As many people have pointed out, looking at simple cluster count or ratio is not enough to predict success or failure, although it works most of the time.
If 1:1 works for Danbait clusters, and Danbaite clusters are 66%/33% Zinc/Copper Then the actual ratio of Copper to Zinc in these 1:1 smelts would be 2:1. However, I have not successfully tested Cuprite/Matraite yet at a 2:1 ratio.
I tested
| 3 Cuprite | 3 Danbaite | 5 brass |
| 3 Cuprite | 3 Matraite | 5 brass |
- Chichis
I tested
| 3 Cuprite | 3 Danbaite | 0 brass |
| 3 Cuprite | 3 Danbaite | 0 brass |
I tested
| 10 Cuprite (in 7 stones) | 10 Danbaite (in 6 stones) | 0 brass |
| 10 Cuprite (in 7 stones) | 10 Danbaite (in 8 stones) | 0 brass |
| 10 Cuprite (in 10 stones) | 10 Danbaite (in 10 stones) | 17 brass |
- Aqba
UPDATE: with additional data, it is clear that elemental ratio does not adequately predict success vs. failure. See Aqba's Alloy Experiments page.
OUT OF DATE: With the limited data Aqba has, elemental ratio may predict success vs. failure. So far, elemental ratios above 1.8 and below 1.2 have failed to produce brass, even when the cluster counts and cluster ratios predicted that there should have been success. However, Aqba needs more data points -- please send Aqba data on your failed brass smelts! (Include exact cluster counts for each different mineral, by name.)
(To learn about calculating element ratio, see Aqba's Element Ratio page.
It doesn't appear to have anything to do with elemental ratios in real life. What the failures do seem to be based on are the AVATAR. Skrug and I did tests using 3 Danbaite + 3 Cuprite. If HE dumped the cart into the furnace, it failed. If I dumped the cart into the furnace, it worked. We repeated this multiple times. - Chichis
Since there appears to be no difference between mineral yields (1.2 clusters per metal) then what is the purpose of having 5 minerals per metal. I think the purpose has to do with gem yields (thanks for the idea Azhrei!). I think that for each metal, the 5 mineral types have a set rarity. The list below appears to be sorted in mineral rarity. The most common mineral for the metal is on the left, the least common is on the right. I know it holds true for the orders for Zinc, Copper, and Iron. Going with Azhrei's theory on the forums, I'm going to label the column names with the gem size that mineral produces, should that set be gem-bearing mineral for that area.
- Chichis
| Name | Dirt | Small Gem | Medium Gem | Large Gem | Huge Gem |
| Titanium | Osbornite | Rulite | Brookite | Anatase | Kleberite |
| Zinc | Danbaite | Matraite | Ashoverite | Sweetite | Wulfingite |
| Magnesium | Periclase | Fluborite | Kotoite | Suanite | Brucite |
| Gold | Maldonite | Yuanjiangite | Auricupride | Weishanite | Calaverite |
| Anitmony | Paradocrasite | Valentinite | Senarmontite | Bystromite | Berthierite |
| Aluminium | Akdalaite | Corundum | Diaspore | Hibonite | Painite |
| Tin | Romarchite | Abhurite | Berndtite | Cassiterite | Stistaite |
| Copper | Cuprite | Tenorite | Digenite | Covellite | Cupalite |
| Platinum | Yixunite | Luberoite | Hongshiite | Braggite | Genkinite |
| Lead | Litharge | Blixite | Asisite | Laurelite | Shannonite |
| Strontium | Tausonite | Celestine | Acuminite | Ohmilite | Jarlite |
| Tungsten | Wolframite | Ferberite | Sanmartinite | Rankachite | Jixianite |
| Iron | Magnetite | Kamecite | Hematite | Bernalite | Fayalite |
| Silver | Pearceite | Jalpaite | Eugenite | Nauamannite | Proustite |
| Lithium | Gricite | Liberite | Sicklerite | Tavorite | Tiptopite |
| Dirt | TC+ | YM+ | SB+ | YB+ | AS+ |
| Sand | TS- | GS+ | TB+ | SM- | GB+ |
| Limestone | GM- | KM+ | SS+ |
That looks good. Now if only it will catch on. :)
- Tepemkau
Something like that Tepemkau ? The first one knowing all short names (or even all long names) wins nothing.
| ABU | Abhurite | CAL | Calaverite | GRI | Gricite | MAL | Maldonite | SHA | Shannonite | ||||
| ACU | Acuminite | CAS | Cassiterite | HEM | Hematite | MAT | Matraite | SIC | Sicklerite | ||||
| AKA | Akdalaite | CES | Celestine | HIB | Hibonite | NAU | Nauamannite | STI | Stistaite | ||||
| ANA | Anatase | COR | Corundum | HON | Hongshiite | OMI | Ohmilite | SUA | Suanite | ||||
| ASH | Ashoverite | COV | Covellite | JAL | Jalpaite | OSO | Osbornite | SWE | Sweetite | ||||
| ASI | Asisite | CUA | Cupalite | JAR | Jarlite | PAI | Painite | TAS | Tausonite | ||||
| AUR | Auricuprid | CUR | Cuprite | JIX | Jixianite | PAR | Paradocrasite | TAV | Tavorite | ||||
| BEA | Bernalite | DAN | Danbaite | KAM | Kamecite | PEA | Pearceite | TEN | Tenorite | ||||
| BEN | Berndtite | DIA | Diaspore | KLE | Kleberite | PER | Periclase | TIP | Tiptopite | ||||
| BET | Berthierite | DIG | Digenite | KOT | Kotoite | PRO | Proustite | VAL | Valentinite | ||||
| BLI | Blixite | EUG | Eugenite | LAU | Laurelite | RAN | Rankachite | WES | Weishanite | ||||
| BRA | Braggite | FAY | Fayalite | LIB | Liberite | ROM | Romarchite | WOL | Wolframite | ||||
| BRO | Brookite | FER | Ferberite | LIT | Litharge | RUL | Rulite | WUL | Wulfingite | ||||
| BRU | Brucite | FLU | Fluborite | LUB | Luberoite | SAN | Sanmartinite | YIX | Yixunite | ||||
| BYS | Bystromite | GEN | Genkinite | MAG | Magnetite | SEN | Senarmontite | YAN | Yuanjiangite | ||||
I've found three pairings of these: YM+ (dirt/kamecite), TC+ (dirt/magnetite), and RB+ (tausonite/celestine). I can't tell them apart by sight, even accounting for O/D or Short/Long. This shoots down the fifth pattern idea, since it would have to be one that could be confused with mottled, cloud, and black patterns, which are sufficiently different for that to be unreasonable. And O/D or Short/long distinction would give 9*4*2*2 = 144, way too many, if it applied to all of them (even applied to half, it's 108, still too many), even if the distinctions mattered to the above examples. 3 more colors would precisely account for 18 more minerals, though, and it's not impossible there are subtle differences we're just not seeing.
Anyone else seen any apparent duplicates? If they prove to be all Yellow, Teal, and Red, we can pin it on three subtly different additional colors; if not it may just be simple randomly selected duplicates, which has the "advantage" of Teppy being able to add more minerals later (*cough*magnesium*cough*). -Nefer-Bast
It is important to determine what the yield is based on: crystal count or cluster count. The new skill has provided us with cluster count, but not crystal count. The data I have so far supports cluster count being what yield is derived from.
- Chichis
To see graphical data on yield vs. cluster count, cluster ratio, stone count, stone ratio, and elemental ratio, visit my Alloy Experiments page. With the data I have, it appears that none of these are 100% reliable in predicting yield, including cluster count. Some counts can lead both to high and low ratios.
- Aqba
| Mineral | Rocks | Crystals | Clusters | Iron |
| Magnetite | 25 | 63 | 39 | 34 |
| Magnetite | 25 | 52 | 37 | 31 |
| Magnetite | 25 | 52 | 36 | 30 |
| Magnetite | 25 | 57 | 35 | 29 |
| Magnetite | 25 | 50 | 33 | 28 |
| Magnetite | 25 | 33 | 28 |
It seems that, at least with iron, all 5 crystal types give the same yield per cluster. With all 5, it takes 1.2 clusters to make 1 iron.
- Chichis
Another important part of this theory is the furnace the minerals are smelted in. I will be conducting tests with higher purity minerals in the fragile furnace. Also if having a dirt/mineral stone affects yield as if it were added as a pure stone.
- Tepemkau
| Mineral | Rocks | Crystals | Clusters | Output |
| Magnetite | 25 rocks | 48 clusters | 40 iron | |
| Kamecite | 25 rocks | 132 crystals | 40 clusters | 34 iron |
| Magnetite | 25 rocks | 63 crystals | 40 clusters | 34 iron |
| Magnetite | 25 rocks | 52 crystals | 37 clusters | 31 iron |
| Magnetite | 25 rocks | 52 crystals | 36 clusters | 30 iron |
| Magnetite | 25 rocks | 57 crystals | 35 clusters | 29 iron |
| Kamecite | 25 rocks | 108 crystals | 35 clusters | 29 iron |
| Hematite | 22 rocks | 90 crystals | 35 clusters | 29 iron |
| Magnetite | 25 rocks | 50 crystals | 33 clusters | 28 iron |
| Magnetite | 25 rocks | 33 clusters | 28 iron | |
| Tenorite | 6 rocks | 12 clusters | 10 copper | |
| Cuprite | 8 rocks | 12 clusters | 10 copper | |
| Cuprite | 5 rocks | 6 clusters | 5 copper | |
| Magnetite | 5 rocks | 6 clusters | 5 iron | |
| Bernalite | 4 rocks | 19 crystals | 6 clusters | 5 iron |
| Cuprite | 2 rocks | 4 clusters | 3 copper | |
| Fayalite | 3 rocks | 3 clusters | 3 iron | |
| Magnetite | 3 rocks | 3 clusters | 3 iron |
2 Hematite, 2 Bernalite, 1 Magnetite, 1 Kemecite = 5 iron
- Chichis - I might have been off by a cluster or two on some of the results, it was getting late. I apologize if this is has screwed the results a bit. Most of the data seems to support that the yield is rounded. It's definitly not floored. Two of the points show it as ceiled, but I'm almost positive those are errors.
- Roen - I'm trying to find the formula to get yield from clusters. If yield is 5/6 clusters as stated then it would give the following chart:
| Clusters | Floored | Rounded | Ceiled |
| 1 | 0 | 1 | 1 |
| 2 | 1 | 2 | 2 |
| 3 | 2 | 3 | 3 |
| 4 | 3 | 3 | 4 |
| 5 | 4 | 4 | 5 |
| 6 | 5 | 5 | 5 |
| 7 | 5 | 6 | 6 |
| 8 | 6 | 7 | 7 |
| 9 | 7 | 8 | 8 |
| 10 | 8 | 8 | 9 |
| 11 | 9 | 9 | 10 |
| 12 | 10 | 10 | 10 |
| 13 | 10 | 11 | 11 |
| 14 | 11 | 12 | 12 |
| 15 | 12 | 13 | 13 |
| 16 | 13 | 13 | 14 |
| 17 | 14 | 14 | 15 |
| 18 | 15 | 15 | 15 |
| 19 | 15 | 16 | 16 |
| 20 | 16 | 17 | 17 |
| 21 | 17 | 18 | 18 |
| 22 | 18 | 18 | 19 |
| 23 | 19 | 19 | 20 |
| 24 | 20 | 20 | 20 |
| 25 | 20 | 21 | 21 |
| 26 | 21 | 22 | 22 |
| 27 | 22 | 23 | 23 |
| 28 | 23 | 23 | 24 |
| 29 | 24 | 24 | 25 |
| 30 | 25 | 25 | 25 |
| 31 | 25 | 26 | 26 |
| 32 | 26 | 27 | 27 |
| 33 | 27 | 28 | 28 |
| 34 | 28 | 28 | 29 |
| 35 | 29 | 29 | 30 |
| 36 | 30 | 30 | 30 |
| 37 | 30 | 31 | 31 |
| 38 | 31 | 32 | 32 |
| 39 | 32 | 33 | 33 |
| 40 | 33 | 33 | 34 |
| 41 | 34 | 34 | 35 |
| 42 | 35 | 35 | 35 |
| 43 | 35 | 36 | 36 |
| 44 | 36 | 37 | 37 |
| 45 | 37 | 38 | 38 |
| 46 | 38 | 38 | 39 |
| 47 | 39 | 39 | 40 |
| 48 | 40 | 40 | 40 |
| 49 | 40 | 41 | 41 |
| Cart contents | Level | Result | Furnace | Submitter |
| 35 Yixunite, 7 Dirt | 2 | 24 Platinum | CF | Gustav |
| 40 Yixunite, 21 Dirt | 2 | 15 Platinum | CF | Gustav |
| 31 Yixunite, 21 Dirt | 2 | 9 Platinum | CF | Gustav |
| 32 Yixunite, 25 Dirt | 2 | 5 Platinum | CF | Gustav |
| 29 Tausonite, 15 Dirt | 2 | 12 Strontium | CF | Gustav |
| 27 Tausonite, 15 Dirt | 2 | 10 Strontium | CF | Gustav |
| 6 Magnetite, 4 Sand | 1 | 2 Iron | CF | Sord |
| 3 Magnetite, 3 Sand | 1 | nothing | CF | Sord |
| 5 Magnetite, 3 Danbaite, 3 Sand | 1 | 2 Iron | CF | Sord |
Using this chart I can find a formula for non pure yield.
We don't know if that yield is for all the metals, or only iron. We need to get more data on other metals before we generalize.
*Y is final yield
*N is normal yield
*S is side clusters (here it is dirt)
*M is main mineral clusters (here Platinium and Strontium)
The formula is Y = N*[1-(S/M)]
With the above chart
* 33*(1-(21/40)) = 15,675
* 26*(1-(21/31)) = 8,58
* 23*(1-(15/27)) = 10.22
Not very precise, but close enough to pay attention. We still don't know how normal yield is rounded (even if 5/6 is the proper yield ratio).
- Roen -
Theory: The yield when smelting different metal clusters together depends only upon the two highest cluster counts (at least when no dirt or sand is present.) The highest metal cluster count determines the type of metal received with the yield being equal to 5/6 of its cluster count minus 5/6 the cluster count of the second highest metal.
Equation: (N1 > N2)
Yield = (5/6)*(N1 - N2)
1st Test:
49 Fe (43 Magnetite) + 4 Cu (3 Cuperite, 1 Tenorite) -> 38 Iron
yield = (5/6)*(49-4) = 37.5
2nd Test:
38 Fe (6 Kamecite, 30 Magnetite, 2 Hematite) + 5 Cu (3 Cuperite, 1 Digenite, 1 Tenorite) + 14 Sn (9 Romarchite, 4 Abhurite, 1 Stistaite) -> 20 Iron
yield = (5/6)*(38-14) = 20.0
3rd Test:
5 Kamecite + 5 Romarchite -> nothing
yield = (5/6)*(5-5) = 0
(note: All tests were performed with crafting furnace.)
Obviously a few tests with only Iron/Tin/Copper prove nothing. But my quasi-results intrigued me none-the-less.
Can anyone else support or refute this?
-Seamubis
Hey Seamubis, i tested your calculating method, but it differs a bit:
1st Test in Craft Furnace:
49FE (24 Magnetite,21 Kamecite,4 Hemacite) + 7 Dirt (mixed with Fe clusters in stones) + 7 other mineral clusters (mixed with Fe clusters in stones) -> 35 iron
yield = (5/6)*(49Fe - 7 Dirt) = 35 -> first test showed me that other mineral clusters doesnt matter
2nd Test in Craft Furnace:
23 FE + 2 Dirt + 3 other mineral clusters = 17 iron
yield = (5/6)*(23FE - 2 Dirt) = 17,5
in my opinion when smelting in CF only dirt decreases the smelting result
-Gspusi-
Description: During the day a mine will primarily produce one mineral crystal type, but at night will primarily produce another type of crystal.
Reason: We have looked at our own favourite mine (Iron/Zinc) that gives high purity Iron from 5am to 5pm with smaller amounts of Zinc crystals present, and high purity Zinc from 5pm through to 5am with smaller amounts of Iron cyrstals present.
Other Evidence:We also have a member who added that his iron/lead mine also appears to do this, but the switching times are not confirmed. We'll be keeping an eye on it, but we are not dedicated researchers. If anyone else wants to do further research on this, thoughts are most welcome.
- Caine, on behalf of The Rising Phoenix Guild (Lower Nubia)
Questions:Any other people looking into this? Sounds odd, but interesting. How many days did this last for? Is it still true?
I think this may have merit. My iron/copper mine "seems" to do this also. - Ryzlar
Munkee Current Theory we have that is proving somewhat true is the Theory of Elevation, the higher you go. The better the ores. At Mid High level we have hit Lead. However at Sea level its Sand, and Ground level(mid) we have hit Dirt/Sand. Also to explain the changes in the Mines on people, we have another theory of digging. The mine digs down, every pull it goes deeper, attributed to the fact of an Extraction Bit later, and the seemingly changes from Metal to Dirt back to Metal, as if it is digging through a 3d map.
Leetah: I'd like to see more information backing this up. Given that I've been told directly from the gms that the concept of "elevation" is almost non-existant in game (although drying pappy matters, so I suppose that must have been a mistruth...) I feel like this is highly unlikely. My understanding since the very first time Teppy mentioned the 3D ore map is that mines would have "levels" and the levels would be accessing different layers of the 3d map. I don't think elevation is likely to have any effect. But I am open to hearing other proof/disproof:)
Amanesus As a counterexample, I have a serviceable copper mine which is right next to the river. Given that it couldn't be any lower without being underwater, I find it hard to believe that its low elevation is affecting its output significantly.
Shadus: Built a mine on top of mountain, 10 coords away bottom of mountain. Same output, I'd say nix this theory.
Gada: In my experience, elevation does matter, but you can't assume that higher = better. Given the notion of "3D mineral maps", it would make sense that an ore vein only goes so high/low. Mine levels seem to be using this map as well.
Leetah: Gada, the important question is "what's the chance that your results are just coincidence?" -- if nobody else sees any correlation, then I'd say it's pretty high. It does *not* make sense to me that Teppy would code into mines that they take the height they were built at and subtract some amount from that depending on the depth of the mine you are digging at. It's possible, yes. But 90%+ of the data shows that it doesn't matter. The idea that you're just changing veins with your mines as you are moving up a hill is much more likely. If others start to notice, it would be worth reconsidering. But for the time being, I say we remove this from this page. Perhaps we should set up an "old theories that people don't think are true" page, so that things like this don't die out, and new miners can read it and consider it and see if they may be able to bring it back?
Gada: I'll have to take it back. The example i was using as an argument does seem to be a coincidence. I have seen other examples hinting that elevation doesn't matter.
Shadus: Just as a point, the mineral map isn't truely 3d the way teppy explained it. It's like 50 paper maps stacked on top of each other. Each time you dig down you land on a new map.