Factorio Needs More Recipe Types

[Yoshifizzle]

Every time that something new is added to Factorio, I get excited because that means I can automate something new. It means that I have to develop my supply chain and factory to be ever more complex. However, that excitement is often fleeting as it is very easy to automate most new items. In this simplest case, you already have all the ingredients in your logistic network and you just set up an assembling machine, a requester chest, a passive provider chest, and a couple of inserters. Done. Excitement over. Additional challenge can be added if the new item is at the end of a long chain of crafting steps, using, for instance, modules in its construction. If large numbers of the new item are needed, then you must build out a significant amount of infrastructure in order to fortify the supply chain (more iron to build more circuits to build more modules to build more new item). This helps to extend the excitement of a new item for some time, but boredom will eventually set in.

I decided to look at why a lot of new items seem so humdrum to automate by categorizing all the recipes in Factorio. It turns out that all Factorio recipes can be reduced to a few generalized forms (see attachments). It may come as no surprise that the overwhelming majority of recipes in the game are of the form (176/181 recipes): A -> B, A + B -> C, A + B + C -> D, A + B + C + D -> E, A + B + C + D + E -> F, or A + B + C + D + E + F -> G. If the automation of each unique item in Factorio is to be viewed as a puzzle that needs to be solved/optimized, then any new item utilizing these same recipes will present little challenge to the experienced Factorio player. These are puzzles that have been mastered and most players are likely to already have a blueprint handy that can used to automate recipes of this form. I will admit that there is some variation to be had by incorporating liquids into the recipe, by making the recipe require ingredients in lopsided ratios (i.e. 40 iron gears to 1 fast underground belt to craft an express underground belt) that make it challenging to deliver an assembling machine an optimal mix using belts, or by increasing the number of ingredients relative to the crafting time that make it challenging to insert all items before the crafting cycle is complete. Overall, recipes of the form A + B -> C are a challenge that Factorio players have mastered.

There are several unique recipe forms found in Factorio, all of which relate to oil processing or the upcoming addition of nuclear power/uranium processing (see attachment). Basic and Advanced oil processing have the recipe forms A -> B + C + D and A + B -> C + D + E, respectively. I think most Factorio players can vividly remember stumbling through their first oil setup, trying to route all the piping from more than one refinery without making a spaghetti mess and without accidentally combining oil byproducts. It was a great puzzle, one that they hadn't dealt with before and that made it fun and exciting. Uranium processing presents two new recipe forms: A -> B + C (uranium ore processing) and A + B -> A + B (uranium enrichment). From the limited pictures we have of a uranium processing setup, it looks like players will need to utilize feedback loops and filter sorting of outputs, concepts that someone could go the whole game without exploring. I am sure that the addition of uranium/nuclear power will provide more entertainment value for players than if the crafting chain was uranium ore -> enriched uranium -> nuclear fuel rod (A -> B -> C).

I hope that this can spark a discussion on how to create new puzzles for the player in Factorio by implementing different recipe forms. Ideally, we can find a balance between challenging, engaging gameplay and approachability for newer players.

[bobingabout]

my mod uses A -> B + C and A + B -> C + D quite often.

People curse it, then love it.

And keep asking me to add more uses for Sodium Hydroxide.

[Deadly-Bagel]

I still have too much Sodium Hydroxide! xD And the problem is when you want to produce say the t3 battery and you want to put down a dedicated Electrolyser for it you have nothing to use the Sodium Hydroxide. I love your mods because it encourages breaking away from the bus design but then you go and do something like that to force us back onto it xP

But I digress. I introduced my brother and his friend to the game and most of what I was doing was pretty much trivial to me, setting up circuits and cracking and belt balancing and the like, and pretty much everything I did had them boggling over it for a good ten minutes.

As a master of the game it is easy to forget what it was like when you started. If you need reminding, go play Bob's Mods ^^ as Bob said it has some nicely complex recipes that'll have you scratching your head and cursing the whole thing but once you get going you'll have a lot of fun =P

[dandielo]

Every time that something new is added to Factorio, I get excited because that means I can automate something new. It means that I have to develop my supply chain and factory to be ever more complex. However, that excitement is often fleeting as it is very easy to automate most new items. In this simplest case, you already have all the ingredients in your logistic network and you just set up an assembling machine, a requester chest, a passive provider chest, and a couple of inserters. Done. Excitement over. Additional challenge can be added if the new item is at the end of a long chain of crafting steps, using, for instance, modules in its construction. If large numbers of the new item are needed, then you must build out a significant amount of infrastructure in order to fortify the supply chain (more iron to build more circuits to build more modules to build more new item). This helps to extend the excitement of a new item for some time, but boredom will eventually set in.

I decided to look at why a lot of new items seem so humdrum to automate by categorizing all the recipes in Factorio. It turns out that all Factorio recipes can be reduced to a few generalized forms (see attachments). It may come as no surprise that the overwhelming majority of recipes in the game are of the form (176/181 recipes): A -> B, A + B -> C, A + B + C -> D, A + B + C + D -> E, A + B + C + D + E -> F, or A + B + C + D + E + F -> G. If the automation of each unique item in Factorio is to be viewed as a puzzle that needs to be solved/optimized, then any new item utilizing these same recipes will present little challenge to the experienced Factorio player. These are puzzles that have been mastered and most players are likely to already have a blueprint handy that can used to automate recipes of this form. I will admit that there is some variation to be had by incorporating liquids into the recipe, by making the recipe require ingredients in lopsided ratios (i.e. 40 iron gears to 1 fast underground belt to craft an express underground belt) that make it challenging to deliver an assembling machine an optimal mix using belts, or by increasing the number of ingredients relative to the crafting time that make it challenging to insert all items before the crafting cycle is complete. Overall, recipes of the form A + B -> C are a challenge that Factorio players have mastered.

There are several unique recipe forms found in Factorio, all of which relate to oil processing or the upcoming addition of nuclear power/uranium processing (see attachment). Basic and Advanced oil processing have the recipe forms A -> B + C + D and A + B -> C + D + E, respectively. I think most Factorio players can vividly remember stumbling through their first oil setup, trying to route all the piping from more than one refinery without making a spaghetti mess and without accidentally combining oil byproducts. It was a great puzzle, one that they hadn't dealt with before and that made it fun and exciting. Uranium processing presents two new recipe forms: A -> B + C (uranium ore processing) and A + B -> A + B (uranium enrichment). From the limited pictures we have of a uranium processing setup, it looks like players will need to utilize feedback loops and filter sorting of outputs, concepts that someone could go the whole game without exploring. I am sure that the addition of uranium/nuclear power will provide more entertainment value for players than if the crafting chain was uranium ore -> enriched uranium -> nuclear fuel rod (A -> B -> C).

I hope that this can spark a discussion on how to create new puzzles for the player in Factorio by implementing different recipe forms. Ideally, we can find a balance between challenging, engaging gameplay and approachability for newer players.

Thanks for the great summary of factorio recipes, this gave me a much better starting point while reworking first things in "DyTech"'s mods!
Thanks a lot once again!

[BenSeidel]

And keep asking me to add more uses for Sodium Hydroxide.

Here I thought it was for target practice! I keep running out of Sodium Hydroxide, not ammo. Maybe you're not making enough of it

[Yoshifizzle]

my mod uses A -> B + C and A + B -> C + D quite often.

People curse it, then love it.

And keep asking me to add more uses for Sodium Hydroxide.

I was almost thinking about putting a disclaimer at the bottom of my post to not tell me to play Bob's Mods I do appreciate that there are excellent modding choices out there to increase the complexity of the game. I was merely trying guide the development direction of the base game to include more interesting recipe forms (and to give them props for trying something new with the addition of uranium processing). While I have not played too much Bob's Mods, I have watched quite a bit on Youtube. One puzzle that seemed to jump out at me again and again was the construction of a early-game (before filter inserters) greenhouse area for wood production that did not clog up with saplings/wood. Many Youtubers I have watched seem to struggle with setting up greenhouse production, despite trying a number of different configurations/strategies. I love that a simple set of recipes of the form A + B -> A + C and C -> A can lead to such interesting gameplay.

I really wish that complexity on the order of Bob's/Angel's would make it into the base game even though there are very good reasons such as approachability for new players to not do this. I am very weird in that I love what these mods bring to the game, but I do not usually play modded. I find it hard to get over things like graphical style differences and mod updating/mod compatibility.

[Yoshifizzle]

Thanks for the great summary of factorio recipes, this gave me a much better starting point while reworking first things in "DyTech"'s mods!
Thanks a lot once again!

I am glad I could help! I was scratching an curiosity itch when I came up with this list of recipe forms. However, after the results confirmed my suspicions about the limited variety of recipes, I decided to share it with everyone so they potentially put it to some good use.

On a sidenote, I have attached the file of all recipes including inputs/outputs/generic recipe form in case anyone is interested. I made this by hand, so hopefully I am not missing any recipes.

[Deadly-Bagel]

I was almost thinking about putting a disclaimer at the bottom of my post to not tell me to play Bob's Mods I do appreciate that there are excellent modding choices out there to increase the complexity of the game. I was merely trying guide the development direction of the base game to include more interesting recipe forms (and to give them props for trying something new with the addition of uranium processing). (...)

I really wish that complexity on the order of Bob's/Angel's would make it into the base game even though there are very good reasons such as approachability for new players to not do this. I am very weird in that I love what these mods bring to the game, but I do not usually play modded. I find it hard to get over things like graphical style differences and mod updating/mod compatibility.

Well as you said there are very good reasons not to do this and if you recall how difficult the game was for you initially would you really want to make it more complicated? IMO uranium processing will be the final appropriate production and logistic challenge, anything more would just be needlessly complicated things.

I mean you're at the point where the base game is too simple for you right? Because you've pretty much mastered it. So you're saying that the base game should be made harder to cater for your level of expertise, while admitting that it would make it too difficult for new players, because you just have a thing against mods. I mean I know where you're coming from with the mod thing, I've never really liked using them either (I'm even running a mostly 'pure' Bob's Mods pack, no extra content just Void Chest and Loot Chest which only installed after getting bored of chests filling up with Sodium Hydroxide and then running after thousands alien artefacts), but that's not a reason to add an even taller barrier to entry for new players.

[aober93]

I dont think so. Adding that and more new players wont be able to finish the game. Its already very hard.

[antichaos]

Shameless plug for my new mod Assemblybots https://mods.factorio.com/mods/antichaos/assemblybots
This adds an extra input and 2 extra outputs to every non-smelting recipe. It feels like a very different challenge and requires a whole new set of layouts.

Plus, the self replicating bots have to be managed carefully or they start to cause other sorts of problems.

[Engimage]

Actually all recipes having multiple output results which are not cross-consumable lead to serious game difficulties. They absolutely require voiding mechanisms like Flare Stack as you can never ideally balance output consumation without voiding excess results. You will most likely need some circuit network magic to void only some excess stuff based on buffer fill.

In vanilla the only multiresult recipe is oil cracking and you still have a way to convert everything into gas which is mostly the only useful result. This leads to pretty straightforward designs of your factory. And even now you need to use Steam Engine for voiding purposes early game or just replace filled tanks to empty them which is likely to be removed in 0.15.
In bob's mods when you are getting a lot of multiresult recipes your factory effectively becomes spaghetti with different kinds of circuit networks trying to balance stuff.

So these recipes add a new level of complexity which is fine for mods but is bad for vanilla.

[bobingabout]

And Sodium Hydroxide is the BANE of every bob's mods player. It's almost like a junk item that you struggle to dispose of as a byproduct of Salt water hydrolysis.

The way the base game handles Oil is pretty good. The recipes are designed with a bias towards needing petroleum gas, and you're able to crack down the heavier oils into the gas using only the oil, and water. This makes it more interesting, and provides and easy solution to the issue.

Adding other results (Like Sodium Hydroxide in my mod) that aren't easily interchangeable only makes things worse.

[Engimage]

And Sodium Hydroxide is the BANE of every bob's mods player. It's almost like a junk item that you struggle to dispose of as a byproduct of Salt water hydrolysis.

The way the base game handles Oil is pretty good. The recipes are designed with a bias towards needing petroleum gas, and you're able to crack down the heavier oils into the gas using only the oil, and water. This makes it more interesting, and provides and easy solution to the issue.

Adding other results (Like Sodium Hydroxide in my mod) that aren't easily interchangeable only makes things worse.

Sodium Hydroxide is the worst example.
Another one would be sulfuric acid (which is required in much greater quantities). You can get it as either byproduct of ore processing or from the oil refining. Both create imbalance in generator recipes forcing you to dispose of extra products.
Even more simple things like water electrolysis where you often need only oxygen forcing you to dispose of hydrogen.

[Deadly-Bagel]

Generally these things work best when there is never any need to dispose of anything. For example:

(current oil processing, requires B never be needed more than C)
A > B + C
B > C

(efficiency when produced together with less efficient alternative recipes)
A > 2B + 2C
A > 3B
A > 3C

(all outputs constantly being consumed but at variable rates)
A > 2B + C
A > B + 2C

Sometimes it makes more sense to dispose of something, such as with oxygen, or maybe you have something that is a byproduct of an efficient process.

(infinite resource, vent biproduct)
[null] > A > B + C
B > [null]
C > [null]

(more efficient to vent biproduct)
A > B
A > 2B + C
C > [null]
D > C

The problem with Sodium Hydroxide is it doesn't fall into any of these categories. There is no way of getting the associated products without producing it, and there's no way to delete it. Even if you make the ratio nonexistent it's still going to be a pain to deal with unless you allow something like mixing it back into water or add a void chest.

[Frightning]

I actually really like this suggestion. Most of the game's recipes just have you combining simpler components into more complicated items. But having recipes with multiple outputs radically changes how you have to approach managing the results (note how oil is a step up in complexity to the earlier basic recipes for example). The new recipe for Uranium refining seem interesting to me, although apparently it is pretty broken with Productivity modules atm.

[bobingabout]

And Sodium Hydroxide is the BANE of every bob's mods player. It's almost like a junk item that you struggle to dispose of as a byproduct of Salt water hydrolysis.

The way the base game handles Oil is pretty good. The recipes are designed with a bias towards needing petroleum gas, and you're able to crack down the heavier oils into the gas using only the oil, and water. This makes it more interesting, and provides and easy solution to the issue.

Adding other results (Like Sodium Hydroxide in my mod) that aren't easily interchangeable only makes things worse.

Sodium Hydroxide is the worst example.
Another one would be sulfuric acid (which is required in much greater quantities). You can get it as either byproduct of ore processing or from the oil refining. Both create imbalance in generator recipes forcing you to dispose of extra products.
Even more simple things like water electrolysis where you often need only oxygen forcing you to dispose of hydrogen.

Sulphur and water electrolysis are relatively well executed examples though. The Oxygen might be backed up, but Hydrogen can be disposed of by turning it into fuel. It can also power flamethrowers!

And Sulfuric acid demand will be increased with the new nuclear fuel mining.

For Sodium Hydroxide, I did plan to include more uses for it, reducing it's production is just a stop-cap measure.

[Deadly-Bagel]

I still have a problem with that though. Sulphuric Acid is produced in both ore processing and oil processing, both categories that also use sulphuric acid so it's going to be consumed nearby when planned properly. The process that makes Sodium Hydroxide (can't remember what it's called) is used all over the place at least somewhere in almost every category of manufacturing but it only has a few specific uses. This means if you're producing it where it's needed you're going to be disposing of it all over your base. It forces you to either produce it all in one place (complicating modular base designs) or you ship it from all over your factory(/ies) to the few places it's used. To be a healthy resource it needs to either be ventable or have constant uses everywhere it is created.

I mean it's used as a cleaning agent right? Maybe you could make pollution filters out of it to feed into some sort of pollution scrubber machine. That way wherever it's created you can just pump it into a few of these scrubbers, you see less pollution and you don't have to worry about excess. The scrubbers are also something you'll likely want in various locations around your factory so it works out well. Just an example of a healthy use for it.

[AileTheAlien]

I actually really like this suggestion. Most of the game's recipes just have you combining simpler components into more complicated items. But having recipes with multiple outputs radically changes how you have to approach managing the results (note how oil is a step up in complexity to the earlier basic recipes for example). The new recipe for Uranium refining seem interesting to me

Agreed. It might also be in the game's best interest, to add a recipe to the game similar to, but simpler than the oil recipes. If, for example, there was a recipe that produced two outputs instead of the three in oil processing, it could be a useful learning tool, to prepare the player for the complexity that oil processing introduces to the game.