Profile picture
3D Rotating Puppy @AriaSalvatrice
, 149 tweets, 68 min read Read on Twitter
Hey friends, I'm gonna start & stream a new factorio game with every damn mod. Whether it's an ongoing thing depends if it holds my interest & yours. Gonna post progress & stream announcements in here. Please mute this thread if uninterested! Channel is twitch.tv/ariasalvatrice
I'm using a lot of mods I'm not familiar with that makes it way harder to finish the game, and I'll also be playing on peaceful mode, which I haven't done yet. Settings are mostly default, but it's so modded I'm not sure how the difficulty will be.
Fishing for a good spawn is not crucial since there's no enemies at all, so this is where we'll spend the next 827 hours
Done with the stream! I'm gonna delete stream announcements once I'm offline to keep the thread clean. Starting a new factory is a laborious process where you must babysit everything, so I make a point to automate everything the second I can, even if it's temporary & messy.
In the vanilla game, you make almost everything out of wood, coal, iron, copper, and oil. In a modded game, there's hundreds of resources, and they're rebalanced. I'll need a ton more wood for example, but now I can mass-produce it.
Progress in factorio means either working towards research by crafting science packs to feed your labs, or later by working towards sending a rocket by crafting its components. Many recipes were changed - and are often harder. Still, automating red science was painless.
But I found some circular dependencies (or possible bugs) due to mod interactions that are really hard to break. It seems my only way to make the most basic circuits entail..... growing alien mushrooms to feed a monster, to collect its piss (yes), to then turn into formica, to...
I do not like this much aesthetically and mechanically but I'll stick to it for a little while. In the next session I'll try to use the Black Market mod to have a non-cheaty way to work around circular dependencies: sell product for currencies to buy what I can't make yet.
Anyway, thanks for checking out the stream if you did. If you didn't, here's an endorsement of it from @snus_kin
I spent most of the play time in my thinking emoji pose trying to figure it all out. Also, if you don't care about Factorio at all please mute this thread, I'll keep using it when I do more things.
After some reading around, I've confirmed the level of absurdity of the py suite of mods is supposed to be this high - circular dependencies impossible to bust without hours of handcrafting.
I knew the mod was "super hard" but spending hours making things by hand you can't automate isn't fun or interesting - next stream will be a fresh start without those mods. We'll still deal with super complex production chains like this:
I started another base off-stream, because the first two hours of any game are too boring to watch. Like before I didn't re-roll too much and went with the first reasonable spawn (few cliffs, no terrain slowing me down like snow).
I know this set of mods better. The start is simple once you've done it once. Ramp up your iron production, make a daisy chain of burner mining drill to stockpile some coal, send out a vehicle to chop down trees, set up electricity and manually fed crafting machines.
After that, research nanobots to have an early game alternative to personal construction bots. They are construction bots that are super cheap to produce... but are consumable. It allowed me to copy-paste my iron setup for copper.
Way easier to make simple vehicles with those mods! Since I play without enemies, I made haulers, the most basic truck type, and sent them remotely to explore the world a bit, to see where the resources are hidden. The guaranteed resources at spawn don't last long.
I got started on the bootstrap base / mall. It's the part of the factory that builds more factory part, where I will go to fetch what I need when I run out. It will also make the science packs, used to unlock research and progress in the game. I'm using a simple 2-lane bus.
And that's where I stopped, next time I stream we'll get to skip this boring part.
Also I'm building every new design from scratch. Perfectly optimized designs for everything are a solved problem but it's no fun to just copy-paste someone else's solution.
More progress off-stream! I expanded the base to mine Rubyte and Bobmonium. The latter is the ore named after Bob. Science. Using the simplest setup possible, where it yields only Tin, but made a quick and dirty sorting facility (right) to get a bit of silicon (I'll need it soon)
Expanding often reveals design mistakes. For example, this circuit cannot cope with coal on both sides of the lane and jams (see how black and blue are mixed on one lane). I redesigned the circuit upstream and emptied the belts.
To feed my base in wood, I'm still dependent on chopping down trees with vehicles. It's not sustainable in the long term, so I'll automate this soon. That's why I collected silicon: I'll need it to make glass for greenhouses.
I automated the productions of circuits Mk.II. The vanilla game has 3 tier, but the modded one has 5 tiers, and this 2nd tier is quite difficult to make with the limited early game tools. So I did it full spaghetti. This is a bootstrap base that will be abandoned soon enough.
This is the "Mall" area. The function of this part of the factory is to produce more factory. Belts, inserters, assembly machines.... all the basic building blocks. It also produces the first two tiers of science packs, used for research.
In Factorio, resources are consumed by research, expanding the base, and working towards a spaceship launch. The goal of the bootstrap base: pump out science packs to research tech, and get started on a big project such as logistic bots, solar energy, or a rail transport system.
With a mall and science production automated, the progression is more open-ended. While the base can't pump out every part in the mall simultaneously, it's keeping up just fine if I just withdraw parts as I need them.
But first, to expand, I need more juice. I solved the problem by duplicating the coal power plant: we play on peaceful mode, so there's no enemies to trigger attacks based on pollution spread, and coal is abundant. No rush to move to solar or nuclear yet.
My fleet of trucks explored the nearby land. The world of Factorio is infinite, but the more you explore, the higher the CPU cost (Factorio FPS is CPU-bound), so I'm not sending them too far. For your convenience I labeled the resources. I have no idea what most are used for.
I figured I'd work on the lumber problem. Doing this with my mods requires a feedback loop kinda design. Prime the machines with wood once, then let them do their thing forever. Will my design work? I'll only know one hour from now.
While I wait, I'll start researching architecture, provided by the Factorissimo mod, allowing you to make mini-buildings you can enter to nest factories and production units. Playing on peaceful mode, space is not at a premium, but it helps organize things.
BTW, if you're following at home, do you think I'm explaining game mechanics....
For electricity, mid-term, I'll build wind turbines. In vanilla, the difficulty of solar is that there's none at night. With wind turbines, it's just random. Making them consumes a lot of resources, but it's at the end of the mall's bus (which isn't load-balanced), so that's ok.
The wood circuit jammed! After a revision, it works now. I'm using the circuit network to start and stop the belt in such a way that I will always be able to process wood pulp (byproduct of making wood) but always keep enough wood to keep the circuit running forever.
The starter saphirite patch, used for iron, is half depleted, so I programmed vehicles to mine another outpost. Coal is supplied to the depot. The hauler grabs coal and delivers it to the mining vehicles, takes their mined ores & brings it back to the depot, following waypoints.
Now that everything works, the bootstrap base is going to research everything that can be researched using red and green packs only. There's 8 tiers of science packs, and a few other research items. I'm using a mod to automate it, way too many to pick by hand.
You can find weird alien plants around the world. While they're a finite resource, I think you only need to grab a few of them to bootstrap farming.
An array of 128 wind turbines produces only an average of 1.6 megawatts. That's not significant, but it's good to have a backup source of power available at night. If the coal mines run out, the factory won't die, it will slow down instead. Easier to restart that way.
More updates - remember to mute the thread if you don't like those.

So, before we continue, for increased realism and immersion, I figured I should play as an anime.
The mod I'm using can convert sprites from RPG Maker. I'm using this classic online generator that is slated to become yet another casualty of Geocities dot JP closing down.
As is often the case, the starter factory consumes iron way faster than I can smelt it. I found myself forced to cordon off some circuits to prioritize the parts I need. Upgrading belts and smelters and injecting ore from the mine to the North helped a bit too.
There are increasingly complex ways to process ore with this mod suite. I'd like to skip level 1 to jump to 2 immediately. To do that, I'll require clay bricks to be able to build floatation cells.
Clay bricks are obtained by washing dirty water. The water goes through a series of washers. It's rather simple, but you must manage some byproducts: mud, saline water, and carbon dioxide. The latter two I just trash for now: in Factorio, buffering production is just a trap.
In fact, I intentionally produce carbon dioxide to make circuits, so I could re-use the dioxide made as a byproduct of making clay, but such a setup is unpredictable. Something going wrong at the clay factory would have the power to stop the circuits factory.
The starter patch of Saphirite, from which iron is extracted, will soon be depleted, and the mine using vehicles to the North isn't keeping up. I've changed it to a more traditional mine + truck depot. The truck is still used to transport the ore.
Some ore patches are infinite, but you mine them slower and you require sulfuric acid to mine them at all, hence the weird layout. The machines with a red dot cannot work without acid here. In the long run using a rail network for this will be preferable.
Getting started on a saphirite ore processing plant. I'll use Factorissimo: buildings I can enter that act as their own self-contained factories, that I can reconfigure easily. Inside the buildings, I do not need electricity poles, allowing compact setups.
I ended up with this setup. From this single input there's many different outputs: iron ore, copper ore, nickel ore, copper ore, blue geodes, crushed stone, slag, and sulfuric waste water. Geodes are refined to 6 gems : diamond, topaz, amethyst, emerald, sapphire, & ruby.
Aggressively discarding byproducts & surplus using the Black Market mod, conditional operation of inserters/belts, and overflow valves. I can then buy any item in the game for those coins. Buffering production is a trap unless done in moderation, it conceals upstream issues.
Meanwhile, a lot of iron from the first ore patch went into this wind farm. 1024 turbines in total. But notice that their production varies a lot depending whether it's windy. Green bumps on this graph means the coal plant has to kick in. Orange = solar panels from bio nurseries.
Two more byproducts from the sulfuric waste water: green algae, and sulfur. For now I'm storing them in a chest. Once the chest is full, the waste water will be safely dumped into the aquatic ecosystem of stupid-ass planet I'm ditching once I build my spaceship.
Iron ore is produced in a very predictable 2:1 ratio, while raw gems follow a probabilistic distribution. The next step is to smelt that stuff.
Now that I can extract iron ore, I want to produce some chloric ferride solution, I will need it rather soon. To make it, I need iron ore, and saline water. They're both produced far away from each other, so it's time to bring some logistic bots.
This modded early game logistic network is very underpowered, compared to endgame bots. They move about 4 times slower than I can walk, and the zones they can service are tiny and require many zone extenders. Here, NE and SW are linked by extenders.
The saline water I previously discarded can now be used to make ferric chloride. You can see a bot providing iron ore in the top-left. The solution is barreled: 1 chest of barrels holds almost as much fluid as a small stationary tank. Barrels are re-usable, so I limit production.
Next, smelting the ores to ingot. Iron and copper are very straightforward.
Silicon and nickel are more difficult, they require processing coal a little. Smelting silicon requires carbon, and smelting nickel requires carbon monoxide. Quite a spaghetti setup, but with Factorissimo I can just plop down a cleaner build later if I must increase capacity.
To be able to make some bronze (I'll need a bit, but not much, it's mostly useful to make ammo, which I don't need on peaceful mode), I need tin ingots. I found it easier to use logistics bots and a temporary setup at the relevant ore patch, for now.
This is what the full setup looks like, and that's just for one ore - but the most important one, as it produces iron. There's still a few steps to turn those ingots into useful metal plates!
Meanwhile, the wood production continues just fine, in fact I might already have produced enough wood to last the entire game, since I will probably rely on algae and synthetic wood a lot. And the 2nd saphirite patch won't last very long, so I avoid unnecessary research for now.
Next step is to turn those ingots to liquid. While there are 5 inputs (iron, copper, silicon, tin and nickel), there are two more outputs: steel and bronze. Induction furnace setups are a simple deal. I keep buffer tanks, but just enough to smooth out shortages, no stockpiles.
The last one, nickel, I just do outside, since I wouldn't fill a 3rd building. After that, casting machines finally produce our plates. In the vanilla game, it's much simpler: the ore goes straight from the mine to the furnace, becoming an exploitable product without processing.
And here's a zoomed view of the setup. Now the important question: how to make it faster? I have not calculated ratios, and just went with a reasonable amount of buildings. If a production line can't keep up, there's many ways to intervene without reconfiguring too much.
Of everything I'm producing, nickel plate is proving pretty useless. It's only used to make ammo for military research, I'm making those science packs with an alternative recipe. Only requires simple ingredients... but huge amounts of electricity, 8MW. And it's slow!
I'm going to run out of coal fast using this much juice, so I'm starting to mine another outpost. I'm using a more traditional train setup rather than using road vehicles again. Using the FARL mod, laying large sections of rail is rather quick.
Those cliffs are why I'm doing military research: I need to bomb them to expand my base further. Otherwise I'd be forced to route things around them.
And here's what the base look like right now, without the outposts. It will be time to ditch it soon! Once I can get started on a logistic network, the plan is to simply abandon it and move shop somewhere else.
Since the factory has been dumping byproducts on the black market, I spent all that money on buying items I won't be able to produce for hours: a modular armor with a few modules (exoskeleton to walk faster, nightvision, personal roboports).
Videogames (polite reminder: mute the thread unless you want to read it)

So far, the wind farm is providing enough for power for the whole base most of the time. Building more costs me only iron, and production is secured.
After depleting my second saphirite mine, I'm getting started on the third one. I'm still transporting the ore back using an automated truck.
I decided to get started on processing natural gas and oil. It's a lot more complex than in vanilla, of course. Being able to put down note on items is a huge help to make sense of it. This starter build is not meant to produce much but to help me understand it all.
Here, I'm using the automated belt planner to move things through this mess of a starter base. I'll ditch the place soon, so spaghetti is acceptable.
Being now able to make the normal speed logistic and construction bots, and having personal roboports and nanobots, I get started on cleaning up the space around the base.
Unwanted patches of water get filled using landfills made of mud. The inverse operation, creating patches of water, is much more difficult.

I also use cliff explosives to get rid of obstacles. It's safe to destroy them in an active factory, there's no splash damage.
In another area, I explore farming. The recipes are complex, depend on exploring the world to some extent, have circular dependencies, and unpredictable yield. Clever builds will be required to avoid running out of seeds.
Also my internet barely works right now and it takes 10 attempts to post images. Here's the messy bootstrap base, with pipes and belts all over. It won't be dismantled but allowed to slowly spin down as it replaced with modular production outposts connected by trains and bots.
Almost every technology that can be researched with red and green science has been researched. I'll need to automate blue science to make further progress.

Random research is auto-queued when I have no better plans, there's so many things to research with those mods.
After making starter builds to understand washing and farming, I realized they could be used as a great source of plastic compared to oil. I decided to expand upon those small non-integrated exploratory builds
Here's what I came up with. Gonna explain how it works next:
First, dirty water is pumped and goes through five washing cycles. Clay, limestone and sand are extracted from it to make clay bricks. Those bricks are used to tile the floor of this factory module!
Mud is extracted in washing. More than I can use. For some reason I can't sell it on the black market (probably a bug in a mod), so instead I incinerate excess byproduct. I can't use as much as I produce. Mud is used to make landfills (to bridge water) & soil.
From the cleaned saline water, I grow brown algae, from which I create compost, lithium chloride, and alginic acid. I can't use the lithium chloride yet so it goes on the black market. Remember, buffering too much is a noob trap.
Seeds are produced using the mud and grown into swamp trees, producing bio plastic and lumber. The wood processed into wood pulp, then turned into charcoal, and processed into coke pellets.
The coke is used by this power plant - 200 Megawatts! I've set it up to sell power on the black market when I have over 30.000 pellets stored. I use a circuit network RS latch to build a bit of hysteresis into conditional operation of black market accumulators.
The farms produces wheaton (plants added by those mods are fictional if plausible). One problem: the yield is random, and to sustain the process, I need enough seeds. So I stop the belts going to processing unless I have > 1000 seeds extracted from the yield in stock.
From those, I extract synthesis gas (which I can't use yet), acetone gas, and fuel oil. The fuel oil is directly turned into electricity, it's a byproduct produced in too low amounts to exploit. What I really want out of the process is the acetone.
The acetone gas is used with the bio plastic from the arboretum to make liquid plastic, that is then molded into plastic bars. (I've blanked out an irrelevant part of the build for clarity)
The lime kiln produces a bit of carbon dioxide, enough to foster green algae growth. I still need to automate delivering crushed stone from another factory module - for now it's manual so I've left a note for myself. The algae are turned into cellulose fiber.
Using the alginic acid from the other algae, the brown ones, I'm processing it into cellulose pulp, which I refine into paper, which I then mold into wooden boards, used for electronics.
This factory module is self-sufficient! I can cut it off from the network and it generates its own power. The only input it needs is crushed stone. Later, I will connect it to the rest of the factory using train stations. For now the module is right next to the bootstrap base.
Things. The starter factory has served its purpose and is allowed to spin down: produce less and less as it starves for input. But I won't dismantle it yet, if I need it to produce something it's useful to have existing builds I can feed manually.
But if I just need a few hundreds of specific items made fast, I can also make a quick manually fed setup like this. It's dozens of times faster than crafting by hand.
Moving the base entails chopping thousands of trees. You won't manage to do that by punching them. I have 200 of the highest tier of construction bots on my personal roboport plus nanobots, and a module to mark for death any tree that enters my sight.
After deciding where the main base will live, I tile it with electricity, roboports, and floor covers. The world is internally divided in 32x32 chunks, so I use this grid for my build.
Everything of value the bootstrap base produced was moved to those warehouses, that have a whole lot more slots than my inventory. I can come here to resupply.
All the new modules will be linked via trains. Road vehicles have terrible pathfinding and tend to bump into things and won't be used further except for personal transportation. Here's the first mine linked to a nearby ore processing plant.
Mining this ore requires sulfuric acid. Processing the ore generates sulfuric waste water, which can be processed to acid. So I built a sulfur processing plant to the North, to integrate it all together.
In Factorio, train networks are a simple affair. You program a route and they travel it forever, and you arrange refueling at some stations. But I'm using the Logistic Train Network mod that makes it more interesting. Orders are generated automatically analyzing supply & demand.
In traditional engineering documentation fashion, the mod provides the most useless documentation possible, an example project with no commentary, and a link to a youtuber who made a quick entry-level 45 minutes video tutorial about the basic. So I figured it out the hard way.
There are two design decisions I must make now and can't change later: first, can train length be variable? It's doable but adds complexity to manage. But it would also allow for more compact builds. Still, without enemies, space is free, so compact builds aren't important.
Second, are the trains double-headed? The locomotive that's not in current use is dead weight making the train longer. But without a locomotive on both end, terminus-style train stations (left) are impossible - all must use a larger roll-on, roll off loop design (right).
I think I will standardize on a train length of six cars with a locomotive at both ends. It will make things simpler to build. I'll have to alter the current builds a little to make it work. I could just copy-paste builds from experts but screw that ctrl-v playstyle.
I'm not passing judgement on people who like to share builds! Being able to turn factory parts into easily copy-pasted text strings is really neat. But I prefer coming up with my own inefficient spaghetti builds rather than copying beautiful builds with perfect ratios.
And here's what the entire known world looks like for now. I've been putting a lot of map markers for your convenience and mine.
The train network went through a few revisions. It's difficult to alter it without disruption, but still doable. I went for a chunk-aligned grid, and trains with a single locomotive, making terminus stations impossible.
The network is comprised of these standard blueprints. The intersection I didn't create myself, I used a community blueprint. I'm not opposed to using them for things like that. An horizontal section can be converted to a T, and a T to a +, without having to destroy anything.
Stations are entered from the South. At the edge of the network, I had to cross the rail tracks and use left-hand drive to make it work on this mining outpost.
In addition of removing the rear locomotives, the jam-prone depots were redesigned.
Trips in the LTN are always: fetch resource, deliver it, go back to depot. Because a locomotive carries a lot more fuel than it'll use on a single trip, no need to deliver fuel at outposts.
Another function of the depots is to trash stray junk (for example if the destination station was clogged). If the train isn't empty, the circuit network is set to block the train at the depot until empty. Logistic bots take care of trashing the junk.
However, I don't have a good solution to trash fluids and gases. There's not one single standard (non-cheaty) way to dispose of them. Fluids go into the clarifier. Gases go into the flare stack. Most stuff can be sold on the black market. But not all. And wrong stuff = clogs.
I standardized on using merging chests in stations. It trivializes the challenge of balancing buffers, but the mods bring their own set of challenges completely different from Vanilla.
More revisions will still be needed, as I had a few deadlocks to solve manually.
If you're following at home you'll remember that I let the bootstrap base spin down & starve for input. It was finally time to dismantle it for parts. I posted it out of thread yesterday:
These are the saphirite processing plant, and the iron/steel/silicon smelting plant. The beauty of this setup is that those modules can live anywhere in the network and the trains just figure it out.
With so many different machines, machine tiers, research upgrades, and recipes, calculating perfect ratios is hard. You can whip out the calculator, but spamming machines until the belt are empty then removing unnecessary machines is a viable design technique too.
Here, I'm getting started on breeding puffers, a native lifeform added by mods that makes fart noises because of reasons. I always start with the simplest messiest build that will work to understand, then I try to make it more presentable.
This specific build must cope with variable yield and random yield. While the engine supports those kinds of random recipes, in the vanilla game there's only one probabilistic recipe: processing uranium, which yields a lot of Uranium-238 and 0.7% of Uranium-235.
More updates. There are 6 main ores in a modded "Angel+Bob" game, that need to be processed. There is not one single correct way to do it, as you can process them in multiple ways to obtain many products. Only one constant: you always start by crushing ores, so it's done on-site.
To get more out of the ores, I have to wash them first. This facility takes care of that. They also produce wastewater that can be reprocessed: sulfuric, chloric, nitric and fluoric waste water. Inside each factory building is an array of floatation cells.
The resulting ore chunks can be processed in multiple ways, so I do not process them further on-site - I made another sorting facility, that requests the chunks by train and sorts them. I can decide to assign a specific priority to each request depending on my production needs.
The ore is transported using the time-honored "sushi belt" design for maximum throughput, before being sorted. The various products are: iron, copper, silicon, nickel, aluminum, zinc, cobalt, tin, lead, silver, gold, and slag.
After that, I realized that I still needed some research for convenient smelting. Since I dismantled my bootstrap base, I had to make a temporary research park, and purchase science pack off the black market.
The black market automatically determines prices depending on recipe complexity, and it taxes your purchases. But even with the harshest tax (25%) I can afford a stupid amount of research packs easily. I only grabbed what I needed to produce them myself again however.
The endgame will take the required amount of science packs to an absurd scale, so using the market is unlikely to be sustainable. I'm using it to smooth out base building, not as a primary source of items.
I wanted to revise my train network depots: standardizing on single-locomotive trains means I could build them more compact, and I need to dispose of leftover fluids. So I experimented (using some "creative mode" cheat items) to figure something out...
You can dispose of the content of pipes with a clarifier (liquids), gas vent (gases), or flare stack (flammables), but here's the problem: they will accept the wrong product and clog. The game doesn't expect "drain pipe" builds.
Being a game limitation rather than an interesting problem, I unfortunately went with a cheat item in my build - a universal fluid void pump. Not that it makes a lot of difference than if I were using the proper buildings, or course.
And this is the revised train depot. If any fluid is found, it sounds an alarm, giving me a few seconds to analyze what fluid was left over. Destroying leftover fluids is a failsafe, so I want it to warn me if it happens.
Here's the new train depot (left), compared to the old one (right). Takes a lot less space.
In addition to modular production train blocks, I decided to have a main factory area to the North, fed by both bots and a main bus. The bus is being fed by trains to the East, and I reserve space South of that area to transport items in and out.
After much deliberation, these are the products I settled on for my bus. Going for a fully hybrid train/belts/logistics design hopefully means that this will be enough.
I'm only getting started on this area. It will serve both as a new mall area (produce more factory essentials) and as a science pack production facility. Every item has production limits set via circuit network.
Sometimes I prefer to build items on-site rather than via bus. For example, those basic circuit boards. 95% of them will be used as in basic electronic board. Those 2nd boards, I will produce off-site and ship via train. This building is for the remaining 5% of recipes.
For now, half the main bus is fed by products bought on the black market, and I've spent half the money earned until now. It's not sustainable, but it's OK, the point of the market isn't sustain but managing complexity & circular dependencies.
Video Games. As part of ditching the bootstrap factory, I had to settle on which tier of belts and inserters to use. The modded game is a lot more complex than vanilla (with only 3 tiers) in that regard. For belts, I figured the 3rd tier (blue) would be a good compromise.
For inserters, it's much more complex. I need to decide whether I prefer fast inserters or stack inserters. Which is best depends on my investment in research.
I made myself this reference to understand all the options available to me, and whether they're components of other recipes - mostly as an excuse to learn more about graphviz.
I decided to get started on a new mall & science fabrication area by the main bus I had built last time. It uses a lot of factory buildings, making it easy to optimize throughput later if needed.
All those buildings have varying degrees of compactness and spaghettiness.
Those buildings churn out both factory parts, and the five first tiers of science packs, used for further research.
Since they accept many different kinds of science packs, the research area is fed by logistic bots. You can get away with clever sushi belt builds with a bit of programming, however.
Being able to mass produce new tiers of science packs I had only produced or bought off the black market on a "as-needed" basis so far, I allowed auto-research to do its thing, and boy did it do its thing.
Predictably, the factory ran out both of iron plates and electricity. For now, the factory uses wind turbines, and burning charcoal obtained by planting trees.

Instead of tiling the land in solar panels, the standard Factorio mid-game solution, I decided to go nuclear.
I had not played with nuclear in factorio vanilla yet, and many aspects of it are the same as un-modded in my game, so I made an exploratory build. It's not ratio-efficient nor elegant but it works.

First, I mine pro-galena, a good source of uranium in my modded game.
It then goes through ore crushers, floatation cells, leaching plants, ore refineries, and ore sorting facilities, to be processed into various products: crushed stone, nitric waste water, red geodes, thorium, calcium sulfate, zinc, silver, lead, uranium.
Now for the fun part! To get use out of uranium, you need Uranium-238 and Uranium-235. The yield is random, and you only get 0.7% of U-235. But for nuclear power, you want 5% of it.
So I need to use an enrichment process christened after the lead programmer of the game. Science. This process requires a stockpile of U-235 as a catalyst to produce more U-235 out of U-238. I made a failsafe circuit to stop the belts and force enrichment if supplies get low.
Nuclear reactors get bonuses in an array, encouraging big setups. But since you must feed them, the biggest setups possible can only have 2 rows or 2 columns. In a 3x3 you can't feed the reactor in the center. Here's babby's first 2x2 reactor!
As you'd expect, my electricity problem are solved for a little while. By the time I feel the need to revise this setup, my production capabilities will probably be in the gigawatts, not megawatts.
Yes it's bullying to ask the bot to do a thread this big, and it will only grow bigger.
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to 3D Rotating Puppy
Profile picture

Get real-time email alerts when new unrolls are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member and get exclusive features!

Premium member ($30.00/year)

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!