#math-and-meta
1 messages ยท Page 530 of 1
I've only collected two hard drives so far
As you build, you spot more drives. Dont even need to explore, if you need highways accross the map
Mainly the HMFs so I can build more fuel gens...hence the concrete
On big projects, split the duty. Some fuel gens, some pipe work, some platforms... That way, you dont need all the HMFs right away
Hmm...maybe. have 20 running now with 10 coal plants also, and my first attempt to organize my starter base has gotten kinda bizarre
So I'm using the manifold method to "load balance". However, I'm using the overflow splitter for three separate lines. I have 227.5 iron entering and should be consuming it all. However, the last two constructors in the last overflow just aren't getting enough iron despite the rest being full. Is there a way to remedy this?
The annoying this is I can't load balance proper numbers without tearing up the entire factory and rebuilding it from scratch
I just don't understand because the math adds up but it's just not functioning
ohp we got a "oil is the rarest and most precious resource" pair editing the hard drive alt analysis lol
on the wiki
is that a good thing?
they moved electrode aluminum scrap to the "Situationally useful, but not resource efficient recipes" category cause it uses oil instead of coal.....
Maybe it's one of the same that wants to burn me at the stake for mentioning using oil to conserve another precious resource... like steel.
or copper lol
or bauxite lol
Yeah, I mean, why use a precious resource when you can use loads more of something else that's limited. I'm pretty sure if you just ask them, they'll kindly tell you that oil was put into the game to look pretty, not to be used.
haha
Oooh
nevermind the fact that the moment you start using it for serious production of RCU's, fused frames and cooling units.. it disappears so fast.. So what can you do about it? I was kind of under the impression that the wiki should only offer basic advice on how to do things.. telling someone not to use something sounds like personal preference to me
I just realized how train stations handle partial containers
They lift the FREIGHT STATION MODULE ONTO THE CAR
Instead of making the box to load onto the car
It takes the same time to load a container from empty and loading partially
the wiki has some guide-like parts, including the alt list
the alt list is definitely the most opinionated part
Technically arent all the recipes more efficient than their original counterpart
I think they're judged on the rarity of the materials used in the alt
And the complexity of the facility
I haven't found the page where they're expressing their opinion
I think the heavy residue into turbofuel is probably less efficient at the advantage of simplicity
some alts need other alts to work, well, or at all.
you have to look at the history on that page
I personally would take a less efficient recipe if it cut down on like half the machines
most alts do something, they're usually not all around better or worse
Yeah just the title sounds like opinon.. I read an "indepth" analysis on reddit once... the man called caterium circuitboards trash, silicone circuit boards the best alt, and then on the same page as he touted how great silicon circuitboards were, he took a dump on cheap silica
So you like spending more copper to make circuitboards? That's cool. it's a fast recipe.. but in your opinion I should burn more of my copper AND quarts? what..
lol.. Imma stop reading this anaysis right here as I think anyone who's run the numbers knows this is not true, it's more of a convenience "Steel Screw" is ultimately more resource-efficient"
I don't know, I'd rather trash the section completely. I think it's misleading to try to rank alt recipes rather than discussing their strengths and weaknesses. That's how you get shit where everyone likes steel screws cause it's fast but the meta is to be resource efficient so people conclude it must be resource efficient because everyone likes it.
Rather than just if you're gonna do screws, steel screws is just much nicer than casted or steel rod > regular screws
Yeah.. that's how I feel too, it should have a short description of pros and cons or someting. when someone asks about an alt, I try to tell them it's useful in x factory of y production but uses more of z
"Plastic Smart Plating: Involves Oil which is a limited resource, hurting its utility in late-game, but can be useful in mid-game as it is 5x faster and twice as efficient per Reinforced Iron Plate and Rotor used." LOL... you go make that 50 thermal rocket factory without it, Okay I'm done now
keep in mind people aren't editing this constantly, and a lot of these are from before the fluid update
fair, but by midgame you'd assume one would already have a decent modular engine factory, and before U4 there was no uses for them after that last unlock
Hah u wish
Im still making like 5 motors a minute
Cause I dunno how much i need to make.
SF has an issue of not giving you any clue how big to make stuff
I mean that's frankly cause it's mostly self-driven
I lied.. i might not be done "Coated Iron Plate, Steel Coated Plate, Adhered Iron Plate: All involve Oil which is a relatively rare resource" >.> yeah this whole section should just be trashed
Great idea, i think giving each recipe a stats instead of ranking them would be a good improvement
@topaz hedge perhaps you didn't know the history, the oil was much more precious back then
just a quick pros and cons, uses x or more of x to reduce usage of y, and a note if it's faster or slower than the original recipe.
Recycled plastic and recycled rubber has been around for long enough that plastic/rubber should be considered about as valuable as bauxite, sulfur and quartz.
No..
We have about 70% more oil than at the start of update 3
that's the big difference now
damn.
oil has gone from one of the rarest, to about mid-rare now
so it throws a lot of the old analysis for a loop
I like the rigor this guy puts into his analysis: https://www.reddit.com/r/SatisfactoryGame/comments/f6kova/my_indepth_analysis_of_the_new_update_3_alternate/
That's fair, and I suppose it's somewhat understandable.. kind of for those who played before U3 and can't keep up XD
but he's still working on Update 4 stuff, so no help at the moment
I have read that, i can agree 50% of the analysis statements
Yup that's the man that praised silicone circuitboards and shit on cheap silica lol
I agree though that if there's gonna be an analysis section there shouldn't be any ranking. It should just be factual what the recipe does. It's better to discuss with other people why what it does may or may not be useful than to fight that out on a page.
to be fair, cheap silica was very space and power hungry
Silicon circuit+cat coms is better than the other way round for a long time
Even ranking it by something objective like resource efficiency implies that that's the most important thing to consider.
it still is, but you can chew through so much quartz so quickly that if you can get more quartz out of something.. it's a no brainer.
Seismic Nobelisk : Just no. You don't need more efficiency for your boom booms. And you definitely don't want to deal with crystal oscillators being part of that chain!
Although there are some universal truths lol
no one likes seismic nobs
At update 2, recycled plastic was a trash. Just to name another example.
but we're not on update 2, or 3 anymore!
i feel like everything that adds a reagent of a higher "tier" than the normal recipe is garbage
we've moved on and need new ways of thinking. lol
Do batteries need to meet factory demand before they discharge?
@gusty nexus not for radio unit,
Like if I have a 500MW factory, do I need 500MWh of battery?
no
Cause i cant seem to jumpstart this factory
No
what are you trying to start?
nah, they will spend their charge to bridge power shortages regardless of how much power they are currently holding
Battery discharge as fast as your factory demand
So I have a fuel plant that needs 2000MW to jumpstart
the power storages need to be charged first, they don't come with power
and I have like...12 batteries for 1200MHw power
But when I activate the fuel plant it kills the fuse
They are charged.
what i don't get is how you are able to make a fuel plant that big without already having more than 2 GW
power storages should only be used for emergencies, or (other uses)
Well from what Id expect, a 2000MW factory should run for 1 Hour on a 2000MWh battery farm yes?
rushing fuel. So, even so if your batteries are charged, they should run your fuel plant for... half an hour give or take.. that should be plenty of time for it to start up?
re-check your power connections, i would say
Yeah, and your fuel plant.. be sure to standby most of your generators so the pipes can fill too
The fuel plant has a power switch that says Power Grid A is my main factory
and Power Grid B is the newly made fuel plant
alrighty then. lol well still, standby your fuel gens so the pipes can fill, and make sure your power storages are charged and connected to the fuel plant..
Power Storages can supply any amount of power for a varying amount of time
They finally admitted to the pipes having issues! sempai has noticed us
..part of me wants to reintroduce/undo the minor fixes I made to systems that shouldn't have had issues but did and send it to them.
No matter what, the fixed systems should work even after any potential patches
famous last words lol
For you, which turbo motor recipe is the best ? For me, it's Turbo Electric Motor because it's the recipe which consume less aluminium, right ?
it is best indeed
And the best recipe for radio control unit is the one with heat sink right ?
'cause the limiting factor on the game is the aluminium no ?
it can be.. least alumium rcu uses highspeedconnectors... the next best seems to be the radio control unit/system alt, that uses alclad casing and circuitboards
Most buildings require electricity, or power, to function. Power is produced in power generators (see below) and consumed by buildings. Power is transferred via Power Lines, Power Poles, or Train Stations and Railways. Power is measured in megawatts (MW).
Note: In the UI of some buildings, the unit of power is incorrectly displayed as MWh, which...
it is not surprising that nuclear power is pretty resource efficient
Surprisingly enough... That depends on your end-game goals ^^
If you try to get close to the max pressure cube production possible while adding other space elevator parts to the mix, you may actually need to switch to another RCU recipe just to skimp on the QUARTZ needed :rofl:
(Notice that you can use up a lot of quartz to max out allu ingots production in the first place
Note: the heat sink RCU units is the one that uses most quartz crystal with 12/RCU, the others use just 5 or even 3.3333...
my analysis for ALL possible Aluminium lines
if it didnt use sulfur, I2 would be quite good actually
kinda fun how the best current line is also the biggest
so, Instant Ingots might actually be worth looking into
if you dont need sulfur elsewhere
what machines are counted? the whole process or just the aluminum solution & scrap steps?
all machines needed. constructor, refinery, blender, etc
though i always rounded up
since you can only build whole machines
i should probably list the amount of refineries and blenders next to them
Instant using sulfur really kills any enthusiasm for it
the only thing it's getting you is less machines total?
also bauxite efficiency
thats why its green
the "best /worst" shows which recipe is best in each usage column
green is the best in each column
it's the same efficiency as sloppy/electrode though, and it's basically as many steps so it doesn't feel very "instant"
here, updated numbers. the ones in bracket show refinery and blender total
I2 only needs 1 blender and 1 refinery
and those are both underclocked
if we ever do get more sulfur, this might actually be worth looking into
definitely, it's right on the cusp of being resource efficient, and the saved machines is nice
I mean it's the compact version of an already compact setup, aluminum is already really small
which I guess is cool?
except for the SEI setup, thats the biggest one of all alu setups
i did the same analysis for turbo
I still stand by the opinion that instant shouldn't have a water byproduct. I feel like it would be in the spirit of the recipe to not need a recycle loop.
and guess what: the most efficient setup for turbo is also the "smallest"
heres the entire I2 setup for 120 ingots / min
yeah tiny, and 120 ingots is like nothing now. I'm gonna resetup my aluminum soon and I'm struggling to keep the parts per minute numbers low enough for the storage belts. Like before my 450 bauxite made like 116 alclad sheets, and now I need half that for 100-something of both sheets and casings
Imagine making silica in constructors 
i mean, you could also go an harvest it from plants if thats your thing

compare these 2: Mk5 Input, 1040 output
if the sulfur was comparably low, it would be worth it
buuut the high need just doesnt quite feel right
if we had more sulfur nodes, then it would be a different story
Maybe we will, they do make stuff with an eye for how the game will be in addition to how it is now.
Think of all those oil alts that make way more sense after mk2 pipes and resource wells
id honestly rather use nitric acid here
since nitrogen is a bit more abundant
or simply bump the output of aluminum scrap
that could make it compete with the other recipes
maybe, that'd add complexity though cause nitric acid is a blender thing instead of a refinery thing
sulfuric acid does have very little usage cases....
oh, also, it would bump it to tier 8
so its good that it only uses t 7 stuff
yeah I like sulfuric acid in it weirdly enough, it just feels like either:
A) it's waiting for more sulfur to exist
or
B) it needs to do more, like not have a water byproduct
the water isnt that bad
it can be fed directly into the sulfuric acid refinery
its 1 to 1
oh it isn't, I just think it'd be a cool thing it could do to justify the cost, greatly simplify the setup by removing the recycle loop
this is actually the simplest loop of them all
since the sulfuric acid refinery can run 100% off of the recycled water
oh really?
he just loops it back on itself
that's weird lol, it doesn't need that much right?
well it's just assigning resources to where they are missing in no particular order, it doesn't "magically" figure out good usage of resources
so you just need some seed water to start the process then it feeds itself, that's cool
yeah, thats really neat
funny enough instant scrap is the recipe greeny's tool pulls when you maximize alu on a low scale
it switches to sloppy/electrode if you switch off maximize
I think it just gets to instant first during the maximize solution
nvm, he uses it for maximize even with the entirety of bauxite
remember maximize only solves for most, not efficient, and instant & sloppy/electrode are equally bauxite efficient
heh, you only need about half of all sulfur to maximize alu xd
kinda highlights how bad the sulfur addition is though: half of all sulfur vs 1/10th of all oil for max aluminum
uh nuclear?
I mean yeah there's maybe an argument that sulfur isn't all that precious cause there's not many uses
2100 sulfur for max fuel rods
thats..... kinda cheap
plutonium doesnt need that much sulfuric acid afaik, right?
depends on your recipe for turbofuel, I'm producing 3600 turbofuel with only 1800 sulfur
hmmm, but if you go nuclear you could theoretically just shut off your turbo plants
moar power
oh right and batteries for drones
aah yes
sulfur cap is 6840/min, right?
ye
so sulfur uses:
- turbofuel
- nuclear
- batteries
- munitions (minor usage, don't need much)
- aluminum

oh is sulfur actually used for aluminum?
instant scrap alt
I didn't know the instant scrap recipe was actually good
it can be, it's used in instant
oh it can be but doesn't have to be
that's what we're trying to figure out
I see I see
i guess thats the question here: with all things considered, would going for instant scrap be worth it?
is the reduction in buildings and complexity ( you win @frosty owl ) worth the sulfur usage?
currently turbo needs 50 sulfur per 100 fuel
the others are more (normal is 0.8:1, heavy is 1:1)
heres a full turbo list
what's the meaning of the colors?
makes sense
its column-wise
ahhh I see
normal turbo is actually the worst of them all
diluted turbo blend ftw
but heavy uses the most sulfur
brb two more rows of generators to hook up before I can work any further on connecting this stuff
I love the fact that I have 600/800 generators hooked up and it shows my capacity is 0 MW because there's nothing on the grid to draw power
i finished my 60 turbofuel gens yesterday
60? rookie numbers
270 turbo / min
its a starter for me
was my first dip into some sort of "medium-large" build
9000 MW is nice tho
all my factories are on draw io

for nuclear it's a little more complicated because it depends on what recipes you use for plutonium, but with default recipes for plutonium it's 226.67 sulfur for 4 uranium rods and 1 plutonium rod, and with all non-fertile recipes it's 246.47 for 4 uranium rods and 1.78 plutonium rods.
both use all alts for uranium rods
i guess the question is: how much do you actually need to invest sulfur into power production
because thats kinda all it does rn
produce 1000 nobelisks per minute
only needs 1500 sulfur
well that's between 330.88MW and 383.41MW per sulfur for nuclear. So I'd say probably not much unless you're really maxing out the map
that still leaves more than 5000
also the big question mark is how much you want to use drones and need to invest in batteries
max points, but honestly there's very few people who do max anything
lol, according to the calc, you could do 50 uranium fuel rods/min, 1000 nobelisks/min and maximize alu and still have like 600 sulfur left over
honestly? imma say it:
Instant Scrap is a valid and viable alt
im now absolutely gonna make my next alu factory an instant scrap factory
If I go nuclear and use the waste to make plutonium to sink (no kept waste) what's the NET power back to my grid?
I know the particle accelerator is a bit power happy, just wondered how much the whole plutonium process added.
if you underclock the particle accelerator, its easily like, quadruble the power back
buuut if you sink them.....
well... nothing gained
a single nuclear power plant needs only like 1 accelerator at 13%
and thats basically no power at all
so really you only lose like 10%
just want a "clean" nuke power station, lots of plants with all waste going to make plutonium rods just to sink to eliminate long term waste issues. I think it should be viable ๐
using a switch room so I have the option to stop minors, sink what's on belts and "make safe" too.
everything is a viable alt ๐คทโโ๏ธ depends on a purpose ๐
i mean its a valid competitor to the current "best" alt path
this should give you some more info
no 'need' just want a new power project.
Turbo fuel is dead, I have a huge fuel gen plant already, along with a few coal gen plants dotted around.... wanted a nice nuclear option.
it really compresses the waste down
125 waste / min translates to 30 pellets / min
Wow! nice, it uses lots of waste, looks like it will be perfect ๐
aaaand that translates to 0.5 plutonium rods / min
so you need like 250 waste / min for 1 plutonium fuel rod / min
I should be able to build a decent plant and use waste in "real time" to eventually sink.
Switch room to turn off minor, turn on sink at beginning of factory quickly flushes uranium ore, then wait for the power stations to burn up the last of their fuel and have batteries to power everything to continue to make plutonium rods and sink them.
should be able to shut it down "clean" at the end.
it's 200:1 waste:rods with default recipes though...
37.5/min waste makes 50 non-fissile x2 = 75 waste + 25 waste for the 30 pellets
Is not the easiest ration 2 uranium fuel rods to 1 plutonium fuel rod?
4 to 1
Question: do you think ceiling lights should have more "range"? I do, and here's why, referred to picture (sorry I didn't get the walkways out of the way first)
At night, this factory gets pretty dark as it's the bottom floor. My idea was to use ceiling lights to light up the beltwork between the manifacturers and walls but... the lights can't reach down there (14m) :
||I moved the lights directly above the conveyors, no difference||
Now THAT is convenient 
So 40 uranium fuel rods, making 2000 waste, into 10 plutonium fuel rods?
with default plutonium recipes, ya
Thats allowing Satisfactorytools, to choose recipes
Which for some reason includes coke steel... Yes, it might be more efficent, but what are we going to spend coal on, if not steel?
range already got increased
the reason they dont overlap more is because overlapping is performance heavy
????? are you just being salty over the tool's recipe choices? the numbers don't change
Aluminium?
Just saying, I didnt do the math, just allowed the tool to do it and the numbers works, assuming its 2000 waste
Tool doesnt care about personal preference, it just calculates based on resource abundance
I don't want them to overlap, just to be able to shine to the bottom of a 4 wall high building :
4 walls seems reasonable enough, considering anything more is for refineries (where you could fit the wall lights)
Check out the "spoiler" part
"I moved the lights directly above the conveyors, no difference"
Meaning that even with lights directly above the conveyors, they don't reach
increase brightness? xd
Scrap with coke, uses less bauxite, while for steel, coal is better (2:3, instead of 3:4). But guess I will just uncheck coke steel and be done with it
but yeah the coke steel being preferred in the tool now is a result of it just measuring weighted resource rarity, and not accounting for what the resource could be used for. So in absolute weighted resource numbers coke steel is better, but I would suspect most would prefer solid steel because steel is the main thing coal is used for. So no point in cutting it out entirely.
change color values itself?
...erm... haven't tried. Is that even a thing? o.O
Solid steel is good because you set it up before coke steel
Why overhaul what you already have lol
it takes very little oil to make coke, with the HOR alt the ratio for raw resources is actually 0.75:4
oil:ingot
btw, they did update the light stuff. did you reset the optionn in the menu?
it's pretty hard to come with objective rating of resources ๐คทโโ๏ธ
Cause with most end-game production plans (i've seen) you can finish coal faster then you finish oil, making coke steel a good choice even just to make more iron products (steel coated plates, steel rods, steel screws)
Which option?
This is a screen I took today, btw
oh I agree, I think beyond weighted resources it gets subjective really fast. I think it's fine as is and the user just needs to understand how the tool works and make the best decisions for their own factory. there's a reason you added turning individual recipes on and off
And thats why its an option to select recipes. Just have to remember to check, when I ask why so much oil is needed, while not seeing coal used at all
light quality option in video settings
what i had before the update vs what i had afterwards
I set that to max just in case, bt that's how the screen turned out
Ohh, maybe need a restart even for that?
I think it would be even better when I add the ability to manually change weights
Wait, so they DON'T reach 
they did before
but you need to update the options
Rebuild the lights. The game updates light distance under weird situations
You mean I need to restart the game after changing the quality to ultra or...?
maybe
I think I tried already, but I'll try again after restarting ๐
The end result was what I posted in #screenshots (B) , so without the walkway in the pic earlier the lights are directly above the conveyors
Noticed that when I was messing with glass and lights. The light kept there short light distance even after removing the glass. Had to change color back and forth, to force an update
worked like a charm xd
The fact they don't shine through glass is a whole other disappointment I have >.<
I mean, what's the point of placing lights (referring to screenshot, where the beltwork is below the glass) if I still need to turn on the flashlight to be able to see the beltwork through the glass 
place lamps inside
Too big
Guess its a resource saving. Counts glass as walls/foundation, when it comes to light
It's just 2 walls high below the glass, same distance to the ceiling as the manifacturer's belts
To which I agree... but at the same time, it kinda sucks since belt visibility below glass became even worse with lights
Eg: since they don't shine through, even if you place a light just above the glass, visibility worsen as you now have the reflection in the way but the belts are still "dark"
Turn them on and off, and they'll stop doing so... รน.รน
To be fair, this is the first version of the lights and they didnt know how resource intensive they where
Real world data vs there internal testing data
it seems that the light updates whenever its turned on / off. because WHEN i placed the foundations, the light shape changed.
But when i removed the foundations..... the light didnt shine to the floor again
Turn on/off or mess with the light controller. Both makes the light update there draw distance
That is fair, and it makes me think I should try pushing for this on the QA. It might be doable, considering it doesn't need to affect too much transparency at once (thanks to LOD)
In other words, current lights care about build order. Also does it update on load?
good question
I'll get back to this next time I load the save (and the game should load with max settings on too this time)
buuut anyway, did this fix your short range lights?
besides the glass thing, of course
I'll check after I restart so... next time I boot up the game (probably later on)
I don't usually Discord in game since it tanks performances 
ive just glanced over some more alts, and heat fused frame is a bit odd
you invest more nitrogen and fuel so you consume less aluminum
Hey guys!
So quick question bout the power changes to U4
I know the changes burn fuel at 100% no matter the power requirements, but does that really include if there is 0 power consumption?
If they are hooked up to a grid, they consume resources.
Dude...
So the only way to backlog a system is to manually disconnect each individual gen
thats dumb af
So: if they have no power connections, they can fill up their buffers. You may also be able to set them to standby: I haven't tested that.
yeah but at a large scale? Thats gross
This is why I have gotten into the habit of building everything but the power lines first.
well resources are infinite, so not sure why should it matter that they are being burned for power ๐คทโโ๏ธ
i get that but youd think that if there is no power consumption at all they wouldnt need to turn on.
I do appreciate the nuances of the new system and managing it but this sucks.
I think the complaint is "it's hard to pre-fill a manifold without individually disabling a bunch of machines".
Oh the fact they burn at 0 consumption means my power plant is burning at like 250% production capacity.
I built my powerplant larger than my production cause under the old system i didnt burn more fuel until i actually connected the floors meaning i could set everything up and add floors as neeed
Now instead i need to go through everything and disconnect floors via... Well the pipe mains i guess
You can just cut the power to those floors.
Well the gens still get connected to the grid ON those floors.
Meaning they still burn even at 0 consumption, hence my grumbling
Ah, right, briefly forgot.
Urggggggh thats dumb
That said, there is a QA site where you might be able to suggest "power plants should not consume resources if there is zero load".
You can send Questions, Feedback, Suggestions, and Bug Reports at https://questions.satisfactorygame.com
<3 @iron prairie
was actually heading there now ๐
Thanks for booping the bot though
You can cut off the flow with valves to stop the generators and add some buffers in strategic places to help the manifold fill up when you need to have them run again (eg: add a buffer at the END of a manifold, let it fill up, use that to help refill the manifold when needed
Thats my plan atm.
1 valve / 9 fuel gens is easier than 162 standby switches hahaha
Still a bit annoying ๐
Thanks for the suggestion though!
Any suggestion on the pipework? I'm literally too dumb to find a way to place the belts/pipes in a way they don't block each other on both sides
Stack the belts on top of each other, then above the splitters, throw the piplines above each other.
Just checking some math here: a single maxed Uranium node will produce 180 GW, correct?
600 Uranium : 480 Encased Uranium Cells (Infused Uranium Cell) : 14.4 Uranium Fuel Rods (Uranium Fuel Unit) : 72 Nuclear Power Plants : 180,000 MW ?
I've got a question about the new power changes to U4, specifically with how to deal with fluid networks running at 100% all the time. I am making 1600 turbo blend fuel per minute, divided into 3 pipes at about 533.3 m3 per minute. These go to 3 floors of fuel generators, each with 59 fuel generators overclocked to exactly 246.266%, yielding 300 MW per generator. the fuel consumption per floor is 531, yet there are generators that are perpetually empty. I have a large buffer at the start of each floor, and I filled each up before connecting the power lines to each generator.
really not sure how to fix the issue to get all the generators to function properly. Overall its a somewhat trivial in terms of the fluctuation of power being generated, but it's annoying that I can't get the math to work out in game
Non-biomass generators consume resources as if power draw is at 100% always now
Easy solution: have a buffer at the END of each floor too, and fill that up before starting the generators, then let it help out oversaturating the pipe network
5 buffers per floor work too 
Do be aware that with turbofuel, manifolds can take a long time to fill up.
I recommend using standby (or not plugging them into power) to let the internal buffers fill up.
I've tried that method, letting all of the manifold fill up before connecting any of the generators to power, but given enough time, I get red generator outages at the same places in the manifold
seems to be some sort of flow issue I don't understand
Mkay, that indicates to me that there's either a transportation or production failure somewhere.
To rule out a production failure very quickly: check your oil extractors. Are they running at 100%?
Use that in conjunction with the buffering I suggested. The issue (probably) is that there is not enough fluid in the pipes to buffer for the fluctuations in fluid needs (maybe generators were at 49/50 full and that can be enough to empty the pipe and have the last gens in the row starve for a while)
Though, a distribution issue can't be ruled out yet ^^
There has been talk about pipe manafolds behaving weird, once you reach a certain amount of splits
So maybe running 2x 300 pipes, instead of 1x 600, would solve the issue
That... really shouldn't matter, though. The earlier machines in the manifold will sip small amounts of turbofuel, and each individual pipe section won't be losing much per second.
I will try adding in more buffers
Overall: I'd trace backwards from the failing generators.
Is there fluid in the pipes?
Are the turbofuel blenders/refineries running at 100%?
Are the fuel refineries running at 100%?
Are the extractors running at 100%?
If the gens aren't fully loaded when the system starts, I can see the last ones in the row starving eventually, with that little overflow/min
Turn of first in split and see if a failing one, starts running. Thats tells if its a production issue or a pipe issue
And, for that matter, is there a hidden section of T1 pipe where there should be T2 pipe?
He's reported starting from completely full internal buffers & pipes.
My bad, must've missed that
the blenders making turbo fuel are running at 100%, there are 14 of them running at 250% (112.5 m3 per minute each_, with one running at 26 m3 per minute
"Egon, this reminds me of that time you tried to drill a hole in your head."
Wait, no that's not what I wanted to say.
Egon, I'm working up to that too. Got 4 online now. But I want to do plutoniuim first.
i always will recommend doing this
^^ I endorse thsi
at the halfway point between the 5th and 6th machines, there are valves that limit the flow in either direction, so that the overflow can pass to the other line,
I had tried to set these to the exact values I wanted flowing in each direction, rounded up a bit to allow for extra flow, but the machines dont produce continually, so that wont work
so 2 fuel lines are taking 5 and a bit full blender outputs, and 1 fuel line gets the extra bit, plus the underclocked blender making 26m3 a minute
heres a cutout of the pipes leading to my turbofuel gens
In terms of pipes, the manifold is looped back on itself, and I've tried a number of configurations. I tried making the manifold a continutious S shape that fed all 6 columns of generators, I had it branch out and feed all 6
270/min turbo
You need some buffering to have that split work properly, else the flow fluctuates left/right without splitting the way you want
honestly, just remove the valves
i had systems with valve and they instantly improved once i removed the valves and it just balanced itself
if I remove the valves I run into the 600m3 per minute limitation of mk2 pipes
since they share a pipe
just try it
At least in theory, once the pipe gets overloaded one way, the fluid has no choice but to go the other way.
โ๏ธ
true
I was thinking that that should eventuallly happen with the valves on the 6th machines
since eventually they should balance the network out if they are overproducing, and there are buffers at the start of each floor
valves are wonky
I'll let this run for a few hours and see what happens
and I will add the rear buffer like was suggested, thanks guys
Does any one have a good resource for a starting petrochemical setup?
as in without any alternate recipes?
then you just run 2 refineries per 60 crude oil being produced, alternating making rubber and plastic, and turn all the heavy oil residue into fuel
when positioning each "layer" of the refineries (i.e. the refineries processing the crude oil, leading into the refineries working with its byproducts), leave yourself at least one foundation's worth of space between them
Maybe 2, just to give yourself a little wiggle room.
Yeh without recipes at the moment, ok cool thanks
also since you won't have fuel generators to start, you'll actually want to turn that HOR into petro coke and then put it in an awesome sink or burn it in coal generators
Once you do get fuel generators (which, as a warning, cost HMFs and computers), residual fuel is slightly more efficient than petrocoke -> coal generators, plus being more compact and simpler.
Anyone experienced or heard about weird light issues (light not shining through glass or not reaching far enough)?
Not sure if this is a better question for here or the question section, but would. 5900x CPU have more consistent performance than a 5600x since it has twice as many cores and a slight clock advantage? I know it's been said that CPUs with lots of core don't tend to provide much more uplift, but could it provide better consistency?
doubtful, 5600x is already really good and having extra cores isn't going to make the game use 'em
Technically yes, actually. IF you can get a notably higher clock speed on the CPU, which is technically possible because they bin those parts a lot higher. Practically, however: no. Your bottleneck is going to be the speed/quantity of your RAM/memory, as well as the speed of the drive you're playing the game on.
I'll add your GPU to that list, actually.
Yeah this game likes good hardware all around, wouldn't say it taxes anything in particular
...yeah, actually, now that I think about it.
Sticking to 1080p and having a high-end card is probably the only way to make the GPU not a bottleneck, and that seems a bit of a waste
Mhm. I have a pretty mid-tier system, and with optimized graphics I'm starting to hit lag spikes in my main factory.
Modest Tier 6 starter base, nothing too insane.
I get 60s a lot, but there's trouble spots all over the map that make it dip into the 30s and 40s, including parts of my factory. I've messed with the settings, but honestly they don't really help. I just think my 970 is showing its age.
The problem is I keep running out of VRAM, and that's true for pretty much any game I have problems with.
Two things I did that helped: turn off hyperthreading, and let NVidia optimize my graphics.
Surprised the heck out of me when it maxed my graphics and I got better fps...
What do you mean turned off hyperthreading? Like completely or just disabled it for Satisfactory?
Completely. I have to turn it back on for other things, but that takes me from a bit choppy to smooth.
...I think I can reliably explain why that helps in factory terms if you want.
I vaguely get forcing it to use physical cores instead of simultaneous multi-threading could be a boost. It's hungry for resources and making it share more than it has to probably doesn't help
Mhm. The factory version is that hyperthreading is like two Mk 2 belts instead of 1 Mk 3 belt when it's off. You can move more kinds of items simultaneously, but sometimes you just need to push 270 ipm.
Do you have an intel or AMD chip?
Ryzen Master makes it as easy as two profiles and a reboot.
Also lets you turn on PBO pretty easily and tweak your memory speed.
I wish there was a way to tell it "hey this program needs it's threads on different cores"
or for the program to do that
Same, but that really does need to happen at the BIOS level, so... hitting a button and rebooting is about as good as we're going to get.
How important is having hyperthreading on in general? Like would I really notice turning it off in non-gaming applications if I mostly just play Satisfactory and have Spotify or YouTube on in the background?
the thing is it helps in some games, so it's not like a simple "this will help gaming" kind of thing
it's very game-specific whether it helps, hurts, or does nothing, and most of the time the difference is like 1% or something
I'd definitely not leave it off, the processor was intended to perform mutlitasking best with it on
Jesus christ
First aluminum factory weve made
and for 4 materials, the setup for a 240 ingot facility is surprisingly small
Best part is how it self recycles
I dont know how
But im getting overflow of aluminum ingots
despite the foundries making 240 and these 4 assemblies using 240
"Invisible rounding"
Is the Instant Scrap recipe any good? Comparing it to regular foundry recipe + electrode alumunium scrap seems like its just trading out 78 crude oil for 520 coal
doesn't really feel worth it but i might be missing something
i did an analysis of that yesterday. its as efficient as a combination of the current best alu alt recipes
aaaaand its unlikely that you will run out of sulfur, even if you maximize aluminium production
its the smallest setup for alu possible
Heres a chart.
How to read: the top row shows usage of resources per 120 alu ingots/min
Each column is the overall usage of that resource per row of recipe
Coloring: green is efficient usage, red is bad
i did consider this too. but with the right recipes, sulfur isnt even a bottleneck
like, max aluminum barely even uses half of all sulfur on the map
and not even nuclear uses that much
I'm still wary because of the X factor: batteries. You could spend none or lots of sulfur on batteries depending on how much you like drones.
this is where balancing comes back into play
assuming we did max nuclear and alu and some nobelisk stuff, we have around 900 sulfur / min left over
huh, really
thats enough for..... around 600 batteries / min
that sounds like more than enough for lots of drones
yeah
so.... sulfur isnt actually all that limiting
if you tap into all nodes, that is
but im barely even using one node
so im definitely gonna go for an instant alu factory next
i c
alu is needed a lot in the higher tiers and its more limited than sulfur, and unless you actually try to maximize everything, its unlikely you will run into bottlenecking
actually with the new bauxite nodes sulfur is now rarer than bauxite
but the point does still stand that everything in T8 wants aluminum, and a lot of T7 too
whereas sulfur is just power, batteries, and munitions besides aluminum
and you don't actually need that much sulfur for batteries for drones
that's what I'm hearing at least
45 / 50 sulfur per min for 30 / 20 batteries per min, respectively
each drone needs between 5 and 10 batteries per min right?
4 per trip +1 per km
is that round trip distance for the +1 per km, or is that one-way? I'd assume the former, but who tf knows
@oblique hollow I figured out the light bug. Now I can finally post good screenshots 
ery good
Did you include power usage compariosn, out of curiosity?
hey guys, im running some nuclear reactors and they keep running out of water for some reason. the water extractors are overclocked to 300 water/min , two of them go to one mk2 pipe then that pipe splits in two reactors, have pumps and stuff, no buffers, anyone knows?
just checked real quick, and instant does need more power
I've seen the same, but haven't spent much time to solve it, figured I need to let them standby for a bit to fill up first and then they'll run well. Are yours also feed water from below (i.e reactor input is also highest point of the pipe)?
yeah, waters down, reactors up around 30 ft , the pumps do the job no problem
max throughput on pipes can be rough because of rounding issues
try using two separate mk2 pipes, one per reactor, and see if that fixes anything
generally speaking you don't want to be consuming at max throughput
what will cost my sanity more, fixing pipes for 164 reactors or underclocking them all to like 95~98%
I think knowingly putting yourself below 100% would cost you more.
so is turbofuell worth it now? or can i skip that?^^
I'd argue both turbofuel and nuclear are worth it.
Nuclear has a lot more processing involved, and as such, has higher power output.
well i have only 48 right now, i can try to fix it with more pipes, no problem, but i dont really know. @mossy pike the stanby solution works for a long time? or at some point they will come over the same problem?
okay nice ๐ any alternate recipe for turbofuell for using?
this is what I personally am using in my turbofuel setup I finished yesterday:
https://cdn.discordapp.com/attachments/835944123549876265/835944377121505280/unknown.png
required resources per unit of turbofuel:
normal:
- 0.45 crude oil
- 1.2 water
- 0.8 coal
- 0.8 sulfur
heavy:
- 0.938 crude oil
- 1 coal
- 1 sulfur
blend:
- 0.75 crude oil
- 0.333 water
- 0.5 sulfur
Blend uses more crude oil than normal, but no coal, less sulfur and less water.
However, since they added 1800 more crude to the max cap per minute with update 4, I'd say that added crude usage is worth it.
thank you!
what I find interesting is that you could do that exact setup I provided AND max nuclear and still have enough sulfur for your nobelisks and rifle ammo.
๐
batteries are for chumps, we got nuclear
that setup above: 1800 sulfur
max nuclear: I think somewhere around 3100 sulfur
max sulfur/min on the map: 6840
if anyone who has been looking more into the max nuclear math and would like to verify that number that'd be great
so, for reinforced iron plates, iron wire and stitched iron plates recipes?
for resource efficiency, yes.
Start by limiting iron input to even number (what you have)
underclock
Question on the Non-fissile U alt "Fertile Uranium". It produces same amount of water as its consumable needs (15 N.acid, 25 S.acid -> 40 water). Can I (and should I) rely on the water to circulate, or will rounding errors come and bite me in the rear?
Add sideways buffer, prefill to half and observe if it goes up/down?
was basically what I had in mind, just as in-line buffer that I would pre-fill before turning on.
you can underclock machines to match the percentage as well
yes!^^ this option was never in my head ๐ thanks you again for this^^
for example 3.86 machines means 3 machines at 100% and one machine at 86%
or 386 machines at 1% 
yeah ๐
ultra low power build
ultra low fps build too lmao
^^
Sideways. Inline will start sloshing, sideways "should" even it out
two machines at 193%
Fluid mechanics in a nutshell, was hoping valves before and after would solve the sloshing issue. I'll do the sideways one.
games are poor reality simulators, unless you're playing Unverse sandbox or the like
I have seen several youtube videos that demonstrate universe sandbox has limits lol
usually hilarious limits when you do stupid shit
Realtime simulations are limited by hardware capability
The only accurate realtime simulation of universe I know is The Universe
its simply unreliable because buffers mess with pressure
parallel buffers is always better than serial
with the exception of large pipe manifolds, all other aspects work pretty alright
aaaand the occasional funny bug
Mk 2 Pipe to Heaven, Abused Valves, Wall Holes eating Head Lift.....
I see you didn't really get an answer for this, by far the most reliable way, is to use the byproduct water in wet concrete for encased cells. and give your sulfuric and nitric acid their own freshwater supply
Remind me, its 2 engines for every 3 freight cars right
Thanks
So what's the advantage to fertile uranium? It's not an efficient use of uranium ore, does it decrease other resource usage?
it saves on sulfuric acid too
I dunno if this is more power per plutonium tho
I did, given an amount of uranium ore you get less power if you go the fertile route
I guess its purely just trading power to save on a resource.
you do end up decreasing the uranium rods you need to make so all the resources there (besides uranium) are decreased, though you end up making more plutonium rods
I guess I need to get the numbers for given an amount of power what are the raw resources usage of both systems look like
it seems to be comparatively uranium efficient, actually
it's not
ah, i ran my math for the wrong thing
gotcha
the split for uranium ore btw is 5/11 to uranium rods and 6/11 to plutonium rods
that results in 0 uranium waste
btw it's less uranium efficient compared to instant + fuel unit, I haven't looked at the others
I did that comparison cause instant + fuel unit is the most efficient use of waste so it gets most power
too bad there's not a way to make non fissle directly from uranium. skip the whole uranium chain and jump straight into that sweet plutonium powa XD
that'd be broken ha
Isnt plutonium supposed to be less efficient than uranium anyways
it was. now it's 2x the powa
Coffee made it twice as efficient???
nah, they shuffled some numbers around to make it more powerful per rod, but we make less rods
increased the energy in a plutonium fuel rod by a factor of 4
during experimental
It's a trap.
but yeah, that made it so we have 4x less rods lol
it's literally was just a "it feels good" change lol
Makes lil sense imo but I guess it works lol
As long as its more convenient gameplay wise
I mean they decreased the waste too, cause the way they handled the numbers made it so it effectively halved waste
Plutonium breeders are supposed to be recyclers of uranium byproduct, not an an alternative
yeah their job is to decrease waste and make extra power, and they do that
My honest opinion is that it's not worth it to do plutonium when you can have clean uranium.
I disagree completely, you're giving up a lot of power to do that
if you hate waste go for it, but waste is not actually a big deal
...I'll listen. How much waste does plutonium power produce?
1 rod produces 10 waste, ya
Mkay... grabbing numbers...
remember ISCs store 24,000 waste each
10 per rod, 3.6 from the inefficient line... okay, 36 per minute isn't bad.
yeah the less efficient you are with the uranium waste, the better reduction you get. it's a tradeoff so you can adjust to how much waste you're comfortable with
Mkay... checking more numbers.
even the least efficient waste reprocessing line nets you ~50% more power though, and reduces waste to 5% what it was before
Hmm... that would put my design up to 270 GW, and be viable as a later expansion if needed. All right, I'll keep that in mind.
Also, if the waste ever bothers you, sacrifice a doggo xd
We live and die by your orders, boss
... or other more secret waste removal methods that can remove containers full of waste.
Of course the smart play is to setup the waste storage and power plants, but sink the plutonium until you need the power.
Mhm.
Clearly we just wait until Tier 12 where we just send it into an infinite pocket dimension building or something
That way you have a bunch of power you can just flip a switch (figuratively) on to get, but aren't building up waste in the meantime
before they did the "feels good" balance my powerplant was producing ~1.2 million points a min.. now it makes like 350k lol
probably why they made those changes.. they felt like it was too rewarding to have clean nuclear lol
How much waste and how much power would a full Plutonium Rod Stack yield?
at least three

isnt it two waste per minute and each rod goes for ten minutes?
10 waste per rod
And yes, 10 minutes
so that would be 500 then?
a full stack of 50? could be wrong but... 20,833 MWh and 500 waste
So 1 stack per stack.... Neat
so I've never actually messed with uranium stuff before, and I wanna make sure I have these steps correct:
Uranium -> Encased Uranium Cells -> Uranium Fuel Rod -> Uranium Waste -> Non-fissile Uranium -> Plutonium Pellets -> Encased Plutonium Cell -> Plutonium Fuel Rod -> Plutonium Waste
Yea
How do I do overflow pipes
Gravity
and it should power ~8 reactors for an hour
I got 1166 Alumina Solution toprocess, but clearly each machine does not make a perfect 300
or 50 reactors for 10 minutes lel
(where the 20833MWh comes from)
So I need one of them to slightly overfill one pipe before moving onto another
How do I accomplish this?
Well, the man asked how much power so. XD
Choices were... 500 reactors for 1 miunte, 50 for 10. or reduce it to MWh xD
8 m means 8 m vertically up
how about 123904823749876 reactors for .00000002345435 seconds
i think thats the most relevant value here
sure why not.
im procrastinating doing my homeowkr by saying dumb things online
Seconds per Frame per Stack of Fuel Rods xd
:c bad!
Go do homework
ok
Should work. Blue is the overflow line
yes
If you worry about pressure, put a pump before the junction and 8m bump
And dont worry, it doesnt change the setup
It will still work as expected
Yeah pipes work really weirdly imo
Im guessing on a junction that points up and forward
liquid will bias forward before up
Well, pipes dont work weirdly, they work realistically, but realism and games dont always mix
Its a half mix
so i gotta ask lol.
that and conveyers defy gravity
*mostofthetime
*allofthetimeexceptwhenyouscrewitup
WhicheverIsAwkwardForYou
is the alternate copper motor recipe worth it?
You mean copper rotor?
yes ^^ sry
Im more a fan of steel rotor
Buut it doesnt seem too bad
But copper sheets do need more constructors for the right amount / min
hm oki^^ and thanks for your pipeline guide on reddit, i like it!
I never used it, but it does seem quite ok
ofc you lose the advantage of making both rotors and stators out of the same things
how does copper rotor compare to steel rotor?
that's the reason I liked to use it
It uses copper sheets and screws
okay it's time for me to stare at numbers until I can make sense of it all, see you guys tonight
i mean, resource wise which uses more
technically best way to make rotors (lowest resource cost):
Aaand maybe copper too
non-steel copper rotors
damn you cut out almost 20 iron by adding in a little over 7 coal
wow
thank you^^
never use water to make pure ingots
(if you want to do stuff yourselves, check the tool I used https://u4.satisfactorytools.com/production )
yes i use it since you released it ๐
normal rotors just way more resource efficient
you can pick any combination of recipes and it'll try to find most optimal way ๐
but never know whats alternate recipes are good and which bad
Is there an easy way to tell how much headlift I got
you can pick all you have and the tool will choose those that are resource efficient
Pump lul.
I think the copper ones are way more efficient, look at the coal usage
never thought about going back to using the default screw recipe because the steel rod recipe is so efficient before though
Refineries and such really need a freaking Head Lift gauge
Pump or tower.
I was hoping like how many conveyer stacks or something
steel rod + normal screw is pretty much the same efficient as steel screw (just slightly more efficient than the steel screw), so can be replaced with that
but this is fine
I recommend tower, itll save on electricity.
steel screw cranks out at silly speed though and thats nice
The default refinery pressure is 10 m, and thats around 4 stacked pipe supports
Right, so I have the aforementioned 5600x, but then GPU wise I have a 3070, then I have 32 Gbs of 3600 Mhz RAM and the game is installed on a NVME M.2 drive. So, aside from having a 3070 instead of a 3080, I have a pretty top tier rig. I'm happy with my FPS, I'm usually always around 120, but I haven't finished my current factory, so everything is off. I know there will be framerate changes once everything is running, so I'm mostly trying to make sure I have consistency down the line. If it get down to 60 fps, that's fine, but I'd like it to be a consistent 60, y'know?
@solar pond steel rod + normal screws
i think all buildings with a liquid output have a default pressure of 10m
steel screws
Second question, how to get rid of excess water from an aluminum facility
use it somewhere else or loop it back to the aluminum facility
Wont the alumina refinery back up cause theres nowhere for the water to go
thats a pretty marginal difference for the speed you get out of steel screws
Once the pipe is full
usually when people do this they have separated refineries for the loopbacked water
why would the pipe be full. set your refinerys to consume exactly the output
@umbral harbor Dear person who managed to find those parts and then afford them: you'll be fine. Throwing money at the problem won't help in this case.
@wind spade have you a paypal link on the homepage? ๐
The sloppy alumina recipe and the refinery to make scraps do not match in water usage/output
click the community link on top right, there's paypal and patreon
@loud heron thats what up and down clocking is for
Then theres not enough alumina to keep the water 1:1
I just got super lucky a couple of times. Nabbed the 5600x in a bundle before Newegg changed their policy on returns, and nabbed the 3070 right after the tariff stuff happened, so I think enough people balked at the price increase for me to rush through.
I was more curious at this point about getting the most out of what I have. Maybe overclocking the RAM or changing the hyperthreading settings
I don't really like putting the links in popups or similar. Hinders user's experience
as you can see there, balancing any part of that recipe is near impossible due to the fact that if I process more bauxite, I make more water, and if I process scrap faster, I run out of alumina solution for it to process and it backs up anyways
tho I'll probably put them in a more descriptive menu, like "support the project" or "buy me a coffee"
links?
Your ram should be overclocked from the factory. If it says anything higher than "2400" or "2666" for its speed, they already overclocked it.
was in response to Jeardius' question about paypal/patreon links
Turn hyperthreading off for this game, though.
No point to using the Ryzen Master or whatever AMDs RAM program is to push the RAM a little further beyond XMP?
maybe hardware discussions could go to #off-topic-tech ๐ค
^
No prob.
I still really need help with excess water
I thought this counted as Meta regarding game performance, my bad
find any process that needs water (pure ingots/whatever), put the water there
I suppose I could send the extra water to an ingot facility or something
Too bad we cant just sink water lol
gives a bit of complexity to late-game recipes I guess
@loud heron just use all the water from the output then set your water pumps to supply the rest
Im making 186 extra water
@wind spade , so how i can split the iron ingots with the amount what i need to get into the foundrys and constructers? hope you understand ^^
this is not gunna go well
Thats the point though Im making back more water than the pumps themselves combine into
186 m3 extra water just form the loop of alumina to scrap
how much excess water? small amounts <240 can be delt with via 1 limestone node, and 1 awesome sink, and 1 refinery making wet concrete
You just let it overflow
usually people just do manifolds
or you can figure out how many machines you need to produce 300 ingots
This makes sense.
Ill see if theres anything nearby
but! if it's best to figure out how to use valves, and get it to work, because larger lategame setups produce too much waste water to just deal with in that mannor
As you can see the scrap refineries return more water than the pump themselves combined
The sloppy alumina recipe be awkward
Excess? From what. Aluminum?
4.65 refineries of sloppy alumina consumes 930 water
Wait
4.65 refineries of scrap outputs 558 water
Is greeny's tool telling me that the return water is making up the rest of it?
yes
It was not obvious to me lol
I thought the system said it only takes 372 water to start lol
Greenys tells you how much fresh water you need.
the out put is 558 and the pumps are supplying the missing 372
Gotcha
The feedback water is already taken into account
Thanks guys, very confusing to me
water generator
More like perpetuum mobile
But an actual one, with water
yeah, this confused a few people already. Though not sure how to make it better
for me it was pretty clear when I made it, but I guess that's the classic "I made the tool so it looks good for me"
Maybe somehow make the water input from pumps tie into the <--->?
the library I use to make the graphs doesn't support that and also would be pretty hard to make it work with the ability to move the nodes
Extra water
So its clear that this is the new / extra you need to supply
if I'll make changes in this, I'd like to make them generic, not just for this specific loop
And add "recycled" to the feedback water
I mean its hard to tell without distinguishing the two conditions regardless.
Recycled and Fresh works for anything
because if new loop like this appears in U5, I'd like to have it ready for it
Acid, plastic, Rubber, Water, canisters, etc
Yeah. Recycled plastic is the machine and recipe node and "Plastic (Recycled)" is the item arrow
Same works for water and Acid and canisters
fresh biomass
Just so people know "ah, this is being reused!"
Maybe even change the color of the arrow
I'm wondering if it's even possible to tell what is "reused"
If the machine isnt an extractor, its reused
like how does it differ from just making a recipe that produces something to multiple places
With some basic Computer Science ingenuity, you could make an algoritm that traverses a graph and detects loops
see: plastic
Then mark the nodes and lines that involve recycling
yeah but it's a loop, so it'll get detected as "reused"
I guess you could maybe note the exceptions in the code, for instances where rubber feeds into plastic or vice versa, for example
Do a traversal to see if they match each others inputs and outputs
A loop should be equal parts going in a circle
Oh, yeah, sum checks
I could, but as I said, I'd like a generic solution, so that I don't have to manually do stuff every update
right
I still think the (if its not a production building, its recycled) works
there's a lot of stuff where I could do some specific code for a single recipe or something to make the tool slightly better, but I don't like that approach, as I'd rather have it generic
Since only Extractors dont need input to produce parts
can you clarify this?
once again, see: plastic
I still think its possible with that in mind..... Give me a minute
traverseLoop();
for(loopInGraph) {
//do some sorta sum check for all parts in the loop to see if they saturate each other
if(conditionPass) {
markNodesAsRecycled()
}
}
Probably a lil too simple of a pseudoblock to get
looks more complex than python so I think you're fine there
.... Right, if a machine's output target is the same as the one that gives input, its a loop
Buuut thats for immediate steps
3 steps are a bit harder
recycled plastic/rubber
Like, A to B, B to C, C to A
This would be found with a graph loop algorithm regardless
so, while this may work, there's also no guarantee, that this will work
Recycled plastic and Rubber are easy those are immediate
A to B, B to A
Same for water and stuff
When you do a graph loop detector, it can find a loop no matter how large it is (with increasing latency)
Then its just a matter of choosing which node to start from in the loop and going in a circle while you count the items
the main issue is that the arrows on the graph may be pretty random
How do you detect if you need to produce "fresh" parts?
and user can connect them differently
for example, I can make the loop go away if I produce some extra pure ingots
Via the loop detector.
If the item sum is negative, there is a deficiency.
Find the building with a deficiency, travel down its node and mark as fresh
well afaik this graph is generated only once per User's Settings
I dont think you can edit this graph other than turning nodes on and off to mark as "finished" no?
If you add extra pure ingots, doesnt your tool recalculate this graph?
basically when I construct the graph, I don't have the information of which items go where, I just have the information which recipe is used in how many buildings and I construct the graph from that (iterating over each recipe's inputs and connecting them to other recipe's outputs). So while you usually get the arrows in the same place, there's no guarantee that they'll stay that way
There must be some final graph Object that presents this graph to the user no?
what I'm saying is that there's more than one way to construct those arrows
another example - the plastic/recycled loop, where usually it's better to connect the residual rubber to the loop to kickstart it
Any time you detect a resource being needed and a resource being produced, and those match, its a loop.
Like,
B: Produces 10 Y, needs 5 X```
True, but it shouldnt matter if the recycle check happens AFTER the arrows are baked no?
If you only detect a demand, but no production, then you can do backwards construction to add production lines
Loops always are made of a double check: Is my Input the same as someone elses Output?
yeah, I'm just saying that technically existence of a loop is determined by semi-random code rather than by actual existence of a loop
so there may be cases where a tool reports a loop when there's none or vice versa
I think you can always do an easy math check, just with raw numbers
what about three-way loops?
Theres gotta be a better way to find a loop and add exceptions to prevent false reports.
(diluted packaged fuel comes to mind)
Im about to try and math that
yeah, but I don't really want to add code that detects 2-[number of nodes] sized loops
that'd have big performance impact imo
You could show the graph first as is then request the user if they wanna see more details (like the recycling)
That way the original graph arrives on time
and the performance impact is optional to the user
Isnt this just a 2 way loop of canisters to packaged water
nah theres three machines there
Canister In, canister out
canister + water -> packaged water
packaged water + HOR -> packaged fuel
packaged fuel -> canister + fuel
In this condition, youd probably want to check the recipe for packaged fuel and packaged water to see if it includes canisters
Since all 3 do at 1:1:1 ratios
its a perfect recycled loop
This can be generalized
yeah that's another thing that users have been "complaining" about. That these perfect closed loops need some amount of items to be inserted first
and the tool doesn't tell them
that might be why its a closed loop lel
That honestly should be their own fix
Usually you just add a storage container and shove a fuckload of canisters into it haha
indeed, but with the generic loop-detecting code, this will be pretty much a free QOL addition
I think my facility needed like several thousands of canisters to properly prime
obviously the canisters come from the fuel unpackagers, i dont need to add in my own
I guess to find out how you would make a loop detector, we need to find out how greeny's tool truely works xd
that's why you do 1:1:1 closed loops with like ~50 canisters per loop ๐
but thats a lot of cargo containers
oh I can talk about that for ages ๐
If a loop like this is detected, you should assume that the buildings involved need a primer item (which can be checked) and say "Hey, I recommend you prime this loop with x amount of y"
I mainly wonder how the consumption and production is calculated. Do yoi start with the desired product and then work your way backwards?
I'd most likely just do "needs y to prime", since the amount cannot be properly calculated in any way
You can make a reasonable guess
or a rule of thumb
could go with some "at least this many" number?
This is preferable
like you would need at least the same number of items as go through the machines per minute
reasonable guess for diluted packaged fuel loop would be "anything from 30 canisters per 3 machines in the loop to several tens of thousands of canisters because their belts are super long"
also, it doesnt bother me that the calculator sometimes does silly loops
To circumvent the loop length, just add primers to each building in the loop
because you can just do your own plumbing anyway
black magic matrix math called linear programming, a way to optimise a given set of constraints and variables for best outcome based on value function
my favorite now is the instant scrap that feeds water to itself
If you prime the refinery with packaged fuel and the packagers with packaged water
and the unpackagers with canisters
tada, perfectly primed and ready for balancing
Is there really such a thing as too many empty cannisters in a loop?
Yes, they can block lol
yes, at some point it'll clog at least some machines
Oxygen Not Included taught me all about overstuffing loops
Fill till it clogs, then start removing till it flows... Not the fastest, but garanties its enough
I could talk about this in more depth, but probably in DMs rather than spamming this channel. It's a really complex topic ๐
i put just enough fluid canisters into my loops for the industrial storage container in my loop to not be empty
Yep
(or half a stack if your loop is small)
i mean
that would mean sending conveyor lifts down below my machines or above them if i have my refineries all next to eachother
probably even 2:2:2 could work, as you'd have belt between those two sets of machines with a splitter and a merger at end/start of the belt
which is how i like to do it for space reasons
Yeah DM is fine. I just read up on linear programming and that theres a few ways for it like simplex and what not
plus imo its easier to spam manifolds than the same thing with belts going around your refineries for every refinery
anyway, let's get back to the topic, since we have already like 4 topics and diluted packaged fuel optimisation doesn't have to be one of them (especially because we have the way cooler brother called diluted fuel)
facts
"you vs the guy she tells you not to worry about"
indeed, could be a worthy addition to #satisfactory-memes
on it lmao
Now that I have a blender
I feel like I could downsize my fuel generator intermediary refineries
in typical shitpost fashion I have made it with paint @wind spade
Oh no nvm
I dont know if this is the right place to ask this, but what is the best way to maximize a fuel generator plant? Are certain recipes better for converting oil/ more power efficient?
step 1: always use the heavy oil residue alt (crude oil -> heavy oil residue + resin, resin is byproduct)
step 2: always use diluted fuel (either packaged or not, people seem to prefer not now)
step 3: turbofuel (most efficient use of resources seems to be turbo blend fuel specifically, instead of turbofuel or turbo heavy fuel)
This is what I'm personally using, for example
And turbo fuel is specifically an alt recipe right?
We dont package it anymore cause we got blenders to avoid the packager loop
I know
I believe they all are
correct
all variants of it
I'm actually unsure if turbo blend is locked behind compacted coal or not
but yes, all variants of turbofuel are alts
I dont think so

