Observations and Questions About the Prospecting Pickaxe in Node Mode

[Liu Bei]

I've been doing empirical studies about the ProPic to understand how it works and what is the best way to use it in the field.  I'm going to post over the next period of time my findings starting with finding out about the use of the ProPic in node mode.

Most of the work was done in survival exploration mode in order to spend more time on mining studies on many world configurations and instances.  This first exercise is pretty simple though. I created a world with the seed 'mining test world' or -921744543.  It is a beautiful, limestone, chalk heaven with small mesa spires scattered about. (I might even go back and play it up to iron at least.)

I went into /gamemode 2 and built some granite cobblestone platforms and then built up a mining test rig of poor native copper ore in granite.  I cleared out all the local surface copper deposits which were in the ground out to a distance of 37 blocks in all directions to avoid interference from them. I was in /gamemode 1 while doing actual tests.

Here is a pic of the final test rig: 6 5x5 layers with a group of 13 blocks on top of that.  This is a total of 163 blocks.

 

Face of 6 layered 5x5 test rig with 7th layer and bore hole in face

 

I then built a little step-up to the center of one face and took out the three blocks toward the center and then put a granite test block at the center. 

 

Top of test rig for 160 ore blocks, layer 7

 

This means there are now 160 Cu ore blocks in the test rig. I set the ProPic 'radius' to 4, which means that the ProPic will scan all blocks in a 9x9x9 cube centered on the test block.  In this case all the Cu ore blocks were within the 9x9x9 cube. Cooly enough, the ProPic posted a 'huge' finding.  I removed one block on the top and redid the test and it posted 'very large'. I then tested settings from 5 to 16 and got the same results. Apparently, the value in the VS wiki is wrong.  I also tested the 'radius' settings with 160 blocks at 3, 2 and 1.  I found, as expected:

Setting Cube Size Cube Volume Expected Blocks    Posted Values
   4          9x9x9           6859              159, 160         'very large', 'huge'
   3          7x7x7           3375              159, 160         'very large', 'huge'
   2          5x5x5            125                   123                  'very large'
   1          3x3x3             27                     26                    'medium'

From this test I concluded:
 1) the values in the wiki for prospecting results for 'very large' and 'huge' are probably wrong and should be updated. I'm skeptical that a world class programmer like Tyron would pick 120 over 160 in a sequence like that...
 2) using a similar approach I found the other values in the results table to be correct.
 3) who knew you can go up to a 16 block 'radius'?  I've been doing this for 10 months now and it is very helpful when exploring for ores that are deposited shallowly (not as surface ores).  BTW, your mileage may vary on whether your game rig can handle 16 block 'radius'.  I have an Intel i7-8550U CPU at 1.85 GHz and an NVIDIA GeForce GTX 1050 video card laptop and it's no problem.
 4) in survival exploration mode you don't need to restart the game when you switch node 'radius' values.  This is very important for developing excellent strategies for finding ores in good locations. I'll talk about this in a later post.  I assume this applies to survival standard mode?  Of course, the first time you enable the node mode you have to sign out to activate this change...
 5) I don't know if switching the 'radius' value is a global setting in multiplayer mode or not.  To me it is incredibly valuable to be able to set it personally since this really reduces the time to find ore lodes in a changeable way. I hope this remains a default condition.

Some things I might do in the future but probably won't, because there's other stuff to get done...
 1) didn't test that poor, medium, rich, bountiful make a difference.  Pretty sure they don't.
 2) didn't test that different ores make a difference, since my general experience is that this doesn't matter.

[Streetwind]

7 hours ago, Liu Bei said:

1) the values in the wiki for prospecting results for 'very large' and 'huge' are probably wrong and should be updated. I'm skeptical that a world class programmer like Tyron would pick 120 over 160 in a sequence like that...
2) using a similar approach I found the other values in the results table to be correct.

Looks like 120 is the point where it goes from "large" to "very large" then, not from "very large" to "huge"?

Which is odd, because then you have an enormous range somewhere in there. Assuming "medium" still ends at 39 as the wiki says, "large" would then be 40-120. That seems... a little much? Can you confirm that range?

 

7 hours ago, Liu Bei said:

3) who knew you can go up to a 16 block 'radius'?

Not sure what the upper limit is, but I've seen people talk about doing debug experiments with a range as high as 50.

 

7 hours ago, Liu Bei said:

4) in survival exploration mode you don't need to restart the game when you switch node 'radius' values.  This is very important for developing excellent strategies for finding ores in good locations. I'll talk about this in a later post.  I assume this applies to survival standard mode?  Of course, the first time you enable the node mode you have to sign out to activate this change...
 5) I don't know if switching the 'radius' value is a global setting in multiplayer mode or not.  To me it is incredibly valuable to be able to set it personally since this really reduces the time to find ore lodes in a changeable way. I hope this remains a default condition.

It is a global world setting, and thus cannot be configured on a per-person level in multiplayer. This is unlikely to change, due to the way worldconfig settings work. They quite literally edit the world itself, not the player. And even if it was configurable on a per-person level, it still would not be useful in multiplayer, because worldconfig commands require admin permissions to use. In singleplayer you are always admin; in multiplayer, that is certainly not the case.

As for the fact that the range can be changed without a restart once the mode is activated... that may well turn out to be classified as an exploit. I mean, I'm not a developer working on the game, but I know the history of node search mode. Tyron was very hesitant to add it in the first place, and to this day, it remains inactive by default in all presets except for Exploration despite the fact that it's a useful tool for prospecting. Indeed, perhaps it is too useful for his tastes? And if you can now manually circumvent the inherent tradeoffs it comes with, too...

Your choice in node search range is meant to matter. Choose a higher number, and you can detect ore in a higher radius, but it becomes very tedious and work-intensive to triangulate the ore vein's position. Choose a low number, and you can easily pin down the position of a vein with just a small amount of effort, but you'll naturally detect fewer of them.

Being able to change your range on the fly is doubly bad, because 1.) it circumvents an inherent and intended tradeoff in a feature that the developer is already very hesitant to offer in the first place; and 2.) it results in degenerate gameplay behavior, in which the player makes entering admin commands a core part of their gameplay. That is the antithesis of immersion, which one of the primary design goals of Vintage Story.

So yeah, sorry to shoot down your excitement, but I would bet money on 20:1 odds that this is an exploit.

[Liu Bei]

Thanks Streetwind, some good comments.

I went back and redid the medium to large boundary and the large to very large, and found 39-40 and 79-80 respectively:

Test of medium to large with 39 vs 40 ore blocks

 

Test of large to very large with 79 vs 80 ore blocks

So the sequence is 1, 10, 20, 40, 80, 160. Not quite a geometric sequence but close enough.  I think the geometric nature of the sequence may have a lot to do with the choice of the optimal propic radius setting for one's particular way of playing.  More importantly, it also affects the way folks might choose a good setting for multiplayer worlds.  I have to think about this more. I'm aware this is a topic for polls and debate.

I enjoyed your comments about the intent of use behind the tools and in the main I agree with you.  Of course, I know that Tyron is putting the 'uncompromising' in Vintage Story at every turn and that is a good thing.  The architecture of the standard game is clearly meant to provide a difficult though rewarding experience for the player as one progresses through the technology levels.  I have the highest praise for the devs to have provided a really good platform for enabling players to pick the level of difficulty and style of play.  Good on them!

Yeah, and I really do agree with you that changing your propic setting on the fly is not immersive, although I'm not sure I am quite as degenerate as you suggest.  Even adding a little button to the propic to change settings would not really be immersive for what is essentially a manual technology tool with magic properties.

BTW, you used the word vein for ore lodes.  I think that the ores come in several varieties that make searching for ores more complicated. Two are the most common.  One style is the iron ore style where there is truly a vein structure: large flat, thin lodes that extend over many blocks.  The style of finding a high ore probability density location on the map and drilling down works pretty well.  The strategy for finding and mining the lodes is reasonably clear.  The second common style of ore distribution is more like raisins in a bread pudding, for example cassiterite, bismuth and sphalerite.  The ore bodies are more compact and salted throughout high probability density areas.  Drilling down and hoping to find a lode is not so likely.  This requires a different strategy I think.  Given your view that changeability of the propic setting is not going to be a play mechanic of any use, I am going to rethink and test new strategies for finding these ores in standard gameplay.

 

 

 

 

 

 

[Streetwind]

13 hours ago, Liu Bei said:

I think that the ores come in several varieties that make searching for ores more complicated.

The neat part about Vintage Story defining most of its game mechanics in JSON files is that you don't have to just "think" - you can look it up

There are currently two distinct ore deposit generation styles, although one exists as a collection of subvariants and the other is more of a special case that ultimately refers back to the former. In practical application, you can eyeball it to "almost all ore generates roughly the same way". That's all a player needs to know, really, because it means that there is no need for different strategies. Digging a straight shaft down will pretty much always be the optimal way to encounter deposits, since ore blocks only ever get displaced along the y-axis. Viewed top down, every ore deposit is some form of disc.

(Also note that I only said 'vein' in the other post because it's a commonly used word and shorter to type than 'deposit', which is what the game calles them.)

If you want to get into the nitty-gritty details of each generator method:

• disc-followsealevel: This is how (for example) iron ores generate. The game determines a spawn point, which becomes the center of a horizontal, disc-shaped deposit of a certain specified radius. The thickness of the disc can also be defined in detail, with average and variation values and even the distance function used; but many ores in Vintage Story just default to a thickness of 1. This particular disc generator follows the sea level of the world, which generally means it is going to be largely flat. Only rarely will you encounter an elevation change in the disc.
• disc-followsurface: This is how (for example) cassiterite generates. Ores with this generator still produce a disc exactly the same way as above, but the individual blocks will be displaced along the y-axis according to the elevation of the surface. So if there is a massive cliff face up above, that might mean that one half of the disc might be pulled up dozens of blocks higher than the other half. If it spawned close enough to the surface, this may even cause the ore to be exposed in the flank of said cliff. But most of the time, the effect is far less dramatic, and you might just get a somewhat slanted disc following the slope of a hill.
• disc-followsurfacebelow: This generator doesn't follow the rock layer above it, but rather the rock layer below. Why bother with the distinction? Well, because this is how things like peat and clay generate: they layer themselves on top of a surface. But, in the end, this is still the same disc generator.
• disc-anywhere: Quartz and olivine use this. Still the same disc generator, but this one just ignores pretty much all other worldgen factors and just sprawls itself in there in whatever way it pleases.
• disc-alluvialdeposit: A special variant that involves proximity to water in some way. Used to generate gravel.
• childdeposit-pointcloud: This is the only generator that is not directly a disc variant. And it is only used for things that spawn inside other ore, such as gold and silver. The way it works is that it attempts a defined number of times to spawn a tiny deposit (which by the way use the disc generator again) in the parent material. As a result you can have a quartz layer which is peppered with groups of 2-3 blocks of silver ore every couple meters.

And now, veterans might say: but what about halite salt domes? Those generate as huge vertical pillars, so that's got to be different generator, right? But no: halite uses disc-followsurface. It just has a thickness value defined that is way higher than the disc radius. So the result is a cylindrical shape. And the dry lakebed variant of halite uses disc-followsurfacebelow.