Factorio Forums

It's important to get out of the early game as fast as possible - every minute you spend without researching and producing steel is a minute longer that it will take to win. As I recall, I spent about 15 (in-game) minutes in this phase, but probably closer to 45 minutes real-time because I was pausing *a lot*. Basically, any time you need to think about what's next, or really any time you are not currently performing some action, you should hit pause.

Your first priority is coal. Start building an iron pick and run to the coal deposit. Drop your burner drill there outputting into the furnace, which makes an effective improvised chest. Dig one coal to get things started, then dig ~5-10 more coal by hand, and use that plus what's accumulated in the furnace to fuel the drill for a while. Take this time to run to the stone patch (in my game, it's half-hidden in grass, northwest of the big copper deposit.) Mine 15 stone so you can make 3 more furnaces, and construct those while running back to your drill. You should have enough coal now to mine and smelt a reasonable amount of iron, so go set that up. While that's mining, set up a furnace by a nearby copper deposit (doesn't have to be a big one) and hand-mine some copper to smelt. Your goal here is to get enough iron and copper to get basic electricity up. Don't forget that you'll need some trees for power poles, but prefer to chop trees that are either going to be in the way of your eventual base, or dead trees (which cut pollution much less than living trees).

Once you have basic electricity, you can start bringing out electric mining drills. Early-game tech requires way more iron than copper, so one copper mine going directly to one furnace is all you'll need for a while. In my early game I had two mining drills going to four furnaces, with extra ore hand-swapped from the two directly-connected furnaces to the spare ones. You could also set up a minimal belt+inserter setup here.

Once you replace your coal mining with an electric drill or two, run your burner drill out to the stone patch and have it feed a chest. You don't need a ton of stone, but you will need around 1800 over the course of the game, which can be covered by running 50 coal up to the burner drill at various points when you need stone.

Finally, build a research lab and lay it down somewhere convenient. Your initial research vials will be hand-crafted, but building the initial part of your lab area will be a high priority.

Up until this point, everything you've done will be pretty ad-hoc, designed to minimize the distance you have to run around to attend to everything. The only thing that ought to be in its final place is the offshore pump. But from this point on, the goal should be to build everything else only once, without needing to move it. Here's the overview of what you're building towards:

I'm going to cover both iron and copper mining here because they're so similar, but when playing the game you should focus on iron first. My build order was iron to about 1/3 capacity, then copper to ~1/4 capacity, then the rest of the iron and steel production, then the rest of the copper. But there was a lot of stuff interleaved in there - I think I didn't completely finish out the copper until after I had my oil up (among other things), because you only need the full capacity for running chip production. As you can see, they're very similar, with both having 24 furnaces per side, for a total of 48 furnaces. This is slightly more than what's needed to max out the 1600/min double-sided capacity of basic belts. The copper block uses 26 drills to meet this capacity, while iron requires 34 drills to additionally feed the steel production.

The funny blob of belts and splitters to the left of the drills is something I call a "decoupler," although I doubt I'm the first to invent it. It decouples the 4 output lanes from the 4 input lanes, so that each output is equally likely to come from any of the 4 input lanes. It operates by using a splitter first, where one of the output belts is lane-switched using an underground belt, and then recombined with the non-switched belt. (It's easiest to understand the copper version, which has less extra stuff around it.) This device isn't strictly necessary, and indeed I omitted it while building my initial production, only adding it back when going full-scale. But it helps a lot in making sure that all the drills get evenly loaded, and allows you to add new ones anywhere in the design without worrying about which lane they will go to.

The U-joints in the output belts exist so that both sides of the output belt get used equally by the inserters. Without them, the final inserters can't find enough space to drop their plates reliably.

The inserter and chest coming off the iron line aren't necessary, but it makes it easy to quickly grab iron when you need to refill your inventory. I limited the capacity of the chest to 5 stacks, although 8 might be a better target.

Lastly, you will note that the design here only uses low-tech: Small power poles, stone furnaces, basic inserters, etc. The stone furnaces are of particular note: I never upgrade to steel furnaces, because there's no reason to, since the stone furnaces already supply the full output I need and they have the same pollution footprint. And upgrading to electric furnaces would require substantial redesign and a substantial amount of steel for a modest reduction in pollution. This follows the overall theme of the design: Build once, re-design never.

In parallel with building up your iron production, you need to be building out your research. Red research is thankfully very easy to produce, but it's important to lay down your assemblers, labs, power poles, etc. in their correct positions now, so you don't have to move them later. I started out with the gear factory, its output belt, and the rightmost red research assembler, connected to the northwest lab. These initial assemblers started out as grey (basic) assemblers, but then I upgraded them in-place (a very useful thing Factorio lets you do) to blue assemblers as soon as I got the research. Initially you'll be hand-carrying copper plates to the assemblers, since you shouldn't have built any serious copper production yet. Once you get copper production going in the far north, run the long belt down down here to automate production. After that, you should build the first quarter of the electronic circuit line, in order to supply the inserter assembler and start on green research. Somewhere along the line, it will also make sense to finish building out the steam power - see the next section for the other relevant screenshot.

Once you have the red and green research sections finished, you'll want to get at least 12 labs, to maximize research speed. You only need the full complement of 20 when researching the final techs that take 60 seconds per research pack. Also, it's time now to finish off the iron production so that you build enough steel. If the steel doesn't get started soon enough, you won't have enough at the end and it will stall the rocket launch.

This is a good place to mention research order:

• Automation
• Logistics
• Electronics
• Automation 2
• Steel processing
• Optics
• Turret crafting (these last two are fillers until you get green science up)
• Oil processing
• Fluid handling
• Sulfur processing
• Battery
• Plastics
• Advanced Electronics
• Modules
• Efficiency Module
• Speed Module
• Productivity Module
• Other green filler techs (Solar Energy, Laser, Laser Turret, Engine, etc.) until your blue research pipeline is ready
• Advanced oil processing
• Automation 3
• The remainder of the prereqs and techs needed to build things (Concrete, Electric energy accumulators, Alien technology)
• Everything else that's directly required for the silo, in whatever order you choose.

You can research additional things if you like, but it will delay the launch. "Speed" techs such as lab research speed are actually counterproductive, since they take time to research but only decrease the number of labs you need, and the layout is already designed for the higher number of labs assuming you don't get the tech. This is also why Advanced Material Processing is not a priority (although you need it eventually for Concrete.) The only not-strictly-necessary techs I researched were:

• Advanced Oil Processing, because it improves oil production by so much
• Automation 3, for yellow assemblers in certain key locations
• Turrets, Laser, and Laser Turrets, for laser turret creep

The last of those is optional and somewhat costly. However, I find laser turret creep to be the fastest and most hassle-free method of getting the alien research needed. You could use standard turrets with regular bullets, or even AP bullets if you prefer, although beware the extra steel needed. Researching the car or tank is probably too expensive.

Once the iron block is finished, you will probably have the techs needed for the oil processing area. You can skip most of the chemical plants in the first pass - four of them are useless until you get cracking via Advanced Oil Processing, which is a ways down the road. The lubricant plant is only needed for electric engines, which also aren't important yet. However, do get the two solid fuel plants going now, since you need to accumulate a lot of solid fuel and it helps use up the light oil that you can't crack yet. Don't forget that you will need to periodically move solid fuel into he adjacent storage chests, so that the output chests don't fill up. You will need a total of 8600 solid fuel, which is a little less than the capacity of these 6 chests.

Do not build the storage chests that draw from the plastic transport line just yet. If you do, it will essentially "use up" your budget of petroleum, since you will fill up your storage of heavy oil and that will block your oil refineries. By letting the outputs of sulfur and plastic back up instead, you ensure that you'll have enough of both to research Advanced Oil Processing, and after that you can start cracking so that this constraint goes away.

The oil processing area requires lots of pipe-to-ground, which will eat up all your free pipes when you hand build them. Because of this, it helps to pre-build lots of pipes in your inventory before you start on the oil section, so you aren't wasting time waiting for pipes to build when you need to place them.

Once basic oil processing is built, it's time to build the rest of the infrastructure needed for blue research. Because of the number of circuits needed, this will also end up limited by your copper production, unless/until you build it out further. Don't build the assemblers that make modules and processing units yet - they will eat all the advanced circuits that you need to research Advanced Oil Processing. Once you get that tech, the chains come off and you'll need to do a number of things:

• Switch your oil factories to advanced processing
• Build the chem plants needed for cracking, or assign them to cracking if they were already built
• Build the storage chests that gather excess plastic
• Build the module and processing unit assemblers

You'll notice an extra assembler next to processing unit assemblers that's not doing anything. This is a spare assembler that's used for producing small runs of items that take too long to build by hand. For instance, you'll need 100 solar panels and 100 accumulators to build the satellite - those are produced here. It can also be used for supplemental efficiency module production.

By this point, most of the base should be built. If you're on the same pace that I was, this comes at about the 2:45 mark of played time. However, don't worry too much if you're slower - as long as your full base is completely functional by the 4:00 mark (halfway), you should be OK, assuming no disasters (i.e. alien attack or base power death spiral).

So, how do you prevent those disasters? In my winning run, I did experience a power death spiral at this point, because I had not yet built any solar, nor any efficiency modules, and I hadn't built up my coal production either. When I finished constructing the full copper capacity (one of the last things to get finished), the additional resources enabled the rest of the factory to ramp up production, and the power drain almost crippled me.

There are several solutions to this. One (which I recommend) is to start building solar sooner than I did. Around 9MW of solar will power the whole base in the endgame (when you've fully efficiency moduled everything), and it will get you over the hump until you can build all those efficiency modules. Another simple solution is to build more coal mines, although beware that the coal patch is surrounded by copper and so it's easy to get copper contamination in your coal outputs. Eventually efficiency modules on the mines and factories will reduce the power demand to a more reasonable level. (Build these in one of the two speed module factories; the other one should build speed modules to be used in making yellow assemblers.)

Theoretically you don't need any solar or efficiency modules at all; you could just build more steam and coal production. I'm worried that this would lead to alien attacks however, and this factory is ill-equipped to handle that. I don't think the pollution cloud would quite reach the initial alien bases (that would be disastrous, since there are so many at around the same distance), but the aliens form expansions closer in that end up in range of more modest pollution outputs. This is why it is safe to spike higher in pollution in the early game, but by the end you want to bring it back down.

Once you've solved your power issues and upgraded the necessary assemblers to yellow, it's time to focus on the module picture. Because the yellow assemblers are such power hogs, it's important that they get three efficiency modules. But it's also vital that they each get one productivity module. This serves two purposes: First, it reduces resource demand by creating "free" intermediate products. Without this, the copper production wouldn't be quite enough to supply chip production going full blast. It also reduces the assembler's speed, which is actually important because otherwise you will get shortages in the pipeline; the build is tuned around the assumption that the productivity modules are being used. Even though it's slow, I found it useful to hand-build the productivity modules (and, automatically, the required advanced circuits) and let my module assemblers churn out the efficiency and speed modules. Once you've got efficiency modules on everything that produces pollution (essentially everything except the labs), switch over to full speed module production, since that's what's needed for the rocket.

Speaking of the rocket, it's time to start monitoring your progress to ensure that everything will come together at the same time. The rocket nominally needs 100 rocket parts; however, you are going to load the silo with 4 productivity module 2s so that it only needs 81. This is the only time it makes sense to use level two modules in this scenario; they are so expensive that the payoff is almost never worth it. However, rocket parts are so costly that even a single one needs more circuits than a 2nd level module, not to mention the fuel and steel required.

Because of that, you'll need 8100 steel and solid fuel, 4050 plastic, and 810 processing units and speed modules to build the rocket. In addition, you need 300 extra processing units for the satellite and the silo itself. You'll also need 1000 extra steel and 500 extra plastic for the low density structures needed for the satellite, and 500 extra solid fuel for the same purpose. Finally, you'll need 2375 steel for blue research. This factory is designed to accumulate these resources over the course of the whole game, and then burn them all in a frantic 40-minute spurt at the end, once you unlock the Rocket Silo.

In my winning run, I noticed that I hadn't started on the processing units quite soon enough, so I ended up tweaking things slightly: I upgraded one speed module assembler and one processing unit assembler to yellow, and then disabled the other speed module assembler. This increased pressure on advanced circuits a bit, so my blue research was slightly short and I researched the silo a little later, but by doing this rebalancing I reduced the overall time slightly. That's a general model for how to tweak this build: If you increase in one area, be prepared to take a hit in the other areas.

In addition, there are some extra tasks that can't be forgotten, but if you start them in a timely fashion they'll be done before you need them:

• You need 200 electric engines for the silo. Engine production is slow, so start this out before the other two, and make sure you've supplied the chests with enough resources (especially electronic circuits) before you leave it unattended.
• You need 1000 concrete for the silo. This takes 1000 stone, which is easily supplied by the initial burner drill. As long as you don't leave this to the last minute, it should be fine.
• You need 260 alien artifacts for purple research. But you only need purple research for the final techs, so you can filibuster with other techs until your base is complete and leave the alien hunting until after.
• You need 5 radars, 100 accumulators, and 100 solar panels for the satellite. These can be built in parallel with the rocket if need be. In a pinch, you can even cannibalize your solar farm at the last minute.

In my planning game (where I designed and laid out this base on the same map), I got attacked from the north at around the 10:00 mark, so I hunt to the north to get rid of any alien expansions that might risk getting in my pollution plume.

If everything went according to plan, the last 3 hours are honestly a bit boring. You'll have completed all the supplies needed for the rocket and satellite, and verified that your resource production is on-track to be ready for when you finish research. The only thing left is to build the rocket staging area. The layout here is designed to process all the raw resources into rocket components in around 40 minutes. More assemblers could speed things up a bit, but you start running into diminishing returns, and after a certain point you lack belt capacity. (Steel is the limiting factor here.) However, if you're right against the wire it can be worth trying.

You'll note that there is one extra assembler for fuel, and two extra for low density structures. This is to account for the additional materials needed for the satellite. Pilfer low density structures and fuel out of either the buffering chests or the silo itself, until you have enough to build the satellite.

You'll be doing a fair bit of scurrying when rocket construction is going on, because you have to keep the resource chests filled. The solid fuel is the worst here, because it only stacks to 50 instead of 100, and thus runs out twice as fast. Steel can also be tricky - in my game it was the long pole (ever so slightly), and I had to keep the amount balanced across both chests as it ran out. A single fast inserter isn't enough to supply steel or fuel, so if one chest runs out then your production will drop.

You can do a lot of the prep-work up front: Building all the factories, loading the chests, making sure that all inserters are facing the right direction, etc. However, there are still a number of things that happen at once when you finally research Rocket Silo:

• You start building the silo itself. This takes 30 seconds, so do it first.
• Assign all the assemblers to their correct production. This is kind of annoying, which is another reason not to overbuild on assemblers.
• Place the silo and load it with 4 production module 2s.
• Connect the silo to the buffer chests with 2 inserters. Do this after loading the modules, so that you don't waste production!
• Load all the assemblers with efficiency modules; otherwise you'll brown out the factory at night.

Technically, you can take care of the last step ahead of time, and by hand-assembling modules you can do it without slowing down the rest of the factory. However, if you forget (or don't want to) there are many assemblers that you can scavenge modules from now that you're no longer researching. Besides the obvious candidates in the research chain, your demand for circuits will go down, meaning some circuit assemblers and cable factories will go idle. This extends up to the mining drills too.

Last but not least, don't forget to put the satellite in the rocket! You don't have enough resources to build another one in time. (Although auto-saves will probably let you rewind from this particular mistake).

Here's a screenshot of the coal production area, which wouldn't fit in the main post. I only needed 9 drills in the end, but at peak (before building sufficient solar) I needed more than this. This also shows how copper connects to the rest of the base.