I tend to get stuck mid-game

[wizcreations]

I struggle to figure out what I need to do next once I get chemical science completed. Early game makes sense to me even though I am very slow. Any advice to help me both play faster and not feel overwhelmed not knowing what to do next?

I also seem to hit a wall where I suddenly need a lot more of everything all of a sudden. In my current game, I got to the point of setting up advanced oil processing so that I could make lubricant for electric batteries only to discover I now need to set up a train system to go get copper, iron, and coal (for smelting). Do you all have a list of actions you complete each time you play so that you know what order to build and when to expand resources?

[angramania]

I hit this wall so many times. And I even know answer to it - research directly to construction robots, replace chests in mall with passive provider and build everything with bots, gradually increasing logistic network.
You do not need rail system for 45-90 SPM base, red belts and pipes are enough. Just triple space you have planned to have between sub-factories. You will need a lot of it for roboports and belt spaghetti. Start with increasing circuits production, so search places with both iron and copper, even better if it has coal not so far, you will need it for plastic. Oil can be far away, it is easier to transport oil and water by pipes than iron/copper/coal by belts.
Use factory planer mod or factoriolab site to plan your factory. It will save hours.

[wizcreations]

I want to get to robots next. That helps me build easier. But it doesn’t solve the problem of not knowing what I’m supposed to be done.

Up to blue science is really straight forward with minimum intermediary products to worry about, mostly just gears, pipes, copper wires, and circuits. Advanced oil processing is very confusing, and everything after it seems to take so many products with batteries, electric engines, and robot frames.

[Alfonse215]

Yes, the game gets more complicated with more possible things to do. So take them one at a time, just like you did before.

You cannot build electric engines; do not even try. Instead, make engines. Then make lubricant. You've already made green circuits. So all of a suddenly, you can make electric engines.

You cannot make flying robot frames. But you can make electric engines. So make batteries. You've already made steel and circuits. So... you can make flying robot frames.

[Tertius]

If you don't know where to begin, since there is so much to do, you're at the same point I was some years ago. I don't know how your brain works, but my brain works by having a goal, then work to fulfill that goal. Seeing a goal, no matter how far away, is easy. Seeing all the possibilities currently available makes me lost instead and not even start.

My goal at the time was the first rocket into space. So I created a new map in map editor mode (sandbox), where all items are just available and free to use. I set all tech to unlocked. I replaced all environment with lab tiles to get everything clean as a lab. And to signify this is not the real map but a development environment.
My goal was launching a rocket, so I placed a rocket silo in map editor mode. Getting started is the most difficult thing, and this was the start.

And with this final building I started to build the corresponding factory in reverse.

I investigated what things I needed to operate the rocket silo, which were 3 ingredients. I used the editor-provided infinity chests to provide these ingredients, and to provide the satellite. The silo worked, so I saw what is required to operate the silo.

The next step was to replace every infinity chest with assembling machines that produce the the item previously supplied by the chest. I supplied them with the next lower set of ingredients from new infinity chests. Then I did the same with the new infinity chests, working my way through all the manufacturing tree down to the smelters and mines. Replacing one infinity chest at a time with the factory that creates this item was a task small enough to tackle and not getting lost.

The crucial difference between map editor mode and real game is the ability to directly copy+paste buildings without intermediate blueprint (called "directly build"/ "directly deconstruct" or something like this in map editor options). With this, and with the free availability of every item no matter how expensive, it's possible to freely explore and refine builds with no delay.

I did the same thing for all of the science packs, since without it's not possible to unlock all the recipes.

And I created a mall to produce all items to build a factory. Belts, inserters, assembling machines etc. I didn't know at the time which item was required and which just nice to have, so I just built them all to have them all available just as in the map editor. I made several malls, for example I learnt it's unwise to design a mall requiring bulk inserters and blue belts if I didn't yet unlock this tech.

This made me realize the scope of a complete factory. How much space, how many items to mine, how many smelters, how many assembling machines per item to get everything built in a sane amount of time.

After everything was working fine in map editor and starting rockets, no infinity chests any more, I made blueprints from all the parts I developed and started using them on a real map. Before I unlocked robots, I set the blueprint as draft and built the items manually, and after I unlocked robots, I let the robots built it. Of course the blueprints didn't work out exactly as envisioned, but they were a start and a thing to work with, so I was successful to build a real and complete factory on a real map this way, and with my own blueprints and not the blueprints from somewhere else.

By the way, in software development this is called "top down" strategy.
Building everything from the ground up (and getting lost) is "bottom up".
In Space age, I used parts of both strategies. Work with small goals that solve the challenges of the corresponding planet. After that, build a whole system that incorporates all the features from all planets to get to the final goal - the final infinity research. Some software developers call this strategy "meet in the middle". You split the final goal into smaller goals better manageable, then work from the ground to reach the smaller goals.

[NineNine]

I want to get to robots next. That helps me build easier. But it doesn’t solve the problem of not knowing what I’m supposed to be done.

Up to blue science is really straight forward with minimum intermediary products to worry about, mostly just gears, pipes, copper wires, and circuits. Advanced oil processing is very confusing, and everything after it seems to take so many products with batteries, electric engines, and robot frames.

Advanced oil refining is tricky because you'll end up with different amounts of light oil, heavy oil, and petroleum gas. Some of those things you'll need more of at some times, and other times you'll need more of something else. Once you figure out one trick to dealing with those different levels of those three items, oil processing becomes straightforward. A hint is:

notice that there's a recipe to create solid fuel from each of those three products.

.

But, in terms of everything else in the game, you just take it one step at a time. If you need an item, then create a process to make that item. If that item needs a particular component, then you create a process to make that item. You don't have to do more than one thing at a time (except for the oil refining I mentioned above). If you just take it one step at a time, the game unfolds in a pretty straightforward way all of the way to a rocket launch.

[Mr Wednesday]

I want to get to robots next. That helps me build easier. But it doesn’t solve the problem of not knowing what I’m supposed to be done.

Up to blue science is really straight forward with minimum intermediary products to worry about, mostly just gears, pipes, copper wires, and circuits. Advanced oil processing is very confusing, and everything after it seems to take so many products with batteries, electric engines, and robot frames.

You can build things up from smaller components. For each science, work backward from the ingredients. For example, with purple science, the ingredients are productivity modules, railroad, and arc furnaces. Prod modules require red and green circuits, which I have from earlier science, so I can make them on site. Arc furnaces required red circuits (see previous), steel, and stone bricks. I have the steel on bus. Stone bricks just require a small smelting line from stone, then I can belt them in and make the arc furnaces on site with the purple science. Railroad is the most complicated, and requires a smaller block to make it (machines making iron stick to feed into the machines making railroad, which I belt to the purple science production.

I follow a similar strategy for yellow science. I build up the drone frames inline. I do the blue circuits and low-density structure separately because they're also needed for rocket components.

Advanced oil processing isn't too difficult as long as you consider cracking the heavier oils. Excess heavy oil (after making lubricant) cracks to light oil. Excess light oil (after making solid fuel and rocket fuel) cracks to petgas. Petgas is mostly what you need to use (for plastic and sulfur).

I would add that I don't find it that uncommon to have some spots where I need to spend an hour (or three) doing a major reconfiguration. Changeover to rail is one of the big ones, and it's not surprising to find that you need to bring all of the raw ingredients in via train as the base builds (in fact, it would be more surprising if you didn't, unless you're running nonstandard settings with rich and/or large deposits).

[wizcreations]

Advanced oil refining is tricky because you'll end up with different amounts of light oil, heavy oil, and petroleum gas. Some of those things you'll need more of at some times, and other times you'll need more of something else. Once you figure out one trick to dealing with those different levels of those three items, oil processing becomes straightforward. A hint is:

notice that there's a recipe to create solid fuel from each of those three products.

.

Do you typically replace all your coal consumption with solid fuel to keep your oil processing running?

[J-H]

I don't. Once mining is set up, the coal is free except for the cost of power. Solid fuel usually ends up going to rockets instead.

[wizcreations]

I went to take some engines off of the line making them for science so that I could use them for bots. Sadly I discovered most assembling machines were idle. I noticed that my factory is using more steel than I produce, and I don't have a nearby location I can use to make more. The one iron patch near the starting location is only big enough to feed my green circuit production. The next nearest iron patch is very, very far south. All my coal patches are to the north. (Copper and stone seem to be everywhere, though.)

I think I have a fundamental flaw in how I play. I think most of my initial setups for producing circuits, plastic, and other oil products are far bigger than I actually need, resulting in me using more resources than I can properly afford to use to feed those setups.

The only solution I see to my current setup is to demolish and rebuild my materials so I can better prioritize their usage. That just sounds like such a massive process to have to do by hand that I almost wonder if I am better off starting over.

[angramania]

I will repeat myself. Use factory planer or factoriolab to get idea of how many assemblers/refinery/plants/drills/etc you will need to fulfill your goal. And goal usually is SPM(science packs per minute).
Let me show by example. I have chosen 2.0 version but not space age to avoid recipes like iron from lava. Assembler machine 2, yellow belts and 20% mining bonus. 45 SPM for red/green/black/blue is enough for first run.
https://factoriolab.github.io/2.0/list? ... mi=20&v=11
What resources requirement we can see here.
Iron: 1200 per minute or 1.4 belts or 34 drills. 34% of it go to steel
Copper: 0.7 belt
Stone: 0.5 belt all of it to bricks
Oil: 241% or 1.5 refinery

Starting resource patches are enough to feed such factory.

Now lets remove everything and add purple science
https://factoriolab.github.io/2.0/list? ... mi=20&v=11
Iron: 2.7 belts. 79% of it go to steel
Copper: 1 belt
Stone: 0.6 belt. 57% to bricks, rest to produce rails
Oil: 428%

So to add purple science you need to increase your iron production five times and most of it will go to steel. Also you need to double stone and copper production and tripple oil. Purple science is very iron hungry.

What about yellow?
Iron: 1.7
Copper: 2.5
Stone: 0
Oil: 668%
Now we see almost reverse for iron/copper. Yellow science is copper hungry. Also it need even more oil.

Yellow and purple together https://factoriolab.github.io/2.0/list? ... mi=20&v=11
Iron: 4.3 belts or six time of what you have for red/green/black/blue
Copper: 3.5 belts or five times of what you have for red/green/black/blue
Stone: 0.6 belt almost same you have for red/green/black/blue
Oil: 1096% or 4.5 times of what you have for red/green/black/blue

Now you have an idea about amount of resources you will need in near future.
Also you can see how many assemblers you will need for every product. For example for circuits you will need 13/28/7 for green/red/blue. And this will produce 1170/210/30 green/red/blue per minute. So number of assemblers is not the same as number of product. Advanced products require much more assemblers.
You can open any product ans see how it is consumed. For example 51% of green circuits are going to blue one production, 36% to red one and only 13% for something else. Now you understand why factories need so many green circuits while you seldom see them in final recipes. If you open copper production you will see that most of it is going to copper wire and most of copper wire is going to green circuits. So most of your copper is going to circuits. Now it is time to explore iron consumption

[wizcreations]

That’s really helpful. Thanks for the examples!

How do I interpret the beacons and speed module info on FactorioLab?

[NineNine]

Advanced oil refining is tricky because you'll end up with different amounts of light oil, heavy oil, and petroleum gas. Some of those things you'll need more of at some times, and other times you'll need more of something else. Once you figure out one trick to dealing with those different levels of those three items, oil processing becomes straightforward. A hint is:

notice that there's a recipe to create solid fuel from each of those three products.

.

Do you typically replace all your coal consumption with solid fuel to keep your oil processing running?

Most definitely, yes. Because

1. Oil is more or less unlimited, and a lot of coal is needed for plastic as you progress.
2. You'll need to burn up that solid fuel (or store it) until you start building lots and lots of trains or rockets.

[NineNine]

I think I have a fundamental flaw in how I play. I think most of my initial setups for producing circuits, plastic, and other oil products are far bigger than I actually need, resulting in me using more resources than I can properly afford to use to feed those setups.

You don't have to overbuild everything in advance, or use any sort of calculations if you don't want to. Since you're just trying to get to a rocket, maybe consider just building the absolute minimum of what you need (ie: one assembler every time you create a new item). Then, as bottlenecks happen along the way, just build more of whatever you need. Some people love to plan everything out in advance, which is a perfectly fine way to play, but it sounds like you're getting hung up on that midway through the game. Maybe consider *not* doing any planning whatsoever, and just build whatever you need when you need it. I find that to be much simpler, and still effective.

The only solution I see to my current setup is to demolish and rebuild my materials so I can better prioritize their usage. That just sounds like such a massive process to have to do by hand that I almost wonder if I am better off starting over.

You never really need to start over. You don't even ever need to demolish anything unless you have some sort of space limitation. Just build more of what you need, and don't build what you don't need any more of at the moment. Again, I think you're getting hung up on the idea that you have to plan everything in advance.

[J-H]

^^ what he said. It doesn't have to be perfectly optimized.

[wizcreations]

You’re right on what I’m getting hung up on. That was also good advice to only build a small number of machines at a time and not expand until I see the need to expand.

[teatime11]

For me, I even drop this game by 2-3 years because of complex blue science and stuff after.
Then, One day I decide I want 8 Hours speed run archeivement. So I start learning them again. I try build stuff in peaceful mode. And, after days of trying to design early-mid game BP for a bit. I decide to try it. Fail like 6 times in the first run(Try to reload previous save and do better.) and many more later. I reload my save like 50 times. So, I took a break and watch some guide and finished the game in 7. I know it wasn't fast. but damn it felt good.

Here's the some guideline if you want
Burner Stage(After Crashland):
Burner Miner+Furnace
Starter red science and coal powered
Prop down some radar about 6-8 of them

Red to Green Science:
Clear out nearby nest, use shortgun to attack nest and craft light amour, Some fish.
Hand feed some gear, copper wire, circuit belt to kick start thing.
Make 2 furnace stack about 12x2 have to leave space for 4-6, 24x2 stack 1 copper 2 iron 1 steel or 2 copper 3 iron 1 steel.
Make steel buffer chest grab them if you need it.
Expand your Boiler/Steam power about 20 steam in total.
Start the Main bus I use 4 belt to start then branch up to 12 in the end.
Put down about 2 set of green circuit assembler 4 in totals
Set up Red- Green Science and put down about 30 labs.
Make some Buffer for belt & Inserter with this
Make some Solar panel buffer need about 300-600 of them.
(You can set up a mall but it could take extra 1 hour or 2 to fix things up so depend on what you want)

Prep Blue Science:
Expand Furnace stack to 1 copper 2 iron 1/2 steel(12x2 stack for now)
Research into Car and Military Science setup quick Military science and buffer your negotiation tools.
Find Oil, if you look and your map preview you should remember where the oil are. Some of it should show up if you have many radar on the base.
Clear biter nest around your base. I used 1 solar panel 1 power pole 1 rader and put them on the nest that just clear and area around base to have clear picture of what going on.
Drag power pole to oil patch and start pumping then drag underground pipe back to base.
Start oil production 4 refinery for now expand up to 9 later make sure to lay down piping for Advance oil later.
Make 2 sulfur& 8 chem plant for plastic for future.

Blue Science:
Expand funace stack to 1+ copper 2+ iron 1 steel (2x24)
Make some buffer chest for red belt need around 3000
Put down all your solar panel that you keep.
Expand Green Circuit to about 4-6 set (8-12 Assembler) for upcoming red science.
Make Motor, Red Science and Wire all of them into blue science (I used 10 Moter 24 Red and 10 Blue Science Assembler.)
Green module and buffer chest for it. Green can decrease pollution cause by miner by 8 from 10>2 which make then less likely to be attacked
First Blue Tech Advance oil processing. then power armour Roboport, Legs.
Expand oil product need 4 light oil cracking 4 Heavy oil cracking set some circuit condition so it wouldn't overuse them.
Set up 1 Lubricant, 1 Surfuric Acid, 2 Battery for prep yellow science. I like to put battery and Sulfur onto the same belt on the bus.
Find more mining spot for iron&copper remember to put 3 green module to each miner and drag 1 full red belt of input back for each resource.
Make around 22 prod module2 for Expensive Yellow-Purple Science

Purple& Yellow Science:
By now upgrade Furnace stack to red and change stone furnance all stacks to Steel Furnace
You should have a bit of roboport, use Blueprint to prop down some build
By side of your motor production should have space for electric motor and next flying robot frame for bots.
Blue circuit stack about 8-12
LDS about 16-20
Yellow Science &Purple Science (about 45 SPM each should do) Put prod module2 into these. can also put beacon into it if your power can afford them

Rocket
Make Rocket Fuel (if your oil input is lacking you might have to find richer oil patch (It should sum up to 1200-1600%) or do Oil Liquefaction
Buffer for Rocket part(LDS Blue circuit Rocket Fuel)
Prep Rocket Building mat.(1000 steel , 1000 concrete(Might need to hand feed them), 200 Electric Motor, 200 BLue circuit, 100 pipe) and around 3 beacons.
Use everything to lanch the rocket.

Tips:
Turrent isn't nessary for default setting as you can attack nest before it came into pollution cloud
Tank, Personal laser isn't need just granade and piercing mag/Normal shot gun and fish with 3-4 damage tech is good to go.
Red circuit and LDS are really copper devourer, you can buffer green circuit for red and some copper to hand feed if you want more speed
Don't overdo on bots, 1 bot craft is arond 1.2 science less. you need around 1000-1200 to finish the game? so if you craft 300 bot you get 360-400 science less.
Them legs are good equip those sexy legs.

[wizcreations]

That’s super helpful! Thanks!!

[angramania]

Red circuit and LDS are really copper devourer, you can buffer green circuit for red and some copper to hand feed if you want more speed

Blue not red. One red requires only 5 copper. One blue is 40(and 30 of it directly from green). LDS is 20.

Don't overdo on bots, 1 bot craft is arond 1.2 science less. you need around 1000-1200 to finish the game? so if you craft 300 bot you get 360-400 science less.

One construction bot costs 12.5 copper and 23.8 iron. One yellow pack costs 49.9 copper and 33.4 iron. So bot at least two times cheaper. But more important thing is, with more bots you can grow production faster.

[teatime11]

Red circuit and LDS are really copper devourer, you can buffer green circuit for red and some copper to hand feed if you want more speed

Blue not red. One red requires only 5 copper. One blue is 40(and 30 of it directly from green). LDS is 20.

Don't overdo on bots, 1 bot craft is around 1.2 science less. you need around 1000-1200 to finish the game? so if you craft 300 bot you get 360-400 science less.

One construction bot costs 12.5 copper and 23.8 iron. One yellow pack costs 49.9 copper and 33.4 iron. So bot at least two times cheaper. But more important thing is, with more bots you can grow production faster.

That's right. Still I wouldn't recommend doing a lot of them maybe 150-300 bot and 50-100 logistic will do. Problem is It gonna require a lot of power untill you get to nuclear power or have a really big accumulator and solar block. Which cost more material to start up.

Expanding power with steam is also solution but biter not gonna like them. Harder to predict where pollution would spread if you expand too much. At that point you'll need a thick defense like a hundred flamethrowers and thousands of walls.

Generally, I was trying to encourage to build a scale down base just to finish the game. So less polution less complex and less biter. But, what you mentioned are all correct.