Content: Slate Blackcurrant Watermelon Strawberry Orange Banana Apple Emerald Chocolate Marble
Background: Slate Blackcurrant Watermelon Strawberry Orange Banana Apple Emerald Chocolate Marble
Pattern: Blank Waves Notes Sharp Wood Rockface Leather Honey Vertical Triangles
Welcome to TerraFirmaCraft Forums

Register now to gain access to all of our features. Once registered and logged in, you will be able to contribute to this site by submitting your own content or replying to existing content. You'll be able to customize your profile, receive reputation points as a reward for submitting content, while also communicating with other members via your own private inbox, plus much more! This message will be removed once you have signed in.

  • Announcements

    • Dries007

      ATTENTION Forum Database Breach   03/04/2019

      There has been a breach of our database. Please make sure you change your password (use a password manager, like Lastpass).
      If you used this password anywhere else, change that too! The passwords themselves are stored hashed, but may old accounts still had old, insecure (by today's standards) hashes from back when they where created. This means they can be "cracked" more easily. Other leaked information includes: email, IP, account name.
      I'm trying my best to find out more and keep everyone up to date. Discord (http://invite.gg/TerraFirmaCraft) is the best option for up to date news and questions. I'm sorry for this, but the damage has been done. All I can do is try to make sure it doesn't happen again.
    • Claycorp

      This forum is now READ ONLY!   01/20/2020

      As of this post and forever into the future this forum has been put into READ ONLY MODE. There will be no new posts! A replacement is coming SoonTM . If you wish to stay up-to-date on whats going on or post your content. Please use the Discord or Sub-Reddit until the new forums are running.

      Any questions or comments can be directed to Claycorp on either platform.
The Shadow

46c Experimentation: Ores

115 posts in this topic

Thank you for doing this! Now everyone can clearly see just how badly broken the new ore gen is.

Has anyone looked at how the new maps distribution percentages compare to maps from older Pre's?

Also, since ore gen is specific to certain rock types and 350x350 blocks can easily miss some types, can you expand the range out to at least 63 chunks (1000 blocks).

0

Share this post


Link to post
Share on other sites

Sorry that took a bit, those of you who want to do some testing can find the file you'll need here. (http://depositfiles....files/0xtom4zei) This file does not bring in -all- the new block types from TFC, just the ores on block ID 213, 214, and 215. I wasn't concerned about the dirt block types or fruit trees or any of that, and I was just ditching all the entities anyway. If you want to add them though, it should be pretty straight forward.

Along with that you'll want to go grab MCEdit 0.1.1 (assuming you don't have it already)

Also, since ore gen is specific to certain rock types and 350x350 blocks can easily miss some types, can you expand the range out to at least 63 chunks (1000 blocks).

I'll give it a shot (EDIT: Probably later tonight or tomorrow). The possibility I might miss some rock types is why I was generating multiple worlds though. My thought was that 23 chunks by 23 chunks was an area a group of players could search in a not totally unreasonable time, but if I went much larger than that the time to search the area would get pretty silly, pretty fast. The multiple worlds on the other hand were to ensure I was getting a mix of rock types and biomes.

0

Share this post


Link to post
Share on other sites

Awesome. Thanks for the details.

I haven't tried checking the surrounding stone types yet. For these tests I generated a world, allowed every chunk within my render distance to generate, then saved the world and opened it in MCEdit. In MCEdit I selected the area that had generated (a block of land 23 chunks wide and long), hit 'select chunk' to make sure I was getting everything from floor to ceiling, then used Analyze to get the total number of each block type (which I saved and imported into Calc, where I did all the checking of percentages and such).

Yeah, I'll definitely try to take a very wide sample in terms of blocks and let it run. Thanks again for starting the search going.

0

Share this post


Link to post
Share on other sites

Awesome. Thanks for the details. ... Thanks again for starting the search going.

De nada, glad to see some other folks getting interested in this. Woo~ Numbers! Spreadsheets! Probabilities!

Oh hey, one thing. If you want to check the stone types in MCEdit, you'll have to edit that yaml file to add the right data values. If you're not comfortable doing that, I'd be happy to, it just might take me a bit.

The TFC stone types are on block ID 209, 210, 211, and 212. If all you want to know is what class of stone they are, that's enough right there (209 is Igneous Intrusive, 210 is Sedimentary, 211 is Igneous Extrusive, and 212 Metamorphic). If you want the specific stone types though you'll have to fool around in something like Too Many Items to find the right data values to go along with the block IDs.

EDIT: Updated the minecraft.yaml file to include the stone types. The new file can be found here. (http://depositfiles.com/files/nz7xri5du)

1

Share this post


Link to post
Share on other sites

Thanks for the .yaml config. Worked like a champ.

The below table was generated from my build43 world I've been playing for comparison to the b46 data. It's a sample of an area of 560*544*256 blocks and at least part of 5 biomes - I tried larger but MCEdit memory errored and CTD-bombed either in purely rendering or during analysis. It looks like, barring someone with a better method, doing a comprehensive study of a very large area will be very time and labour intensive.

Statistically, I'm not fond of these small snapshots so I believe limited conclusions can be drawn. Sure, it's 78M blocks in the sample but without specific expectations from Bioxx for validation, I'd rather have at least two more orders of magnitude of data before trying to draw any global conclusions. Obviously the total quantity is VERY small at 0.067%

It'll take some time to gather more, non-overlapping data. I've not explored TOO much of this world beyond this sample, as yet.

Interesting that Jet is on the list - I thought Bioxx had killed that before b43 but I must be mistaken.

Posted Image

When in doubt, .png to the rescue. I could not get the damn text to format cleanly so I apologize for my forum naivety.

2

Share this post


Link to post
Share on other sites

(214:13) Saltpeter 891 1.70 0.001

A save that doesn't have massive deposits of Saltpeter.. madness! Was this world generated in b46c, or in an earlier version?

0

Share this post


Link to post
Share on other sites

A save that doesn't have massive deposits of Saltpeter.. madness! Was this world generated in b46c, or in an earlier version?

This was a b43. I did my current world for comparison. This weekend I'll do some more areas in the b43 and then generate some b46c and see if we can detect any significant delta on a similar set of samples.

0

Share this post


Link to post
Share on other sites

Very impressive work. I'm currently attempting to try and provide my own data for this thread. I even figured out how to generate a TFCraft world in MCEdit so that I could predefine the chunk size. I started at 100x100 chunks which gives me a world that's ~1800x1800x256, but I keep getting a memory error when I try it. :unsure: Tired now, will try more tomorrow.

On topic with this post, I noticed the change log for 46d says they'll be a ~10% increase in ore generation rates. New data awaits!

0

Share this post


Link to post
Share on other sites

Double checked in a vanilla world, I got 0.06% Gold Ore, 0.54% Iron Ore, and 0.02% Diamond Ore (which is pretty much what I expected).

right, and again nice work, but i see i wasn't too far off, i mean heck based on your numbers just iron ore alone nearly outspawns the entire gambit of ore by itself in tfcraft. by about 3 hundredths actually to be more specific.

0

Share this post


Link to post
Share on other sites

right, and again nice work, but i see i wasn't too far off, i mean heck based on your numbers just iron ore alone nearly outspawns the entire gambit of ore by itself in tfcraft. by about 3 hundredths actually to be more specific.

Well, to be fair, that's because Iron is pretty much the only tool-making ore in vanilla minecraft (other than diamond, which is extremely rare). If you really wanted to equate them, I suppose stone/tin would be the equivalent of wood/stone in vanilla, copper (and bronze) would be the equivalent of iron, and iron/steel would be the equivalent of diamonds (though because of the number of bars required to manufacture steel, iron probably should not have the same spawn rate as diamond).

The overall spawn rate seems fine to me, the issue I had with b45/46 was in the distribution of ore, and how difficult it made it to find anything. In vanilla, if you dug a hole straight down to the bedrock almost anywhere, you'd be sure to hit (or pass within a block of) at least one or two blocks of iron... in the last few builds of TFC, there were vast, vast areas where you would have no chance of coming within prospector's pick range of anything at all... which is an issue, since it makes finding the higher tier ores a bit too much of a gamble (in my opinion at least).

All of that may have changed for the better as of this last patch though... I haven't finished testing anything in 46d yet, but just looking at some early results it looks promising.

0

Share this post


Link to post
Share on other sites

A distribution of many but small ore deposits near the surface, and few but huge ore deposits deeper down, or in the center of extreme mountain regions, would seem appropriate.

Enough ore on the surface for 10+ players to progress up the tiers, and create infrastructure needed to slap down a large scale mine to get at one of the rare superveins, so they can get enough raw materials to fuel individual or group projects.

A big issue with the whole locating the large deposites deep down thing, is that (in my experience anyway) you tend to find that cave systems are sliced off by the next layer down, thats when you get those cave passages that 'sink' into solid rock.

It seems that generation of the passages and spaces is having problems maintaining cohesion across the layer boundaries (you can also see this occasionaly on the sea floor, when ocean terrain generation it hits the layer of rock and flattens out). This means finding a way down to the lower strata where you might pro-pick a motherload is pretty hard.

Personally Im in favor of a less honeycombed underground than you find vanilla. But if there are fewer cave systems, the ones that do turn up need to work down the layers.

0

Share this post


Link to post
Share on other sites

Updated data for 46d, this time with a larger sample. Last time I generated five 23 chunk by 23 chunk sections using five different world seeds, for a total of 2654 chunks in the test. This time I generated three 45 chunk by 45 chunk sections using three different world seeds, for a total of 6075 chunks being tested.

World I Results

Posted Image

World II Results

Posted Image

World III Results

Posted Image

Observations: Ore generation is looking much, much better (though it could maybe still use some tweaking).

Iron was present in reasonable amounts in all three worlds, as were copper and tin.

My only major complaint was the distribution of ores near the surface. Every world had at least some starter metals near enough the surface that you might find them with a prospector's pick, but the amount of ore was pretty meager and only one world had any copper or tin near the surface. Still, that isn't game breaking, just a matter of preference.

The amount of saltpeter still seems suspiciously high, but since there is next to no info on where (and how) it is supposed to be spawning, there's not too much I can say about that. It's possible the large amount of saltpeter is intentional? Perhaps intended for the eventual production of gunpowder? Maybe we're all supposed to be mining with TNT? Dunno.

1

Share this post


Link to post
Share on other sites

The fact that gold was missing from two of the three worlds you generated would seem to be a large problem to me.

0

Share this post


Link to post
Share on other sites

I'm not so worried about gold. Once someone can make steel tools finding the gold you need to make the higher tiers of steel would get a whole lot easier, just because of how quickly you can mine/dig. Although... that does give me an idea for something interesting to test later... "How far on average would someone have to travel to find all the materials necessary for high tier steel?"

0

Share this post


Link to post
Share on other sites

Well, to be fair, that's because Iron is pretty much the only tool-making ore in vanilla minecraft (other than diamond, which is extremely rare). If you really wanted to equate them, I suppose stone/tin would be the equivalent of wood/stone in vanilla, copper (and bronze) would be the equivalent of iron, and iron/steel would be the equivalent of diamonds (though because of the number of bars required to manufacture steel, iron probably should not have the same spawn rate as diamond).

The overall spawn rate seems fine to me, the issue I had with b45/46 was in the distribution of ore, and how difficult it made it to find anything. In vanilla, if you dug a hole straight down to the bedrock almost anywhere, you'd be sure to hit (or pass within a block of) at least one or two blocks of iron... in the last few builds of TFC, there were vast, vast areas where you would have no chance of coming within prospector's pick range of anything at all... which is an issue, since it makes finding the higher tier ores a bit too much of a gamble (in my opinion at least).

All of that may have changed for the better as of this last patch though... I haven't finished testing anything in 46d yet, but just looking at some early results it looks promising.

correct, although you overlooked a few key points and ores in vanilla, coal, lapis and emeralds (emeralds depending on build of course)

0

Share this post


Link to post
Share on other sites

Loving the work you're doing. Your data has caused me any my friend to decide to take the plung and regenerate our world. One question though: how are you determining "near surface"? Is it anything above y=90, anything <12 from grass or what?

The fact that gold was missing from two of the three worlds you generated would seem to be a large problem to me.

It does seem concerning given the data present, but as more worlds are generated and their results posted, the data will hopefully show more gold on the average world.
0

Share this post


Link to post
Share on other sites

While gold isnt too critical to TFcraft, pretty much every other big/popular mod uses a fair bit of gold (and redstone for that matter). Just worth keeping in mind.

0

Share this post


Link to post
Share on other sites

Loving the work you're doing. Your data has caused me any my friend to decide to take the plung and regenerate our world. One question though: how are you determining "near surface"? Is it anything above y=90, anything <12 from grass or what?

I'm being fairly generous in my definition of 'near surface' and including everything from y=119 up (25 blocks from sea level).

At some point I might try and write a filter for MCEdit that selects and copies just the blocks within 25 blocks of actual ground level, but that would be fairly complex (and memory intensive), so this works for now.

0

Share this post


Link to post
Share on other sites

I've been trying to create a few worlds and generate data for this thread. I've managed to bypass my memory error by using an unoffical 64-bit build (using the offical 32-bit always would cause it to run out of memory before completing a 45x45 chunk anaylsis). My results though are troubling. I actually logged into the minecraft_server.jar I created to make sure I was using 46e. Sadly, I was.

My process is this:

1.) Create a 45x45 chunk world in MCEdit using the 46e minecraft_server.jar as the world gen. Check the simulate world box.

2.) Hit CTRL+A once it's created to select the entire world and then hit the anaylze button

3.) Hit save file... when anaylizing is complete.

4.) Load comma seperate results into Open Office. Trim off anything above Granite as well as anything in the entities and below.

5.) For each ore, add up the entire column of stone and ores and divide the ore for a given row by that amount to get the percentage of ore per any given block

6.) For each ore, add up the entire cloumn of just ores and divide the ore for a given row by that amount to get the percentage of that ore in relation to all ores generated

I generated 6 worlds before I decided to stop, thinking something was wrong. Here is my results:

EDIT: Apparently html isn't allowed. Shame, simple <.table><./table> could of made this all much easier.

Posted Image

BTW, in regards to your lack of sulfur, sulfur seems to have its own block going by the config file and MCEdit when you look around lava. It's 208:0-208:3. 3 is the highest I found anyway.

EDIT2: BTW, I noticed in some of the worlds I was generating that certain rock formations didn't appear AT ALL. World 5 for instance had 0 blocks of Phyllite and a few other worlds were like that too. Anyone else having worlds generated that do that?

1

Share this post


Link to post
Share on other sites

1.) Create a 45x45 chunk world in MCEdit using the 46e minecraft_server.jar as the world gen. Check the simulate world box.

Hmmm.. considering the numbers you're getting are quite a bit off of what I've seen, I have to wonder if generating a world in MCEdit is handling all the ore generation correctly. If you would, grab the seed out of the world you generated in MCEdit and see if it produces identical results when used in game.

0

Share this post


Link to post
Share on other sites

I was assuming that's how you were generating a 45x45 chunk world. What is your method for pre-generating a world? Merely running around didn't seem reliable. I will try using the world seeds in SSP, but MCedit seems to put the spawn area in a weird location relative to the actual word (IE it puts me nearly in one corner of the world instead of dead center).

EDIT: It's hard to say for sure, but it appears SSP generates the same seed slightly different. It looks very similar, but there is slight differences (where sand is instead of dirt, where there's exposed stone instead of dirt, etc.). When I made the world with the same seed on a SMP server, the world seemed identical. On the one large hillside I was looking at, SSP had some dirt where stone was, but in SMP, the stone, dirt, and sand were in the exact same spots as in MCEdit. I double checked and the SMP and SSP seeds are the same. Not sure what amount of variation is expected. That said, I narrowed down a spot in the world where some ore was using MCedit. Both SSP and SMP had ore there, though I didn't dig it all out to make sure it was the same amounts. At a loss here. :unsure:

0

Share this post


Link to post
Share on other sites

To generate my worlds I'm using single player commands to turn on /noclip (to avoid falling in the next step), then teleporting to y=355... letting all the terrain in my view range spawn in (11 chunks in every direction +1 for the chunk I'm in == a 23 x 23 chunk area), teleporting 320 blocks along X or Z (/tele 320 355 0, /tele 320 355 320, etc.), and repeating. Takes a few minutes, but this way I can be pretty sure that I'm getting exactly the same terrain as I would if I were just exploring normally.

0

Share this post


Link to post
Share on other sites

keep up the good work guys, i think this is getting above my head though XD :)

0

Share this post


Link to post
Share on other sites

@Brodiggan: If you could try running your seeds in an SMP server and run around to see if you get the same ore distribution, I'd appreciate it.

I took World 1's seed from my previous results and did 2 tests. First I did your method of loading it up in SSP and flying around the area to generate the chunks. Then I started a SMP server and did the same (though slower since I didn't have /fly 15 :( ). Here's the results. MCEdit is the world I generated inside MCEdit with the minecraft_server.jar, SSP is single player with SPC to fly around generating the world, and SMP is starting an SMP server and running around. Seed was -591351512185364480 if you wish to reproduce these results:

Posted Image

EDIT: Nvm, thought that was .04 and not .4. My previous conclusion is irrelevant.

The ore generation seems to require fly overs. Simplying generating the world in MCEdit won't give you nearly as much ore as you should have. Still, the ore gen algorithim seems very strange. For the same world seed, there is a difference of ~.3% of blocks being ores. Almost 50% more ore in the SSP version than the SMP. With it being just one world, it's hard to draw the conclusion that there's a bug in play here, but it's worth further testing for sure. One other interesting bit of info is that, while there was 46155 claystone generated in MCEdit and 138821 in SSP, SMP said there was *0* claystone in the entire world.

0

Share this post


Link to post
Share on other sites

@Brodiggan: If you could try running your seeds in an SMP server and run around to see if you get the same ore distribution, I'd appreciate it.

Maybe at some point; I mostly play single player though, so the process of setting up a SMP server isn't one I'm familiar with. I'd be happy to provide you with the seeds if you wanted though.

In other unrelated news, I come bearing gifts (or at least a script). After fooling around with MCEdit for a while (and teaching myself some python), I wrote an MCEdit Filter that I think might be useful for people having issues with worlds they generated in b45/46. The script is pretty simple, it just walks through a MC world and adds small (3-9 block) deposits of the four vital metals for higher tier metalwork (copper, iron, gold, and nickel) in appropriate stone types. The deposits are irregular, but on average you'll find at least a small amount of all four metals within ~128 blocks of wherever you chose to set up a base. This won't give you enough to set up any serious heavy industry, but at least it can give you a place to start.

If you're mildly comfortable with python, it's easy enough to edit the average distance between deposits... just change the stepDist variable to whatever you'd like the average to be. I would not suggest setting it too low though, halving the step distance more than doubles the time the filter will take to run (and the amount of ore it would generate could get a bit silly if you went too low).

You can download the filter script here EDIT: Here ( http://depositfiles.com/files/gugwv4e67 ) Have fun with it.. but please, back up your worlds first. This worked great for me, but I absolutely cannot promise it won't bork up your save.

EDIT: Of course, this may all turn out to have been pointless once the new version comes out... but oh well.

EDIT EDIT: Assuming it isn't useless as of b47, I might come back and add user selectable options to this so no one has to edit the script to tweak things. That's all going to depend on whether or not it's worth the time though.

1

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.