Can someone help me with basic game theory?

[sea.dart]

I'm struggling to really understand and excel at this game. I have some ideas, but I'm not sure how to implement them. I'll try and explain my understanding, so please let me know if I'm clueless or clarify things I don't quite get.

It seems Satisfactory is a logistics management game. Aside from manufacturing products, the game focuses on the distribution of those material. I feel I need to come up with a method that is both adaptable and scale-able. As such, I need to create multiple parallel banks of storage and distribution systems.
So I need to store and distribute Iron Ore. One storage unit can hold the ore, but the belts can only move them so fast and if I need more than one manufacturing machine, that decreases the overall availability of the ore. How many parallel storage systems should I create? Nine (three banks of three-way splitters)? Fifteen? Thirty? For every resource and product?
Then I need to be able to get that ore or product to numerous machines. After that I need to store the screws, rods, frames, etc. If I need to supply more pipes than I'm creating, I need to be able to duplicate that production line to double production, which requires more machines and more supplies. Even worse when I need to combine iron, copper, and silica, all of which may have their storage in vastly different locations. How do I route all these belts without making an unmanageable tangle?
In short I feel completely overwhelmed trying to accomplish what I think I need to do. Just with the Ficsmas Wonder Star I need like 25,000 candy canes. I'm seriously wondering how I'm even going to be able to get that in any reasonable period of time. In addition to the production of Ornament Bundles and Tree Branches.
I hope you can make sense of my musings and bring some clarification to my confusion. Thanks.

[Ryzilynt]

Over time i've learned that if you can dream it, you can make it happen.

You will make it happen, you will make it run to spec.

[Still B0r3d]

Originally posted by sea.dart:

So I need to store and distribute Iron Ore. One storage unit can hold the ore, but the belts can only move them so fast and if I need more than one manufacturing machine, that decreases the overall availability of the ore. How many parallel storage systems should I create? Nine (three banks of three-way splitters)? Fifteen? Thirty? For every resource and product?

Really once you get fast enough belts and you get a little more familiar with everything, you do not need to store ore because you use it just as fast as you mine it.

Ultimately there is a balance that YOU need to find based on how you want to approach the game. My item storage consists of 1 industrial storage per item, generally they fill faster than I use them. If I am doing a larger project and suddenly I find myself running out of an item like modular frames or concrete, I have to decide if there is something I should do while I wait or if I should spend the day increasing production so it is no longer an issue going forward. Then you may even have additional factors to consider if you are running a dedicated server, because those can continue to produce items/coupons while you sleep.

Ultimately my suggestion is to not overthink your first play-through, it is ok if it is not perfect or even a little wrong. You don't know what you don't know right? There are many factors you can't account for because you don't even know what your going to unlock yet.

Then your second play will likely have many ideas you gained from doing things wrong for the last 300 hours.

[Doc✪Hollywood]

Game theory ... reminds me of some sophisticated math courses.

I think you understand the game quite well. But it makes no sense to store ore. Just use what the miners give you. Later products that you need to build stuff you will want to store.*

All other products, which are not needed to build stuff, you just process.



* Usually beginners store far more than seasoned players. Of most stuff you will not need more than one or two industrial containers. Don't place containers between machines as buffers either. They just need a lot of space and are superfluous if you clock machines correctly.

[RadioIesbian Fluid]

Its mainly down to scale. If you determine you need 25000 of something, and you produce it at a rate of 10/ min then it'll take 2500 minutes to make it. If you want it faster then making more machines to make same product speeds it up. If you're limited in the belts capacity then using more belts is the solution.

Method for scaling up is pretty much down to you. There are a few ways to do it.
Overclocking, up to 250% reduces the amount of machines you need but at the cost of more power.
Vertical building, if you can't place more machines in the area you want, building another floor above is an alternative.
Theres other solutions too, like multiple locations then making them converge the resources. Typically using trains and/or drones to transport them.

The game is a bit of a puzzle where you can choose how to solve it, either with larger factoriers, more factories or just waiting.

[Doc✪Hollywood]

If you are interested in game theory I recommend Erickson https://www.amazon.com/Game-Theory-Introduction-K-Erickson/dp/1494364891

[Padds]

Originally posted by sea.dart:

So I need to store and distribute Iron Ore. One storage unit can hold the ore, but the belts can only move them so fast and if I need more than one manufacturing machine, that decreases the overall availability of the ore. How many parallel storage systems should I create? Nine (three banks of three-way splitters)? Fifteen? Thirty? For every resource and product?

Now im not saying you should not do that but i do want to point out you dont have to.

first i wouldnt make any huge decisions until you get to see what mercer spheres are for from MAM research,

now iron ore is everywhere, why do you need to create a huge depot? if you want to dive on in, but why not make each more complex part from scratch creating the iron plates you need at site

[Doc✪Hollywood]

not only is iron ore everywhere, even if not, it simply makes no sense to store it

you would build more machines than what is needed for what the miner alone can deliver. then when the iron containers are empty these "more" machines would be idling. That is pointless.

[Maehlice]

I feel like you're overthinking this too early -- like you've watched someone's megafactory build and now want to recreate it.

(If that's the case, put it out of your mind. Those guys have literal tens of thousands of hours in this game. Crawl before you walk before you run before you totalxclipse.)

As you play and build your factory piece by piece, you'll develop your own patterns and style that makes sense to you. (That's the important part: it should be whatever makes sense to you. Efficient, inefficient, doesn't matter. It should to you be ... satisfactory.)

---

Read up on the Manifold[satisfactory.wiki.gg] in the wiki and make sure you understand that. That is the defacto standard for distributing parts.

If you want some detailed steps, I go through step by step how to get a starter factory up and running in this guide:

https://steamcommunity.com/sharedfiles/filedetails/?id=2686103223

[churuchuru]

you don't really need to store anything other than stuff you use to build foundations/belts/power poles etc but storage is not the goal of the game and after getting cloud storage you forget about these, the goal of the game is to calculate how much resources you need to produce the next thing on the list and build a factory capable of doing that

[Jack-o-Lantern]

A bit in hurry so I will give only a first quick response on first part of question....

OP speak about moving ore....
Moving ore isn't a good option.
Always try to do a first transformation locally to minimize amount of item that need to be moved around.

Also consider that as you move to higher tiers, you'll get more options for transferring items ..
More in terms of different type (like drones and train) and as same but faster versions.

[Doc✪Hollywood]

Originally posted by churuchuru:

the goal of the game is to calculate how much resources you need to produce the next thing on the list and build a factory capable of doing that

You don't even need to calculate that. The game tells very accuratly how many parts you need for the next HUB upgrade and MAM research, it never requires you to produce a certain rate like "produce 100 Supercomputers per minute".

Personally I find this a pity. Whenever I build something big I feel half proud. And half stupid. Because it wasn't necessary. The slowest facotory could produce the same amount of required parts.

I don't know how other people with ADHD see this, but me personally I would much more enjoy if the game tells me to build 30 Nuclear Pasta per minute, than when I just chose so myself "because I can". The sense of achievement and satisfactory would be much bigger. But I guess most people, foremost the ones without ADHD, appreciate this freedom the game gives you. After all: This is how Satisfactory is. That won't change.

[Maehlice]

Originally posted by Gordon✪Gekko:

... I would much more enjoy if the game tells me to build 30 Nuclear Pasta per minute, than when I just chose so myself "because I can". The sense of achievement and satisfactory would be much bigger. But I guess most people, foremost the ones without ADHD, appreciate this freedom the game gives you. After all: This is how Satisfactory is. That won't change.

+1

Constructing X quantity of a thing is super straightforward and easy. With quantity goals, you can basically turn Satisfactory into an idle game. Even making just 1 ppm, you can let it run overnight and wake up to 500 parts of something.

I think it's good that we have that option, but I would enjoy seeing rate goals sprinkled in -- maybe even as special events.

[robw23]

Just look at the input/output numbers and do the best you can and, before you know it, you will be zipping along. There is no penalty for making a crappy build and you can always build a better one somewhere else and keep/destroy this one. U will slowly build it up as you concentrate on the MAM tree and the first couple Tiers. I also like to explore and get some hard drives and spare parts (don't sink them - they are great for getting you through the tech trees faster). Just a tip - you can deconstruct the HUB and take it to another location at any time. When I first played, I felt like the field I landed in was big. By the time you reach the mid-game, you will be using half the entire map. Don't be afraid to spread out - the factory must GROW!

[Doc✪Hollywood]

Originally posted by Maehlice:

+1

Constructing X quantity of a thing is super straightforward and easy. With quantity goals, you can basically turn Satisfactory into an idle game. Even making just 1 ppm, you can let it run overnight and wake up to 500 parts of something.

I think it's good that we have that option, but I would enjoy seeing rate goals sprinkled in -- maybe even as special events.

+1

Rate Goals and Special Events (like in No Man's Sky) would be awesome.

For me personally it would add a lot to the game and make an already good game an excellent game.

[jvjd]

Simplest approach ??

Focus on the end product , check the resource and nodes available , build the factory at a place that close to all the resources .

That end product might be an input to another product you may need to build in future so now this factory get treated as a resource node by it self

Rinse and repeat , Soon you will have multiple factories all over the map - so you will need transportation . You can pretty much zip line , hover everywhere , teleport every where which becomes very usefully for farming mercer spheres and somergloop in the future

However ...

If you plan bring all the resource to one area and then from there divide it to different units. Then you would have to build your factory in the air or at ocean because that HUGE , that complicated and too much strain on your pc it will become .