Gleba has killed the game for me.

[Tinyboss]

With Gleba, though, no such luck. I have to monitor it, keep adapting ratios etc, and if the system collapses and spoilage takes over I have to spend ages rebooting the whole thing.

It's very possible, though not at all trivial to figure out. I spent a long time doing exactly what you're talking about. But eventually I came up with a small number of rules that allowed me to avoid lockups. My Gleba base has been stable and productive for 10+ game hours now.

I don't use logistics for much, but the "blue inserter filtered for spoilage aimed at a purple chest" is the real hero. Every building that holds anything spoilable needs one.

The few places where I use blue chests, I make sure "trash unrequested" is checked. Likewise, I rarely use provider chests, but where I do, I always go with purple, not red. Probably not as necessary as long as I'm requesting spoilage to be burned quickly enough. But I've gotten into the habit elsewhere, too.

And every belt end needs it. Mostly I try to loop belts back to their start, but there are several places where they have dead ends, and spoilage MUST be removed there. Add in a few places in the middle of looped belts, and you're pretty much good to go.

[schorsch_76]

About the rebooting the production at Gelba: It is possible to maintain a steady production.

In my opinion, if you look at the spoilage handling like a sewage system it makes sense to me. If the sewage system is stuck and cant take all stuff, all sewage producers drown in sewage and nothing is working anymore. My first sewage system was bot based. It worked for surprising long time until all bots were donut like hugging one roboport near the "entry to the sewage treatment plant". The bad reachable points are still handled by bots. My "sewage treatment plants" are 16 Biolabs for spoilage->nutient. The factor 10 spolage->1 Nutient is key for this.

.

After i took some time "off" from the factory, it feels better now ....

[BlueTemplar]

While a dumb working system is already quite rewarding to do, it's even more rewarding that it still has a lot of room of improvement : to maximize freshness : which becomes especially noticeable once you start exporting Agricultural science instead of consuming it there on the spot !

[quineotio]

Having now reached the end screen and thinking back, here's my thoughts about Gleba.

I still think the initial introduction is very poor. Too many concepts thrown at you at once without enough explanation. Spoilage, nutrients, soil types and harvestors are all new things, on top of the new recipes. And it's made even harder because these concepts are interrelated so you can't just learn one thing at a time.

After the initial hurdle I had a really fun time on Gleba, and I feel like I'm a better player because of it, and really enjoyed thinking through new concepts. For a few days. But now I reflect on it it feels tedious. It's a lot of work for... carbon fiber and science packs. And it's non trivial to expand production because there are the added steps of nutrients and spoilage mixed with just-in-time management just to make ore.

I think the basic mechanics are really fun, but utilized badly. I think it would be better if the Gleba production chain was entirely separate and new from the basic materials, so instead of harvesting fruit to process (while separating out seeds and making sure labs have nutrients and managing spoilage) so that I can make bacteria to kickstart bacteria production which I then need to let spoil into ore that I then smelt into plates... if instead there were new resources I was getting from the fruit. Keep the exact same production chain, but instead of producing basic resources I was producing different acids or gels or rare minerals or something.

In this sense, I think the Gleba mechanics would be better on Nauvis, because you could then integrate new resources into your factory. Fulgora has a similar problem, in that once you've solved the problem you only really get one new thing - holmium. The sorting mechanic is interesting, and I think it could be combined with the Gleba mechanic, so that you process the fruit and deal with spoilage, and then have a bunch of different non spoilable things you need to sort. And if this was all on Nauvis you wouldn't need the basic resources and could integrate these more advanced concepts more gradually into your already working factory without having to build everything from scratch 3-4 times (on each planet + space platforms).

I also hate the current Gleba stuff on Nauvis. Biter eggs needing bioflux is dumb. I don't need very much of it, but it's a tedious and punishing production chain to set up. Don't biters eat Nauvis food? Why bioflux and not just nutrients? I was also disappointed that there's still basically nothing to do with trees and only one type of tree. In this sense there was absolutely no need for Gleba when that mechanic could have easily been integrated into Nauvis, and more gradually. The fruit gives buffs when you eat it - this would have been a fantastic way of introducing spoilage early on - before you actually need anything from it. And then tree planting - why do I need to go to another planet to learn how to plant Nauvis trees? This could be green science tech, and introduce the harvestor early before anything critical relies on it.

I feel this way about the whole expansion. Vulcanus tech could have been on Nauvis, and the foundry is an interesting enough building without needing a new planet to manage - it still requires a refactoring of your production. And why can't I build them on Nauvis? The Fulgora lightning mechanic is trivial, and the islands aren't really interesting, just annoying. You only need one "ore" patch. Building space platforms is interesting... until you've done it, and then it's a burden every time you want to build a new one.

I think the expansion would be better if it was only two planets and most of the game was spent on Nauvis and the space platform, with the new mechanics being split between these two. The endgame could be arriving at the new planet. Because as it is, the new stuff is spread very thin and with a lot of "work" between these things, and so much of this work is for little reward. I barely touched any of my factories after building them, and they barely interacted with each other apart from science.

The post-game situation is not good. I basically realized yesterday that I was finished after automating Aquilo, and this was the least rewarding science to unlock. All it really gave me was the end ship, which I launched before I'd properly finished it because it was getting late. It took a couple of asteroids at the end, but made it first time. It feels like there's a bunch of content missing. What is the point of quantum processors? Personal fusion factor?

There are some awesome things in the expansion, and a lot of really well designed mechanics, but they don't mesh well as a whole. There a were multiple times when I became overwhelmed - notably the first space platform, Gleba and Aquilo. It's the oil problem x10, because at least with oil you have a working factory next to you. Gleba is conceptually overwhelming for little reward. If it was done as I suggested above the challenge could be added more gradually, with tiers of recipes. At the moment, it's really hard to get started and all you really get is carbon fiber, so the middle is fun but the beginning and end aren't.

I also became annoyed at having to build the four basic sciences. I think there should be productivity for them, or perhaps a recipe that takes red, green, blue and purple and some advanced material and makes 2x a new science pack (Nauvis science?) - so basically you get double production if you design a more complex setup. Doing it this way, you could eventually phase out the early science packs entirely in the end game with a more advanced recipe.

But in summary, I like the new mechanics, but I think they're introduced badly (too many at once with too many new dependencies) and the end products aren't rewarding enough, the progression isn't integrated well (a bunch of isolated tasks gated behind "work") and the planets require replicating a lot of already done work. You're doing a bunch of new stuff to end up in the same place.

I think the game would be better if the planets were smaller challenges, like Aquilo, but you still spent most of your time building one factory. That way you can still have unique challenges but you don't need to learn new production chains to build inserters...

One last thing... on quality... I initially liked it, but when it comes down to it it's just a waiting game. I think a better implementation would be to make quality items have different recipes - a reason to use the more advanced resources you'd get instead of iron ore and plastic. For example, you can make a "green" quality mining drill that has built in efficiency. Or a blue one that works faster, or orange with productivity. Or a "legendary" one with all these combined. Like imagine you get fruit, process it, separate out various minerals, combine it with iron to create "efficient" or "hard" or "lightweight" iron, which you can then use to craft higher quality machines.

Adding to that, I'm not convinced the relationship between efficiency modules and biochambers is healthy. I just included two mark 2s in every building as a matter of course, and I see not point in doing anything different. If there were tiers of recipes, perhaps module slots for biochambers could be something you unlock later, and the different modules effect the output percentages of the mixed products you get from the fruit.

I hope the devs keep working on the game, because there's a lot of untapped potential. A lot of good work was done, but I don't like the current implementation.

[BlueTemplar]

Non-Gleba spoilers, please ? (Including, possibly, for

the Nauvis things that require Gleba

.)

[quineotio]

Non-Gleba spoilers, please ? (Including, possibly, for

the Nauvis things that require Gleba

.)

You are annoying.

[EustaceCS]

We all are.

On-topic, I have to admit, I'm guilty.
FFFs jebaited me into thinking that three mid-game planets are equally challenging.
Gleba clearly isn't. It's a game on its own.
I... honestly, don't know how to feel about this.
From one point of view, even without spoilage mechanics risk management, which is introduced by egg related technologies and some slightly related lesser problems, makes Gleba experience too damn different from the rest of the game.
From another point of view, I'm thrilled by necessity to unpack and apply my college degree in sales (which included production & supply chain management for food and related goods, a little) just to beat Gleba with real-world science. Which doesn't quite look right - but hey! pre-2.0 Factorio also had some soft requirements of knowledge of real-world science!
From third point of view, the game should be relaxing or fun - and so far Gleba doesn't look like either.

At least we don't have to tackle six times more plant varieties in release version

[quineotio]

At least we don't have to tackle six times more plant varieties in release version

I don't think that would be a problem necessarily. As I said, I think the problem is the beginning and end - learning the mechanics and the rewards. I don't think six new plants is necessary though.

Imagine that initially you just process Yamako into mash, but later you get a recipe to process into mash and fiber (as a percentage), and then later it's mash, fiber and "Yamako minerals" - and then you process the Yamako minerals into different things. So there are tiers of complexity as the game progresses, kinda like how you start with basic oil and then advanced oil.

As it is now, you kind need to learn the entire planet before you get anything, and then things don't really progress past harvest > process > add bioflux > end product. And most of the end products are things you already have more easily on Nauvis. I understand the desire to have some new mechanic for ore so that it's not just repeating the exact same thing you've already done, but this is why I think it'd work better on Nauvis to begin with. I forgot to add power generation as another new thing. I think this could also be added to Nauvis - introduce it way earlier, before Nuclear. Then you'd already know how the turbines work when you get Nuclear.

Imagine the difference if you arrived on Gleba and you already understood how the power worked, how the tree harvester worked (and seeds), and how spoilage worked. They could also introduce the concept of different soils - but you'd need some reason to plant trees other than just aesthetics - a Nauvis agricultural tech. Like a single fruit that has one basic process to produce one basic thing and a basic nutrient/spoilage system. If you arrive on Gleba with all that information there'd be no problem.

[EustaceCS]

Imagine the difference if you arrived on Gleba and you already understood how the power worked, how the tree harvester worked (and seeds), and how spoilage worked.

I've mentioned my college (technically TVET - but... ugh... who cares) degree. I have some fieldwork experience in related areas.
+ I've read thoroughly EACH FFF since Space Age announcement.
And still Gleba perplexed me to the point where I've just landed for the first time, did reconnaissance, saved the game, closed the game, thought about what exactly to do next for two days, only then tried to do something.

And failed, lol.

It's not necessarily a bad thing.
Seasoned Factorio engineers wanted something completely different - and hell! they got something completely different!
And newcomers could use some extra variety after turning Nauvis inside out for 100 hours straight and standing still on Fulgora for 50 more hours
Still, as Vulcanus was moved out of the way to Aquilo - Gleba probably had to be moved closer to Aquilo.
Current-state Gleba is much closer to Aquilo's endgame challenges anyway.
Tripling Nauvis<->Gleba distance and slightly bumping asteroid belt's peak density will be enough to fend off unprepared people.

[WeirdConstructor]

As I stated before, I think the biggest part of overwhelm could be fixed with a little more explanation or a tutorial. I would've welcomed a little tutorial, showing me where to get started and progress for every planet, of course especially Gleba.

The tech tree is already a good start, it was helpful to go through it step by step. There are just a few too many things to wrap your head around at once. It broke my flow after I easily progressed through Fulgora and Vulcanus. Aquilo was a fun to puzzle too.

Maybe offer the player a safe space to experiment around with the Argicultural Tower and the Biochambers, like the current Tutorial for the main game? I know there is /editor, but it almost feels like an escape hatch/cheat one is not supposed to use - especially new players will not think about this option.

[quineotio]

And still Gleba perplexed me to the point where I've just landed for the first time, did reconnaissance, saved the game, closed the game, thought about what exactly to do next for two days, only then tried to do something.

I don't think that problem exists once you know how it works. It took 3 days for me to wrap my head around it - the first 2 being very frustrated. But once it clicked it was fun. It's an onboarding/tutorial problem.

It is harder than anything in the base game due to increased constraints, but conceptually it's actually not that complicated. But getting to that point of understanding is unnecessarily difficult.

Here's some things I figured out:

* use filters on ALL your inserters. Not only does it reduce the likelihood that something will go into the wrong place (like spoilage), it's way easier to see what's going on

* get used to using both sides of the belt for different things and priority splitters

* Use a single belt for nutrients and spoilage (because nutrients spoil quickly), and always put nutrients on the inside of the belt so that the inserters always have a place to output spoilage

* Use loops for everything that spoils - no dead ends

* always run your loops in the same direction - either clockwise or counterclockwise - makes it easier if you need to combine/crossover belts with splitters

* have an output for spoilage for EVERYTHING that can spoil, normally a priority splitter, because it will spoil

* try to only use one splitter per loop so that you can easily measure the contents with the circuit network

* don't transform anything into something that spoils faster until you need it, including only harvesting when needed. Circuit conditions and radars can communicate to agricultural towers

* use spoilage and an assembler to kickstart things automatically - use a circuit condition to only work when needed (i.e. that section doesn't have enough nutrients on a belt). You can store some spoilage for this - at least 200 but I recommend at least 600 to be safe.

* place this "kickstarter" assembler right before whatever your main nutrient producer is so that it gets the nutrients first

* have a dedicated spoilage "bus" lane to remove spoilage from all sections - easier than routing things from every area individually

* bioflux lasts a long time, so you can bus it - easier than producing in every place you need it

* everything that spoils quickly should be produced where it's needed

* you can have a main bus, but it should be a loop. You can measure how much is on it with the circuit network and only harvest when you need to.

* don't even try to manage excess wriggler eggs - just burn them immediately if no machine has picked them up. They are easy to multiply. You can decrease the liklihood of unwanted spawns by setting inserter hand size to 1 and only picking up if there's none in the machine.

* the order of inserters is important (also applies to space platforms). If the machine uses the output of itself, the input inserter should be after the output so that it can pick up it's own output.

* when placing soil, use a 3x3 grid and use a placed harvester as a guide on where to place.

* keep things as simple as possible

[quineotio]

Maybe offer the player a safe space to experiment around with the Argicultural Tower and the Biochambers, like the current Tutorial for the main game? I know there is /editor, but it almost feels like an escape hatch/cheat one is not supposed to use - especially new players will not think about this option.

I spent a few days in the editor. I actually think it should be integrated into the base game as a PDA or Sim computer - a way to test designs and make blueprints.

[BlueTemplar]

People were complaining about the removal of the Sandbox mode (/editor being hidden).

I guess we might get /editor also from a main menu button ('Sandbox' ??) once it's deemed ready enough, maybe for 2.1 ??

[EustaceCS]

Here's some things I figured out:

Radars... that's something new. I'll look into what these pieces of low grade high tech can do now... oh! They work like extra-long-range signal relays now, right? Daaaamn, I wish I knew this few days earlier...

As of Gleba tricks...
Loops are very overrated.
The only storage&distribution solution which is completely immune to clogging while providing acceptable thoroughput AND coverage is

stationary solitary cargo train wagon, with tanks serving as pricey yet very convenient substitute where either smaller footprint or full power or personal roboport is required.
Remember! Inserters can grab things out of tanks...

Actually, screw the former. The latter also gives the option to throw 90% of drone-dedicated power supply out of the window for at least small scale production. With an option to remove (and add back) whole sections of your base from logistics network in a jiffy. There're not many Uranium applications until lategame anyway...

(hmmmmm, for Gleba it helps with spoilage management, for Fulgora (and Aquilo) it helps with making tad more compact bases on its famous teatable sized islands... for Vulcanus it probably also does something useful... something is about to be nerfed sooooooooo much!)

[Khazul]

It irritated me to start with, but now I have a mostly belted base there. Spoilage isnt an issue in terms of clogging up the works now. Eggs hatching out of the blue is kind of funny, but a bunch of tesla turrets around the relavent bio chambers sorted that out. Now just hear the hatching noise followed by sizzle, splat now and then

I think it will be a hard time is you don't do Fulgora first. My main issue was to start with in not being sure what the pie icon meant (ie needing nutrients as 'power'). I turned up on gleba ready to do my usual-ish thing - ie productivity modules and speed modules and beacons etc - until I realised that gobbles up nutrients at a ridiculous rate.

Now I tend to have a beacon with power modules in and a couple of productivity and a couple of power modules in each. The huge nuclear power plant I bought along in the ship initially didnt end up having to provide much power so basically run everything on minimum nutrient 'power' and just stick down more factory tiles as needed.

If people are stuck with dealing with spoilage - use a sushi belt type setup for handling nutrients and spoilage. Use filtered inserters to explicitly direct what goes where and have spoilage collector inserters at the ends of belts to pick out spoilage to stop them blocking and transfer away (via active provider or on to the sushi belt).

All recipe builds end up following a similar build pattern for me and once I had that sorted out in terms of spoilage and nutrient handling it ceased to be a bother.

[BlueTemplar]

My main issue was to start with in not being sure what the pie icon meant (ie needing nutrients as 'power').

But while this isn't spelled out literally in the Gleba briefing, if you hover over [Entity : Biochamber], the popup says « Consumes Nutrients», how much Watts, and what Nutrients are accepted (there's only one kind in vanilla SA) !

A sushi belt setup mixing nutrients on the same lane as other ingredients seems like a recipe for disaster. (Not that it's impossible, but Gleba is hard enough...)

[quineotio]

My main issue was to start with in not being sure what the pie icon meant (ie needing nutrients as 'power').

But while this isn't spelled out literally in the Gleba briefing, if you hover over [Entity : Biochamber], the popup says « Consumes Nutrients», how much Watts, and what Nutrients are accepted (there's only one kind in vanilla SA) !

A sushi belt setup mixing nutrients on the same lane as other ingredients seems like a recipe for disaster. (Not that it's impossible, but Gleba is hard enough...)

Every recipe has at least 4 inputs/outputs, and some 5. Not easy to have a row of biolabs if you don't use both sides of the belt. You could use a lot of undergrounds I guess, but the more belts you use, the more outputs for spoilage you need. It's not difficult if you're consistent - the belt has nutrients on the inside, spoilage on the outside, and an output for spoilage.

[The Phoenixian]

Went Gleba first, and actually very much enjoyed it, even if I got the approach right as much by accident as by intent, in retrospect.

I have some thoughts on the principles at play, and advice for dealing with it. Quite a lot of thoughts, really.

Working with biochambers in a belt based factory was interesting. The extra input and output for nutrients and spoilage seems very clearly intended as a way to to add more challenge on just how many inserters are needed to supply a 3x3 building.

As far as principles of the place and tools for dealing with it goes...

-Everything must be used or sorted: Gleba's biggest overarching principle. If something is biologic, it's going to spoil, and the fact that it transforms, not the result, is the problem. If Gleba is the biologic planet, then the arteries of the factory must continue to flow. It's a bit like Fulgora in a way, but Fulgora feels a bit heavier on complexity of the the sorting to be done and Gleba is a bit heavier on the constancy and throughput side of it.

(Looping belts are a nice way to deal with that, but at the core it seems like they're really a tool for making things available to machines and granting the entire belt access to whatever sorting equipment is going on, be that a filtered splitter or otherwise. A good tool, but one of many.)

To a degree it feels like spoilage is it's own solution. Especially with heating towers in the mix. You don't even have to burn it "for" some purpose. Just using the towers as an item sink with no power generation is useful, especially when testing sections of builds.

Similarly, even before before rocket shipments start, Agricultural Science is very useful. Specifically, it's a way to transmute pentapod eggs into spoilage, which is easier to manage than hatchlings.


-Trees are storage.
Since everything spoils once it's been made, one of the best ways to store things is to just not make it in the first place. Picking fruit only when the supply belt was running low seemed to work quite well.
As a companion point to that, fruit is one of the most important items to use up. If it isn't nearly completely processed, then seed returns will be negative, which is undesirable. I didn't realize until much later that I had overbuilt my fruit-to-bioflux plant by a significant degree, and probably had a much easier time as a result of doing so.

Tangentially, Overbuilding processing facilities and running them slow also seems generally useful for dealing with science and eggs. As long as it's limited by the rate of production eggs don't even get particularly stale, let alone spoiled, unless the whole base broke. (Ten too many tesla turrets) I've even lost the entire bioprocessing sector to a pentapod attack once and had it cold start to full production before a single egg hatched.

-Chests are ideal for managing fermentation.
If spoilage is desired. Chests are a compact place to enable it. Storing items to spoil is a great way test out Bioflux production, for instance, and far more compact for bacteria transformation than a belt. If a chest contains multiple stacks of items, it immediately becomes a tool for sorting more spoiled from less.

If two chests contain the same spoilable item, then the ratio between their storage sizes can be made into a very rough ratio of their relative freshness. If one chest holds 2000 bioflux, and one holds 4000, then an inserter pulling the most spoiled from the first to the second will generally leave behind the freshest 1/3rd of items, assuming some care not to average out the freshness of the stacks in the first chest by inserting new items at the wrong time. (Useful if, say, the former is a storage chest or passive provider for rocket usage and the latter is a chest with no roboport access.)

- The tank is great for dealing with early problems.

The three most valuable damage types on Gleba are Electric, Fire, and Explosion. Of them, Gleba doesn't have access to local flamethrower fuel until after coal synthesis and at least calcite processing, electric weapons seem to be more utility/anti-swarm than DPS, and explosives have to be shipped in until rocket turret research.

All that said, explosives are incredibly strong against early pentapod enemies, and I've found it well worth the expense of shipping in. Not even uranium rounds, just sets of basic cannon shells. I haven't played with landmines on Gleba, but I would expect them to be similarly useful.

Just as much, I've found that bringing a tank with an equipment grid was a handy way to fix problems remotely if the base ever goes down or meets something it can't handle, up until spidertrons came online. Parking one just outside the bioprocessing plant saved me a lot of hassle once the stompers came knocking.

- Walls are almost pointless.

Not only is difficult to stop pentapods from going where they please, but it also feels like you barely even need to keep enemies away from the base. I've had stompers big and small walk straight over solar panels without a scratch after being angered by artillery and only start doing damage after getting right up next to the turret encampment. If they can't it eat it, and it isn't hurting them, they really don't seem to care. Once they find a target they rampage, by until they reach it they're very chill.

In fact, I suspect that the turrets defending my factory may well have caused more damage than they prevented, at least a couple times, by keeping pentapods angry longer or making them angry sooner.

- Stack inserters are better than they look. Even if they already looked good.

I did not realize how good stack inserters effectively multiplying belts would be, but Gleba feels very clearly built around them being made to free things up, between how much Jelly, Yumako mash, and Nutrients the machines tend to need. It rapidly went from filling the red belt I built the system around to having enough spare capacity that it felt empty.

[BlueTemplar]

My main issue was to start with in not being sure what the pie icon meant (ie needing nutrients as 'power').

But while this isn't spelled out literally in the Gleba briefing, if you hover over [Entity : Biochamber], the popup says « Consumes Nutrients», how much Watts, and what Nutrients are accepted (there's only one kind in vanilla SA) !

A sushi belt setup mixing nutrients on the same lane as other ingredients seems like a recipe for disaster. (Not that it's impossible, but Gleba is hard enough...)

Every recipe has at least 4 inputs/outputs, and some 5. Not easy to have a row of biolabs if you don't use both sides of the belt. You could use a lot of undergrounds I guess, but the more belts you use, the more outputs for spoilage you need. It's not difficult if you're consistent - the belt has nutrients on the inside, spoilage on the outside, and an output for spoilage.

Yes, you just described some of the many solutions without having to resort to a sushi belt, the easiest of them probably

to have nutrients on one lane of the belt, and a single ingredient on the other one, just like with burner furnaces at home.

.

but the more belts you use, the more outputs for spoilage you need

What do you mean by 'outputs' ? Just put it on a single spoilage-only belt.

[MechBFP]

I feel like Gleba is almost a litmus test for “fluid intelligence”. You need to be able to think ahead, and then think on the fly.
If you are the type of player who wants perfect ratios, no spoilage, and wants to test everything one step at a time, you will be in absolute pain because that is virtually impossible.

Don’t attempt that.