#math-and-meta
1 messages Β· Page 562 of 1
Yea I know how it works. But meh its just annoying
Yup I maddenly color piping
Recycle has a downside - all breaks down when your primary material supply is uneven, and you didn't ensure the primary machines have priority for whatever resource you use.
[found it the hard way with aluminium plant - not enough bauxite, primary refineries underfed and make no output, secondary can't process because of no water.]
How much you making? O0. I only needed 16 for mine..
Checking the safety/accessibility of my factory makes me wonder - what are those tilted extensions on the integrated walkway grating? Are these for protection?? Looks more like a trip-bump for me and not likely you catch yourself on them if you fall. Is this real life thing or just flair/decor?? If they are meant as extension of walkway, why not just make the walkways wider?
Maximum. 3360 non fissle material
Did anyone tell You you are crazy yet?
Yes
Yes I am crazy
Probably more mascist
I see your point and how I misunderstood the meaning behind your first message 
Sorry for taking so long to understand you π
A sincere FYO that might interest you about the more practical conveniences (I found) of belt mixing: if you don't mind managing a few overflow belts (that can also work to refill your storage), those actually come really handy when dealing with manifacturers and most high end productions, where you can merge lots of machines together before needing to start worrying about throughput. The saving in space can then allow you for more flexible or tighter designs
With U5, would you replace (or wish to have done differently) the walkway-around-pipes part?
I'm not implying it looks bad, I like it, just curious if you had different involving U5 about that sort of design
Actually kinda had U5 in mind. I will be using soft clearance to full advantage and closing off the holes in the patheay pieces
Even tho kinda like the holes as is may just add single side rail pieces, dunno
What were you thinking of substituting them with, glass or foundations?
Nah just the catwalk pieces
can 2 nuclear power plants last on one mk2 belt? because despite perfect flow rate, I don't have enough water for both
eh theoretically, but it's finicky and depends on your fps. Best to just not combine them and give each a dedicated 300m^3/min pipe
will I run into the same issue though
no, especially if you don't run mk1 pipes
and you let the pipes and machines fill before you turn them on
a full pipe is a happy pipe
no
ok, why would someone to make 1200 iron ores per minute in only 1 machine, (i donΒ΄t know the name, iΒ΄m spanish)
Ore Miner*
@marble adder the max clock speed on pure mk3 miners is 162.5%
Please clarify
Max for conveyor saturation?
yes, that clock speed gives you exactly 780/min
Thanks
Guess, my 'santa' wish this year - would be a mk.6 conveyor
a mk3 miner with double output has been requested for years
but its always "nuuuu" response
They have solving that on their list, but won't be for a while they said
and im pretty sure they said that they cant get too much faster with the belts/pipes because it will make it more inaccurate the faster it gets
but if you "need" faster belts then theres probably a mod that adds mk6 belts
there is, and it does the max of 2000ppm
not sure how well it does it mind, never used it myself
Gonna try and build a bauxite train loop, any recommendations and/or tips before I start?
just dont do anything that will fuck up in U5
Hey.. so.. I'm new to using the Satisfactory Calculator website's production planning.. how can you change it to use alternate recipes ? (I don't mean for the final step, I mean for example if I were to use stitched iron plates for modular frames.)
The second one
it appears you can change alt recipes here
and other alt recipes can be picked in the options tab
Alright... is there any way to assign overclocking to specific machines by any chance?
I was to use the calculator for Heavy Encased Frames and only overclock the manufacturer to 250% and possibly the miners with a maximum of 200%
Maybe use https://u4.satisfactorytools.com/production, its the most useful and efficient calculator
That's the first link he got ^^ xD
Any "production node" can be divided into single machines with clock of your preference as long as the numbers add up.
Eg: 4.5 screws constructor = 4 constructors at 100 and 1 at 50% or 3 at 100, 1 at 150, or 2 at 225%....
Has anyone calculated if we were to make same amount of every craftable item per minute how many would the map allow?
define same amount of everything, not counting as materials going in bigger ones i assume? probably wont be many, as the high-end items are quite limiting
yikes
i just hope i wont come to a point where i will never be able to receive low tier items because high tier ones used them up
i reckon base materials are more ample then the fancy ones, so you would always have spare iron to make early stuff π
I doubt that... What kind of production are you aiming for? You can usually spare at least a few ores to make a small factory for all the basics
i am building up from the basics, the goal is as close to a full mk1 as possible for every non equipment item without clocking
Oof. Pasta and TPR are quite hard to get both to 60/min for starters
Even just 60 TPRs would not leave enough nitrogen or alu for 60 pressure conversion cubes or even fused frames maybe @stark bronze
well again this is just basics, i might find a new threshold for different tiers
i feel like a genius for doing 10 pasta/min and have drained half the desert from copper. hard to see how 60 would be even thinkable
30 pasta and 30 TRP would allow much more leisure in getting 60/min of every other thing
i might just end up using the most confortable math for everything regardless of item
and end up with one machine per item runs 
For anything before blending, I think 60/min each is feasible with plenty of spare
It's fused frames and pressure conversion cubes that would really strain the plan
60 heavy frames would be a big project, as well as 60 supercomputers
im thinking 30 for anything involving assemblers 20 for manufacturers 10 for blenders
even that might be too ambitious
i just want a pretty number 
Right 
Lucky me I'm aiming for 58 
||Iirc||
Numbers might turn out prettier if you allow for some 12s, 15s and such to slip in 
Depending on what output is convenient to get given the final machines' clocks
Eh, preferences π
I think one can still get however item/min he wants below 30/min of each. Excluding SE parts
The clock's precision allows for it within an error one shouldn't be able to notice "visually" (simply by watching the resulting output belts)
no it sometimes matters, i'm making 20 fuel rods. it's the first time in the game i had to build a load balancer to divide them between reactors. same would be with pasta. if you decide to make like 13 pasta, then good luck
on a bright side, it got me into all this load balancing stuff, which is pretty interesting
Naturally, but if one prioritizes numbers for the output I assumed he'd be willing to have the inputs overflow, as it's impossible to get 10, 20 or 30 min without getting very weird input numbers
A whole new world of possibilities

If I'm making a 1600m long 2 foundations wide skybridge, I'll need 400 foundations which is 2400 concrete, right ?
just take more with you, it's only 5 stacks
but yeah, your numbers should be correct
But this site says it's for Update 3- am I missing something
Yes. U4 toggle upper right corner
π΅βπ«
it says experimental but that label is just a little old
there is no reason to use the U3 version because i don't think U3 is even available to download anywhere
also Jace said there will be no recipe or manufacturing building changes in U5
Now I gotta learn how this one works
it's really simple and tbh the best tool that i've seen from this amazing community π
the Codex tab in the menu is really useful too
i use it more often than the wiki
So how can I overclock one machine in specific on this here flow, if at all
this flow shows how much production you need, so it if it says 1 manufacturer that means 1 machine at 100%
you can do 2 machines at 50% each if you want
sometimes it will say stuff like 12.33 constreuctors, so you gotta figure out for yourself how many actual buildings that is and how much to overclock them
you can do 10 at 100% plus one at 233%
I'm hoping to keep this as just one 250% manufacturer for heavy modular frames since I don't really need that many per minute, this is just a quick placement so I have frames to build aluminum stuff with
the best way to do that is do a simple setup with whatever number of output, then go to the codex and see how much output by default for the recipe that the tool chose for you and multiply that number by 2.5 so you can go back to the calculator and use the result for your output target
Yes, 7/m
yep, so go back to your production tab and switch this dropdown menu to "items/min" so you can type a number
I'm slightly worried that I might have to set up very poor aluminum first because these need up to 500 item/second conveyors unless I get weird
if you want it all on 1 conveyor, sure
I don't wanna have to come back to this in five seconds and fix all my spaghetti
π
Either one fast conveyor or many slower ones
I've heard about that using the water + ore recipes is better yield but more power - perhaps worth it for this ?
Such as wet concrete which I already intended on using
the best thing about this tool is it automatically saves your work so you can always come back to refresh your memory
Or you share it / save it
if you decide you only want to use that recipe for concrete you can go to the recipes tab on the left and untick the one you dont want
Important: The calculator prefers certain alts over base recipe if you have them enabled. So if you want to use base recipes or a certain alt but the calc doesnt show it, disable the other alts.
This was the plan
Seems solid
Careful that the power consumption doesn't increase linearly, it's exponential
200% clock takes about 4x the power
Same if you go below 100%
I was only going to overclock the manufacturer and possibly the miners
manufacturer at 250% is gonna take A LOT of power
Oooor you get 2 manufacturers
I have a 5000MW I'm not using
The manifacturers are the worse (power-wise) to OC
Best are constructors or assemblers, low power machines
A 200% OCed manifacturer is over 120MW 
nearly 240MW
Bruh 
compared to 55MW normally
I still have 5000 to use that I don't so I might as well
5000 is gonna get used up super fast later on
Yeah but you wont need it all for just one production / manu, will you?
Better save it for other production lines too
Eh, it's not like OC is a permanent solution, usually
Makes sense xD
I always end up OCing HMF manifacturers, sometime in the playthrough
It's a line I'll probably have on a breaker and turn it off once I've got like half a container's worth or so, same thing I did with motors
anyway, from your production chart, i dont see any problems. you can build two of these for 14 heavy modular frames/min with 2 overclocked manufacturers and if you need the power back later on you can just add 3 more manufacturers later to reduce the power consumption
I reckon 7/m will be simple enough considering I have a bunch of old outdated factories to go clean up and aluminum to start, it'll gather well enough
~250 MW to save on 3 machines
Worth it 
I'm still trying to battle this out in my head which is the best route to take
I got a sheet for you
He's got the good sheets π π
5*55=255, 2*238.27=476.54
it actually costs you less than double the power to use 250% overclocking
TFW you realize OCing isn't so bad when you don't take underclocking into the equation 
Sometimes I underclock, sometimes I just awesome sink excess because I'm lazy and want more lights
Though, with how cheap power is, I don't see that many use cases for underclocking to save power 
I have several constructors in my motor line that are at 0.2mw of power usage lmao, it can add up
Sinking is life
Oh right, underclocking with biomass burners is a thing too
I can imagine somoene spamming down assemblers at 0.2 clock to get SE parts quickly ahahah
ig a better example would be the 8% coal miner I have for that motor factory
That's much better on power
i've saved up some alien parts. i plan to use them for making packaged liquid biofuel for my trucks lol
u5 is gonna be fun
So.. should I use pure copper and pure iron for this or just roll with normal ?
pure copper and copper alloy are great. pure iron is also great
Imagine being like: I give up, trains without a decent UI are too much of a pain to deal with, they eventually take the wrong route anyway so...
sees trains U5 teaser
... Everything's gonna be fine, I guess 
use copper alloy when you have a lot of copper and iron in close proximity but not much water
I presume the trade off is that pure ingots take longer to make but produce more at once for cost efficiency
As most refinery recipes are slower
the downside of copper alloy is that it produces so much per machine that you need to be careful with your belts
1 pure copper node can output huge amounts of ingots
True
Still better than the smelters recipes at least π
upside is power consumption and floor space
I'm building over water with a lot of power to spare
I believe my most praised pure so far is the pure quartz
good idea is to raise the floor high enough that you can put the refineries directly over the water extractors. you will need pumps, but it's worth it.
pure caterium is the only choice for caterium ingots
so...
you can drag the boxes around to make it easier to read if you want
they do
Alright great
they try to at least. dont count on them actually reaching 600/min continuously
see the thing is machines dont pull fluid out of the pipes at a constant rate so the fluid speed would need to fluctuate above and below the desired rate to average out at the desired rate, but they cant go above 600, so if the desired rate is 600, it can only fluctuate below that line
Weird question but I've never really considered it ; how do the mega builders in beyond end game get all the parts for their stuff? I never see a small depo off to the side that makes the basic parts for them
Do they use excess production of other lines to make their construction materials ?
if theyre smart they make basic parts at small production sites all over the map
like always make ingots near the mines
No I mean like where do they get the massive amount of say, rotors they will need
You need a lot of simple parts, do it's not a small depo, either a large factory or usually a whole outpost
rotors can be considered basic parts depending on your experience
I make motors and control rods on the other side of the world and import via train, for example
Fair enough, I've started doing this recently
for example you can take something like 9 resource nodes that are all relatively close together and use all of them just for making motors
This. This is the job of the Northern Forest for me.
Distance stops being a problem in the late game thanks to trains, drones and fast conveyors, so it's only logical to make remote outposts
northern forest also has a lot of quartz and caterium so maybe consider computers too
I'm that alien species that likes rigour motors
I would split up my motor and computers, probably
Rigour motors are reasonable if you have good alts for oscillators
If you manage to mass produce a lot of them, it has benefits for sure
rocky desert is my pick for stuff that requires a lot of iron because the nearby oil can produce plastic and rubber to improve yield of stuff like heavy modular frames
It was more of "I only needed to siphon 7% of the production of my other busy miners in order to run a motor factory at 100% efficiency might as well slap a splitter down and call it good"
you can also take all of the really crappy resource nodes from grass fields, load it up on a train and ship it over to rocky desert to use in refineries on the beach
im gonna put a train station there and some truck stations. it's gonna be fun.
trucks load up the train, train takes it to rocky desert, all that oil and water added onto that will mean a huge production facility for anything that doesnt need quartz
I haven't messed with trains at all
Never even researched the monotrail tech before
could even use the blue crater instead of rocky desert so grass fields and blue crater combine resources... hmmm....
Some have dedicated factories for stuff, so they don't need to make massive amounts of rotors
So, after some quick calculations, starting from 480 crude oil and using the heavy oil residue alt recipe, as well as the diluted packaged fuel alt recipe and turbofuel, I got to a total of 840 sulfur/ min needed. Is it right ?
Granted, that's for 237 fuel generators but still
That's a lot
a bit more, but yeah
"using all the crude oil from 1 oil extractor makes enough turbofuel for hundreds of generators and that's a lot" sounds about right lol
With all alt recipes, oil is quite imbalanced
with diluted fuel recipe you make 4x as much fuel from the same crude :S
if you also use heavy residue alt
oil goes hard on the added complexity for extra efficiency
it's fine, it's a good tradeoff
yeah and it's beautiful
its a lot of work but it feels so good to get it all done
i havent decided how much HOR to keep for petroleum coke (alu/steel) and how much to make into fuel for plastic, rubber and turbofuel π
i dont even know how to decide
off topic but what app/program is that? or is it something in the game idk about
thx
Aluminum is paining me more than I expected
I've narrowed it down to either using Alumina Solution --> E-AS --> Aluminum Ingot, Or Sloppy Alumina --> E-AS --> Pure Aluminum Ingot, but I can't tell which one is the more effective
π΅βπ« I can do this with MK4 conveyors and the bauxite of my build region, but this doesn't factor for the fact I would immediately have to tear down and start over with mk5 conveyors and mk3 miners
build it with mk5 conveyors in mind and later replace
Why would you have to tear down?
Would need too many new machines for optimal functionality
You can just build it all with mk4. You don't need to rebuild it afterwards
More bauxite moving in
You'd just have to replace miner and small belt section in front of miner until first splitter to two mk4s
Build the setup just like you already have mk3 miners and mk5 belts and then just upgrade the miner(s) and belts in-place
The factory will run inefficiently until then but you don't have to do any rebuilding
Yeah like the others have said, build the machines and production lines with mk5 belts and mk3 miners set up that way you can just upgrade the sections without tearing it all up.
https://u4.satisfactorytools.com/production?share=QqHY34nPCdkD3ooxBwlA
im looking forward to build this, does anyone, have a better idea or does anybody see an issue
im rebuilding the oil production, i dont like the aditional outputs
"Some" being... Uranium? 
yeah..... that oil production looks..... bad, honestly
you already have recycled rubber and recycled plastic
so the Heavy Oil and Polymer Resin wastage is unreasonable
the calc always seems to break down when you maximize
once i switched from "maximize" to "items per min" it immediately optimized it again
so, in summary: when you want to ACTUALLY build a production line, switch from maximize to items / min
hoping for this to be fixed for U5
thx :)
ik, its actually 22.4 (ups) and 50.4 i calcutated it
it is 22.4, that's the part maximize always gets, since it only solves for most
it's just the production line is just whatever was there when the most was found, so you can end up with undesirable production lines
Sloppy->electrode->Pure ingot gives 1:1 ratio of bauxite to ingots. Other recipes as I checked give less.
guys I am confused with how to approach my math issue. I have a single conveyer transporting 300 ore/min. I need to divide the 300 ores per min to deliver 30 ores/min to 10 smelters. essentially i need to evenly divide 300 by 10, however since splitters only allow you to divde by 2 or 3, im so confused as how to get the correct combination of div 2 splitters and div3 splitters to get to 30 evenly
so either you do the "balance everything out" approach, or you can also just manifold
manifold is usually the easier option, it looks like this:
--S--S--S--S--S--S
| | | | | |
X X X X X X
S = splitter
X = any building (smelter in your case)
but how does that solve the issue? am i confused as to how splitters work
it works because the first smelter will get more resources than it needs and so it will fill up and the belt to it will also fill up. That means that all remaining resources will overflow to the next smelters and so on. After some time (usually a few minutes) most of the smelters are filled up and all smelters will be running at 100%
you can also remove the filling time by just putting a stack of items into each smelter
splitters try to output an item to the three outputs they have in some order e.g. A B C A B C A B C. If a belt is full (or not connected), it gets skipped. So if e.g. belt B would be full all the time, the splitter would output A C A C A C
basically: you flood the first machine with items until it cant accept any more. then the belt backs up and the splitter starts moving all items that dont fit on that belt to the next belt
then it floods the next machine. then the next. repeat until you are at the last 2 machines
aka how i learned to stop doing maths and love the manifold
I wasn't really sure what to do with the water byproduct of aluminum production until just a moment ago. What do you suggest about packaging and sinking water from the process?
Maybe a wet concrete factory instead? Instead of feeding packagers plastic to get rid of water, you feed a refinery limestone to get rid of water.
Perhaps, but one issue is that I also have no use for concrete right now
Right, I meant to sink the concrete instead of sinking the packaged water.
Oh true
Another issue is that my mega plans for an aluminum factory are far from any other deposit
The best I have in proximity is still extremely far away, and the nodes are normal
Any pure ingot recipe will work as well. Basically, find the closest resource node that has a water recipe, and sink everything.
Region where I am producing all of my aluminum (easy source of coke for electrode with building space and water.)
That's actually where I have my aluminum production as well, nice. Though I feed the wastewater back into the process.
can someone go check screenshots and see if the thing im about to set up is good or if i should do any changes
Your setup goals are beyond us and thus we cannot make suggestions
If you showed the recipes we could make suggestions on efficiency, else.... Well, just build it
Iβm being stupid here. Reinforced plates made at 5/ minute but only require 2/ minute for crafting next item.
Will I have 100% efficiency if I just split the 5 into 2 outputs? Is 2.5 close enough lol
At some point it will back up
Usually the solution is to unlock overclocking (and also underclocking)
Underclock it maybe?
@wind spade your tools says that a nuke plant produces 5/min of waste, the plant itself says 10/min, the latter seeming plausible after some testing
the wiki agrees, its 10 waste per minute, or 50 per rod consumed
but make sure the tool wasnt showing you half a plant or such
yeah it's a bit outdated π€·ββοΈ
is there any other known outdated information taken from the game?
I turned the entirety of those islands into turbofuel π
(Though I will have to rebuild it all in U5)
the problem with waste is that it wasn't in the data file that devs gave us so I had to hardcode it, but then they changed it. Most of the info should be accurate
pretty sure the waste display in nuclear powerplant was wrong in game too
and it was changed like two times iirc
I remember the same thing, iirc there was a display bug which was fixed plus a rebalance during experimental of U4, but since EA it's been 10
Sry if this has been asked a million times but what is the designing graphical interface people have been using to make factory layouts?
there are some in #math-and-meta , depending on which one you mean
if you mean this one, that's mine and it lives at https://u4.satisfactorytools.com/
Oh I will check this out!
In terms of lag, 3 conveyor poles/stackables vs 1 conveyor wall. Do we know if there is a difference?
The least lag would be to build the belt then dismantle the supports
It depends on where its built. In some places, you might need to build it more then once
But that would only affect lag if you are near the object limit
Guess I will wait till we have data on U5, as the engine we have tested, is probably not the same one as we get tomorrow
how bad pc you have if you need to optimise this deep lol
The less objects per project, the bigger you can build, regardless of PC
then remove the supports after you're done
walls and supports are trivial though, probably barely a dent
Biggest dent is probably to shard everything
as long as it doesn't increase gens too much, though I guess you can shard those too
not sure if there's a net gain there or not, cause 250% means 4x the power, which means 4x the gens, and gens can only be about doubled, so you'd still end up with about twice the gens but divide your production machines by 2.5
so probably best balance is whatever overclock is 2x the power, so you keep the same number of gens
250% means 2 machines replaces 5. So even if you double the power system, you still save 1 machine
Need help/opinion for alu plant module.
Using sloppy alumina and electrode, so 6 to 8 refinery ratios. Currently my modules seem to be blocked on water/alumina sol. Am I seeing it correctly that if two rightmost sloppy refineries run all time, they make "infinite" alumina solution, and recycling refineries simply cannot recycle water, because they have nowhere to put alumina solution to? Would removing the red pipe piece solve it, by breaking it into two systems, of which left would empty the right one from water?
Dont bother trying to recycle water into the same system. That only works if you do the numbers right and keep it at 100% efficiency. You go down to even 99%, you start getting a backup, which needs dealing with. So send that water into another system, like coal gens or wet concrete
or have a few refineries dedicated to process "waste" water
Still requires you to get the numbers right. With a 2nd system, you can overload the other resource, so 100% of waste water is used, regardless of efficiency of main loop.
Limestone is everywhere, so dont take much to get wet concrete going
LIES!
Not a single damn node in red jungle π€£
There is 1, on the outer reaches
But there you can just send it to a few coal gens instead, as you have 8 of those nods in that area
I also generally prefer isolating the freshwater from wastewater lines (usually by sending it to wet concrete), as rounding errors, the slightest PEBKAC, or other factors can cause mixed pipes to back up and stall the whole thing out.
yes, plenty of coal
it doesn't, you can just have more refineries than you need
Guess it depends on how you connect the waste water refinerys
I think that split setup above "should" work with red pipe removed? the set on left will just run on recycled water so it cannot stall on alumina solution being full?
Friendly reminder that you can make one way pipes. π
If memory serves, the flow valve, is not 100% accurate
For myself, I basically gathered up all the bauxite in the red forest/jungle/generally-evil-place, and belted it all to some waterfalls where there's a reasonable convergence of everything needed (crude oil, copper, limestone, water).
As to @wind spade 's suggestion, I suspect what he means is "have a bit more bauxite consumption than you actually have bauxite production, and set it up so bauxite+recycled water has priority on the bauxite".
When in doubt, slap valves at random places. Kinda like diodes.
Also, in the case of this system, actual valve value does not matter, if it stalls, it will overfil anyway. But valves enforce direction, so I'll try that
but yeah burning water in gens is probably easiest
Nah, that would require extra extractors and coal gens.
Personally, after mixed lines stalled out one too many times on me, I decided "send the entire lot of recycled water to a different process, e.g. wet concrete/coal gens" was the path of least resistance, but hopefully it works out for you.
V2, hopefully this should work? Separated the output of sets
Aaaah!
[will report if that works after fixing a pipe...]
Think that will work, but you might need to have a water tank to supplement in the beginning or it will take a long while for it to produce close to 95%
Shouldn't be much of an issue, the left side is running surplus water. Will observe how it behaves.
I have 4 another modules to fix in various ways π
One solution to that is to standby the downstream aluminum foundries/smelters, let the whole thing fill up, then start the aluminum smelting.
Or just chat for 10 minutes while the system balances itself. I'm seeing all my refineries green now, and all outputs are being emptied. π
I did mine this way for 1680/min Bauxite input
sorry it's a bit messy because this factory sits in the void, and there's a foundry level underneath
you can use the scrollbar thing on the right side of the map view to filter out things above and below what you want to see
that's what it looks like from the top/corner
Oh yoooooo thank you i told you guys feeding the water back into refineries doesnt work now i finally have proof π
just use a VIP for the water
btw this would work if you use a VIP for the alumina solution to prioritize the alumina solution from the recycled water over the fresh water alumina solution
thats why ive been saying that using the waste water in the refineries doesnt solve the problem it just moves it from the water to the alumina solution
I don't understand what you're saying.. I'm feeding the water (output from alu scrap) back into my refineries (to make alumina solution) and it works fine
some people are also just connecting the recycled water to the input water and it works fine
except sometimes it doesnt
I can share my experience
I limit the amount of water coming in to a precise amount
I didn't before, and it backed up
I spent about 100 hours trying to solve this particular factory (that I posted above)
so I have a valve at each subfactory controlling a very precise input of water to each triplet of refineries
you can see the first 4 subfactories only get 330/min water, this is controlled by a valve because Mk2 pipes can transport more, and the combined Water Extractors make the flow fluctuate
there is also another way around this, which is to make a priority input pipe junction
let me get the link for you... I only learnt about priority pipe junctions after I finished finetuning this factory
thats what i mean with VIP
variable input priority
I think that's the VIP thing you're talking about?
yeah
but I did mine without VIP, and it works
ive tested all these layouts for multiple hundreds of hours and the only one that doesnt break is the VIP
it hasn't backed up for about 300h now and the aluminium ingot output is more or less perfect
I don't use buffers in my setup
they cause too much fluctuation
I tried that too in my experiments
Ive used a total of 3 different layouts and they all worked
#math-and-meta message im using buffers and i have no fluctuations at all
One where the scrap refineries were above the solution refineries and i just limited extractor input with a valve, one where they were all next to each other (an i also valved) and then once with a VIP
extractor input valves sorted everything out for me
you canh search in questions and help and theres 100+ people that use valves etc. and still have it breakdown
I've been toying around with instant scrap in bigger planetary builds and I really want it to work, but you need that sulfur for classic battery -> supercomputer -> ADS.
Unless you maximize ADS sulfur wont be an issue
I'm not saying everyone else is doing something wrong
I'm just saying this is how I did mine and it hasn't backed up for a long time
yes and im just saying youre lucky
then you don't have to quote everyone else saying they can't...
Right, this is the point at which you're doing some sort of build to make the final SE load in 30~ minutes and are also doing max uranium for power.
I could manually inspect every personvs factory and possibly find a reason for why things work / dont but i wont because i have better things to do. Using a VIP or burning the water in a coal gen are the 2 safest solutions
Technically packaging too but people who package and shred water are lunatics to me
you can package and unpacakge
that actually works 100% aswell
its just big and uses alot of power
Package water and send it to diluted fuel production xd
btw ive been working on writing this all up but im lazy and suck at english xD
https://docs.google.com/document/d/1nN7lceueOQClan9NDvEnfBso8sHd1DBlG-qDaMXUrY8/edit#
Dealing with the Water output from Aluminium Aluminium scrap refineries produce water as a byproduct, which can cause the refinery to shut down if the buffer fills up, so it must be dealt with. There is multiple ways to deal with the water: 1: Use the water in different parts of the Factory 2: P...
My solution is complicated and not really recommended.
- I ise sloppy solution + electrode scrap method
- Each set of refineries only uses 600 a min bauxite ore.
- Each set needs 600 water. 420 is produced as byproduct.
- So, I use a water input from external source. I only pump in 180m3 water from external and loop back the byproduct water.
- I use overflow sinks at the aluminum ingot output end to ensure nothing ever clogs up.
So far no issues at all
I mean I know there are easier ways like using VIP piping
Used the same setup, and it worked as long as you had everything running at all times. Had an overflow pipe, aka pipe with higher height as fluid flows to lowest point, sloppy refinery. Overflow water went to a residual rubber that took polymer resin being made as a coke byproduct already to sink.
Helped out when trains didn't supply enough bauxite or didn't use all scrap fast enough due to blockage in supply chain after
You already need copper to make aluminum sheets, so pure copper ingot makes some sense for aluminum wastewater. Whatever copper you don't use for aluminum products you could turn into steamed copper sheets and ship it out.
Mine has no overflow piping. Let me use some screens.
-
This is the water input from exterior. Each mk2 pipe has a full supply and is divided into 3 mk1 pipes. Which are valved to 180
-
Big project picture. Each zone is zegmented and fully isolated besides shared petro coke input. Each mk1 water pipe feeds only one of these modules
This is the front facing piping. Each segment has mk2 piping in front of the refineries and the wastewater comes from the other end of the pipe
I know its mk1 in the screen but that got fixed. Limited options on my phone
Hm. That's no small addition. The 6846 wastewater from processing all 9780 bauxite could process 10269 copper ore into 25672.5 ingots. Far more than what you need for aluminum products.
Update 5 tools! https://www.satisfactorytools.com/
Changelog:
- updated visualisation
- tooltips when you hover over recipes
- new colors for byproducts and inputs
- better display of numbers (more decimals, removing of trailing zeroes, etc.)
- better calculation
- all (or most) of the bugs regarding maximise + input + items/min should be fixed
- the tool still doesn't optimise for raw resources when maximising though
- removed maximise ratio (will be replaced by a better system very soon)
I've also rewritten major part of the code which will allow me to do many more cool features very soon, so stay tuned!
sounds great, many thanks for your great work
sure, go for it π either here or on my tool's discord
Ta. Great work, as always. You're a huge help.
aw my stored factories are gone again, well, most importantly the one i had my recipes setup in π
not gone π
u4 version still has them π
(and since recipes didn't change, you can keep using the old tool)
you just won't have any of the new cool features
but I'm planning to eventually add a way to bring U4 stuff to U5 tool
its just my unlocked recipe list, i can just tick the boxes again
just slightly tedious π
the other factories in there are all build..
not really that worth fixing probably
Any way to import plans from the U4 site?
importing is broken on U5 tool, but after I fix it (hopefuly soon), there should be a way
worst case it will be replacing u4 with www in the link
(but again, it doesn't work yet)
@wind spade good updates! Nice stuff
how many of y'all are gonna start having your conveyors clip through absolutely everything for maximum space effeciency
Uum...never
I think I 'fixed' it by dividing the alumina solution into separate flows. need couple hundred hours uptime to verify.
just figured that it's not gonna be that easy. The save data format changed between the two tool versions, so I'll have to write a convertor (which I will at some point, just not today, too tired). So if you're in hurry, it's probably faster to just set up new production lines. Otherwise you'd have to wait till tomorrow.
I can wait till tomorrow, was just gonna start a new game anyway, but worst comes to worst I'll just manually recreate them again
Isn't that the basic method and not the complicated one?
Already did that with the Smart! mod XD
Clipping is heresy!
Perhaps, could be but even simple stuff gets less simple at scale
anyone who played with train signals can verify how they work compared to Factorio?
Based on available information, it should be like this:
- block signals are the same in both games
- path signals reserve a path through a block only if the train can safely exit the block
- if reserved path exits through another path signal, then it "chains" (like Factorio's chain signals) and it has to be able to go through the next path safely as well
huh?
π€·ββοΈ
its just like this, it forces all to be path or block
or it errors
oh, it allows you to place the signals, but they don't work, right?
yeah, they turn into yellow !
I thought it doesn't allow you to place them and I was a bit confused on how that would work
there are no limitations on building
the key difference with factorio is that trains follow fixed path
they only compute path when leaving station and then stick to it as stubbornly as possible
well they do recalculate
or when they stop
they do if they wait at chain signal for longer amount of time
they dont tho
oh I thought you were talking about factorio
right
the key difference with factorio is that trains follow fixed path (in satisfactory)
i guess it should be "difference from factorio"
oh yeah, now it makes more sense π
english is not my first language
no problem, was just a bit confusing π€·ββοΈ
also, in this situation, how much does the path signal reserve? just till A, or till B or till C?
so it doesn't check if the whole train can exit the path block, but only if it can start exiting it
yup, so it would seem
yeah, good to know, thanks π
you likely need to space A and B far enough that whole train can fit
(would be helpful when I try to help people with trains π )
yeah, that makes sense, I just remembered that Jace mentioned "Path signal will make sure that the train can exit the block", but it's more like "path signal will make sure the next signal is also green"
yeah
i guess this might need more testing, maybe it does reserve whole path
but from my testing, it doesnt seem to
its problematic to test due to schedule ignoring stop conditions and train collision boxes being random/bugged
so its not even possible to tell if some segments are merged into block or not
https://www.youtube.com/watch?v=gNHT04xCPM4 someone also posted this which should not be possible at all
ie. it somehow collided with train in different block (block had to be different coz it assumed it can go to the sliding and hadnt stopped on the signal)
unless there's a way for a train to ignore a signal (like e.g. in openttd where you can press a button for a train to ignore next signal)
it feels like there might be some issue with calculation of boxes on junctions or something
no clue honestly
there is also a question of how much vertical clearance trains need/use
its completely not obvious
this test may be useful then
(run both trains at the same time, so left train enters first block, but then top train enters second block)
if the top train does enter second block, then only one block is reserved. Otherwise whole path is reserved
clearly something is fishy with vertical clearance
why the next signal doesn't light at all?
its blinking yellow, just screenshotted at black phase
ah
there is no path beyond the screen
https://questions.satisfactorygame.com/post/61784bfe831c85205235232d left a comment on the path recalculation here
also https://questions.satisfactorygame.com/post/61792a4c831c8520523532ee we need this even more now π¦
clunk-ness of track building goes over 9000
@wind spade my understanding of path signals:
Draw line from path signal to next block signal.
Reserve that specific rail track as a new "sub block".
Act as a block signal.
They don't care if something is in the path, they don't care if the terminal signal is red or green. They just draw the line then turn into a block signal for that specific line.
After which, if something is in the path or the terminal signal is red, normal block signal logic applies and they will not allow passage until a block signal would.
so a path signal can reserve a path to next block signal even if the block signal is red? that goes against what Jace said in the video
hmm
i think they wait till exit signal is green and allow trains to cross in meantime?
Jace said they reserve a path, and it allows non-intersecting trains to cross.
You can still reserve a path and have intersecting trains, they just won't both go at the same time.
I think he said something about allowing the train to safely exit the block
This is why we can't build stackers as we discovered last night.
The path signal doesn't "try to find a way". It uses the existing "what is the fastest route" logic for trains and locks it in. If the route it blocked it just stops the train.
It does not have a "find alternate path if blocked" logic to it.
Yes. The path is locked in, then if it is safe to exit or enter the light turns green.
"Safe to exit" also deals with the scenario of oncoming traffic.
Where if you solely use block signals you could end up with 2 trains facing each other and neither is able to move.
A path signal will check that "safety" and stop the train before it even enters the converging block, allowing the oncoming train to clear, then allowing the pathed train through.
ah
so it reserves a path but doesn't let the train come in until next signal is green
Indeed.
I can already see how that may (and will) slow down traffic on complex junctions
When the next signal is green that is the true "reservation" moment.
but the "soft reserved" path is also blocking other path reservations, right?
Definitely had to build with it in mind, that's for sure.
Soft reservation wouldn't block soft cross-reservation.
Like I said, it just draws the line then acts like a block signal for the subblock it just created.
So in the case of crossing path signals, you're looking at a "who goes first?" scenario
to my understanding, the trains follow fixed path, so it stops until it can exclusively reserve path + exit block
other trains can cross if exit block alone is the problem
This, but clarifying it will not attempt to find alternate paths. It just waits until the best path isn't blocked.
yup
if there are multiple ways to cross an intersection and part of it is blocked, it wont attempt to avoid the obstacle
That's why the best practice is usually to block signal an area first, then go back and replace with path signals when you know where you want subdivisions.
Reason stackers are impossible.
its extremely unfortunate really
it also means that multi-rail tracks are not possible either π¦
They are but that has to be handled at a multi-station level, not a signal one.
as in multiple parallel rails in same direction
yeah, you would need to somehow force stations to feed into particular rail, but it wont auto balance
Yup. The more I think about it though, the more I am fine without stackers.
It sort of forces you to consider the other transportation options in some cases instead of doing 100% trains.
Which I think is a good thing.
not sure honestly
the stackers allow solution like making multiple waiting bays before multiple stations
which massively decreases footprint
From the beginning I haven't been a fan of trains making trucks completely obsolete.
they kinda dont tho
They still can tbh, but as you say it takes a much larger footprint now.
trains are super clunky to build, so you will probably use trucks whenever possible
I believe like 90% of the community doesn't care how clunky they are and switches everything to trains immediately and never uses trucks again..
the core advantage of trains is that you can reuse existing infrastructure by expanding it a bit
if you already have north/south rail somewhere, its much easier to just add stations on both sides and have free connection ready
i use a lot of trucks in my save, but mostly for short-ish distances where putting trains dont really make sense π€
its fine like this imho
but this implies that, for example, trains should be long (i use 1+4 mostly) and travel large distances
You and I have similar opinions on truck's place in the world.
They aren't shared by many though from past conversations I've had. Lol
but in such case, you really want stackers to work, as you cant control how many trains will arrive at once
for example i feed two sulfur mines into turbofuel factory in blue crater, there is significant distance between them so i use two trains
now i would kinda want a stacker so that if both sulfur trains arrive at once, they wont block other stations in the refinery area
So the "solution" to that is having 2 stations.
Also curious as to why you need 2 trains for it?
My turbo plant used the entirety of 2 overclocked pure sulfur nodes, and I just had 1 train roll it through, never skipped a beat.
it was just easier to build like this π€
Indeed. Not saying you are wrong.
Also the only thing trucks need now imo is the smart-loading option they added TO TRAINS....
right now the backbone looks like this
two sulfur stations feed sulfur into the refinery on right, then it feeds plastic and turbofuel left
and there is new alu factory to the north that also gets some oil from crater
on the calculator is it more efficent to us items/min or maximise
doing max uranium and i wana know which is cheaper and both options have different factory's
I feel like there is so much more unused potential for trucks
Once you add smart-loading the things they can do explodes in number.
i was thinking about adding stacker here to handle that complex station, but its not possible right now π¦
I guess, but I think the true potential lies in a major pathing revamp
dual I/O and the queue system are fantastic, so just that 1 measure of control in combination leads to massive possibilities without any other changes.
Pathing has been fine for me since their inception.
Imho people just don't like having to drive the initial path.
I wish you could record paths and have them all join in a single road map so to speak
Which trucks would then be able to use to get from any place to any other place
That's what trains do though.
As much as I love my trucks, I want them to stay in their niche.
Copy/Paste pathing was a Godsend π
I know it's what trains do, but why not have trucks do it?
Because
As much as I love my trucks, I want them to stay in their niche.
I think the limits of what you can do with a certain technology shouldn't be ceiled off by simple programming
as it stands right now, trucks are vastly superior to trains, over large distances too
The limits of all technology is based on programming?
the only demerit is that you need item fuel for them
Making logistics smarter will make things incredibly cool without making it feel cheaty
If you had a road network, you could build fuel stations all over the place
Oh you're going to start a fight with some people...
I think they are too btw
just trains need to be buffed a bit π
Even in their current state
its more like trains are kinda mediocre for their costs and complexity
need better acceleration and top speed at very least imho
i still love them but yeah
I disagree tbh.
I think both have their place.
If you're trying to use a truck for train things, train better.
If you're trying to use a train for truck things, truck better.
π€·ββοΈ
xD
Same way in its place, Drone > All.
But don't use drones for truck things or train things.
You can really draw quite some useful conclusions from real life, where trains are incredibly powerful for massive distances and large quantities of resources, and trucks are really good if you need smaller amounts of resources in a more local area. But in satisfactory the map is relatively small, and the bigger struggle is often the wide range of variety in items, rather than the quantities. Using a more dynamic approach with smaller vehicles that you can run more of at the same time and need less infrastructure for is really quite powerful.
This is about as simple as I could make the list for signal logic rules: https://www.reddit.com/r/SatisfactoryGame/comments/qgy9z0/signal_logic_rules/
is it possible to put a splitter directly on a machine therefor being able to output 3x more since you have three conveyors?
no
f
would completely defeat the point of belt limits
it wouldn't really
you still have limits
you just have creative options around them
until the next guy asks if you can put a splitter on a splitter, etc π
oh
well i already knew you couldn't do that
i was asking about machines specifically because tbh i've never tried
the only actual limit is OC Mk3 miners, and they are thinking about a solution for those
yea it seems weird that it's impossible to use the miners to their fullest potential
It would allow for pretty much infinite throughput
how would that be infinite
i was never stating it as a suggestion btw i just wondered if it worked
Machine - splitter - ISC - splitter - ISC - splitter - ISC - ...
if it were allowed you could stack them .. lol
oh i guess you could use other machines
meh, i dont think it would be that game breaking anyway kek
Also the whole game is balanced around mk5 speed being max, so tripling that wouldn't help
oh
Instant transportation of infinite items seems a bit game breaking
i was just thinking you could big brain and get full use of things before you have max tiers, or actually use maxed out miners
instant?
i said nothing about instant
ISCs and splitters have no travel time
oh, idk what an ISC is i guess
Industrial storage container
ah
Industrual storage container, you could do the same thing with normal storage containers, we just don't have a neat acronym that's easy to type for them π
And yeah the limitation of low lever tiers is exactly why it's there in first place
Splitters on splitters attaches to ISCs with more splitters attached to them. π
Keep chaining until the entire map is just 1 giant storage container.
They finally fixed the smart/programmable models and updated UI
I just checked if I could slide below the belts connected to them and didn't even remember to check the UI π€¦ββοΈ
Is there a copy - paste for their settings yet?
An observation about path signals:
- Trains do not reserve a path until they enter the block before the path signal.
- Path signals are red until a path is reserved.
- Trains approaching a red signal in any future block will slow down in order to be ready to stop at the signal.
This means that train throughput around path signals is affected by how long the block in front of the signal is. I tested this with a short (2-foundation) block, with a block signal and then a path signal, and the appproaching train slowed to around 20 km/h before it passed the block signal, then immediately sped back up when it reserved a path.
--->>slowing>>---green block---red path-------
This is going to have some interesting consequences for interchange designs.
Yes.
It's also why you don't want block signals immediately before stations.
Having them after is fine, but right before will make them crawl into the station.
All of these behaviours are completely reasonable! I was just surprised to find my trains slowing down before an uncontended intersection, and dug into it a bit
I actually moved away from path signals in station manifolds entirely
which solves that problem, as all signals in the manifold are green most of the time
Trains waiting on "fully unload AND wait" conditions sometimes seem to reserve their exit path even though they won't go anywhere for potentially a very long time
so path signals out of the station into the exit manifold ended up blocking all platforms, because one train had the path to the manifold exit reserved
I do think "docked trains hold path reservations indefinitely" is probably not the ideal behaviour; it'd be reasonable for docked trains to release reservations and re-acquire them after they undock. I agree with you about that specific configuration, though.
Put in a suggestion: https://questions.satisfactorygame.com/post/61799c4c831c852052353bf9
Hmm, do derailed trains hold their reservations?
Probably just them not expecting path signals after stations.
The logic is working properly, but you'd have to make an exception for the station case due to how it affects the logic.
yeah
Path signals controlling an N:1 junction is meaningless, but for people with bidirectional exit manifolds I can see them getting surprised by this
N:2, two nonconflicting trains want to proceed, only one gets to move even though they're "using" path signals
I think? Will test.
Which is a case to not use path signals if I'm understanding you.
Oh, agreed. I'm only saying that the behaviour is likely surprising, not that that configuration is the right choice
I believe I just thought of a simpler way to describe signals that may help more people understand them.
Trains do all of the pathfinding and routeplanning.
Signals tell them to stop or go.
Signals do not tell them yes or not to the route.
Train has already said "yes" to the route that the Train decided it wanted.
when you come back to a server only to find your other players cheated to unlock everything (Kind of OK), had no discernable organization (kinda not OK) and was trying to automate quartz tier stuff when the ENTIRE POWER GRID was 4 coal generators and, no joke, 45 biomass burners
You need better friends.
I do have them. they just played on a server where the owner was a dick and blocked me from the discord where we passed the save around
The thing was i was trying to get them on that world too...
Some people honestly enjoy cleaning up after dumbassery like this
Though I agree, it sounds like the save is the least of NemoAmet's problems here π
Alfred was a person who enjoyed cleaning up after Batman's dumbassery.
π
I can't think of a simpler solution tbh. The only difference I have with my system is that I have refinery sets of 600 and 750 Bauxite. I linked you the shot of my processing plans for all 9780 bauxite. I think it's 27 refineries total for all the scrap. I honestly can't think of a simpler way.
Probably the simplest solution is boiling it all the way down to feeds of 60 into 163 refineries. Then do 163 electrode refineries feeding 326 pure ingot constructors. It's a higher build count, but you do get good symmetry.
Good power efficiency too, if that matters. Sloppy alumina refineries would be running at 30%, electrode at 40%, and constructors at 100%.
@vapid gorge my setup is actually 13200 ingots. Im using the micro manage mod to get all 1200 out of the pure nodes
So its only groups of 600
And one 300
Ahhh yeah. I guess with that I'd just have some extra 600 lines of prod. The 750s are just because I split some 300s to 2x 600s at the train line to save on platforms.
The 780s have 30 over flow merging to a single refinery.
Nice!
why not just make a tower into the sky?
Mine is 100% belt fed. Like this
I'm trying to keep it as vanilla as possible. I just have Smart, AA and some functional support stuff like the 3 wide belt supports. They make my life so much easier
Heh heh, much of the time I'm using lower tier belts, it's because they're skinnier and I can avoid clipping.
quick question, besides materials that require lots of screws like heavy modular frames and reinforced iron plates and whatnot, is having a lot of screws really necessary? i was building a quick factory for iron plates rods and screws but i thought that if i have to build a factory for heavy mod frames or so for example, id just make the screws onsite instead of making a lot at this factory and bringing it all the way over
i was planning to just have like 60/min iron plates, 25/min rods and 20/min screws
it's not necessary at all, all recipes have screwless alts
most in fact have a preference for screwless alts, as dealing with screws can be a pain
that or they use steel screws to avoid transporting them
and just transport steel beams instead
id understand, was trying to set up a factory for reinforced iron rods and the damn amount of constructors i need just to get 300/min screws was painful
yup, default recipe sucks
I mean it's used in the technically most weighted resource efficient combo (steel rods + default screws), but it's such a pain no one bothers
ive already gone down too far into the process that it would be more work to replace all the machinery for an alternate recipe
and also because my steel factory is like really far away and i dont yet have access to locomotives
eh you'll build more anyway
or not if you go screwless
I don't get caught up in trying to plan ahead for intermediaries personally, I find it a bit fruitless since I haven't planned end products that use them yet so they may or may not be enough, I just build as needed
i see
i was just concerned that i wont really need screws at the moment and then suddenly late game rolls around then everything is like "pls give me 5000 screws for literally like 1 of this product thanks" and then i have to deal with screw factories
nah, I mean you can if you want, but not really
alts let you do anything, even torture yourself with screws 
also
is there a good spot that isnt far from the grassland whatevers thats good for heavy modular frames
I was just gonna set up a mini factory near my modular frames with locomotives in the future but im already running out of space because of the gas thing near the factory
yeah any of the places in grasslands good for making steel works for HMFs, they only need limestone, coal, and iron
so like somewhere around here?
yup, looks like a good spot
not sure if i have to automate it right now though
im gonna be moving into tier 5 & 6 so i dont know if its something i should have a lot of right now
you can setup a little mini-production that's not too much ppm, HMFs are used a lot
oh and by used a lot, I mean for building, not like other parts, it doesn't get used as an intermediary for a while
ok thats great
so i wont need to connect a bunch of locomotives 
are there any other materials i should set up factories for before entering tiers 5 and 6 though
you can make 4 satellite factories, each with truck stations that transport all their stuff to a central train station that connects blue crater to gold coast and rocky desert, then connect blue crater to swamp/rocky desert eventually making a big loop around the map with trains
yeah that is probably something i am going to do
but fun thing is
that pure caterium node on the waterfall can also have a train station for supplying water to the grass fields
i am planning to process 6000 iron/min in refinerys
i use that one pure iron node in blue crater for pure iron -> solid steel and then use the steel with plastic and rubber to make other things
at the desert there is so many pure iron and copper nodes it is insane
atm it's just loading all the steel onto a train because i know im gonna have some extra steel that can be shipped off somewhere else
lots of caterium there too
im afraid thats probably too complicated for me at the moment
what tier are you at the moment
3 and 4
i would recomend just getting to oil as quick as possible, and just afking the space elevator parts
ive already got my space elevator parts set
i just feel like i dont have a good source of a lot of materials to really go to the next tier yet
like what stage are you at rn
just do one factory at a time on small scale and use truck stations to bring stuff to your space elevator production
you can always dismantle parts of factories to upgrade them or transport ingots etc somewhere new
i dont have an efficient
- iron plates, rods or screws factory
- steel pipes, beams or encased industrial beams factory
- caterium parts factory
- reinforced iron plates and modular frames factory
my "base" is also a mess
everything is just spread around and its an eyesore
only thing im missing atm is quartz transport. once i have that i can unlock tier 5
-where did you start
-Make a bus to get resources that is mega
started in the grassland whatever its name is
ah tough
its my first save
oh ok, i did the same
trucks are better than a bus tbh. can reroute trucks so much easier than buses
he only has access to tractors though
tractors also better than bus
hmm depends
at that early stage can just use coal as fuel
oh yeah that too
i havent even reached a quartz node yet because last time i attempted i didnt have access to the xeno basher yet
and htere was an alpha whatever spitter coming after me so i couldnt set up a factory there
you seam to know a lot of knowledge to a start out, how many hours you got
1137.7 π
oh my go-
i restarted so many times lol
i mean my alunium setup took like 70 hours and i have to max it out
i mean complety redo it and spend 150 hours on nuclear and fuel power
honestly trucks are kinda annoying for me rn
i always had everything close by and if it wasnt in somewhat walking distance (or zipline distance) then i would just build a hypertube
i just produced stuff onsite and stored it there
i never felt like i really needed to bring the storage to one area
i got this new game from skipped tutorial up to tier 3 in 1 day but that was like more than a month ago and still not unlocked tier 5 because im lazy and space elevator things are annoying lol
i just afk the space elevator parts
but now it seems like i should probably do that, but trucks are a pain for me to set up, also fuel
so i was just gonna wait until i unlock locomotives to do that
ya but i need to get my quartz from that really far away cliff to automate the space elevator parts
i also never automated the tier parts to get to 7/8, just used storage crates in manufactor
i dont have HMF or computers automated
really? what do you need the quartz for?
i got the silicon circuitboards alt
i got a factory making about 8 computers/min, 4 floors
i wanna use quartz > silica > circuitboards > computers
get the alt recipe for the assembler recipe for computers
i like using recipes that require oil because oil is cheap
oil is easy just annoying for me rn
with all the alts i can make soooooo much plastic and rubber itrs insane
did use those nodes by the northern forest to make 600 rubber and 250 plastic/min
oil is everywhere except where my base is
blender fuel recipe into recycled stupid op
didnt go there yet, im using blue crater for oil, and after that gold coast and then i'll go to northern forest
but polymer into recycled into fuel is even better
like are u serious
most oil on the map is up north lmao
HOR into diluted fuel into recycled plastic/rubber loop with some of the byproduct polymer to kickstart it is my favourite
something like 3000 and something oil fully overclocked
yeah
i gotta head to sleep but i sure will be coming back tomorrow because i will probably be confused about literally everything in tier 5-
i used just heavy oil residue as the byproduct though but yeah that method is better
im gonna take all that oil in this order, blue for plastic and rubber, orange for fuel
4 probably turbofuel
turbo ain't really worth it anymore
and 3 maybe petroleum coke for aluminium/steel
you need most sulfur for batterys and nuclear
i used blue crater for pretolium because my alunium setup is right there
but i need power to begin nuclear and i can repurpose that fuel when nuclear is done
the northern can generate 9200 fuel btw
all the bauxite nodes are across the equator; you can do aluminium production anywhere on the map. i think i might do it on gold coast
yeah you got 4 pure and 3 normal and 1 impure right close
imma go nuclear top left area
the left and center bauxite for aluminium, the ones in the swamp for batteries maybe?
yeah because you have easy access to oil on gold coast so yeah
thers also that lovely coal lake in northern forest if i wanted to use coal for aluminium
i might do that aswell because i need to process it
waiting to get mk3 miners rn, so turbo moters, but they can suck my dick
i'm scared
so many plates rofl
1.8m ficsit points per minute
jesus
the amount of space it would take up
like how do you even store that
Gotta power it with biomass, coal, or nuclear.
max fuel and nuclear can't forgot about max coal and biomass
Thereβs no max biomass
i know
Other than as limited by your sanity
like if you want to refill every 30 fing mins
Biocoal being more MW efficient than Liquid Biofuel still hurts my soul...
you can automate biomass really well if you have a couple storage containers full of animal parts and make liquid biofuel π
Yes, but LiquidBF is still just... terrible.
im gonna use liquid biofuel for my grass fields trucks until it runs out and then use regular fuel
i have so much solid biofuel and animal parts saved up lol
leaves and wood nearly full, solid biomass full, 2 personal storage full of solid biomass and one industrial storage in the back full of biomass ready to be turned into solid biomass. just dont have any use for it right now
also hub storage is half animal parts :S
anyone have a relatively compact 3:8 split i can use?
eh can i get context for the splitter
doing steel, with 360/min of coal and iron
alt recipe or normal
normal
yeah don't have one, but i will be on tommorow, so i can help you tommorow if you want
this link may help https://www.reddit.com/r/SatisfactoryGame/comments/f4rhbb/anyone_know_how_to_create_ratio_of_38_with/
0 votes and 17 comments so far on Reddit
Using faster belts along with smart splitter with overflow usually solves a lot of problems.
just had a thought, whats a good ratio for steel pipes and beams for tiers 3-6?
N:M splitters aren't needed most of the time, and they're not even fun cause there's no clever solutions to them, just a brute force method where you split the N outputs M times and then merge one from each set into M belts.
Since resources are unlimited you don't need them because throughput will always be the same. So you can just do manifolds or separate 1:M balancers.
Nuclear fuels rods going to power plants are probably the only time you might want to do balancing to skip the loading phase of a manifold.
Initially you want more beams than pipes, specifically 3:2. Later on you need more pipes than beams. As beams are exclusively used to build railways and some buildings only, while pipes are used for bulk quantities
And feeding multiple drones for throughput... if you value sanity
All depends on your playstyle
Placing a few more power storages can be an appealing solution too
I don't follow, how do power storages relate to the initial feeding of the reactors? I agree with having power storage just on general principle.
Assuming one would balance to get power sooner (or even just "stable" power sooner), one can get a similar result by having batteries smooth out the warmup time
Build 1k storage... or 1 balancer.. lol
Though, I guess balancing would end up being easier for over ~12gens
Tbf, "1 balancer" could still mean a room full of stuff depending on the case π
Luckily fuel rods have ez ratios
Most can be done on a 4x4 grid.. :/
Usecase π€·ββοΈ
I usually balance in bulks, so it's hard for me to guess my average for each
The way I balance fuel rods is just one manufacturer per one nuclear power plant, but I will manifold the feeds to the fuel rod manufacturers and let those prime.
That makes for one toasty rods factory xD
I'd gladly split directly from the manifacturers, but that gives some weird outputs for waste, imo (for skme reason, I prefer balancing rods than waste)
So I split to feed sets of 10, 20, 30 or 60 gens each
balancing nukes is cancer because you have the waste/water going everywhere aswell
where you can just spam the power storages
but for me the real reason is that with a balancer the nuclear reactors actually have 0 radiation
i dont really care how long the start up time is as i only have to do that once
manifold everything; build uranium fuel rod factory and npp's and get all that setup, standby your NPP's and turn on the fuel rod factory. By the time you finish your waste processor the NPP's will be full, spicy and ready to go.
So I'm going to take my "haven't played the game in months and I see they made a change to Coal Gens" question to here.
I'm hearing they run at max power now, and are no longer focused on fuel conservation?
...Why was that made as a change? Was it because Coal Gens are "too easy"?
Tbh i dont know the exact intention but i guess it is because of the batteries that can now catch the overproduction and send it back in in case your consumption is higher than your production
(Saved me while building my big nuclear plant, else i had to build another oil plant xD)
Before I had two Tier 2 Extractors at standard speed, and they handled my 32 coal gens perfectly. Now only a handful of my coal gens are working.
Haven't automated batteries yet.
Which purity?
Both Normal.
So 240 production, 1 coal plant needs 1 coal every 4 seconds so 15/min.
240/15=16
So its should be luck that it worked with 32 before, because it looks like you were under the half of your max produciton so that every coal generator just like "ran on 50%"
but as they work on full power now they do not do this anymore, unless you set them manually to a lower percentage
Negative. They're all at standard 100%.
Yeah but before (U4 i think) they did not do 100% unless needed.
So if you just use 30MW with one coal plant, it just produced 30MW
And so used less coal
Since U3 i always use 1 pure coal MK2 (240) for my beginning setup and this is always 16 Coal Generators, even before the overproduction update
Yes. That's what saved me. So, MK4 is the highest belts I got.
How many slugs am I going to need to get things normal again for these two extractors? Currently I have MK3 belts as it was enough at the time.
You need 480 Production, so on 2 normal nodes you need both to overclock to 200%, so 2 shards each = 4 shards
(ps.: i always recommend to use underclocked water exractors (to exactly 75%), as they than can feed exactly 2 coal plants each)
you just use 3 extractors for 8 coal gens
Yeah, I got plenty of water. Just not enough coal.
Anyway, so.. 480 / 2 belts = 240. So I can keep my MK3's. Perfect.
I honestly dont like that in terms of building. Imo its way more better looking with 75% ones. But this is personal opinion π
Yes π
Thank you very much.
It was changed to make it easier to deal with oil (byproducts) and nuclear (waste) as now hyou have guaranteed output instead of variable
does anyone have tips on what i should have on my main factory bus? or at least what are the higher volume items i should have on bus?
as a side effect it also makes it easier to verify your setup is actually working properly, before you could have a sub-optimal setup that would fail if your load reached a certain level, which was tedious to debug
I personally wouldn't build bus at all
It's imo better to make factories connected directly to nodes with products going to storage
Belt throughput is very low in SF and you don't reuse too many intermediates
But then lets say you making computers, you then need to bring all the items to that location to then ship to a main location, does it not make sense to ship to main location sooner?
but i do get what you are saying, I am finding trying to calculate how much i need and the choke point of belt speed to be a real pain
You can pick a location that has all the required nodes
Or use trains to move raw materials around
Again tho - it's just my suggestion, but I have reasons to think that bus approach isn't ideal in SF
if you are building more complex products, a singular spot that has all of the materials is likely not to be found - however transporting one or two materials in is still far easier then transporting everything, especially since those rare materials are often much lower volume, especially if you refine them first (eg. ship caterium ingots instead of ore, etc)
i have everything on my main bus
if you dont like that most important are probably wire/quickwire/steel pipes/rubber/plastic
Well i have just gotten trains so am now trying to plan how to expand from here, maybe i will rethink my "main bus main factory" idea 
you can just belt/pipe everything to 1 spot though
you can, but you dont have to π
btw this isnt really an issue since you can just have multiple belts stacked
theres designs that balance those quite easily
Luckily the game is rather flexible. Personally I subscribe to the idea of satellite factories that build complex products from raw resources only, and you would ideally place those in a spot that already has most (if not all) of those raw resources
it's the amount of belts needed to be built that's horrible
and the balancing as well, apparently overflow is bugged and eats items
got any examples for me? that i may steal look at
(unless they fixed it in U5)
#old-questions-and-help message
im using the vertical one in the back
wait really? never heard of that
balancing belts isnt too bad, as long as you dont have a rather odd number of belts that doesnt fit merger/splitter patterns
the 9 items in internal buffer apparently never get out
somebody told me that back in U4, not sure if it's still valid
isnt that just a bit of startup cost, unless you do weirdly mixed belts?
oh so it just fills up then its 100% efficient? i mean thats fine by me
it's a bit problematic if it eats your rare items 
anyway, if you're doing the bus, it's usually better to not balance it
anything you produce isnt exactly that rare, but if you use it to sort your personal inventory maybe
https://satisfactory.fandom.com/wiki/Balancer
has this under Belt compressor
Balancer may refer to Load Balancer or Belt Balancer. Splitters are built in a nested way, such that all downstream belts or buildings receive an equal amount of material, regardless if the supply belt is providing sufficient input. A factory that is built this way tends to start up faster, as there is no need to wait for the internal storage to...
What is this about overflow being eaten?
Overflow setting on smart splitter causes the internal buffer of 9 items to fill and never be used again
At least it worked like that in U4
wasn't it 3 items?
Iirc the buffer has a size of 9, but that may have changed as well xD
Interesting
if you block all outputs it fills up to 9, but overflow maintains it at 3
Well, as long as you're not making long smart splitters manifolds I think it's okay
Overflow outputs were intended to go to sinks, or at least that's what most people do with them
Yeah but for e.g. sorting systems this may be problematic
It can eat hdds or slugs or something like that
Didn't think of that. We always tell people to scan those on the spot for good reason
Slugs and powershards probably have different importance to different people. I cheated several containers full of shards because I use them in everything. So slugs are just decoration to me. I think most other people use them in miners
And miners only. Iono. Storages and autosorters arnt my thing lol
ye just tested it still goes to 3 and the limit is still 9
new ui is fancy though xD
ok interesting if you have 2 overflow outputs it actually waits until it has 4 and then outputs those 2 at the same time so the buffer is 2
i always thought it goes left-right-left-right
oooooooh @frosty owl i finally understood thats what you meant with prog. splitters for sushi xD
I think everyone initially thought programmable splitters were different from what we actually received.
i dont even remember what programmable offers over smart
Smart is 1 item sort per direction.
Prog is up to 3 per direction iirc.
wiki says programmable can have 64 rules in total, i suppose that helps with making sorters
@fierce ruin do you already have some knowledge about signals? the current wiki page on them is less descriptive that their ingame description
Let me get my reddit link.
https://www.reddit.com/r/SatisfactoryGame/comments/qgy9z0/signal_logic_rules/
That people have downvoted because accurate information has no place on reddit.... π
@deft lichen if you need anything more complicated verified/explained just shoot me a dm.
I tried to keep that post as simple as possible, so it is the basis on which all more complicated things are built, and you can deduce more complicated aspects from those basic rules - but if you need a more complicated rule constructed for easy of wiki understanding I can do that for you.
I heard you can actually sort for up to 24 items per output...
Dunno where the 3 can have come out from 
That's why I put "iirc" lol
Which apparently I don't. 
I just remember it being a "more options smart splitter"
And with the way I do sorting, I've never needed a prog splitter tbh.
Everything is just chained smart splitters.
It is, it's just not straight forward what possibilities the "more options" can give
thank you! it's already useful as-is and can be expanded upon
I mean... I would say any "expansion" is just combining what is there into more complex things like I mentioned.
Those are the base rules upon which everything is built. That was my intention when writing it.
The biggest one people from Factorio seem to need to keep in mind is that trains do the pathfinding.
I don't get how that relates to the overflow setting discussion though π π
@frosty owl I am still trying to figure out a place where I could "optimally" use prog splitters.
Because they accomplish nothing that chained smart splitters don't do.
And unless you're combining resources into a single container, you're going to chain anyway.
"Normal" usecases I can think of would be to sort items for storage or to take items from a mixed like to feed a manufacturing building (all cases in which you're saving almost 1 splitter per item type)
Are you asking for curiosity or trying to find some usecases right for your playstyle?
Like I said though, unless you're mixing items in the same container storage sorting will always need to be chained.
@deft lichen this Message is also helpful to understand the system in place #math-and-meta message
Depends on how you set things up.
Example: I had ~5 belts coming in my central storage, but only 4 inputs into the sorting system, each input connected only to a few items' containers. So I had to split-merge my lines to be prapared for the sorter and programmable splitters were just right for the job
Also keep in mind trains passing a signal calculate the distance to the next signal.
So putting one immediately before a station and immediately after a station causes it to crawl very, very slowly into said station.
Eg: one line of the sorter dealt with iron products, so a few prog splitters were set to merge all iron products from the input mixed belts onto the "iron sorting" lane
Oof. All of my storage is on a single line that just loops through the splitters behind each container.
Splitter shoots appropriate item into the container and sends all other items forward.
I disliked the throughput limitations of that approach, thus I made 4 lanes. I sinked everything from the factory, so it had ti be a decently fast system (up to 4 full mk5 in throughout 9
But outside of sorting, if you ever need to split off materials from a mixed belt to feed manifacturers, it's always: 1 prog splitter Vs (minimum 3 smart splitters)
That's just to get the items from the main belt into the manufacturing building
I don't mix my belts outside of storage so that won't ever happen π
You're missing out :P
High tier items rarely are enough to make your belt-usage optimal (they leave belts emptier than fuller) without mixing
Manifolds for me are about shooting items down the line as fast as fucking possible.
Which is why the main line is always a mk5 and I do "appropriate" belts coming off the line to the machine.
I want the main line to be as empty as possible tbh.
Imo, that's what it's intended for (mostly), but people keep "misunderstanding" it since programmable splitters (that offer the extra value of easily avoiding the overflow "issues") are generally looked down upon
I mean I use overflow for "fill this thing and then send overflow and everything else forward"
Doesn't always sink.
Like for concrete where I have multiple ISC's of it, the splitters behind them are set to Conc ->, Overflow ^
So the containers fill sequentially.
Of you do overflow-mixing, you usually want both speed and for your main kjne to be as empty as possible (since overflow would get "wasted" in a sink)
Not trying to force this in you, just saying that your objective might not be incompatible with belt mixing ^^
If I am reading this properly, you just gave me a new way to think about it. So I will definitely test in U5.
Only issue would be when the sum of the items is greater than a mk5 could handle.
Yeah. Honestly, I find that part of the fun, but I can see why the extra care needed for your systems can turn one away from the idea
But belts have throughput limitations regardless, wether you encounter them by merging the same items or different ones, you just encounter the same issues with a different flavoring
You'd still want to split the input at the machine though, right?
You wouldn't want to be putting coal and ingots through the same hole on a Foundry, for example.
Unless you're looking for a doctorate in sushing or really in need to save on belts, I doubt anyone would want that
If we can attach splitters directly to machines.... the possibilities.
Machine inputs, not outputs.
i thought that was the point of mixed belts
Attaching to outputs would break things.
If splitting such belts was as easy as splitting normal ones, I would answer differently though. The point is that the time taken to both balance the 1:1 coal:iron belt and set the programmable splitter to then split it would make setting up individual lines faster/more convenient
If prog splitters had copy-paste functions though...
How long does it take to make a 1:1 belt though?
It sounds simple in concept but I know it cannot be.
I've passed the past few months testing if the "1 belt to feed them all" sushi approach is viable... 
It's a thing~
Because if there is a simple way to make a 1:1 then there are a host of recipes that can easily be reduced to 1 belt.
In the simplest case it can be a merge from 2 miners with the same clock. Else you first need to balance the lines to make sure the belt always has the same rateo of items on it, which can lead to complicated situations (eg: you rely on coal from trains, those collide and your coal gets cut off, belt jams on iron. You rely on overflow iron but your overflow reduces for whatever reason and the belt jams up on coal...)
There (might) be ways to make systems that only let items through if all the required ingredients come in in full amounts, but that's like SF fluid computers: nice concept, nobody will use it
(Can look for "Crazy Machines" on reddit for more on this)
Same clock part makes sense.
I'm saying unless they are the exact same distance you'd have a degree of "more of this one" for a bit, no?
So you just have to waste the first few cycles of runtime until it becomes 1:1?
Unless you made your beltwork taking that into account, you'd kinda need to "prime" the belts leading up to your "sushi merge", yeah
π€―
So many ideas for so many different recipes dropping.
Simplest ofc is steel production.
i think mixed belts is probably best for high level items, at low level you run into belt throughput issues, unless your scale is very modest
But this still leads to some convenient things... Eg with steel: iron is 50m away, coal 500, both going to a merge and a foundry. If you set them all on the same grid and turn on the power, the foundry would fill on iron (duh), but once you get the system running, it'll be impossible for it to fail unless
-The output fills up
-You manually interfere with it
Even cutting out the power balances out (to a certain degree) as items piling up on the longer belts will mean items lacking later when they'll be coming all the way from the miner
It's limited by mk5 capacity ofc, but even like Solid Steel Ingots can do 19 machines on a single line.
Which isn't bad tbh.
yeah, but i wanna pull that belt out of the miner, so i don't want to split it then
Bruh, you can have a modular frame factory where you take in 1 belt of ore, merge output ingots on the feed belt (if they're smelters that's a 1:1 anyway, throughout issues avoided ez), and keep doing the same until you get to modular frames ^^
Pardon my bad grammar, a bit in a hurry
I was with you after "bruh" π
Or whenever you have amazing compress ratios to make up for the inconvenience of keeping track of your mixing
Eg: mixing caterium and copper on trains for rather then shipping quickwire