I found the best "Nutrient from Spoilage" Setup

[Lindor]

Hello fellow factorio engineers

This topic is about the "Nutrient from Spoilage" recipe which turns spoilage into 50% spoiled Nutrients.
This recipe can either be crafted in an assembling machine or in a biochamber.
The caveat is that the biochamber has an innate productivity boost, but it also consumes Nutrients during the craft (instead of electricity)
I don't care about electricity or the amount of Nutrients gained per second.
I don't care about anything else but the amount of Nutrients gained per recipe.
So the question is: to biochamber or not to biochamber? And which modules to choose?
This does not seem like a "generalizable" problem to solve, i don't think there is any other way other than to brute-force it, so i wrote some code for it:

Open or Close

Code: Select all

local moduleEnergyEffectTable = {
    tier1 = {-30, -39, -48, -57, -75},
    tier2 = {-40, -52, -64, -76, -100},
    tier3 = {-50, -65, -80, -95, -125}
}

local moduleSpeedEffectTable = {
    tier1 = {
        energy = 50,
        speed = {20, 26, 32, 38, 50}
    },
    tier2 = {
        energy = 60,
        speed = {30, 39, 48, 57, 75}
    },
    tier3 = {
        energy = 70,
        speed = {50, 65, 80, 95, 125}
    }
}

local moduleProductivityEffectTable = {
    tier1 = {
        energy = 40,
        speed = -5,
        productivity = {4, 5, 6, 7, 10}
    },
    tier2 = {
        energy = 60,
        speed = -10,
        productivity = {6, 7, 9, 11, 15}
    },
    tier3 = {
        energy = 80,
        speed = -15,
        productivity = {10, 13, 16, 19, 25}
    }
}

local nutrientEnergy = 2

local nutrientsFromSpoilageRecipeTime = 2 -- in seconds

local biochamberBaseEnergyConsumption = 0.5 -- in MJ
local biochamberBaseCraftingSpeedTable = {2, 2.6, 3.2, 3.8, 5}
local biochamberBaseProductivity = 50

local beaconBaseEfficiencyTable = {1.5, 1.7, 1.9, 2.1, 2.5}


-- @param tier:                     for modules
-- @param quality:                  for buildings & modules
-- @param halfSpoiledNutrientWorth: how much do you consider halfly spoiled nutrients worth in fresh nutrients?
local getEfficientSetup = function(tier, quality, halfSpoiledNutrientWorth)
-- get constants
    
    -- buildings
    local biochamberBaseCraftingSpeed = biochamberBaseCraftingSpeedTable[quality]
    local beaconBaseEfficiency = beaconBaseEfficiencyTable[quality]

    -- modules
    local moduleEnergyBonusEnergy             = moduleEnergyEffectTable[tier][quality]
    local moduleSpeedBonusEnergy              = moduleSpeedEffectTable[tier]["energy"]
    local moduleSpeedBonusSpeed               = moduleSpeedEffectTable[tier]["speed"][quality]
    local moduleProductivityBonusEnergy       = moduleProductivityEffectTable[tier]["energy"]
    local moduleProductivityBonusSpeed        = moduleProductivityEffectTable[tier]["speed"]
    local moduleProductivityBonusProductivity = moduleProductivityEffectTable[tier]["productivity"][quality]
    
-- function for calculating the nutrient gain per recipe for biochamber setups
    local calcNutrientGainPerRecipeForBiochamber = function(bonusEnergy, bonusSpeed, bonusProductivity)
        local bruttoGain = halfSpoiledNutrientWorth * (1 + bonusProductivity) -- how many nutrients are crafted per recipe cycle?
        local consumption = (biochamberBaseEnergyConsumption * (1 + bonusEnergy) * nutrientsFromSpoilageRecipeTime) / (nutrientEnergy * biochamberBaseCraftingSpeed * (1 + bonusSpeed)) -- how many nutrients are consumed per recipe cycle?
        local nettoGain = bruttoGain - consumption
        do return nettoGain end
    end

-- calc the best setup
    local bestGain = halfSpoiledNutrientWorth * (1 + (4 * moduleProductivityBonusProductivity / 100)) -- base assembling machine gain
    local winner = {
        input = {
            tier = tier,
            quality = quality,
            halfSpoiledNutrientWorth = halfSpoiledNutrientWorth
        },
        machine = {
            machine = "assembling machine 3",
            energyModuleCount = "irrelevant",
            speedModuleCount = "irrelevant",
            productivityModuleCount = 4
        },
        beacons = {
            beaconCount = "irrelevant",
            beaconEnergyModuleCount = "irrelevant",
            beaconSpeedModuleCount = "irrelevant"
        },
        gainPerRecipe = bestGain
    }
    
    for beacons = 0, 12, 1 do
        local beaconEfficiency = beaconBaseEfficiency / math.sqrt(beacons)
        local beaconModuleSlots = beacons * 2
        --for beaconEnergyModuleCount = 0, 0, 1 do               -- NO ENERGY MODULES IN BEACONS
        for beaconEnergyModuleCount = 0, beaconModuleSlots, 1 do -- ENERGY MODULES IN BEACONS
            local beaconSpeedModuleCount = beaconModuleSlots - beaconEnergyModuleCount
            
            local beaconModuleBonusEnergy = ((moduleEnergyBonusEnergy * beaconEnergyModuleCount) + (moduleSpeedBonusEnergy * beaconSpeedModuleCount)) * beaconEfficiency
            local beaconModuleBonusSpeed = moduleSpeedBonusSpeed * beaconSpeedModuleCount * beaconEfficiency
            
            for energyModuleCount = 0, 4, 1 do
                local leftoverSlots = 4 - energyModuleCount
                for speedModuleCount = 0, leftoverSlots, 1 do
                    local leftoverSlotsForProductivityModules = leftoverSlots - speedModuleCount
                    for productivityModuleCount = 0, leftoverSlotsForProductivityModules, 1 do
                        local slotModuleBonusEnergy       = (moduleEnergyBonusEnergy * energyModuleCount) + (moduleSpeedBonusEnergy * speedModuleCount) + (moduleProductivityBonusEnergy * productivityModuleCount)
                        local slotModuleBonusSpeed        = (moduleSpeedBonusSpeed * speedModuleCount) + (moduleProductivityBonusSpeed * productivityModuleCount)
                        local slotModuleBonusProductivity = productivityModuleCount * moduleProductivityBonusProductivity
                        
                        local energy = math.max((beaconModuleBonusEnergy + slotModuleBonusEnergy), -80) / 100 -- at least 20% energy consumption
                        local speed = (beaconModuleBonusSpeed + slotModuleBonusSpeed) / 100
                        local productivity = (slotModuleBonusProductivity + biochamberBaseProductivity) / 100
                        
                        local gainPerRecipe = calcNutrientGainPerRecipeForBiochamber(energy, speed, productivity)
                        if (gainPerRecipe > bestGain) then
                            do bestGain = gainPerRecipe end
                            do winner = {
                                input = {
                                    tier = tier,
                                    quality = quality,
                                    halfSpoiledNutrientWorth = halfSpoiledNutrientWorth
                                },
                                machine = {
                                    machine = "biochamber",
                                    energyModuleCount = energyModuleCount,
                                    speedModuleCount = speedModuleCount,
                                    productivityModuleCount = productivityModuleCount
                                },
                                beacon = {
                                    beaconCount = beacons,
                                    beaconEnergyModuleCount = beaconEnergyModuleCount,
                                    beaconSpeedModuleCount = beaconSpeedModuleCount
                                },
                                gainPerRecipe = gainPerRecipe
                            } end
                        end
                    end
                end
            end
        end
    end
    do return winner end
end

local stringifyRecursive = true
do stringifyRecursive = function(obj, indent, ignoreFirstIndent)
    local result = ""
    if type(obj) ~= "table" then
        if (ignoreFirstIndent) then
            do result = tostring(obj) end
        else
            do result = indent..tostring(obj) end
        end
        do return result end
    else
        if (ignoreFirstIndent) then
            do result = "{" end
        else
            do result = indent.."{" end
        end
        for k, v in pairs(obj) do
            local newIndent = indent.."    "
            if type(k) == "table" then
                do error("stringify: table keys are not supported") end
            end
            local s_key = newIndent..tostring(k)
            local s_value = stringifyRecursive(v, newIndent, true)
            local append = "\n"..s_key.." = "..s_value..","
            do result = result..append end
        end
        do result = result.."\n"..indent.."}" end
    end
    do return result end
end end

local stringify = function(obj)
    do return stringifyRecursive(obj, "", false) end
end

print(stringify(getEfficientSetup("tier1", 1, 0.5)))
print(stringify(getEfficientSetup("tier1", 1, 1)))
print(stringify(getEfficientSetup("tier1", 2, 0.5)))
print(stringify(getEfficientSetup("tier1", 2, 1)))
print(stringify(getEfficientSetup("tier1", 3, 0.5)))
print(stringify(getEfficientSetup("tier1", 3, 1)))
print(stringify(getEfficientSetup("tier1", 4, 0.5)))
print(stringify(getEfficientSetup("tier1", 4, 1)))
print(stringify(getEfficientSetup("tier1", 5, 0.5)))
print(stringify(getEfficientSetup("tier1", 5, 1)))

print(stringify(getEfficientSetup("tier2", 1, 0.5)))
print(stringify(getEfficientSetup("tier2", 1, 1)))
print(stringify(getEfficientSetup("tier2", 2, 0.5)))
print(stringify(getEfficientSetup("tier2", 2, 1)))
print(stringify(getEfficientSetup("tier2", 3, 0.5)))
print(stringify(getEfficientSetup("tier2", 3, 1)))
print(stringify(getEfficientSetup("tier2", 4, 0.5)))
print(stringify(getEfficientSetup("tier2", 4, 1)))
print(stringify(getEfficientSetup("tier2", 5, 0.5)))
print(stringify(getEfficientSetup("tier2", 5, 1)))

print(stringify(getEfficientSetup("tier3", 1, 0.5)))
print(stringify(getEfficientSetup("tier3", 1, 1)))
print(stringify(getEfficientSetup("tier3", 2, 0.5)))
print(stringify(getEfficientSetup("tier3", 2, 1)))
print(stringify(getEfficientSetup("tier3", 3, 0.5)))
print(stringify(getEfficientSetup("tier3", 3, 1)))
print(stringify(getEfficientSetup("tier3", 4, 0.5)))
print(stringify(getEfficientSetup("tier3", 4, 1)))
print(stringify(getEfficientSetup("tier3", 5, 0.5)))
print(stringify(getEfficientSetup("tier3", 5, 1)))

There are following general questions to consider:

1. Nutrients energy capacity does not depend on its spoiledness. So do we consider half spoiled Nutrients worth 1 Nutrient or 0.5 Nutrients?
2. Do we want to put efficiency modules into beacons?
3. Which tier modules do we want to use?
4. Which quality are our modules and buildings? (Assuming they have the same quality. I did not run the code ot check for any for mixed quality setups)

Here is a summary of my findings:

• if we don't use any modules, it is better to use assembling machines than biochambers
• if we don't want to put efficiency modules into beacons, then:
  • it sometimes matters how much we consider half spoiled Nutrients worth
  • we will always have worse gain than with efficiency modules in beacons
  • the difference is noticeable using low quality setups, but it gets really small for high quality setups. Interestingly quality matters way more than tiers here, even though the difference shrinks with the tiers as well.
• if we want to put efficiency modules into beacons, then:
  • we always want to use biochambers with full productivity modules
  • it does not matter how much worth we consider half spoiled Nutrients, the most efficient layout is always the same per tier and quality
  • using rare quality tier 3 modules is the most interesting edge case, as it is the only place where using 11 beacons is the most efficient setup. I have no idea why that is the case

Also, and this i think is the most interesting finding of mine: Speed Modules actually do increase the amount of Nutrients gained per recipe in the biochamber! Even though they increase the overall energy consumption, the reduced creafting time actually matters more, it leads to the biochamber consuming less energy per recipe (but more per second).

I will share the full list now, don't get triggered by the "woth" spelling mistake please


Full List without using Efficiency modules in beacons:

Open or Close

grafik.png (55.82 KiB) Viewed 418 times

Full List with using Efficiency modules in beacons:

Open or Close

grafik.png (63.67 KiB) Viewed 418 times

[Tertius]

Stupid question: why is this relevant and worth researching?

As far as I understand, this recipe is meant for just for a few crafts (20-30?) for cold starting a stopped factory, so any kind of boost or efficiency isn't relevant. Afterwards the vastly more efficient "nutrients from bioflux" and "nutrients from biter eggs" recipe in biochambers will be used as soon as the corresponding biochambers have their nutrients. For safe cold starting bioflux production the intermediate recipe "nutrients from yumako mash" is useful as well.

Balance without productivity:
- with nutrients from bioflux you get 8 nutrients per bioflux
- with nutrients from biter eggs you get 600 nutrients per bioflux

If you use the nutrients from spoilage recipe, you need 10 spoilage per nutrient. To get the 600 nutrients (per bioflux) as from the nutrients from biter eggs recipe, you need to somehow produce 6000 spoilage (per bioflux), which I don't see a way to achieve.

In case you want quality nutrients, the "nutrients from biter eggs" is still the most efficient recipe. It doesn't allow quality modules, but recycling after production yields 600 / 4 = 150. If used with productivity modules the ratio is even better, since with nutrients from bioflux you have only 4 module slots (1 production machine) compared to 8 module slots for nutrients from biter eggs (4 modules for production machine, 4 modules for recycler).

All this calculation doesn't even take productivity into account, so with productivity it's even more efficient to not use the spoilage recipe.

So what's the use case?

[Lindor]

Stupid question: why is this relevant and worth researching?

As far as I understand, this recipe is meant for just for a few crafts (20-30?) for cold starting a stopped factory, so any kind of boost or efficiency isn't relevant. Afterwards the vastly more efficient "nutrients from bioflux" and "nutrients from biter eggs" recipe in biochambers will be used as soon as the corresponding biochambers have their nutrients. For safe cold starting bioflux production the intermediate recipe "nutrients from yumako mash" is useful as well.

Balance without productivity:
- with nutrients from bioflux you get 8 nutrients per bioflux
- with nutrients from biter eggs you get 600 nutrients per bioflux

If you use the nutrients from spoilage recipe, you need 10 spoilage per nutrient. To get the 600 nutrients (per bioflux) as from the nutrients from biter eggs recipe, you need to somehow produce 6000 spoilage (per bioflux), which I don't see a way to achieve.

In case you want quality nutrients, the "nutrients from biter eggs" is still the most efficient recipe. It doesn't allow quality modules, but recycling after production yields 600 / 4 = 150. If used with productivity modules the ratio is even better, since with nutrients from bioflux you have only 4 module slots (1 production machine) compared to 8 module slots for nutrients from biter eggs (4 modules for production machine, 4 modules for recycler).

All this calculation doesn't even take productivity into account, so with productivity it's even more efficient to not use the spoilage recipe.

So what's the use case?

I wanted to see if someone else answers because it might have come across as defensive. I admit that i expected a different kind of reaction to my analysis, but i don't want to attack you here.
But it's been long enough and nobody answered, so here i am

For me it's all about Gleba.
Nutrients from Biter Eggs are not available on Gleba, so that's off the table here.
Yes nutrients from bioflux, yumako mash and/or jelly is far more efficient than letting them spoil into spoilage first.
But if you happen to have leftover spoilage and don't need to craft like t3 efficiency modules or smth like that, turning it into nutrients is simply the best you can do with it. Or at least i consider it the best so far.

On Gleba it's all about efficiency.
You might think that it's always better to craft nutrients from bioflux than from yumako mash or jelly and always better to do that than to craft it from spoilage. Well that's not entirely true. It is not as strong of an "always" as you might think. If your production happens to be unbalanced and you produce more yumako mash than you need so the jelly becomes the bottleneck, it's better to turn the leftover yumako mash into nutrients first before you turn bioflux into nutrients. This way the bioflux (and therefore the bottlenecking jelly that it's made of) that would have been turned into nutrients can be used for something more useful instead.
The same goes for spoilage. Turning leftover spoilage into nutrients means that whatever else would have been used to craft nutrients can now be used as an ingredient for something more useful instead.

And it is completely unavoidable to have an imbalance in production because the balance depends on demand which is highly varying. Only at the very very end when your factory won't ever change except for tiny tiny differences here and there it may be possible that your factory is completely balanced if you're some kind of factorio superhero. We will always have something leftover that can be turned into nutrients. And if it's spoilage, you now know how to do it most efficiently.

And then there's one last final point which however i deem most important, and that is that it's simply an interesting problem to solve. If we only ever do something if we expect usage value from it, then that comes with tons and tons of issues. That's a very capitalistic thought process - and we see where that has led us. It's alright to focus on things that are very urgent if there are such things, but always take at least a little time to do just interesting things. Even if i wouldn't see any practicality right now it might become useful in the future for me or for someone else. I have done these kinds of analyses many times not only here but also on different forums for different games, and sometimes after years of silence i suddenly get notified on a reply of someone who used it to build on my work and do something good with it. I'm very strongly against utilitarianism.

[Tertius]

I'm sorry if my comment offended you and I apologize. I was just surprised how there was such a thorough research about a thing I never even noticed, and I was curious what's it about.


We seem to have approached Gleba in different ways. I don't see any imbalance between yumako and jellynut and its products. Yes, we consume more yumako than jellynut (I think), but that's not an issue. My factory starts with bioflux production, which does jelly and yumako mash production first. This fuels the bioflux biochambers, and the overflow is being put on a belt that leads to the secondary jelly and yumako mash consumers: bacteria, rocket fuel, lubricant, and whatever. From what's left from all this, is directly put into the heating towers and being burnt. I don't even let them spoil.

Fruits are constantly being harvested, mashed, products consumed, and the leftover is being burnt. So there is never any imbalance. If there is some for you, you need to add more production on the lacking side. And if your farms are not able to deliver enough fruit, craft artificial soil and overgrowth soil and fertilize your farms to increase the harvest. This is what you should concentrate on, not on how to squeeze as much out of the spoilage as you can.

About biter eggs. You don't seem to have it on Gleba, but you should have it on Gleba. For overgrowth soil and for t3 productivity modules. Gleba is the perfect location for all your T3 module factory, because everything is free. You just need to plant and harvest enough fruit, and fruit is infinite. I import biter eggs from Nauvis with the same platform that brings agricultural science packs to Nauvis, on the return flight. And I actually use the "nutrients from biter eggs" recipe for nutrient production on Gleba, with a fallback to the bioflux recipe in case of egg shortage. I could just increase bioflux production instead, but since I import eggs anyway for modules and for soil and I'm destroying them before they hatch, I can use them for nutrients just as well. A circuit contraption chooses the appropriate recipe for the biochambers and switches between bioflux and egg supply. Hey, that was my first use for the new splitter circuit control. Using this recipe avoids having to build a bigger bioflux production, which would require a redesign, because the current size relies on the capacity of the green belt.

So for me spoilage is for a few recipes like carbon that have spoilage as mandatory ingredient, but I never felt the need to recreate nutrients on a regular basis from it. There is always enough nutrients left from the machines that feed the bioflux biochambers and the pentapod egg biochambers. Because of the high nutrient demand of those recipes I use direct insertion, so there is much overflow as byproduct that can be used for feeding all the biochambers.

Consumption is quite dynamic, because it has to run even if there is no research with the agricultural science pack. Or if there is nothing to craft at the moment so no bacteria production.

[Lindor]

I'm sorry if my comment offended you and I apologize. I was just surprised how there was such a thorough research about a thing I never even noticed, and I was curious what's it about.

That's what i tried to say, you didn't offend me and i don't want to be perceived as being defensive. Just because i expected a different kind of reaction doesn't mean that you hurt my feelings. There's really no need to apologize to me. On the contrary i'm glad that you're so interested

About biter eggs. You don't seem to have it on Gleba, but you should have it on Gleba. For overgrowth soil and for t3 productivity modules. Gleba is the perfect location for all your T3 module factory, because everything is free. You just need to plant and harvest enough fruit, and fruit is infinite. I import biter eggs from Nauvis with the same platform that brings agricultural science packs to Nauvis, on the return flight. And I actually use the "nutrients from biter eggs" recipe for nutrient production on Gleba, with a fallback to the bioflux recipe in case of egg shortage. I could just increase bioflux production instead, but since I import eggs anyway for modules and for soil and I'm destroying them before they hatch, I can use them for nutrients just as well. A circuit contraption chooses the appropriate recipe for the biochambers and switches between bioflux and egg supply. Hey, that was my first use for the new splitter circuit control. Using this recipe avoids having to build a bigger bioflux production, which would require a redesign, because the current size relies on the capacity of the green belt.

Now that is quite interesting. I craft the soil on Nauvis. Having biter eggs in space is too dangerous for me. And i'm not worrying about t3 modules until i have legendary quality. I don't know, in my head it kinda seems like a waste to do non-legendary t3 modules.

[angramania]

But if you happen to have leftover spoilage and don't need to craft like t3 efficiency modules or smth like that, turning it into nutrients is simply the best you can do with it. Or at least i consider it the best so far.

From my point of view, it is the worst you can do with spoilage. Spoilage is the only local source of carbon and sulfur. Sometimes I have to produce and recycle nutrients to get enough spoilage, and almost never vice versa.

And it is completely unavoidable to have an imbalance in production because the balance depends on demand which is highly varying.

Most of gleba's products are crafted from bioflux. So it is the only thing you really need to balance. And it is quite simple, just have twice more yumako than jelly.

[coffee-factorio]

My interest this topic is using spoilage as an efficiency optimization. It's part of a set of strategies to extend the life of nutrients. If you can throw them in a recycler before they spoil, a nutrient is made from 10 spoilage, so it recycles down to 2.5. So anything I don't use ends up getting rebuilt on the justification that 1 bioflux is an awful lot anything interesting.

I found that jelly->iron bacteria makes at least as much spoilage as you'd normally get or a little more once you develop the system with productivity. Yumako can be sunk into nutrients or projects more easily. The extra iron is turned into ammo before you have recyclers. After you have recyclers, it's a small boost to parts of a facility. So instead of waiting around for nature to take it's course when I clearly have too many fruit, this is one of two options I use if I can't make something useful.

The other option is to directly burn the jelly product to run beacons (or if nothing else be rid of it).

I also have some use for it on Nauvis because several metric tons of bioscience spoil when I inevitably turn my attention away from cool things you get at Gleba. I only need so many eff. 3 modules and once biter eggs get rolling, that's going to cover that too. The idea there is to maximize the amount of fish I have.

As for things like carbon and even iron... I can guarantee a bulk line of those coming through a cargo bay. So that one is going to be a playstyle choice until it isn't. I have messed around with some weird concept lines for making plastic out of it but... it's iffy?

[Tertius]

Now that is quite interesting. I craft the soil on Nauvis. Having biter eggs in space is too dangerous for me. And i'm not worrying about t3 modules until i have legendary quality. I don't know, in my head it kinda seems like a waste to do non-legendary t3 modules.

That's really interesting, because artificial and overgrowth soil can only be crafted on Gleba. You have to import biter eggs to Gleba to fertlize your farms with that. The required soil is not just landfill.

Biter eggs are cargo like everything else, they just must not hatch on a platform. This can be achieved by making sure the loaded eggs are reasonably fresh, the platform is flying straight to Gleba, and on Gleba there must be a permanent request for more biter eggs the platform is ever supposed to carry. On Gleba, you just need a few laser turrets to keep any hatchlings under control (you need them anyway against hatched pentapod eggs) and a few more near the storage chest that's filtered for biter eggs.

Bonus: create a timer that counts the time since the last egg delivery and pulls all existing eggs into a heating tower if it was more than 10-15 minutes ago.

Or better, continually destroy biter eggs at a slow rate while eggs are above some upper threshold and use the "spoiled first" filter for the inserter that's pulling eggs from the storage chest into the heating tower, so fresh eggs will come in from the platform and stale eggs are slowly being removed and burnt. 5-6 simple laser turrets nearby are able to cope with 2-3 stacks of biter eggs hatching simultaneously. They're no real threat, especially if you have roboport coverage and repair kits, which you should have anyway for any hatching pentapod eggs.

About normal t3 modules: their bonus is too good to wait until you're able to craft legendary modules in acceptable quantity. The reduced factory footprint and increased production will help immediately and will help everywhere. The earlier you use them, the longer you profit from them. If you're afraid you waste resources crafting them: you need them anyway to upcycle modules to higher quality, so you can later collect them again and inject them in your legendary module factory.