Factorio - Blog Posts

This is interesting coming from a post-expansion perspective. I do feel Space Age in some ways tried to approach this problem, and succeeded in some ways. Ultimately someone else pointed out very well that this isn’t a problem inherent to Factorio. It’s a problem that stems from how humans approach problem solving as a whole. We like to find one-size fits all solutions that we can apply over and over again. Ultimately what the expansion tried to do to solve this problem was to introduce new mechanics to act as new constraints. Spoilage is a constraint, constraining your throughput by time. So is Aquilo’s increase power draw for bots. Likewise Fulgora’s inverted crafting tree and Vulcanus’ lava do force you to rethink how you approach certain problems and they don’t reward a one-size fits all solution. Sure, a bot base WILL WORK for EVERY planet, but…

Spoilage will cause a lot of unnecessary bot work, and bots do NOT take freshness into account, which I think is intentional. Aquilo requires a lot of bots to get anything done in a timely fashion and they drain power like crazy too. Fulgora’s biproducts likewise introduce more jobs for bots. They end up becoming very unscalable on 3 of the worlds, and I think that’s a good thing. Ultimately it won’t stop you from just building huge bot bases on them, but it definitely works to discourage that. Each planet tends to have different optimal solutions, and we’re currently in a time where we’re free to explore those. Admittedly there’s still some of the old problem as the “LDS Shuffle” presents a new endgame homogeneity for solving the Legendary production problem, but something like it would evolve regardless. I admit… I also turned around and went to modded playthroughs after finishing the vanilla game for similar reasons. And I still am doing that, but now it’s less of to explore the fun of the base game, and more to explore new mechanics because I like seeing how people try to create their own challenges for mods. Like I’ve been meaning to do a playthrough of Ultracube myself

there's something kind of amazing about this. that you can take an obviously terrible design approach on purpose as a challenge and then on some level it turns out there's still a one-size-fits-all solution that is... maybe not 'optimal', who knows, but, highly optimised? the whole factory is in large part the same basic building block stamped one time after another. the design constraint prevents the already-existing standard solutions from working but then you find there's a new kind of standard solution, even more uniform.

and on some level you'd think that was an artefact of this run, but no. i've seen this guy's other challenge runs, like the beltless one and the all-burner one. they all end up with 'yeah turns out there's a standard solution i am just going to keep implementing over and over'.

i am reminded of what @definitelynotplanetfall was saying about how the main bus architecture and more broadly the factorio 'meta' of standard arrays for doing things means it's very easy to just take The One Tool That Solves The Problem and implement it and it feels... a bit like drudgery? idk i don't want to put words in their mouth that's the impression i got from what they were saying. and like at the time i pushed back a little because, like, i am having fun playing.

which i am, but. idk. there's something there. it seems easy sometimes to take the tools that simplify your life in this game a little too far and simplify the fun away. but at the same time it's also the case that i hate it when i grow used to a tool and it goes away, like when i started a vanilla playthrough for reasons a while back and noticed how much lacking simple things like module inserter and autodeconstruct was annoying me.

Factorio Space Age: Gleba

Since I recently reblogged a post about Gleba, I figured I should go into more depth about it. In a week or two when I graduate I’ll go into more detail about it, but I’ve probably spent the most time of the expansion on Gleba, exploring things like Quality, the circuit network changes, and sushi belts. I admit some of the tech I learned on Gleba ended up being essential for a later rebuild of Fulgora & stuff I learned on space platforms went to Aquilo, and then what I learned there got brought back to space.

Regardless I figured I could share some of the overall lessons I learned on Gleba & beyond during the DLC that helped me “master” Gleba.

-Identifying where spoilage & freshness matters

There’s a total of about 13 items that can spoil, and they spoil into one of six things: iron and copper ores, spoilage, and enemies. As you want iron/copper ores, we can ignore spoilage here, they become the thing you want typically so except in the production loop, this is a good thing. For spoilage, it’s an item, and you should generally assume anything that stores a spoilable item in it, is going to at some point have spoilage in it, and it will need to be removed. EVERYTHING, including things like biolabs. Lastly enemies, they don’t leave behind items so you don’t need to clean out the machines, but you probably don’t want them wandering around, so you likely want some defense to kill them if they show up. They can wreck havoc on things like space platforms or power plants if they get there, but generally they’re more a nuisance than a threat, so long as you don’t let a massive amount spoil at once… (Most I did was let 100 biter eggs spoil in a chest surrounded by lasers. Didn’t even notice it happened).

So clean up spoilage, and handle “Hazmats” (eggs & spawners) with military or disposal methods (pentapod eggs can be burned & biter eggs mulched into nutrients). Another tip for the Hazmats is to not store them in chests unless necessary (rocket silo loading), and to not put them in assemblers/biochambers unless they are the only missing item. I seriously recommend setting agricultural egg inserters to hand size 1 & to only insert if they have bioflux. (Wire the biochamber to the inserter, use read contents & enable/disable). It might slow down your science slightly but it does make it so you don’t need turrets by the science area.

Spoilage itself can be easily disposed of by either converting it inefficiently into nutrients, or burning it in a heating tower. Personally I use nutrient crafting as a spoilage upcycling system to produce high quality spoilage for efficiency modules, or high quality carbon for coal synthesis w/ asteroid mining for the matching sulfur. (I.e it’s a supplement for Legendary plastic for red circuits and LDS shuffling…)

As for where freshness matters, it actually only matters in a few places. Not all recipes do actually inherit their freshness from their parents (bacteria & pentapod eggs are top of my mind, but I think Fish also don’t), nor does freshness matter if the finished product isn’t spoilable. Ultimately freshness only really matters in the items directly connected to lime (agricultural) science as it’s the ONLY item in game that freshness impacts how useful it is. 5% fresh bioflux will feed a biter nest, as will 5% nutrients a biochamber & so on. Freshness only really matters if you need to move something or if it’s for lime science. So generally with that in mind you can send all your near rotten fruit and other spoilables for producing things like ore, rocket fuel, plastic, lubricant, sulfur or carbon fiber.

Another key idea is “shelf-stability”. You generally want to move raw fruit, bioflux & lime science around because they have long shelf lives. You don’t want to move jelly, mash, or nutrients around because of their short shelf life. It’s much easier to move jellynut, yumako, or bioflux instead and all of them are more space efficient to move as well.

-The Spores, Simplicity, & Quality triad

From what I found it is impossible to create a base that is simple, spore-efficient, & produces quality. At best you can do two, and I suspect it is genuinely impossible to do all three because of how they interact.

So I suppose I should define what I mean by these things. Spore efficiency is basically a measure of how much of the fruit products you make turn into spoilage. A more spore efficient base has less products rot. Why? Because the less products that rot, by definition, the more of your harvest WAS used for production rather than was wasted. While spoilage has its own uses, it isn’t ideal for most production in your base, unless you plan on mass producing only coal. Simplicity is how much of a headache setting everything up is. The more circuit conditions, belt priority shenanigans, and other complexities in the build, the less simple it is. And quality production, I mean large scale quality production, which usually relies on inter-step processing to roll up the products.

But wait you might be asking yourself, this implies it’s possible to build a quality base that’s easy without it being a major headache? Yes! Quality lime science is arguably one of the easiest sciences to produce in quality, only truly rivaled by the easy of quality space science in the late game! My first rocket silo of lime science was 1k rare science. This is because pentapod eggs are a catalytic recipe that can take quality modules, and so rolling up a high quality egg once is super easy, and then you just need to keep feeding it with high quality nutrients… which comes from bioflux, the other item you want to raise in quality! And you can even use a spoilage upcycler to supplement this to prevent the eggs from going off. I’ll show off a surprisingly easy to design base for producing rare quality science in the early Gleba game sometime later when I show off some Gleba designs.

However I do need to point out that the triad does inherently conflict. Trying to reduce spoilage amounts by simply reducing fruit in caused quality to stall. Trying to get quality up again caused it to become more complex, making a newer less-complex design required me to gut quality… you have to decide WHAT you value going in.

So for my MP base I decided I would cut quality, and focus on spore efficiency, as I wanted to produce the most science with the least spores, as I couldn’t rely on Tesla Turrets from Fulgora to protect me.

-Spore efficiency maximization

One of the best ways to actually improve spore efficiency is to start at the fruit production itself. Every second a raw fruit is waiting around, it is getting one step closer to rotting. Why harvest if you don’t need to? Keeping planting going without harvesting is simple if you keep in mind that agricultural towers prefer to plant first, then harvest. So if you wire the tower to anything, and set it to output inventory & only work when seeds > 0, it will only work when it has seeds in it, and will prefer planting first. Which means it will only harvest if there are no plantable spots and you put seeds in. Which means you can control harvesting by controlling when an inserter loads seeds in. So have the inserter only put seeds in when you need more fruit (you can use a circuit condition, like fruit less than 50 (one harvest) to determine when to start a harvest and load seeds in one at a time, if this is your only condition, you’ll probably produce 2-4 stacks at a time depending on your inserters).

Likewise… if you control when you process the raw fruit into jelly/mash then you can again further reduce spoilage. You can use a similar method to the harvest, but by turning off the biochambers for those lines. This reduces fruit usage, which will decrease tower usage & spore output… yet since you only produce the jelly/mash when needed, the assembly line shouldn’t actually slow down. What might happen though is that your power production dips because you’re not burning as much spoilage.

Well that’s a very easy fix. Gleba has the CHEAPEST rocket fuel recipe in the game, especially if you look at the fuel values of its ingredients. It is the ONLY power positive rocket fuel recipe in the game without productivity, and it has a default 50% bonus to it too! Literally no other rocket fuel recipe can get that bonus except the base recipe which requires exported biochambers to Nauvis! (Or Fulgora technically, but why would you do that? Oil is free there) Which is its own nightmare. So you can actually just burn rocket fuel for power in a heating tower! Which has a 250% fuel efficiency, meaning 100 Mj of chemical energy (one rocket fuel) becomes 250 MJ of electrical power! Excluding startup costs for the heating towers.

Well I’d recommend against burning all your rocket fuel because that’d just gobble it all up, but what you can do is measure the temperatures of your heating towers, and if they drop below a certain threshold (I recommend at least 600 degrees) to feed in rocket fuel.

Since I hooked up this failsafe to my power plant the only blackout I had was when I accidentally burned all my yumako seeds and stalled the entire factory, and it took almost an hour for it to begin to get close to a brownout, and it hadn’t when I found out the problem (I had an alarm if the power plant went critically cold (all towers below 600 degrees), so I could intervene before power goes out)

How you decide to reduce spoilage from here is up to you. I decided on my second run to just dead end belts and extract spoilage rather than run them all to the incinerator, so that lines could just pull half rotted mash/jelly for things like lube and ore. Only bioflux has a flowthrough section, and I overbuilt lime science & eggs so it never backs up there either (I’d much rather have rotting lime science than make half rotten lime science)

-Finally… solving the “How do I load my freshest items into a rocket?”

This is much easier than people think. It takes 2 chests, a logistics provider of some flavor (I use red) & a steel chest (wood/iron would work too). I then place the chests a tile apart, and have the inserter wired to the logistics chest. It’s set that if I have more than my desired storage amount (usually one rocket’s worth, sometimes two rockets) it grabs the MOST SPOILED item from the provider and puts it into the steel chest. This removes the item from the logistics network (and the rocket silo therefore) which will turn the inserter back off if the chest no longer has more than enough for the rocket launches requested & reduces the average spoil time of the chest. This is key. The individual items are still spoiling, I’m not managing to magically remove spoilage, but I am reducing the average spoil amount. When a rocket comes, it takes the items from the provider chest and it will gradually fill up again. Since only the freshest 1k items are typically available, this means I always load the freshest items I have. I could then feed in some items back from that “rotting chest” if I wanted to, but I find it’s more trouble than it’s worth, and I’d rather just produce a fresh 1k usually… I might play around with feeding it back in, but I only do this with science… extracted bioflux in this system gets fed into production elsewhere, instead of into a secondary chest. The whole point of the chest was just to act as a large storage vessel for composing science to spoilage.

Anyways, as someone who actually liked Gleba I talked about everything I can without getting into the specifics of like… how to build Gleba with bots, belts or trains… which I would love to cover at some point, because I do think there are too many content creators out there that don’t do Gleba justice… (Looking at Nilaus… I died inside when I saw he just plopped down a bot-base and a parameterized biochamber mall essentially. Dosh likewise also disappointed me on his OG Space Age run with his Gleba (import based) and Fulgora (bot based). I did love Doc Jade’s nightmare scrap train Fulgora though. He understood that the most fun can be had in the creativity of a solution, not necessarily the efficiency)

So as someone who recently finished Space Age, I have to say, I had to return to rebuild, unclog, or fix a bottleneck manually on Fulgora, Nauvis, & Gleba. I’ve had to do it the most on Gleba, but at no point did I find it so hopeless I had to reload my save. I did rebuild my entire base 3 times roughly, and I’m still not super happy about it, but I really did enjoy it, and it was one of my favorite planets. Spoilage was and is my favorite mechanic, and I’ve built 2 completely different Glebas on a singleplayer save, and a multiplayer save. Gleba is a beast and it rewards approaching it differently. All my Glebas have been belt-based for almost every “core” resource (My mall is bot-based and that’s it, it matches all my other bases except Aquilo, which has no bots at all, not even to load the one silo I have). I think overall I can agree with several of your points, like the terrain is very unclear, and I think a major part of it is due to how Gleba got trimmed down in scale over development. I suspect the reason why Gleba has roughly 4-6 different biomes (Green marsh, Red Marsh, Grey marsh, Rock highlands, yellow highlands & lakes) is due to it originally having more fruit lines.

I don’t know when your post is from, but one of the patches from a month or two ago MAJORLY fixed a huge gripe I’d had with Gleba. And it was BAD. I’d noticed while playing on my first save that the spore clouds spread far and fast, but not densely like they do on Nauvis, and came to the hypothesis that spores aren’t absorbed by terrain (which was a bug or balance change). Wube has since changed this, giving the terrain spore absorption, and since this change happened, I’ve found the same base produces far less spores on the map, and attacks massively scaled down.

On power… a lot of people seem to ignore that rocket fuel is a very viable fuel source for the heating towers on Gleba, and it’s very efficient to do so. I circuit control the heating towers to only input spoilage/spoilables if they’re below 950 degrees, and if they drop below 600, I flood my incinerator line with excess rocket fuel. Since I set this up, I’ve only had a blackout once, and it was after the whole base crashed due to accidentally incinerating all my seeds once, over an hour before it blacked out.

But yeah, I do agree you have to have most of your base built out, like a bare skeleton of the production line to actually hook it all up. Ironically, as I come from software engineering, it felt like having to do all the documentation first, before getting to “code”, because I had to have a roughly finished setup first, but… you can do that with roughly 6 biochambers to set up biochamber production, and lay it out in an easily expandable setup. I’ve tried a few different methods (all belt-based) so far for my Glebas and I still have ideas for refining and experimenting with new layouts. I’ve tried actually liked my second game’s Gleba better because I was more ambitious with setting up my infrastructure, and made it much more expandable/experimental. My OG base has a direct belt connection from farms to processing, uses quality, and basically has struggled ever since I got epic quality unlocked. Like I’ve redesigned it more than a few times and every time felt really proud. It challenges me in a way the overhaul mods never did before.

My MP base is built almost exclusively with just Gleba and Vulcanus tech. I do have turbo belts, foundries, big drills, and EM plants, but no recyclers, Tesla turrets, or artillery. (I’m technically waiting on our Fulgora player to export those, but we don’t have them researched and our Fulgora is out of rocket fuel, and I don’t want to touch someone else’s planet. Our Vulcanus player hasn’t set up a production line for artillery either, so I’m actually just defending with only gun turrets atm). Like the power plant is just a 3 heating tower setup, I only burn spoilage, eggs, carbon & rocket fuel (burning fruit is more fuel efficient, but I’m using a main-bus that cleans itself instead of a flowthrough design, the reason was to reduce spore production rather than maximize freshness like my other base. I only have a flowthrough section for bioflux, as I determined only bioflux & lime science freshness matters). I say this Gleba is more expandable, as if I wanted to, since I used a rail network, I could just add more farms to the network, move to a new highland area, design and setup a new base, and then start sending in the trains once it’s ready and either keep two bases, or decommission one. I’m legit thinking about designing different bases spread across the map for different products, so I can have them specialize how they handle the chain to take more or less advantage of the different build styles. It’s funny because now Gleba is the planet I look forwards to because I like trying to figure out better ways to handle it. (I am doing similar on Fulgora, trying to find ways to optimize my scrap production for science, modules, and holonium products simultaneously while reducing waste… it’s an interesting challenge, but it pales in comparison to Gleba.)

Ultimately, I do think Gleba is ironically the hardest planet in the game, by far. The only part of the game I have found can even compare with its difficulty is the shattered planet, and I do think more people should try to keep a more open mind about Gleba. I do really appreciate the people who say “I want to like Gleba” because it is worth liking, but it does have its flaws. Gleba tests you in ways that very little else does, and I admit some of that is frustrating. Especially since Gleba’s rewards are unfortunately not helpful outside of research or combat. The Biochamber & Biolab do not compete with the possibilities Foundries & Big Miners or Electromagnetic Plants & Recyclers offer to bases. Like I love the biolab, but… it doesn’t warrant a base redesign on any planet other than one build on Nauvis… the EM plant and Foundry offer worthy redesigns on EVERY planet, AND SPACE PLATFORMS. Meanwhile… outside of Gleba the agricultural tower and biochamber ONLY have uses on Nauvis, and BOTH are really niche… like 50% bonus to oil cracking, rocket fuel, & fish breeding and infinite wood… cool if you’re doing legendary wooden power poles, but if you’re not, completely useless. Likewise fish breeding is ONLY useful for quality spidertrons, which… is unfortunately a quantity beats quality area… there is no real time I can think of when one legendary spidertron would be better than 5 common ones. It just feels like bragging rights, and we already have mech armor for that.

The problem with Gleba

There's a game I'm a big ol' fan of and don't write about enough called Factorio. It's an interesting beast of a game. There's a lot of RTS DNA in it, and a lot of grand logistics puzzle/progammer-brain game. The main appeal is that as a the player, you are running around setting up a giant tangled mess of machines, conveyor belts, and little robot arms to produce large amounts of stuff to feed into research machines, teching up to more on more complex stuff, requiring you to scale up more and more until eventually hitting a win condition, but the more you expand and produce, the more the resulting polution causes your basically-Zerg neighbors to become larger and more aggressive. There's a really great inherent push and pull to this where if you're new to the game and just kinda struggling along, you generally have a lot more leeway on enemy aggression, and if you're really confidently rushing through (or just seriously overbuilding all your production), big deadly attacks roll in super early and you'll have to be way more aggressive about defenses.

Back in October, Factorio got an expansion, which I described while streaming it as the sort of expansion that's for "real Factorio sickos only." It makes the game significantly longer and more difficult, mainly in that normally, you advance through 5 flavors of science packs, each more of a challenge to produce at the rate you'd like, then head off into space. In the expansion, you can get into space with just the first 3 science flavors, but to hit the new victory condition, you need to be producing the original 5, plus an additional 5, one produced on orbital space platforms and the rest each coming from setting up bases on 4 new planets, each of which basically require you not only to start your big setup from scratch, but have their own resoruces, tech trees, and obstacles, meaning you end up playing 5 variations of the base game, simultaneously, and an extra logistical challenge in tying their science outputs together.

As a real Factorio sicko myself, I love this, for the most part. I have long since mastered the base game to the point where it's fairly trivial for me to get a thriving base going on what's now just the starting planet, and set up defenses that won't hold up INDEFINITELY without any further input from me (places to mine up the most basic resources do eventually run dry and one must push out into the map to set up new outposts now and then). So hitting a point where I have to just step away from my primary base and spend several hours setting things up on new planets is a cool change of pace.

And of the new planets, three of them are just fine. There's a volcanic planet where there's no water with which to set up the usual early game steam power nor the late game nuclear plants, nor can you mine for the iron and copper you need to produce basically everything in the game. The big challenge is figuring out the new tech tree and how to get the basics set up, then in realizing just how incredibly generous this new tech tree is with everything, and how much more efficiently you can set everything up, and the normal enemies that would be harassing you have no real equivalent. There ARE stupifyingly large and tough new enemies, but they won't come to you. They camp out around the map, guarding their personal territory, and requiring you to essentially handle a boss fight every time you need more territory to set up your stuff or harvest finite resources (but honestly, in practice, you'll need to expand in this just once, most likely).

Another planet's main hook is that literally the only resources to work with come from setting up your mining drills on the ruins of a long-dead civilization, pulling up an odd slurry of what in the base game are end-game resources. Complicated electronics, fuel, and superstructure materials just come out of the ground, and need to be broken down in recyclers for the actual base resources, which is just sort of hilarious. And the real puzzle is you have this mixed slurry of all these resources you need to sort out, then also deal with the incredibly unbalanced ratio, and find some way to keep the resource pipeline flowing and not getting gummed up with all that concrete and super advanced electronics you don't actually need that many of. And the final planet, only unlockable after mastering the rest, needs a good interplanetary logistics network as you need to important damn near everything from elsewhere.

All of this is great. Head to a new planet, spend a couple hours puzzling out it's quirks and how to set up a new rocket platform, its required inputs for perpetual rocket launches, and how to produce each planet's science flavor to send home. Then since it's been a few hours since you've checked on your main base, you head back, do some maintenance, maybe move some mines, maybe take a moment to make upgrades everywhere as each planet also has some infrastructural stuff that can't be made anywhere else, giving you better production structures and faster conveyor belts and so forth you might want to use everywhere. But then there's Gleba.

The gimmick of Gleba is it's the biological planet. There's no metal to work with (technically). No oil. Solar power doesn't even work particularly well. So like the volcano planet, you have to reinvent the wheel with everything using a new tech tree where you harvest two types of fruit, throw them into a series of goop-filled tanks powered by "nutrients" rather than electricity, and various combinations of byproducts your tanks spit out let you make literally everything you're ever going to need. In fact, a properly set up Gleba base becomes a perfect closed system, circulating seeds back to the two fruit farms for an infinite suppy, producing all the nutrients required to keep everything running, and enough surplus production of some form or another to feed into incinerators to provide electricity for the few things that still need it (basically just the inserters moving things from one tank to another).

And then there's the downsides. First, and this is a real serious problem for anyone dealing with this for the first time, Gleba has a real serious problem of "what the hell am I even looking at?" Everywhere else, there's pretty clear divisons between flat open ground, cliffs, some sort of liquid, and whatever useful resources you can harvest, without anything else really factoring in. And then here's Gleba.

I love the visual variety, but for comparison's sake, the base game looks like this:

It is very clear where the water is, it is very clear that there is a big patch of copper you can mine up. Meanwhile in these Gleba screenshots, you can't make out where the important resources are (a bit of a cheat because I didn't actually include the biomes where either of the plants that matter grow), and it's honestly quite hard to tell where the water is (I'm PRETTY SURE there's some in every screenshot, and probably a lot more than you'd think as it looks real different when very shallow)... oh and almost all water on Gleba is shallow to various degrees so you can't even go by what's walkable, you'll only really notice an area is flooded when you try to place stuff on it. It will probably take you quite some time before you can even successfully identify what's important, where it comes from, and where you have enough dry land to set your base up. And during that time you'll probably start dealing with the second complication.

Everything rots on Gleba. Well, almost everything. Stuff you build is fine, but the two important fruits, their intermediary peeled forms, the main intermediary material you make from mashing them together, the nutrients that power everything, the bacteria that you need to breed for your basic metal supplies, the one ingredient I haven't mentioned, and even the science packs you're eventually going to be exporting decay over time. Fresh picked fruit spoils in an hour. Peeled fruit and nutrients only last a few seconds. And once stuff rots, generally, you have this completely different item called spoilage, which is going to gum up all your automation by blocking conveyor belts or the input slots of machines and it can be pretty difficult to clear out.

Also as some things decay VERY quickly, any number of problems can cause something vital to spoil in transit, like say the nutrient supply to getting fruit initially processed, or the nutrients powering your production of nutrients, and everything's going to grind to a halt. Including the little inserters that move stuff to the burners providing power to those very inserters. So it's not at all uncommon when setting stuff up on Gleba that one tiny thing will be wrong, maybe as you cut off a belt to reroute it for a change in your overall design, everything rots, the whole base dies, and you have to go around clearing out rotted gunk from literally everything by hand, hand-produce a few nutrients from said rotted gunk, and slowly manually restart everything. Meanwhile we have the last issue to worry about.

Gleba is the one planet other than the one you start on with aggressive enemies to worry about. And there's a lot more to worry about from them. As the above sizzle real shows, they're significantly tougher on an individual level, but also, having these cool stretchy legs, they ignore all terrain. So you can't funnel them to choke points with walls, and they're likely to skim over water you can't build on in their approach. So you just sort of have to have a huge amount of standing firepower where they're likely to attack, which will only be your tree farms (and the path they need to take to them) which will be two very remote locations that are more or less completely flooded out... and your defenses most likely will require a lot of electricity, which is hard to get.

Also that last ingredient you have to worry about rotting? These things' eggs. Yeah both the buildings you use to produce everything on Gleba, and the science packs you eventually export, require the eggs of the local monsters to produce. Good news is, you really just need to risk your life attacking their nests to run off with a couple to start with, since you can make more eggs from eggs without too much trouble, but if one sits around for a few minutes without being processed, it hatches, and now there's a bunch of baby monsters freaking out in the middle of your base. And more importantly, after you clean that resulting mess up, you have to go on another super dangerous safari to get fresh eggs.

Now, individually, I actually love all this. There's some delightful cruelty and the puzzle of working out how to keep everything from rotting and clogging everything up in a fail-safe way is pretty neat. But putting it all together, there's two big things here that just feel real real bad.

First there's the pollution system that makes me love the base game so much. If I'm barely mining and producing stuff, I'm not causing a lot of pollution, so enemies aren't getting big and scary. If I make some huge mistake like, oh, running my whole base on coal power, scaling up a ton, and forgetting that I'm just plain not bringing enough coal in to sustain that, and my entire base de-powers and grinds to a halt, that's pretty bad, but I am producing zero pollution until I get it back online. If some small part of my factory stops working, because I'm massively overproducing something or I'm under-producing something, some machines are just going to stop doing anything until they get what they need, or have a place to dump their stuff, and even mines will stop mining if their output backs up.

Gleba... doesn't work that way. When anything goes wrong in any way, you go from having a ton of stuff you've produced to having a ton of spoilage. Or if you have some safety valves, you are suddenly tossing a massive overproduction of eggs or science or something straight into the furnace. But you're always going to be planting and harvesting the important plants (unless all your fruit rots on the line and there's no seeds to plant) whether you're really doing useful things with that fruit or not, and that's the one and only thing that generates "pollution" (officially it's spores that smell really delicious as a byproduct of harvesting). So catastrophes that end up being more of a full reset than a pause still leave you with jacked up pollution and much deadlier attacks, and that self-balancing difficulty just doesn't happen.

The other big problem, and this may be a bigger one, is you're really discouraged from tweaks and experimentation. You really are just sort of forced to fully design and deploy your entire Gleba base, with every emergency pressure valve and contingency, and the full production line to producing the final products you're shipping offworld before you even "plug it in" and start the actual plant harvesting. You can't really slowly build it up as you go (largely because you kinda get all your power by burning overproduction at the end), making a tiny change is going to make something start starving or backing up which can cause a disaster within seconds, and you either need to really really carefully manage ratios, or commit to massive overproduction and burning everything (spiking the difficulty).

So the first time you ever set up a base on Gleba, you're probably going to spiral into a failure state and need to reload from when you first landed there, maybe several times. But once you know what a functioning base looks like, either from your own trial and error or copying from someone else, you're going to have a nice little blueprint saved of this very nice compact efficient closed-loop base you can just stamp down on future play-throughs, hook up, and basically never have to look at again, ever. I was prompted to write this because I'm doing my second run of the expansion, got set up real quick here, and it's going to be a couple hours still before my defenses even get tested. Meanwhile I have basically all the Gleba research done already. There's no middle ground here between overwhelming and frustrating and a totally dull turn-key setup. Which is a huge shame!

Of course I'm also saying that before testing my defenses. The other inherent problem with Gleba is that from the moment you set foot on it, you do inherently have two planets with a steadily increasing difficulty modifier. Plus the science rots. So you are always going to have to divert SOME mental processing cycles to babysitting it at least a little bit even after you've solved the planet, even if it's just remembering to clean rotting science out of the labs on your starting planet here and there. And that really makes it into something you're still going to want to put off visiting for as long as possible even after playtest response to it being such a nightmare lead to the developers locking all sorts of cool researchable goodies behind it.

And then thing that really bothers me about all this is I can't really think of an easy fix for it. The closed loop where overproduction gets burnt is too conceptually foundational to really mess with. The cascading difficulty spike you could maybe fix by tying it to space launches and not basic production (rockets ignite methane in the air and freak the locals out)? Make solar work OK or take inserters out of the equation maybe by just letting belts feed directly into and out of the important machines here? If nothing else it'd certainly help if coastlines were more obviously marked in some way.

Also like... I'm not an outlier griping about this. Everyone hates Gleba. I just want to be the weird contrarian who thinks no, rotten planet is super rad, you should head there first even, get all that cool stuff to use elsewhere but... no there really are problems with it that are always gonna suck.

i think everyone should be at least a little bit fixated on a niche video game that came out like 10 years ago and has never been relevant enough to discuss with people you know in real life. for your health