 January 2021 (UTC)

My evidence isn't anecdotal! I provided seeds and coordinates where you can see for yourself, how much more empirical can it get? Way too many, and too regular, examples to be the result of vein merging. (Also examples entierly inside a single chunk.) And could you please explain how to interpret your data? (It's not obvious what the tabel/graph show.(Ok, I figured that out, but how are min and max generated?)) Can you point to how the "spheres" are generated as single blocks? And how do the varying sizes of the "rings" affect things? Hexalobular (talk) 09:34, 2 January 2021 (UTC)

Can confirm in 1.16.4 that diamonds can generate up to 10 in Large Biomes. In Default, IDK. The Great Spring (talk | contribs) (Tagalog translation) 10:15, 2 January 2021 (UTC)

Lol yeah that's not how it works. Regardless, I'll be back after I do it in Java so it's easier to understand and we'll go from there. Scimiguy (talk) 10:25, 2 January 2021 (UTC)

My examples _are_ what the code outputs, when it's used in the actual game. "If the map doesn't agree with the ground the map is wrong." Hexalobular (talk) 10:35, 2 January 2021 (UTC)

@Scimiguy You might want to take a look at TheMasterCaver's simulation, he did something similar to you but modifying the code to place 1's in an array, instead of ores in the world, and count the numbers of 1's. His result was veins of 0-10 ores. https://www.minecraftforum.net/forums/minecraft-java-edition/discussion/3048635-what-direction-s-do-diamond-veins-grow-when-they#c7 I don't know that I have the programming knowledge to figure out if your "spheres" or his 1's are a better representation of ore blocks in the game, all I know is that the game absolutely does generate veins with 10 ores. I again urge you to check out my examples higher up on this page (Note that the first batch are for an older version (1.8.9) from 2016 and the batch for 1.16.1 are in a Large Biome world.) Screenshots at https://drive.google.com/folderview?id=0B5c6qRTAo-4nSjI2WUhzOXprQkE&usp=sharing and another at the bottom of the Wiki page. Hexalobular (talk) 16:24, 3 January 2021 (UTC)

All good, I commented on that post as well with newer results which confirm 1-10, so we're good :) I posted results and source on that same MCForums thread, so they can be viewed there if needed Scimiguy (talk) 18:56, 3 January 2021 (UTC)

Did a more comprehensive test with more detailed results, can be found here: https://www.minecraftforum.net/forums/minecraft-java-edition/discussion/3048635-what-direction-s-do-diamond-veins-grow-when-they#c11

This can pretty much be our source I think for this Scimiguy (talk) 11:37, 4 January 2021 (UTC)

Range
I can verify that in snapshot 21w06a diamond ore generates down to at least Y = -62 (but should info about features in snapshots be on the main page?) In seed "-8762690768343294453" (without citation marks of course) I found a vein at -65 -62 -121 And one at -93 -63 -27 Hexalobular (talk) 11:54, 14 February 2021 (UTC) Diamond ores at Y=-63 in 21w07a, seed "-740008907303374303" (without citation marks of course), 5 -63 208 and 154 -63 212 as well as 476 -63 209 Hexalobular (talk) 09:45, 22 February 2021 (UTC)

level
A diamond ore article requires a layer of discovery.hi i,m is to the finland (talk) 11:08, 30 May 2020 (UTC)

Clarification for generating 'once per chunk'
Under "Natural Generation", it mentions that diamonds will spawn once per chunk, but it isn't clear if it means a 16x256x16 space (16 in x and z, but extending all of the y values), or a 16x16x16 space (16 in all directions, sometimes called a 'sub-chunk'). This leads to some more confusion because the Chunk page strictly defines a chunk as 16x256x16, but then in the the in-game F3 menu defines a chunk in terms of 16x16x16 and the chunk coords are defined as "x%16 y%16 z%16 in x//16 y//16 z//16", clearly showing that the in-game definition of a chunk is 16x16x16. So if the number of "once per chunk" was determined by looking at minecraft's code/update logs/other official sources, it may follow that definition instead. If the numbers were determined experimentally, then it should be reflected such as "once per 16x256x16 chunk". Given that Diamonds spawn from 1-15, this could present the opportunity for diamonds to spawn twice as frequent as mentioned in the article. --Ximmrspyx (talk) 05:23, 22 June 2020 (UTC)

What? Surely 1-15 fits in the lowest subchunk, 0-15? --Hexalobular (talk) 11:18, 30 June 2020 (UTC)