• Dear forum visitor,

    It looks as though you have not registered for a forum account, or are not signed in. In order to participate in current discussions or create new threads, you will need to register for a forum account by clicking on the link below.

    Click here to register for a forum account!

    If you already have a forum account, you can simply click on the 'Log in' button at the top right of your forum screen.

    Your Elvenar Team

Cauldron Recipes

Porcinus

Member
Is there a recipe that increases my chances of getting a polar bear evolve token to craft in the magic academy?
Since we have to drink them right away and cannot save them, i want potions that make permanent changes instead of effects that only last a few hours.
 

Katwick

Cartographer
Best General Mayhem Recipe
4 Mana Shell,
4 Flying Iris,
4 Ear Truffle
4 Cottoncloud Bud
For the next week or so I'll be plowing through the recipes that Folks have found to be useful, and doing a side by side comparison. The naming convention will be a simple reference to the Post in this Thread, plus the Author's Name and the Date of Creation.

Once we have a useful bunch of Recipes, and folks have gotten interested again, I'll formally release a Beta Version of my Google Sheet. For now the stuff that I'm accumulating is Viewable (and downloadable), in the directory at Cauldrons 102 Beta 2023C23, including a PreBeta Release of my Google Sheet, if you'd like to follow along.


Exactly as posted, except for the Standard Recipe FormatAs close as we can get with an Optimized Recipe
Cauldron Recipes #1a - AJ  2023B03.png
Cauldron Recipes #1a - Katwick  2023C24.png
Cauldron Recipes #1b- AJ  2023B04.png
Cauldron Recipes #1b - Katwick  2023C25.png

I'm only in Chapter 11, so I can't test and didn't include the Diplomas for Chapters 13-20, but if an endgame player is willing to do some rigorous testing, it will only take a couple of days to plug in the endgame chapters.
 

Katwick

Cartographer
I have two recipes for Trader's Basic Plea - increase Tier 1 production:
54% nut, 23% iris, 23% leaf
(for example: 7 nut, 3 iris, 3 leaf)
OR
8% shells, 24% eggs, 36% blossoms, 32% buds
(for example: 2 Shells, 6 Eggs, 9 Blossoms, 8 Buds)

and one for Magical Tax Increase - increase coin production:
52% nut, 48% iris
(for example: 13 Beetlenut, 12 Iris)
Cauldron Recipes #5a - David 2023B03.png

There's probably a typo. This is a T3 Recipe.
Cauldron Recipes #5a - Katwick 2023C24.png
Cauldron Recipes #5b - David 2023B03.png
Cauldron Recipes #5b - Katwick 2023C24.png
Cauldron Recipes #5c - David 2023B03.png
Cauldron Recipes #5c - Katwick 2023C24.png
 
Last edited:

Katwick

Cartographer
For monster's mayhem this week, you can try something like this with non-diamond ingredients:

5 part beetlenut
5 part nightshade blossom
5 part bitterhair leaf
4 part mocking tongues
2 part eye sprouts

6429 wp used, 91.7% success chance, 50.2% monster mayhem trigger for me.

The mix will probably work if less or more ingredients are mixed in roughly similar proportion, but I can't say for sure :)

The potion success rate can be made higher, but then the monster mayhem trigger will be lower.

I'm still trying to figure out ingredient interaction, and I'm not sure if people with different numbers of diplomas and available effects per ingredient are going to get widely different results, or can still use some kind of template. Hopefully there will be some commonality :)
Values and Factors - Katwick 2023C08.png

It's not a trivial consideration. EVERY Ingredient influences EVERY Effect in some way.

If you look at the 3s on the main diagonal, you'll soon realize that DOWN or LEFT from any #3 gets you EXACTLY the same vector. So EACH 3, which is the join of an Ingredient and an Effect, sees an Identical local environment. CONSEQUENCE?? If you have ONE good recipe you actually have TWELVE good recipes. You just bump everything up left (or down right) twelve times. The recipe space wraps around (tessellates), so you don't run off the edge of the map. You can think of the space as a Cylinder both Horizontally and Vertically.

Let's look at the bottom right corner. Notice that you can go DOWN OR LEFT and you'll see the same vector. So let's just go left across the bottom, and recognize that 12 Effects are influenced.
  • The first seven numbers are in the Value column of the vector cell.
    • ___ 1 1 2 3 are POSITIVE VALUES. They're a Primary ADD for that Effect.
    • ___ -2 -1 -1 ___ are NEGATIVE VALUES. They're a Primary SUBTRACT for that Effect, which is IMPORTANT because it's a zero sum game, and knocking them down a bit means that you'll get more of the pie.
  • The next five numbers are in the Factor column of the vector cell.
    • ___ 1 1 2 ___ are POSITIVE FACTORS. Whatever Value Points (zero or less is ignored) you've accumulated will be COMPOUNDED by +10% for each Factor point.
    • -2 -1 ___ are NEGATIVE FACTORS. Whatever Value Points ..... will be TAXED by -10% compounded for each Factor Point.
So you should dump GREEN on what you want, and RED on what you don't want. Rinse and Repeat.
Caldron Recipes #15a - Inyri 2023B05.png

Those bottom two rows are the HEART of the Brewery.
Cauldron Recipes #15b - Inyri 2023B05.png
Cauldron Recipes #15b - Katwick 2023C25.png
 
Last edited:

Katwick

Cartographer
I've just used five ingredients to obtain a (nearly) 100% Trader's Basic Plea effect for (almost) 11 hours:

View attachment 14741

Cauldron Recipes #33a Moho 2023B09.png
Cauldron Recipes #33a Katwick 2023C26.png
This is a wonderful recipe. The model shows slightly better results than you got, but the IMPORANT concept is that NO OTHER EFFECT shows up at all .... until you get that 8th Diploma and then Settlement Goods will knock everything into a cocked hat.I cheated a bit and allowed six Ingredients, and only 6 Diplomas, which gave me a 4 more percentage points, but both of my Ingredients, 23 and 24 get clobbered as soon as Diploma 7 is activated, while Caterpillar Lime hangs in there for an extra chapter.

In the next screenshot, right under 08_GS, you'll see 7 Value points plus 1 Factor point. It's not that the Effects interfere with each other, because they don't, but rather that everybody is sharing the same sqrt(N/25) space and there's suddenly another Big Dog in what used to be your exclusive territory.
Cauldron Recipes #33b Moho 2023B09.png


This is also a good opportunity to highlight two known anomalies in the Recipe Space.
Cauldron Recipes #33b - Katwick 2023C26.png

We've switched to 12 Diplomas, and we've added one of each Ingredient, so the bottom row SHOULD have identical numbers, right?? If you've been following along, we've introduced the two highlighted anomalies into model, which diminishes 06_T1 and primps up 09_OR.
 
Last edited:

Katwick

Cartographer
I tried to come up with the ultimate recipe for military boosts ignoring diamond cost so here's my version:

View attachment 15065

Cauldron Recipes #77a Silly Bubbles 2023C12.png
Cauldron Recipes #77a Katwick 2023C26.png
My Cauldron model is within 1 Witch Point of what you're seeing in the Brewery, so the model seems to be OK. You're showing lower Percent Success Rates, but that's merely because I'm using Diploma level 12 while you're probably using Diploma level 20 where the Sqrt(N/25) space simply has more folks who deserve a slice of the pie.But then I ran an Optimization to see if we still had some wiggle room. I designated 03_SB as my target, and constrained the three Military Strength Effects to be >= 0.25 and no solution was available using GRG Nonlinear. So I backed off 5% and got a pretty nice solution that saves 25% on Witch Points for an additional 50 Diamonds. Not bad at all as it will be accessible to more players.
Cauldron Recipes #77b Katwick 2023C26.png
Cauldron Recipes #77c Katwick 2023C26.png
This is one of the three sorts (as indicated in Blue) that my model automatically generates. This green bordered sort, in Amount Order, is the one you'll usually see in the documentation, and you can plug it directly into the Brew Kettle.An Effect/Diploma sort is provided mostly for completeness. I've emphasized HOW MANY Witch Points the user is saving by using the Check Box. If the User want's to avoid spending Diamonds altogether, they can Include a zero constraint, or they can remove that effect from the list of inputs that the Solver is allowed to manipulate.
For Completeness, here's the Recipe Space as well, and the Solver configuration.
Cauldron Recipes #77d Katwick 2023C26.png
Cauldron Recipes #77e Katwick 2023C26.png
 

Silly Bubbles

You cant pop them all
@Katwick I just have couple of points. How is it any better/different to Karvest version? It just seems the same to me, just presented differently and you can't choose which diamond ingredients you get, they are set by the game.
Also, at the level you're playing, it is subject to personal preferences. For example, I'm happy to sacrifice variance to get higher probability but at the end it's just playing with 2% difference, it doesn't make a big difference. Have a look at this recipe:

Katwick.png
 

Katwick

Cartographer
@Katwick I just have couple of points. How is it any better/different to Karvest version? It just seems the same to me, just presented differently and you can't choose which diamond ingredients you get, they are set by the game.
Also, at the level you're playing, it is subject to personal preferences. For example, I'm happy to sacrifice variance to get higher probability but at the end it's just playing with 2% difference, it doesn't make a big difference. Have a look at this recipe:

View attachment 15196
I'll take a close look at your recipe tomorrow. For now, I LIKE Karvest's approach quite a lot, but he's so fascinated with avoiding Diamonds that it's pretty much the core of his program, which is just a black box by design. The choppiness of the Witch Points progression forces him to use an Evolutionary Solver, which can take several minutes to find a solution, and it only runs on Excel. My program is a Vanilla Google Sheet, which anybody can access and easily share and it uses the GRG Nonlinear Solver, and takes 15 seconds at the most.

I've also run several recipes through your application, and it's pretty dependent on step-by-step user participation, following the suggestions of the application, which is OK but doesn't say much about Why. I prefer to have the bones showing, so that folks have an opportunity to develop their intuition, and avoid most of the Gotchas.

Cauldron Recipes Witch Points Katwick 2023C26.png


Plus, nobody else is collecting a Recipe Library. This data is driven by a LIBRARY, not an equation.

If you want to boil it down to a single design difference, I WANT the Users to see HOW MANY Witch Points their Diamonds are worth, this week and, equally to the point, I want them to go after some recipes that they simply can't touch with just their weekly Witch Points allotment.

But yeah, we're all solving the same problem, so our results should all be the same, including all of the fiddly tradeoffs. One of the important questions is "Can we make it more accessible?" Right now the Cauldron is "too complicated" for most players, so there's plenty of opportunity for user interfaces that are well designed, and accessible.
 

Silly Bubbles

You cant pop them all
I'll take a close look at your recipe tomorrow. For now, I LIKE Karvest's approach quite a lot, but he's so fascinated with avoiding Diamonds that it's pretty much the core of his program, which is just a black box by design. The choppiness of the Witch Points progression forces him to use an Evolutionary Solver, which can take several minutes to find a solution, and it only runs on Excel. My program is a Vanilla Google Sheet, which anybody can access and easily share and it uses the GRG Nonlinear Solver, and takes 15 seconds at the most.

I've also run several recipes through your application, and it's pretty dependent on step-by-step user participation, following the suggestions of the application, which is OK but doesn't say much about Why. I prefer to have the bones showing, so that folks have an opportunity to develop their intuition, and avoid most of the Gotchas.

If you want the bones showing, have a closer look at the Karvest spreadsheet. It can't be presented in any more clearer way. Also, my spreadsheet is based on that and is showing all the formulas if you're interested. This is as pure as it gets, no hiding anything or complicating it. And yes, my spreadsheet has never been intended to replace Solver, it's for people that don't want to use Solver. And both can be used with or without diamond ingredients.

But I do agree that it would help to use the Solver in Google Sheets if people are happy to learn how to use it and I don't think they are. It does have to be user friendly. People that already know how to use Solver are already using it so it's not much for them, it's for the people that don't know how to use it and want to learn. This will be quite a challenge because the criteria aren't fixed, they need to be adjusted to each recipe to get the most out of the ingredients.

Plus, nobody else is collecting a Recipe Library. This data is driven by a LIBRARY, not an equation.

You do realise that there's 20 diplomas, 20 different effects and it's possible to get up to 5 different effects per brew. That's a lot of recipes to include in your library.
 

Katwick

Cartographer
You do realize that there's 20 diplomas, 20 different effects and it's possible to get up to 5 different effects per brew. That's a lot of recipes to include in your library.

In a year or so, I hope to have a Recipe Library that's nicely indexed and linked. It will only take a half dozen folks who enjoy the WHY of it all to support a really nice library. Google Sheets has some excellent tools for filtering and sorting, and if somebody downloads the Google Sheet just to look at the Best of the Best list, then we've got them on the slippery slope that ends up with some of them using the Solver to "improve" things.
Cauldron Recipes #77f Silly Bubbles 2023C26.png
Cauldron Recipes #77f Katwick 2023C26.png
I don't know which World you're in, so I didn't try to guess which Herbs will be needing Diamonds this week. But the recipe comes out to 9437 Witch Points.In Khelonaar this week, as indicated by the Short Date and the conditionally formatted Gold Bars, 07_ST requires Diamonds while the other two battle buffs require Witch Points. So I optimized for 07_ST.
If we just consider Witch Points, the solutions are within 1% of each other, in the low 9000s, which is out of reach for most players. While it doesn't make any difference to the program, I've made a POINT of allowing the user a simple Check Box, AFTER they've seen the Recipe Results, to indicate that they'll happily spend 250 Diamonds to knock the Recipe down to a viable number of Witch Points.

Cauldron Recipes #77g Katwick 2023C27.png

If a player DOES NOT want to spend the Diamonds, then a glance at the 3s on the main diagonal, in the recipe space, suggests that it will be a pretty good week for 01_C_, 06_T3, 03_SB (without the other battle guys,) 08_GS and 12_GD.

This is where Karvest and I don't agree, as he prefers to present the recipe space in Effect/Diploma order, rather than in Ingredient order, and then use the Black Box optimizer, just to be stubborn. ;) My preference is to look at the Recipe Space and realize
  • That's NOT going to happen this week or
  • OH LOOK!! This week I can dump some Diamonds into 05_T2 or 07_ST, and punch WAY above my weight class.
 
Last edited:

The Fairy

Scroll-Keeper, Buddy Fan Club Member
You do realise that there's 20 diplomas, 20 different effects and it's possible to get up to 5 different effects per brew. That's a lot of recipes to include in your library.
And if you like me (and probably like most other players) don't want to spend diamonds then there are 495 different combinations of which 8 ingredients that are available each week. So you can multiply your number of recipes with 495.
 

Katwick

Cartographer
EffectRateDpl#WPs$$?Title/Link
01_C_50.3%1025high0TRUECauldron Recipes #5c - David 2023B03
01_C_91.1%1025high0TRUECauldron Recipes #5c - Katwick 2023C24
03_SB21.0%12252994200TRUECauldron Recipes #77a - Katwick 2023C26
03_SB22.6%1225high0TRUECauldron Recipes #77f - Silly Bubbles 2023C26
03_SB29.4%12253916150TRUECauldron Recipes #77a - Silly Bubbles 2023C12
04_T10.0%101329250TRUECauldron Recipes #5a - David 2023B03
04_T144.7%753050TRUECauldron Recipes #33a - Moho 2023B09
04_T149.2%1025high0TRUECauldron Recipes #5b - David 2023B03
04_T184.8%1025high0TRUECauldron Recipes #5b - Katwick 2023C24
07_ST22.1%12253755250TRUECauldron Recipes #77f - Katwick 2023C27
07_ST55.9%121316690TRUECauldron Recipes #5a - Katwick 2023C24
There are plenty of easy ways to present Library Recipes, including this simple Cut&Paste from a Google Sheet into the Forum which, delightfully, carries the Recipe Links.
 
Last edited:

ajqtrz

Chef - loquacious Old Dog
For the next week or so I'll be plowing through the recipes that Folks have found to be useful, and doing a side by side comparison. The naming convention will be a simple reference to the Post in this Thread, plus the Author's Name and the Date of Creation.

Once we have a useful bunch of Recipes, and folks have gotten interested again, I'll formally release a Beta Version of my Google Sheet. For now the stuff that I'm accumulating is Viewable (and downloadable), in the directory at Cauldrons 102 Beta 2023C23, including a PreBeta Release of my Google Sheet, if you'd like to follow along.


Exactly as posted, except for the Standard Recipe FormatAs close as we can get with an Optimized Recipe
View attachment 15154View attachment 15155
View attachment 15157View attachment 15156

I'm only in Chapter 11, so I can't test and didn't include the Diplomas for Chapters 13-20, but if an endgame player is willing to do some rigorous testing, it will only take a couple of days to plug in the endgame chapters.
I put the optimized recipe (upper right one) in (using diamonds) and got the result of 33.6% From where does the 59.47% come? Is that based upon the number of trophies? I mean since each trophy gives you another chance, (I have 5) does that mean I have the overall chance of 59.47% of getting the 03_SB effect?

AJ
 

Katwick

Cartographer
I put the optimized recipe (upper right one) in (using diamonds) and got the result of 33.6% From where does the 59.47% come? Is that based upon the number of trophies? I mean since each trophy gives you another chance, (I have 5) does that mean I have the overall chance of 59.47% of getting the 03_SB effect?

AJ
That recipe was brewed at Diploma Level 10. If you're in a later chapter then additional Effects will be stealing some of the common space, and any particular Effect will therefore be squeezed tighter to fit into the Sqrt(N/25) space, where N is the total number of Cables. At Sqrt(16/25) = 4/5 = 80% the Bridge Troll is likely your main competitor, with minor Effects soaking up the rest of the space.

33.6% isn't an impressive number, so you can probably do much better with a different recipe that uses more than 16 total Ingredients.

After you brew, you'll have 5 tries (Goblets) to hit that 33.6% chunk of the space, so your odds of success, overall, are 5/3. You'll probably get what you want. Once something is selected, there's no Brewery multiplier. Rather the Boost and the Duration are determined by your previous Studies of that Effect.
 

ajqtrz

Chef - loquacious Old Dog
Among the questions I have, some of them revealing how little I understand, are these:

1) What's a "cable?" If it's a critical concept to understanding recipes, can you help me understand it?
2) How do you improve your diploma's. I think this has to do with the chapters and research done (I'm on chapter 19 and thus would have 18+ diplomas -- which are the effects I can seek.
3) How do you get more goblets? (I think of them as small trophies, but goblets will do). I average about 7500-8000 tournament points and top the Spire but I see (or at least I think I see) you can have a dozen or more. So what do I need to do?

AJ

ps. These questions are probably answered elsewhere. Just point me to the relevant thread if you'd rather.
 

Katwick

Cartographer
Among the questions I have, some of them revealing how little I understand, are these:

1) What's a "cable?" If it's a critical concept to understanding recipes, can you help me understand it?
2) How do you improve your diploma's. I think this has to do with the chapters and research done (I'm on chapter 19 and thus would have 18+ diplomas -- which are the effects I can seek.
3) How do you get more goblets? (I think of them as small trophies, but goblets will do). I average about 7500-8000 tournament points and top the Spire but I see (or at least I think I see) you can have a dozen or more. So what do I need to do?

AJ

ps. These questions are probably answered elsewhere. Just point me to the relevant thread if you'd rather.
Cable should be Ingredient or Herb. My history is showing. Don't ask me about Spanning Trees unless you have the time for a lengthy discussion.

You don't improve Diplomas, per se, but after you Brew something you can study the related Effects by throwing Witch Points and excess stuff at them.

You can get up to five goblets (rolls of the dice) each week, use them or lose them, based on your Tournament score.
 

Smooper

Well-Known Member
To avoid my eyes glazing over, I have been looking for recipes which look like this: 3 whatchamajigs, 3 doodads, 2 thingamabobs. Done. If any of them cost diamonds I will try a different recipe. So if I have on a post it note, 3 or 4 recipes in this form and they are sorta working, I am done with cauldron for the week.
 

ajqtrz

Chef - loquacious Old Dog
Sadly, I used to know about spanning trees but for the most part have not used that knowledge in decades. Now I barely know about them, which pretty much tells you I know a lot of stuff (nodes) about a lot of stuff (other nodes) but sometimes I don't know how this batch of stuff (these nodes) interacts (spans or connects with) with that batch of stuff (other node sets) and thus, everything appears pretty jumbled even when it isn't. Like a spanning tree.

AJ
 
Top