The Alchemist Cookbook

[Kandarim]

Well, I only have the one thing to share, but since it is with rye flour, I guess i'll put it on here:

Humble Meat Pie
Rye + Bear:

Code: Select all

Multiplier: divided by the multiplier
Regen: 17.5, 17.5, 17.5, 17.5
Salt: 31.3, 2.3, 15.4, 15.4
Mercury: 3.2, 32.3, 14.9, 23.9
Sulphur: 8.9, 2.9, 40.8, 24.7
Lead: 5.2, 33.5, 15.4, 29.8


[gnox24]

Aztec Abattoir
Rabbit

Code: Select all

Multiplier:  1.01
Sa 24.08 / 3.5 / 14.0 / 3.5
Me 14.0 / 2.8 / 13.6 / 2.8
Su 17.6 / 3.5 / 8.3 / 3.5
Le 15.1 / 3.0 / 13.5 / 3.0


Aztec Abattoir
Beaver

Code: Select all

Multiplier: 1.05
Sa 14.6 / 0.0 / 20.1 / 9.1
Me 11.7 / 0.0 / 20.6 / 9.4
Su 14.6 / /0.0 / 9.1 / 4.1
Le 12.5 / 0.0 / 20.1 / 9.1


Aztec Abattoir
Deer

Code: Select all

Multiplier: 1.01
Sa 21.1 / 3.5 / 21.1 / 8.8
Me 17.7 / 2.8 / 21.7 / 9.0
Su 19.2 / 3.5 / 9.6 / 4.0
Le 18.6 / 3.0 / 21.1 / 8.8


Aztec Abattoir
Domestic Turkey

Code: Select all

Multiplier: 1.01
Sa 21.1 / 17.6 / 21.1 / 0.0
Me 16.9 / 18.1 / 21.7 / 0.0
Su 21.1 / 8.0 / 9.6 / 0.0
Le 18.1 / 17.6 / 21.1 / 0.0


Aztec Abattoir
Bear

Code: Select all

Multiplier: 1.02
Sa 24.08 / 0.0 / 35.5 / 8.8
Me 20.7 / 0.0 / 36.5 / 9.1
Su 22.9 / 0.0 / 16.2 / 4.0
Le 21.8 / 0.0 / 35.5 / 8.8


[Erunildo]

Those abattoirs looks interesting..Thanks for the effort guys;)

[Inotdead]

argh why didnt i see this thread earlier - could have posted mushroom and berry pies will do once I get my crops up unless someone does it before.

[gnox24]

I'm trying to find some pattern behind the change in the glutony values, but i don't have any clue until now. The first idea was, that there could be some kind of linear:

newValue = a + b * oldValue

or quadratic behavior:

newValue = a * (b + oldValue)

With a and b beeing 'any-food' specific parameters, but that doesn'n work. There are examples where equal inputs give different results, which can't be solved with such simple approaches.

Maybe the devs created distinct values for every food combination, but i don't believe that. For example 'Fish in the Reeds' requires 'any plant' and 'any fish'. With 16 plants and 14 fishes there are 16 * 14 = 224 different combinations - and that's just one product. There are some recuring patterns, like bear seems to give better results than rabbit or white gabbage beeing better than coleworth. But thats the only thing i got so far. Maybe someone else has an idea.

[lachlaan]

I'm pretty sure your first idea is the closest to reality. I think it's something along the lines of oldValue * modifier1 + modifier2 * oldOtherBileValue. The easiest way to test the modifiers is to try and cook stuff that only has a single ingredient's buff. Like humble meat pies with generic flour and bear cuts and shreds. Or Dry Pie with each of the different flours. And then trying to get a variety of recipes with that same ingredient isolated to see its effects on different biles. Sometimes if an ingredient adds say 0.5 of the old YB value to the new phlegm value, if the food you're making has a base 0 to both yb and phlegm, you won't see any change. So you can't really tell what modifiers it uses until you can see changes in all values in different recipes.

[gnox24]

I did some tests that you mentioned, but i give you an example where the first idea does not work:

Lumberjack Frikadel

Code: Select all

Base values:
          9.00             0             0             0
          7.50          8.00             0             0
             0             0             0          9.00
          4.00          6.50             0             0

With beaver:
          7.60             0             0             0
          6.30          9.30             0          4.20
             0             0             0         11.50
          3.40          7.60             0          3.40


Cabbage Rolls

Code: Select all

Base values:
          6.60             0          7.10             0
             0             0          9.10             0
          8.60             0          6.60             0
          7.10             0             0             0

With beaver:
          5.40             0          8.00          3.60
             0             0         10.20          4.60
          7.00             0          7.40          3.30
          5.80             0             0             0


Note: these values are not adjusted by the purity modifer, but for the sake of this example it does not matter.

Now take the last entries from the first row: In the case of Frikadels a base value of 0 converts to a value of 0, nothing fancy so far. But in the case of the Cabbage Rolls, a base value of 0 converts to a value of 3.6. This can't be solved with a simple linear approach.
However, the first three collumns can be solved with such a behavior, but sadly the results can't be applied to another food item, like Aztec Abattoir.

But i agree that your idea:

oldValue * modifier1 + modifier2 * oldOtherBileValue

This could be a possible solution. The difficulty is, that there are 4x4 possible equations for every food item with a total of 32 varying parameters in the worst case. Finding the right constellations is quite a mess. I wrote Matlab code which is capable of solving these equations (or at least minimzing an appropriate measure). But wee need to find the right model, which fits the reality of the game.

[gnox24]

lachlaan's approach seems promising, but i need another sample to improve the result. I can't make anything else than oatmeal at the moment, so it's impossible to create Humble Meat Pies without the oatmeal modifier. Can some please make one of these with just the beaver modifier and post the result? I would appreciate it if the purity multiplier is not removed from the values, and instead the four elemental values are given. Calculating the multiplier from the formula in the wiki gives a better precision.

thx.

[lachlaan]

Lumberjack Frikadel

Code: Select all

Base values:
          9.00             0             0             0
          7.50          8.00             0             0
             0             0             0          9.00
          4.00          6.50             0             0

With beaver:
          7.60             0             0             0
          6.30          9.30             0          4.20
             0             0             0         11.50
          3.40          7.60             0          3.40


Cabbage Rolls

Code: Select all

Base values:
          6.60             0          7.10             0
             0             0          9.10             0
          8.60             0          6.60             0
          7.10             0             0             0

With beaver:
          5.40             0          8.00          3.60
             0             0         10.20          4.60
          7.00             0          7.40          3.30
          5.80             0             0             0


Note: these values are not adjusted by the purity modifer, but for the sake of this example it does not matter.

Now take the last entries from the first row: In the case of Frikadels a base value of 0 converts to a value of 0, nothing fancy so far. But in the case of the Cabbage Rolls, a base value of 0 converts to a value of 3.6. This can't be solved with a simple linear approach.

Actually that's an awesome case where you can figure out what contributes to the various multipliers of foods. If you consider a beaver's multiplier for black bile values to be x*original + y*random_original_humor_value, without knowing either of those multipliers, you just look at the differences between cabbage rolls and frikadel. Assuming it can't get negative multipliers to bring a humor back to zero, since the frikadel's salt, mercury, and lead bb bonuses are originally 0, and with a positive multiplier of the original bb value you'd always get zero. The fun thing to notice is that it doesn't get a value from what i assume to be the secondary multiplier either, in the case of frikadel. So which columns are zero for frikadel? Phlegm and yb for the salt event are both zero, and the outcome with beaver included is also zero. In the merc and lead events however the original 0 value of bb is now 4.2 and 3.4 respectively, and the only thing different in those events is that the phlegm value is no longer 0. So you could assume that if bb now gained a value from somewhere, it's likely a multiplier of the phlegm value. And if you divide the new bb values by the original phlegm values, you'll see that 4.2/8 = 0.525, and 3.4 / 6.5 = 0.523, so the bonus is proportional to the original phlegm values, even if you don't take into account the purity multiplier of the frikadel. On the other hand in the cabbage roll recipe it's important that there not be a colewort, green/white/red cabbage modifier at all, and if there wasn't in your example then you can still conclude that beaver's black bile multiplier also has a yb component, and now you just need to find a recipe to test them both at once. I'll have a look at various recipes and see if i can find a useful one for this situation.

EDIT : Alright, after having a look at a simple sunday steak with beaver in it, I think i have a better idea of the beaver multiplier. Let's call the original values b0, ph0, yb0 and bb0, and give the resulting ones an index number of 1. So b1 = b0 * 0.8 , ph1 = ph0 * 1.1 (roughly) , yb1 = yb0 * 1.1, bb1 = bb0 * 1.23 + ph0 * 0.5 + yb0 * 0.5 , more or less. There's always a chance i missed something on it, so feel free to try it out with whatever food you can think of

[gnox24]

Yeah, my results look similar.

You could even write down the modifier values in matrix notation:

Beaver Modifier Matrix M:

Code: Select all

          0.80             0             0             0
             0          1.10             0          0.50
             0             0          1.10          0.50
             0             0             0          1.20


The values maybe a little different because i used the imprecise purity multiplier value.

Than you take the base values of a food item and write them into a 4x4 Matrix B aswell. Multiplying it from the right by M yields the correct values V:

V = B * M

These values work for Cabbage Rolls, Lumberjack F. and Aztec A. We just need to create enough test data for the other Food items and derive the modifier matrices from them. Great stuff is happening here



Edit:
I wrote a Matlab program which takes at least two combinations of [Base Matrix B, Modified Matrix V] for the same modifier item and calculates the corresponding Modifier Matrix V. I you guys can help me collecting samples, i'll calculate the values and put them into the wiki. The samples must contain the altered glutony values for one modifier item only. And please leave the values as they are without dividing them by the multiplier and give the actual elemental purity values instead.

Thx for the help.