Factorio Forums

www.factorio.com
https://forums.factorio.com/

Determined not to hate Gleba anymore

https://forums.factorio.com/viewtopic.php?t=133119

Page 1 of 1

Determined not to hate Gleba anymore

Posted: Thu Mar 19, 2026 4:46 am
by royce3
The plan is to build small modules that solve specific problems. I won't use bots of long-distance belts to transport quickly expiring items like nutrients, mash and jelly, etc. This will make troubleshooting problems easier. I will likely stick with an outer loop of belt surrounding each module so it's easy to the boundaries of each module.

Also I want to make sure all fruit gets processed so I don't run out of seeds, so at the end of the line I well process the fruit and throw them straight into a heating tower. I briefly considered biochambers for this end processing for the productivity boost but decided against it because it too could go wrong and processing with assemblers is better than letting them rot at the end of a belt.

Technically I could import/export spoilage, but since that's so closely bound to nutrients, it just seems easier to keep spoilage contained in each module. Seeds I will dump into an active provider chest. I have buffer chests to collect them until the farms are ready for them.


First blueprint is a simple one. I needed to craft some bioflux to unlock technologies:




Next is rocket fuel because I am going to need more power than solar can provide. Actually it's my third attempt at rocket fuel but I wasn't happy with the first two. I haven't actually started running this in production yet but tests seem promising. This fits well between substations and can be chained together.


03-19-2026, 12-45-26.png
03-19-2026, 12-45-26.png (1.23 MiB) Viewed 1065 times

Re: Determined not to hate Gleba anymore

Posted: Thu Mar 19, 2026 6:16 pm
by royce3
I seem to have iron bacteria stable. *EDIT* I found some bugs and decided that belting it out instead of chests was better. (Similar changes to copper are coming but it's not running at the moment)


03-21-2026, 01-29-01.png
03-21-2026, 01-29-01.png (1.39 MiB) Viewed 874 times

Re: Determined not to hate Gleba anymore

Posted: Thu Mar 19, 2026 6:17 pm
by royce3
and copper is almost the same:


03-20-2026, 02-17-46.png
03-20-2026, 02-17-46.png (1.13 MiB) Viewed 959 times

Re: Determined not to hate Gleba anymore

Posted: Fri Mar 20, 2026 5:26 pm
by royce3
plastic turned out to be a copy of rocket fuel with small tweaks:


03-21-2026, 01-26-10.png
03-21-2026, 01-26-10.png (1.07 MiB) Viewed 877 times

Re: Determined not to hate Gleba anymore

Posted: Fri Mar 20, 2026 5:32 pm
by royce3
lubricant was fun.


03-21-2026, 01-32-15.png
03-21-2026, 01-32-15.png (1.05 MiB) Viewed 872 times

Re: Determined not to hate Gleba anymore

Posted: Fri Mar 20, 2026 6:39 pm
by mmmPI
royce3 wrote: Thu Mar 19, 2026 6:17 pm and copper is almost the same:
I like the idea and the execution for the blueprints, it look nice and functionnal but on the picture the copper-making biochamber are missing the modules.

Also i think you could refine the process on this one, maybe it's nit-picking but seeing that you use circuits in other parts, i wanted to suggest to deactivate the top left biochamber, when the 2 others making copper are "working", to keep the more expensive recipe as a restart mechanism that doesn't run when not needed, given the layout i think it would only require adding a wire connecting them and setting conditions.

And also maybe you can use long handed filter inserter to remove the spoilage directly in the outer belt with other spoilage and nutrient to get rid of a splitter or simplify a bit the way you sort the output because atm the output inserter are filtered but they take both the spoilage and bacteria/copper, whereas if you have an filtered inserter just for bacteria/copper , it can output in a chest , and be on the side or the bottom

EDIT :


I made a blueprint to illustrate what i meant, because i thought my word could be confusing and i noticed another thing that i think is more important to fix :
V2.jpg
V2.jpg (273.72 KiB) Viewed 840 times
The rectangle shows something that isn't just a fancy detail, it is there to make the system robust, because in previous version, it is possible that the yumako at the end of the belt spoils, and if that happens the system is blocked , whereas with this addition, if it ever happens, then it will be dealt with.

( also added power pole and roboport )

Re: Determined not to hate Gleba anymore

Posted: Mon Mar 23, 2026 3:21 pm
by royce3
yes I eventually found and fixed those bugs too, but I prefer to grab spoilage from the same belt that fresh fruit is coming from. the reason for no power poles in the blueprint is because all my blueprints sit between a grid of robo ports and power stations. I also decided I wanted to buffer the bacteria inside the module to give more overall throughput when chaining multiple together. I'll post updated iron and copper blueprints later showing the changes. Also, I don't mind the uncaptured fruit spoiling at the end of the bus because it will just sit there forever, it's not like more fruit will keep spoiling, it doesn't jam up anything there.
mmmPI wrote: Fri Mar 20, 2026 6:39 pm
royce3 wrote: Thu Mar 19, 2026 6:17 pm and copper is almost the same:
I like the idea and the execution for the blueprints, it look nice and functionnal but on the picture the copper-making biochamber are missing the modules.

Also i think you could refine the process on this one, maybe it's nit-picking but seeing that you use circuits in other parts, i wanted to suggest to deactivate the top left biochamber, when the 2 others making copper are "working", to keep the more expensive recipe as a restart mechanism that doesn't run when not needed, given the layout i think it would only require adding a wire connecting them and setting conditions.

And also maybe you can use long handed filter inserter to remove the spoilage directly in the outer belt with other spoilage and nutrient to get rid of a splitter or simplify a bit the way you sort the output because atm the output inserter are filtered but they take both the spoilage and bacteria/copper, whereas if you have an filtered inserter just for bacteria/copper , it can output in a chest , and be on the side or the bottom

EDIT :


I made a blueprint to illustrate what i meant, because i thought my word could be confusing and i noticed another thing that i think is more important to fix :

V2.jpg

The rectangle shows something that isn't just a fancy detail, it is there to make the system robust, because in previous version, it is possible that the yumako at the end of the belt spoils, and if that happens the system is blocked , whereas with this addition, if it ever happens, then it will be dealt with.

( also added power pole and roboport )

Re: Determined not to hate Gleba anymore

Posted: Tue Mar 24, 2026 12:51 am
by royce3
When I took a closer look, I saw that the bacteria machines were not effectively offloading spoilage when iron/copper was backed up, so I had to fix that real quick. Here's the updated blueprints and a larger screenshot of iron showing the power stations. The new blueprint uses stack inserters since I have that unlocked now, but you can use bulk inserters there easily enough.

Here's Iron:



And Copper:



Here's my power grid blueprint for context:



03-24-2026, 08-46-51.png
03-24-2026, 08-46-51.png (3.39 MiB) Viewed 563 times
royce3 wrote: Mon Mar 23, 2026 3:21 pm yes I eventually found and fixed those bugs too, but I prefer to grab spoilage from the same belt that fresh fruit is coming from. the reason for no power poles in the blueprint is because all my blueprints sit between a grid of robo ports and power stations. I also decided I wanted to buffer the bacteria inside the module to give more overall throughput when chaining multiple together. I'll post updated iron and copper blueprints later showing the changes. Also, I don't mind the uncaptured fruit spoiling at the end of the bus because it will just sit there forever, it's not like more fruit will keep spoiling, it doesn't jam up anything there.
mmmPI wrote: Fri Mar 20, 2026 6:39 pm
royce3 wrote: Thu Mar 19, 2026 6:17 pm and copper is almost the same:
I like the idea and the execution for the blueprints, it look nice and functionnal but on the picture the copper-making biochamber are missing the modules.

Also i think you could refine the process on this one, maybe it's nit-picking but seeing that you use circuits in other parts, i wanted to suggest to deactivate the top left biochamber, when the 2 others making copper are "working", to keep the more expensive recipe as a restart mechanism that doesn't run when not needed, given the layout i think it would only require adding a wire connecting them and setting conditions.

And also maybe you can use long handed filter inserter to remove the spoilage directly in the outer belt with other spoilage and nutrient to get rid of a splitter or simplify a bit the way you sort the output because atm the output inserter are filtered but they take both the spoilage and bacteria/copper, whereas if you have an filtered inserter just for bacteria/copper , it can output in a chest , and be on the side or the bottom

EDIT :


I made a blueprint to illustrate what i meant, because i thought my word could be confusing and i noticed another thing that i think is more important to fix :

V2.jpg

The rectangle shows something that isn't just a fancy detail, it is there to make the system robust, because in previous version, it is possible that the yumako at the end of the belt spoils, and if that happens the system is blocked , whereas with this addition, if it ever happens, then it will be dealt with.

( also added power pole and roboport )

Re: Determined not to hate Gleba anymore

Posted: Tue Mar 24, 2026 3:31 am
by mmmPI
royce3 wrote: Tue Mar 24, 2026 12:51 am When I took a closer look, I saw that the bacteria machines were not effectively offloading spoilage when iron/copper was backed up, so I had to fix that real quick. Here's the updated blueprints and a larger screenshot of iron showing the power stations. The new blueprint uses stack inserters since I have that unlocked now, but you can use bulk inserters there easily enough.
I think the picture is missing some belts where i've put the red circle, so that the iron doesn't jam !
missing parts.png
missing parts.png (3.36 MiB) Viewed 514 times

Re: Determined not to hate Gleba anymore

Posted: Tue Mar 24, 2026 8:06 am
by royce3
No, belts aren't missing. There's an inserter below taking the bacteria/ore directly out of the splitter and putting it into the chest.
mmmPI wrote: Tue Mar 24, 2026 3:31 am
royce3 wrote: Tue Mar 24, 2026 12:51 am When I took a closer look, I saw that the bacteria machines were not effectively offloading spoilage when iron/copper was backed up, so I had to fix that real quick. Here's the updated blueprints and a larger screenshot of iron showing the power stations. The new blueprint uses stack inserters since I have that unlocked now, but you can use bulk inserters there easily enough.
I think the picture is missing some belts where i've put the red circle, so that the iron doesn't jam !

missing parts.png

Re: Determined not to hate Gleba anymore

Posted: Tue Mar 24, 2026 8:49 am
by Hurkyl
As food for thought....

Without modules, a biochamber consumes one nutrients every 4 seconds. With -80% energy costs, that jumps to 20 seconds.

The bioflux to nutrients recipe gives 40 nutrients per craft, for 800 seconds worth of biochamber activity. At 100% freshness and instant delivery times, that can feed 2.666 biochambers running continuously for five minutes.

Every other crafting cycle you get double that, because of productivity. 5 biochambers running continuously for five minutes can't even consume that much!

And not to mention there's another crafting output worth of nutrients sitting in the output buffer because the machine doesn't wait for things to empty out first.

---

Especially with such small blocks, I think it's really worth considering productivity modules rather than efficiency modules (with one speed module to offset the slowdown). IIRC, it reduces the overall fruit costs despite increasing nutrient consumption even if you weren't spoiling stacks of nutrients.

But more importantly, the increased nutrient cost might even be a benefit rather than a drawback, so that the bioflux to nutrients recipe isn't overachieving by such a large margin. I've found my designs seem to run more smoothly and reliably when nutrient costs are higher rather than lower, since they avoid these intermittent large stacks of spoilage.

There is a drawback in that it makes it a little trickier to jumpstart the system from the spoilage-to-nutrients recipe (and you may need higher throughput inserters to move nutrients from biochamber to belt), so definitely caveat emptor.

Re: Determined not to hate Gleba anymore

Posted: Tue Mar 24, 2026 10:08 am
by Tertius
Hurkyl wrote: Tue Mar 24, 2026 8:49 am As food for thought....

[...]
There is a drawback in that it makes it a little trickier to jumpstart the system from the spoilage-to-nutrients recipe (and you may need higher throughput inserters to move nutrients from biochamber to belt), so definitely caveat emptor.
I also recommend prod+speed modules. They reduce factory footprint. You need vastly less of everything. Machines, belts, inserters, item input. It's easier to build a beacon instead of more crafting machines, because it doesn't need any input/output except power: it just sits there and boosts.

I use productivity modules and beacons with speed modules all over my factory, which results in quite high nutrient demand. Advantage is that high throughput of nutrients result in nutrients being fresh everywhere and not spoil in the biochambers waiting to be consumed and not on the belt.

Jumpstarting this factory is a 2 step process in my base.
A spoilage-to-nutrients assembling machine is feeding one of the regular biochambers with the yumako mash recipe.
There is an additional biochamber with the nutrients from yumako mash recipe, which also gets its nutrients from the nutrients-from-spoilage assembler. The nutrients-from-yumako mash biochamber supplies the main nutrients belt.
This preliminary nutrient supply is enough to supply the 1st biochambers in a longer production line for producing bioflux (yumako mash, jelly, bioflux), so bioflux gets crafted. Bioflux from this machine is then being fed to the regular nutrients-from-bioflux machine and the main nutrient production kicks in.

In general I built a different factory approach:
- one dedicated bioflux production line. It also outputs its byproducts: nutrients as surplus from the nutrient supply and surplus yumako mash and surplus jelly.

with belts, this goes to:

- 2 dedicated bacteria production lines, getting bioflux from the bioflux production line and its byproducts, since this is what bacteria production needs.
- one dedicated agriculture science pack production line, using bioflux from the bioflux production line. It has its own bioflux-to-nutrient biochambers which are being jumpstarted by some of the surplus nutrients from the bioflux line.
- one production line for the rest of the necessary products such as lubricant, rocket fuel, sulfur, carbon, carbon fiber etc. It inputs the leftovers of all of the previous production lines, which turns out to be sufficient, and also has its own bioflux-to-nutrient biochamber for its local nutrient supply which is jumpstarted by surplus nutrients from the bioflux line.

About ratios: build 1 jelly biochamber for every 2 yumako mash biochambers.


Example for bioflux and the jumpstart machines (continue to the left, this is 1/4 of the line):
03-24-2026, 10-46-10.png
03-24-2026, 10-46-10.png (1.08 MiB) Viewed 437 times
Example for the bacteria production line:
03-24-2026, 10-48-26.png
03-24-2026, 10-48-26.png (903.09 KiB) Viewed 437 times
These bacteria lines are able to output 45/s (if upgraded to T3 prod modules). I designed Gleba as central mall because everything except stone is free. Not Vulcanus, because on Vulcanus oil products are limited because of coal dependency and the land is fragmented by lava pools, so circuit and module production as well for everything else is free and easy on Gleba but not on Vulcanus. This changes in the endgame, but not in the mid game.

Re: Determined not to hate Gleba anymore

Posted: Tue Mar 24, 2026 1:09 pm
by mmmPI
royce3 wrote: Tue Mar 24, 2026 8:06 am No, belts aren't missing. There's an inserter below taking the bacteria/ore directly out of the splitter and putting it into the chest.
That is correct, i misread the picture , it is functionnal on this part,my bad. There is still one thing that's puzzling me with the picture, and it's the spoilage there :
spoilageherewhy.jpg
spoilageherewhy.jpg (1016.23 KiB) Viewed 384 times
Because how ?

my guess is that it's left over from previous version without filter, and the chests are full of iron,that indicate the jellynut in the end belt will occasionnally spoil but it's taken care of now :)

Re: Determined not to hate Gleba anymore

Posted: Wed Mar 25, 2026 2:24 am
by royce3
mmmPI wrote: Tue Mar 24, 2026 1:09 pm
royce3 wrote: Tue Mar 24, 2026 8:06 am No, belts aren't missing. There's an inserter below taking the bacteria/ore directly out of the splitter and putting it into the chest.
That is correct, i misread the picture , it is functionnal on this part,my bad. There is still one thing that's puzzling me with the picture, and it's the spoilage there :

spoilageherewhy.jpg

Because how ?

my guess is that it's left over from previous version without filter, and the chests are full of iron,that indicate the jellynut in the end belt will occasionnally spoil but it's taken care of now :)
It is leftover. I used to output spoilage onto that belt which is why the splitter filters spoilage to the left, but the machine keeps taking in nutrients and producing spoilage when output is blocked, so I refactored it to not put spoilage onto that belt.

Re: Determined not to hate Gleba anymore

Posted: Wed Mar 25, 2026 2:33 am
by royce3
Good recommendations! I will employ all those things when I come back to Gleba with legendary resources. These blueprints are designed for bootstrapping Gleba when you don't have access to many tier 3 modules or quality. TBH, I don't care for messing too much with beacons or modules (except efficiency) until I start producing legendary. I will probably produce a new set of blueprints for the legendary rebuild. I know that my designs aren't very good with ratios and the order of nutrient delivery on startup is about as suboptimal as it can be. However, every resource on Gleba is infinite except stone as long as you can manage to feed most fruit through a biochamber. So I find I just don't care that ratios are perfect. My goal was making modules that were small and self contained and stable and I could easily copy/paste if I need to expand. Honestly, this is the first time that Gleba has been copy/paste friendly for me, so this is a huge win for me. All my previous attempts were incompatible with copy/paste because I always had a turn-around at the end for exporting spoilage (I should have embraced active provider chests sooner obviously). While yes it's easier to craft beacons, when you setup a mall crafting all the machines and parts you need, those are essentially infinite too. I literally have hundreds of biochambers sitting in a chest begging to be used, and that was before I even started agri science, now that is automatically taking an egg every time it's ready to craft another biochamber and that will continue until the biochamber supply chest is full.
Tertius wrote: Tue Mar 24, 2026 10:08 am
Hurkyl wrote: Tue Mar 24, 2026 8:49 am As food for thought....

[...]
There is a drawback in that it makes it a little trickier to jumpstart the system from the spoilage-to-nutrients recipe (and you may need higher throughput inserters to move nutrients from biochamber to belt), so definitely caveat emptor.
I also recommend prod+speed modules. They reduce factory footprint. You need vastly less of everything. Machines, belts, inserters, item input. It's easier to build a beacon instead of more crafting machines, because it doesn't need any input/output except power: it just sits there and boosts.

I use productivity modules and beacons with speed modules all over my factory, which results in quite high nutrient demand. Advantage is that high throughput of nutrients result in nutrients being fresh everywhere and not spoil in the biochambers waiting to be consumed and not on the belt.

Jumpstarting this factory is a 2 step process in my base.
A spoilage-to-nutrients assembling machine is feeding one of the regular biochambers with the yumako mash recipe.
There is an additional biochamber with the nutrients from yumako mash recipe, which also gets its nutrients from the nutrients-from-spoilage assembler. The nutrients-from-yumako mash biochamber supplies the main nutrients belt.
This preliminary nutrient supply is enough to supply the 1st biochambers in a longer production line for producing bioflux (yumako mash, jelly, bioflux), so bioflux gets crafted. Bioflux from this machine is then being fed to the regular nutrients-from-bioflux machine and the main nutrient production kicks in.

In general I built a different factory approach:
- one dedicated bioflux production line. It also outputs its byproducts: nutrients as surplus from the nutrient supply and surplus yumako mash and surplus jelly.

with belts, this goes to:

- 2 dedicated bacteria production lines, getting bioflux from the bioflux production line and its byproducts, since this is what bacteria production needs.
- one dedicated agriculture science pack production line, using bioflux from the bioflux production line. It has its own bioflux-to-nutrient biochambers which are being jumpstarted by some of the surplus nutrients from the bioflux line.
- one production line for the rest of the necessary products such as lubricant, rocket fuel, sulfur, carbon, carbon fiber etc. It inputs the leftovers of all of the previous production lines, which turns out to be sufficient, and also has its own bioflux-to-nutrient biochamber for its local nutrient supply which is jumpstarted by surplus nutrients from the bioflux line.

About ratios: build 1 jelly biochamber for every 2 yumako mash biochambers.


Example for bioflux and the jumpstart machines (continue to the left, this is 1/4 of the line):
03-24-2026, 10-46-10.png
Example for the bacteria production line:
03-24-2026, 10-48-26.png

These bacteria lines are able to output 45/s (if upgraded to T3 prod modules). I designed Gleba as central mall because everything except stone is free. Not Vulcanus, because on Vulcanus oil products are limited because of coal dependency and the land is fragmented by lava pools, so circuit and module production as well for everything else is free and easy on Gleba but not on Vulcanus. This changes in the endgame, but not in the mid game.

Re: Determined not to hate Gleba anymore

Posted: Wed Mar 25, 2026 5:27 am
by mmmPI
royce3 wrote: Wed Mar 25, 2026 2:24 am It is leftover. I used to output spoilage onto that belt which is why the splitter filters spoilage to the left, but the machine keeps taking in nutrients and producing spoilage when output is blocked, so I refactored it to not put spoilage onto that belt.
Ok thanks for the explanation and sorry again, even the power grid , i realize now it just fit, i don't know what i did wrong last time.

This time i tried to be more careful, i checked again the blueprint, not just the picture to avoid mistakes, and i think you can use a red wire to connect the biochamber making iron/bacteria and their nutrient inserter 3 pairs, then "read content" on the biochambers but just the "fuel", that is the nutrient, and so activate the inserter only when nutrient = 0, i'm sure it works, i made the blueprint to show what i mean, but i'm not sure it's super useful, nutrient spoils in the machine instead of on the belt, i guess it saves a little overtime if you have many machines that do not run, but you can't use the red wire for other things :)


royce3 wrote: Wed Mar 25, 2026 2:33 am Honestly, this is the first time that Gleba has been copy/paste friendly for me, so this is a huge win for me. All my previous attempts were incompatible with copy/paste because I always had a turn-around at the end for exporting spoilage
I literally have hundreds of biochambers sitting in a chest begging to be used, and that was before I even started agri science, now that is automatically taking an egg every time it's ready to craft another biochamber and that will continue until the biochamber supply chest is full.
This is pleasant to read :) I tend to use more robots and logisitc chests to make it copy/paste friendly usually my last Glebase had 0 belt on the whole planet, but now i want to try those or similar in my next game, it's x10 science, it needs something like this. Props to you for solving the challenge this way ! Also you may like having a safety chest of biochamber to recycle anyway, it's a good thing , in case of aciddent and you run out of eggs and you need to restart the process :)

Re: Determined not to hate Gleba anymore

Posted: Wed Mar 25, 2026 10:49 am
by Tertius
royce3 wrote: Wed Mar 25, 2026 2:33 am All my previous attempts were incompatible with copy/paste because I always had a turn-around at the end for exporting spoilage (I should have embraced active provider chests sooner obviously).
My rule of thumb is to have loops if a belt will carry spoiling items, and somewhere at the loop, no matter the exact location, there is a filtered inserter that moves spoilage into an active provider chest. Then, trash removal can be handled elsewhere in the logistic network. Sometimes it's sufficient to not use a loop and pick spoilage from the last belt piece, but not always, since there are different spoiling times so some items spoil on the last piece first, but other items might not and spoil somewhere along the belt.

While you design, you need to think about every single belt piece: "what happens if the belt stalls and all items on this belt piece spoil. Is it possible to remove all the spoilage or will the belt stay congested? What will happen if an inserter wants to grab an item from this belt piece, but everything is spoiled. Is this ok, or is it necessary to keep the belt ever-moving to avoid spoilage blocking this piece?"

Also as rule of thumb, low throughput kind of unwanted byproducts like seeds are extracted from the corresponding machine with a filtered inserter into an active provider chest.
Since spoiling items can spoil in crafting machines as well, there must be some kind of exit for this. It's either extracted by the usual output inserter onto the output belt, where it is taken off by another spoilage-filtered inserter somewhere else, or it is extracted by dedicated filtered inserters that also extract other byproducts. I extract seeds+spoilage from the mash/jelly biochambers this way.

So low throughput and unwanted items are moved to active provider chests and handled by the logistic network.
Normal throughput items such as nutrients are primarily moved by belts.
Finally, high throughput items such as jelly/yumako mash for bioflux or nutrients for pentapod egg breeding are moved by direct insertion. The corresponding production machine is located directly next to the consuming machine. Saves much space, as well as belts and a second inserter, but also needs some design effort for the layout. The items are always fresh, and the throughput of chest-to-chest movement is much faster than chest-to-belt, then belt-to-chest. Since with direct insertion there is usually some kind of overproduction, I have an additional inserter moving things out if the biochamber gets full. These are the byproducts I mentioned in my first post. This also keeps the biochambers reasonably empty, so the products don't start to spoil early just because there is more production than consumption.
All times are UTC
Page 1 of 1

Powered by phpBB® Forum Software © phpBB Limited