#math-and-meta
1 messages Β· Page 545 of 1
no, also read that article I linked
generators don't change production rate linearly with clock speed
yeah
that's fine, just trying to balance it, I'd rather it use too much fuel than not enough (since if it doesn't clear it enough, it will block up plastic / rubber production)
if you use too much fuel your power will jump up and down due to one gen not working at 100%
also use recipes which don't block plastic and rubber
Are plastic and rubber just avoided late game?
nah, there's a closed loop version with enough alt recipes
and people generally avoid mixing their production/power lines, but it's doable if you're careful and have safety measures in place like sinks and fuel gens and power storages
the key is to make sure stuff doesn't stop, or in your case if it's a bit overclocked to have power storages to make up for slightly unstable power
but I'd still figure out what the actual percent you should set it at is, and only slightly go above that if you're paranoid about stopage, no need to make it more unstable than it has to be
yeah, just need to work out what like 13% is
you want the last formula, the one that figures out clock speed by operation rate
your operation rate is 13%, so you can use that formula to get the clock speed from that
If you had 246m3/min fuel to burn you would need {\displaystyle {\frac {246}{12}}=20.5\ generators} which means you would need 20 generators at 100% operation rate and 1 generator at 50% operation rate.
so my math isn't wrong
one should be on 13%?
60 + 60 + 60 + 20.4
that's to get the operation rate, not the clock speed, the clock speed is different than the operation rate
that's the tricky part that's different about power generators, unlike producers 13% clock speed is not 13% production
OH WAIT WHAT
YOU CAN DO % OVERCLOCKING
like with decimals, not just integer
so since it would be 7.04898%
yes, clock speed gets 4 decimals and production rate gets 2 decimals, so do clock speed if you need more precision
just leave it on 7.049 ig
also never do production/power in generators, it's bugged and wrong
I'd actually round up to 7.05 if you want to make sure it doesn't ever clog
oh wait I'm silly, that was already rounded up
the number you gave
yea
7.049 was rounded up from 7.04898
So my power should now modulate a little bit, but very rarely in general
yup!
wait
I can't tell if I should be 40 * 3 + 20.4
So that'll be 140.4, 140.4/12 = 11.7
so one should be 70% effe
so 62.8966%
when doing math for generator clock speeds I keep as much precision as my calculator allows, until I actually get the clock speed
cause that's the only number that needs to be rounded, and then you'll get the most precise answer
I'm just running it through wolfram alpha so
did you get the clockspeed for the last refinery making fuel wrong?
well it looks like you've got the math down good, so you should be set anyway
π
my power seems to be modulating way more now
wait, is there a reason for me not to have all my refineries on max? @bleak coral
uses less power
not meaningfully
It'd stop the modulation, just means that the pipes on that end would be full
power usage is quadratic with clock speed on production machines
but there's no difference between that being full
or the heavy oil being full / slowly filling
may as well leave the refineries maxed out
250% clock speed for example is about 4x the power, while 1% power is like trivially low, like I got smelters to be below 0.1MW low
because I have 4x rubber and 10x plastic
so that's 100 + 4x 20
so 180 total
I guess I can put this down to instead of outputting 20.4
make it output exactly 20
well, I can't do that because it's 1/3rd
if there's backup/modulation, it's cause there's mismatch of input/output somewhere
you could go through the system double checking your numbers, and changing clock speeds to be more precise
What should I treat that 20 as?
20.00004?
treat it as 20.00004 but just round to the nearest X
wait when you said it was 120 + 20.4, that's the 20 you meant? That's the input, not output?
you had the output right the first time: 13.3333.....
haha oops
so it'd be 3*40+13.333
need a cup of cofee or a nap?
yup, looks right
oh a little trick for more precision: when it's obviously a repeating number, like 3 or 6 cause it's 1/3 or 2/3 for example, just add as many of the repeating number as you like or it'll let you
yeah
I just hold the key for like a second or two
but 33.3334% is preferable to 33.3333% no?
no
or does it allow for more
oh in the game, sure yeah
but only displays X
I meant for using the calculator/wolfram alpha
yeah
also it does accept equations in the game, so you can put in 2/3 for example
but it evaluates right to left, so you need to do lots of parenthesis if you do more than one step
or specifically for that: 100/3
so i guess this should be fine
Always round up. The percentage accept 4 decimals, so just round up the last decimal
yeah
..... We can use decimals? I thought we had to use full %, well this makes things way easier.
Exact numbers for a production target are also nice. Set your heavy encased frame manufacturers to something without 4 decimals.
Internally it will still be converted into percentages so when you re-open the machine UI, the number may be off a bit.
I would say that's a matter of finding sweet spots. For instance if you set heavy encased frames to 6 per machine, all components also have an even number. I double checked the math outside the game to be sure there were no hidden decimals or other weird rounding.
44 concrete per minute is still a little weird I suppose, but better than 20.625
Also needs 16 frames, 20 beams, and 72 pipes.
An easy way to find that is just to multiply all the components together and then just divide by 2 or 3 until you can't get nice even numbers any more.
And consider building a set of machines big enough to get you a convenient input/min
I ended up making a 60/minute HMF factory with those numbers. I needed 58 HMF for other production items, so 60 worked nicely for some excess to build with.
So hey, has there ever been an attempt in satisfactory to benchmark the performance impact of certain objects? Like from what I understand you just want to minimize the amount of visible conveyor belts, but has it ever been quantified?
That's going to be a tricky thing to really pin down, I suspect--different hardware configurations are likely going to ding performance much differently.
I mean I'd mostly be interested in a relative impact. Just like a "this will really put a strain on your flops" so I know to keep it to a minimum.
I think multiple glass surfaces in succession (where you see through them all at once) are the biggest offender. Then non-walled big factories, then giant belt buses
Overall, I think your best bet, FPS-wise, is to build with as little stuff as possible for the most output and be smart about logistic so you cut down on the total amount of belts/trains
Eg: use OC to reduce the number of buildings and/or have 1:1 ratios so you avoid splitting/merging belts. Prefer recipes that favor production speed over resource efficiency to build less machines...
This whole paragraph is a kick to the nuts on turbo fuel lmao
man im trying to get drunk and play video games
gj baine
why these scammers gotta mess up my jive lol
sry Baine
its all good π
Drunk on thursday? You have 4 day work week?
Only 40? Pffffft rookie pure ingot numbers π
Kidding aside, nice setup therr
is there a way to prevent the water backup from aluminum scrap production? I'm convinced that I did my math right for efficiency but it still keeps pausing every now and then
u need to either use it or void it
TBH a packager system might work for that
I'm recycling it back into the sloppy refineries but every now and then it'll back up
then either ure making too much water or not enough refineries
I've checked the math out a couple times now and I have enough water to feed the sloppy refineries
if you try to put backup water into the same water supply that take water from extractor things can be tricky because of flow priority
even with valves and buffer systems?
McGalleon recently discovered that if you merge 2 input on a cross junction for a pipe, the bottom input have priority over the top, that can be a way to ensure having no backup.
My usual solution is just to keep by produced water and water from extractor into 2 separate pipes.
https://www.reddit.com/r/SatisfactoryGame/comments/obqjrd/ficsit_pipeline_plumbing_manual_first_edition/
Give this a read
Page 14 or so
yes, i summoned him ! π
Otherwise, Page 12 also has some suggestions for Aluminum
I really recommend dedicating some refineries to run only on byproduct water
as solution 1 can be finnicky.
Otherwise, if the VIP Junction works here, hey, great to hear
that's my usual solution: having one set on refinery using only external water, and 1 set using only by-produced water (i just merge all by-produced water to feed this last set) and i play with OC to have the correct number, worse case scenario i just build 1 more refinery than needed. But i least i'm sure i'm safe (it just may be a bit long to start up too).
But for next time i need that, i will try the merging priority option, at least for science
for the VIP junction, is it necessary to have the pumps? or will valves work the same way?
I recommend at least one pump on the priority input, for good measure
i can work without one, buuut i dont trust it
hey Galleon question for ya the setup u gave me for splitters and mergers will that work on pipes?
thatsh the default for pipes. you cant really do it any other way
go read my manual, page 11
that shows it
ok cause i was wondering if i can do that with pipes cause im trying to get better with pipe/belt layouts
and page 6 explains flow rates in manifolds
Im really glad i got something where i can just say "Go read Page X"
Because having to repeat the same things to many people many times a day can get tiring
will do thank you
also ure setup on the oil to HOR is 1 very confusing and 2 quite intersesting to learn
Gonna hang up this pipe manual on my wall real quick so I don't have to ask anymore questions about it
All i did was use rubber to make HOR
on your save
rubber and plastic technically
well thats what the calculator told me.
its the most efficient way to get rubber, plastic and fuel
all in equal ratios
without alt recipes
idk how or y but the calculator site hates me
no matter wat i try it never gives me a reasonable answer
btw is the Greeny in the main chat the one who actually made the calculator u showed me Galleon?
why do people keep putting big "G" in front of my name π¦
autocorrect AI should already know that my name is lowercased
not a big deal lol I just hate how it looks π
Or the AI knows how you should be called 
and is trying to set its plan into motion
The idiosyncrasies of language. π€·
The concept of capitalizing proper nouns is probably awkward to learn for non-native English speakers.
more like trying NOT to capitalize everything as a native german speaker
Heh, that's exactly what I was thinking of
German be like: "Capitalize all the Things!
"
That's a top tier recipe, when combined with recycled rubber, recycled plastic, and diluted fuel. Obscene quantities of rubber and plastic from a crude node.
^^ combined with other alts, it is the gateway alt that enables insane crude oil setups
Such as 300 oil > 900 plastic or rubber for example
Nice
Basiclly, when you get it and the other oil alts, you never use the base recipes again
The recipe itself doesn't has too much uses, but when combined with some other recipes it become one of the key component
Yup, on its own its not that good, maybe if you need alot of petro coke. But, combines is when it becomes next level
whats the most efficient way to split 120 resource/min into 4 30 resource/min lines using only splitters and mergers?
Pretty much any way is 100% efficient, usually people just build manifolds
!wikisearch manifold
What is more effective:
Producing huge amount of ingots and then trucking them to factories around the map.
Producing basic materials (i.e. Iron plates) and trucking them to factories around the map.
Producing final item near the nodes where whole factory is set up.
mostly for long transport routes you always want minimum materials
So instead of transporting loads of quickwire you want to transport caterium ingots and produce your quick-wire close to your electronics
As long as you keep to that the rest shouldnt make all too much of a difference
Why it wouldnt be better to transport quickwire?
I can make 40 ingots/m or 200 quickwire/m. Wouldnt be better to make quickwire near the node so trucks can get full load?
limited by transport speed
On a small scale it doesnt matter, but later its better to transport a few ingots (which still fit onto one belt), instead of massive amounts of quickwire (which you will have to split onto multiple belts/trucks)
oh I see
Nope, trash. Youβll never need it 
how much coal is needed to have a coal gen to operate at 100% efficiency
!wikisearch coal generator
15 a min @ 100%
Thx
dumb question but is there a way to deal with 720 oil either on site or packaging for off site refining?
you can with several pipes or with train
but i think it's usually better to transform oil on site into solid parts then transport that.
thats wat i was thinking cause the area where im getting it from isnt big enough for everything
build on several level; using verticality is a good way to gain space to build
idk how to do vertical wall pipes well enough yet
i think it's still easier than moving liquid far away
true but unless someone has a setup i can try out to learn vertical pipes lol
if you have wall pipe support, just line them on the wall you want to climg, put a pipe between them, put a pump on the pipe down the wall, then you will have a blue indicator showing where to put next pump and you just have to build like that.
except idk the wall height limits thats the issue i still need to work out
no limit, you will just have to pump more pump along the way if it's higher. But once you place a pump down, the game will tell you where to place the next pump
that's just something to try, worse case scenario, you have built 1 wall support too far and need to move it
Soo, I am planning on redoing my aluminum setup so that byproduct water is sent out to do other stuff.
I need to find a good use for 564 water a min x 5 pipes worth
Either pure ingots, coal gens, etc
Maybe cooling systems? That would at least benefit from the aluminum proximity.
Still, would prefer something with exact or near exact usage. May just do coal gens. Dont care if a few under preform
could do water packaging for coal if u want easier transport
Coal is next door π no need to package
If you have the fluctuating coal generators on a Separate powergrid it wont bother you on your main graph either :)
Unless you need those couple of mw
TBH if u have enough coal and water then they shouldnt fluctuate
I also have a petro plant nearby that can suck up aome of this excess water
No, its not for the power, I got plenty
Wants to ideally use up exactly 564 water/min
Just my pure bauxite ore lines are starting to drive me mad
Which is kind of a painful number
oh
Well, thats the exact output of my alu byproduct
This would be a bigger project, but what about a HMF factory utilizing wet concrete? Requires coal, limestone, and iron.
ooo also something I might try with my excess aluminium water (so far Iβm using it on a few fluctuating coal gens)
Already have a 60/min HMF fab π
Do you have valves in place for proper regulation?
Yup, already went through ideas with Mcgalleon
Hm. Any use for a large diluted fuel operation?
Looks like you're down to pure ingots.
I think that my math is a little fucky
I have 8 Alt Recipe 40 Iron / 40 Coal > 60 Steel
So that should be 8 * 60 = 480
The coal however, should be 8*40, so 320
I killed the birb
I've set a miner MK2 to 320 out, but some of the machines at the end of the manifold still have huge issues keeping up with demand
and I can't quite tell why
Fucking kill me https://cubity.getsharex.com/nLnK52BJbF
Mk3 belts only handle 270
Yeah, I know.
Well, thats your problem
I was running along the thing and found that to be the issue yea
Ooh π
Just didn't want to leave something with no reason, or delete and erase my mistakes :^)
Neat
https://satisfactory.fandom.com/wiki/File:Pipeline_overflow_with_head_lift.png
Might want to add this to your pump line, this will prioritise the recycled water, or have the buildings with the recycled water a few meters up relative to your water supply
Overflow is one thing, prioritizing inputs is a different thing.
I'd think that as long as you have enough head lift for the water to make it over the hump it would have the same pressure on both sides of the hump?
I tried putting the refineries producing recycled water higher and they still backed up.
Adding a valve seemed to help though.
It seems that adding valves resolved the issue, in addition to using the VIP junction
hooray for VIP
junctions only allow priority in vertical configuration
and even then, its the BOTTOM Pipe that gets priority
I'll be interested to see how people end up deploying this. You've got that explained very nicely. Now we need to hope it isn't patched out. π
I guess it had never been tried with the pumps
When I get home on Tuesday, gonna give it a shot on my aluminum fab
I guess the falacy was: " If pipe is higher up, pipe has bigger pressure. Bigger pressure = priority!"
The VIP doesnt give a shit how high up the bottom pipe originally was. Or from how low the top pipe rises up. If its connected to the bottom input, it has priority
Also, this only works with vertical junctions
My first instinct is systems of a single sloppy alumina refinery to produce 360 slurry/minute, and an electrode refinery to process 360. Add a water extractor set to 90, and perhaps a half-filled fluid buffer, and you should be all set?
the pipes resetting headlift probably helps, cause then the pipe system before the pumps don't matter
Actually, after more test: pumps not even needed. But i just prefer using them over valves
oh, why'd you think they were more unstable before?
Because of fluctuations i saw. But it turned out they were negligible
Hi McGalleon, how about adding 'side view' to the top right schematic, like 'Circuit diagram - side view'
Hm. Seems fair
Trying out another "dumb split sushi setup"
@dull bolt This should fail quickly since it uses a 780/min belt, right? (less than an hour?)
The splitter with arrows is a normal splitter
The smart splitters marked purple send overflow to the left so the belt doesn't back up
Said overflow is divided so that all items that shouldn't be on that belt (plastic or rubber on the wrong side) are sent in a container, the normal overflow gets sinked. This way any mistake in splitting doesn't jam the system, but gets "registered" in the container circled in bluish.
Logicaly the vip should not work because it looks like it goes against physics, but in reality what we are seeing is 2 pipes having the same pressure at the point of the pump then the one on the right has lower pressure from the left because it goes up, so the left overpowers the right
Yes, if those belts can overflow but only your main line is 780/min so the lines before the purple splitters will never be a full belt, since it's a combined belt that is splitted into two, so the issue won't be seen.
@vivid lotusBut the one on the right goes back down after going up, so the pressure should be the same.
And I don't think the pressure matters? Otherwise you'd get priority for recycled water by having those refineries higher, and that doesn't work.
And it would work even with the junction horizontal.
Funny enough no
Even if you raise the right one to have more gravitational pressure or whatever
Doesnt matter
Lowest pipe has priority
Pressure literally doesnt matter
Satisfactory treats pressure oddly, see the exploit where you can get free head lift by connecting a low (or no) flow pipe with a higher head lift.
Pressure and Head Lift seem to be seperate entities
Thats why i seperated them in my Manual
thats also why my first batch of translations were messed up
I don't get it, why should it not show up? 
Wasn't the issue with 780/min inputs ruining the splitter's output on the long run?
BTW, the purple splitters are there mostly to intercept rubber/plastic on the wrong line, there should be no overflow since consumption=demand
Is pressure even a thing?
Yep. Pressure is done for flow calculations
Aha!
Pressure differences determine flow from pipe to pipe
So VIP seems to be something "Work Pressure" related
So, any idea what happens to VIP if the junction is at an angle?
Lowest leg gets priority?
Give it a try
What ? No lowest has highest pressure obviusly ? the other one is going UP then down , making it lower its pressure on the upway , then it gains SOME pressure back on the down but its lossy , its rounding DOWN so one pressure can be 1 while the other is 0.99998
to gain more pressure you would have to start from same hight
You dont need the pump and it still works
Even if the bottom pipe is longer or went up 30 m before
Weird fluid mechanics, all my real world knowledge is basically invalid ingame.
Yes
Except Hydrostatics and communicating vessels, at best. But even that can get weird here
Could it be possibly due to lower input is nearer to the junction?
If I have a normal oil node and I donβt overclock it how many refineries do I need?
what do you need them for?
120/30 = 4, possibly
Yes, that's based on the default recipe and normal purity. Things get interesting if you got the complete set of alt recipes
One thing I hate about oil is itβs soooo far away
Does anybody else use this method for setting out splits for assembler lines?
Basically if you arrange your assemblers in 2 levels, at every 3rd foundation-frame hole, you can build this repeatable thing. I think it's cool
Like 2000 m
view from above is clearer
Usually it is trial and error if you want to get the multi-level assembly works. Start from small scale
Will it hold? 
@proven prawn @upbeat tide @prisma hound
@magic shadow Are you certain? ^^
Also, what's the reasoning?
just one fluctuation and it's gone
it might hold, but at this point it's just a matter of time and chance
Tbh, I thought the same about a similar setups with a MK1 belt, but then that survived reloading the area, reloading the save, having someone else load the save 
I dunno if this can act the same since it's at 780, but I got hopes 
It ran 1 hour already btw. No wrong split in the meanwhile
Depends on how many resources you want to use
the most effecient
Ideally you would just make your own using statisfactorytools or satisfactorycalculator to fit your exact needs
theres 2 most efficient setups: one uses turbo blend fuel, the other turbofuel (for refinery)
both utilize the HOR alt and diluted fuel
it depends ALOT on what alternate recipes you have
funny enough the most efficient turbo setup is the one without coal
the most efficient turbo fuel is nuclear xd
Im not there yet π
!wikisearch Turbofuel
that works too :)
DT and DTB are tied for efficiency
i personaly went with turbo blend fuel and didnt have HOR so i dont use water cause its a stepping stone anyway
just dont use heavy if you just want turbofuel 
ez pz rubber HOR to turbo heavy 
YEP
peak efficiency
I guess
How much power it gives you?
i love how well-organized factories look like PCBs from SCIM
whelp SCIM is down
Maybe maintenance
is this a website?
Guess I won't be working on my production lines today
(I know there are other sites for it, but Satis-Calc is the one I use most)
https://u4.satisfactorytools.com/
Much better calc IMO
That one's alright.
I disagree, Greenyβs is much more user friendly. Satis calc gives you a not so usable physical layout
where's greeny's
btw, anyone know any online map ||(except the SCIM one)|| so i can guide myself to where i need to build the oil factory?
Not sure on maps, but oil isnt that bad
- West coast islands
- Blue crater southeast
- North coast
- Middle ish of the map
- Southern part of dune desert
- Oil well south of swamp
- Oil well west of the center map oil area
ohhh. I thought it was someone else's
Nah thats Greenyβs site
has anyone made a layout tool for satisfactory buildings? (like a simplified 2d version of satisfactory)
apart from the map visualiser
I have seen a few reddit posts (fairly old) with gridline representations that could be used in tools like Visio, draw.io, etc but never really tried it
And dont have links to those threads
I see
But I never felt a need to use them personally
I'd like to use it as a novelty tool for creating pretty blueprints
just like oldshavingfoam
I think he uses some sortof graphing tool, but best to ask him directly on his newest thread
there is a tool called SaLT
Tried this and lowest pipe still gets priority
Good to know!
I got stuck trying to think of an easy way to test. I guess 2 extractors and a fluid buffer would be the simplest.
yayyy the SCIM is up again
how much coal do I need per coal generator?
I find it odd that it shows water requirements but not coal
15min i think
ta
Satisfactory helper to calculate your production needs. | Gaming Tool/Wiki/Database to empower the players.
There will be no such map, as it could possibility make many player's setup to become stereotype
Generally anywhere close to oil node and has access to watee
Only one exception, and thats the southern dune desert oil nodes
Theres a water resource well near there
@oblique hollow just read your guide and saw this picture
shouldnt it be the other way around? as in water from aluminum at the bottom and the extractor at the top
i really didnt care for the image orientation there but yea.
ok :D
Dunno if due to "the issue", but it failed in a couple hours ;-;
@magic shadow wins the poll
@oblique hollow I can confirm this design work wonders. The two buffers on the right feed water/fuel to the END of the water/fuel manifolds (bottom HOR -> Fuel, top recycling plastic/rubber)
Step 1: Let the Water buffer fill to 3/4
Step 2: Turn on HOR -> Fuel, all machines start in <3 seconds (4 blenders) ||oil has 2 buffers too||
Step 3: Let the fuel buffer fill to 3/4
Step 4: Turn on the recycling. All machines start in <5 seconds (6 refineries)
Even though the fuel probably doesn't have enough pressure to fill up the entire buffer, it's not even needed for the buffer to fill up completely (no additional pump)
Both water and fuel are at 360/min per pipe system (each buffer)
For my preferencies, yes. I can check on the flow from the machines properly (since the flow between the input and output side of the manifold has "correct" flow) and usually just needs a single small buffer every 10 machines or so. I also love how it frees up the space above for belting
Having the buffer also makes it very easy to check if the system is working ok later on (wether the buffer is mantaining it's level, emptyiing out, or filling up)
It's pretty much the same amount of work as a loop, maybe less depending on how long you need to make the loop pipe 
Oh, I placed the buffers 4m higher than the blenders/refineries btw
confused stonks
i like being right but i feel bad that it's about your factory going to shit 
Luckily, I didn't actually get mad enough to rely on that for my 2k plubber factory 
Just a test~
"Plubber" 'cause I'm too lazy to write "plastic/rubber"
plubber is best name
I just call them both by what they are chemically: Polymers
Polymer Resin is Resin to me
and Rubber/Plastic is "Polymers"
I fear very few people would understand by "polymers" though xD
also plubber if you mention it the first time
"tf is plubber/ polymers"
"rubber and plastic"
plubber reactor is its official name
Plubber-inatorizer 9000 MAX pro
anyone wanna check my math?
what math
nuclear recipe setup:
total uranium: 2100
infused uranium cell:
25 uranium + 15 silica + 25 sulfur + 75 quickwire = 20 encased uranium cell
2100 uranium + 1260 silica + 2100 sulfur + 6300 quickwire = 1680 encased uranium cell
Uranium Fuel Unit:
20 encased uranium cell + 2 electromagnetic control rod + 0.6 crystal oscillator + 1.2 beacon = 0.6 fuel rod
1680 encased uranium cell + 168 electromagnetic control rod + 50.4 crystal oscillator + 100.8 beacon = 50.4 fuel rod
uranium fuel rods consumption is 0.2/min - 252 nuclear power plants - 630000 MW
^ everything look ok so far? am I being a pepega anywhere?
I'm quite pleased with this fluid-sink/water water disposal facility I made to deal with excess water (particularly aluminium & blender recipes). What do you think?
you package and sink it?
basically, but I keep it guaranteed to run by generating the input from a fuel power plant π
i know its a pretty recent discovery, buuut.
maybe try this:
it makes fluid sinking completely redundant
oooh, nice
as the water just gets prioritized
but doesn't that run your plant at a comprimise between outputs?
60 gets through, and the rest gets topped up by the unprioritized input
to make 180
and the scrap refinery never backs up
only the water extractors
calculator says 31.5 fuel rods is the limit
not 40.5
hmmmmmm
I haven't quite understood it, but will think about it some more. One question: does it keep your factory producing the scrap at a steady rate or does it fluctuate?
alright guess I gotta go see where my math went wrong
Looks sophisticated π
ah sorry
i didnt see you used infused cell recipe
was hard to tell
lemme recheck it
right. 50,4
the calculator on the interactive map just breaks so I don't know where to check
oooh thanks!
I think I get the VIP; it's literally the lowest elevation input that is prioritised, because the pump to the top pipe has to overcome the U-bend?
you can also remove the pumps
so neat! Just starting to understand pipes; built my first water tower today!
what is the benefit to different priorities?
idk. but just 1 level of priority is good for aluminum
it just CAN be expanded
but there really arent much uses for that
You can ensure that the bi product that would otherwise clog the output gets cycled away to keep the productive chain clear
got an answer to my question?
you might be surprised
My guess: trick question
They have the same priority. The head lift on the pump is the same as the pressure from the U bend
but I'm guessing π
nope. the one connected to the bottom of the VIP STILL has priority
so pressure doesnt matter
that's nuts
no matter what, the lowest input has priorty
ok, I guess it's because the thing is basically a big oddly shaped bucket and it always fills from the bottom
its hard to say.
Even i dont quite know WHY it works. i just know when and that it works
also, this only works on vertical junctions
junctions that are horizontal dont behave like this
but cool tips mr piping expert,
Avoiding pumps is cool because I just traced 4km of pipes looking for a section of tiny MK1 pipe which was caused by a pump
flow rate was dooing woodly things
I do love the progression in this game.
As it gets more complicated; it'll let you through to the end but if you didn't understand it you will Chernobyl your world.
The progression at each tier is really fascinating to me, too! It all seemed pretty visible and obvious to me through the first several steps, but then trains came along and I realized what the game was encouraging you to do, and the scale of it, and it was like π€―
Aluminium was the major mindboggle for me. Such a pain to set up, but in the end it makes sense.
ever since i found the VIP Junction, it has become pretty much effortless to do Aluminum, except when using Silica maybe
The what? Pray tell.
Oh, I need to try that. I played around with something like that early on, but it didn't behave.
the vertical junctions made all the difference
Does anyone have a good setup for producing heavy modular frames? because the factory builder websites are not helpfull, with splitting iron to 205 units/per min and so on.
You can find the wiki here manifold
<3 @sullen cloud
splitting to 205 is not necessary
Hm why?
Because it effiecent? So the machines doesnt need to be full for other machines to get full too
Okay thanks
but thats exactly what you just said you didnt want
it lets machines fill up and overflow
and what doesnt fit goes elsewhere
Yes i know, but if there is no other way arround it? Then its the only thing i can do about it
another way would be to try weird splitter arrays but those are huge and complex
I already did that with 190 iron/per min
currently, overflow is the most reliable for odd numbers
Yes, than i will try that, thank you ^^
You can always split to what you want, but it often involved complexe a big splitter/merger setup if the ratio aren't good. Overflow is the easy way to do thing, and the drawback of having to wait for machine to be full isn't too important.
Me, wanting to have the extractors run at 100% too: 
Me, not caring: 
why dont you use one o your magic buffers
I do, but doesn't using the VIP imply your extractors are overproducing?
Or just that you generally have too much input water from one side
even when things were supposed to be balanced, when you merge external and by-product fluid you can have some problem if you don't use priority junction
you can set it exactly of course.
Using the VIP just gives you extra safety
really, every Special CIrcuit can be used in combination to make your system safer every time
Ah, gotcha
I just never merged water other than in sloppy factories, so I always had a few things to blame other than the byproduct loop :laughing:
really, the most over the top safety for Aluminum would be:
- Underclock Water extractors
- Slap a Valve on it.
- check the exact flow with an Equalizer or Compensator
- Merge Byproduct and fresh Water in VIP
- Loop Refinery Inputs
- Add a buffer to Input line
Eh, I feel safer sending the water to be consumed by something
Like dividing the alu refineries between by-product fed and extractor-fed
ofc, thats the most stable solution
and i even highlighted that
Point 6 very important, buffers everywhere xD
in the manual
Oh, my bad, haven't read that π
Been fiddling with manifold loading of machines vs perfectly balanced, and finding that when fully saturated prior to starting up, both input and output belts full, they both retain 100% efficiency. However if cold-starting an empty system and all outputs being consumed, the machines at the far end of the manifold never are able to reach saturation. I'm thinking it's because the travel time to the end of the line is longer than the processing time of the first few. Anyone else done similar tests?
You need to prime the system first
Aka let the machines run and fill up internal hoppers, once thats done it will work fine as long as your input/output and machine count math is right
The issue with manifolds is that the splitters do not split in all output slots at once, its staggered distribution. But, if its all hooked up right, it will work after the system is prined
Yea still not finding that true. Running a 16 smelter manifold and the last couple of machines, despite being fully primed before start up, aren't even activating anymore and the output is less than the 480 I'm demanding from it
It it belted correctly?
Something was not filled up and the buffers shifted around then
And are all the belts the correct type? Aka not a mk3 belt somewhere?
hmmm you might be right on buffers shifting then. few more minutes they kicked on; maybe the outputs slots being full on start up did made it wonky
Best way is to disconnect the final output belt segment after all the mergers
There is a manifold fill calculator - check if you waited long enough π https://satisfactory.greeny.dev/machine-fill
A powerful Satisfactory calculator, filled with features, supports overclocking, alternate recipes, live map, power calculations, etc.
Dont see why this aspect of it would be broken, given what I'm inputting. Didn't expect the 18 minute fill time for this set up
It doubles with each splitter, so...
manifold tool should work normally, if it doesn't, there's most likely something wrong with your setup. However the calculation is an estimate, so don't expect exact results
If its one long row, I would suggest this
Machine < splitter > machine
I have done many two sided rows of manifold machines with no issues
Yea I should have set it up like that makes more sense for what I've got next in the chain
also a tip for 'faster' filling - make the main manifold belt the same grade as input, eg if first piece is mk3, go mk3 all the way. It speeds up initial fill
Yea I've got a ton of encased beams backed up that I'm just using them everywhere
Appreciate the tools and tips, definitely learned something today π
Are the supercomputer alts worth it over standard?
OC or super state?
I use super state. Built next to my classic battery factory.
Im looking at OC because cooling systems via the alt isnt that bad but, I can get super state too
I think the higher bauxite use ended up swaying me out of OC? But super state was also convenient, due to classic battery also sharing wire as an ingredient. Both the batteries and supercomputers are using the same node for caterium wire.
Ooh I wouldnt use caterium wire. Just iron wire it π
It was convenient π€·
Understandable
I got my eyes on the standard and Super-State recipes
OC supercomputer is way too much bauxite for my liking (188 ingots Vs 105 ingots for the Super-State for 6 supercomputers) and also needs quartz
Could someone explain to me how a splitter, splits an in even number like 5 .... I know of the chart that shows me how to do it, I just don't understand how it determines 3 one way and 2 the other way.... Hopefully what I'm saying makes sense.
It doesn't, it always split in 2 or 3, by sending one item on each aviable output in sequence
Its simple. You split into 6 even parts, then junk or recycle one of them, having 5 left
The one in wiki just goes via shortcu of deleting one splitter (which isn't needed if you just need 5 parts out of 6 and using manifold to distribute the 5)
Like so reinforced iron plates make 5 per min, splitting it up to make modular frame (3 reinforced iron plates pre min) and smart plating (2 reinforced iron plates per min) .... So splitting 3/5th and 2/5th is what I was saying @versed violet
just manifold that
I mean I got it working, I just don't understand the logic behind splitting an odd number.
split into even number and put the remaining output back into input
It's magic~ 
or just
--S--S--S--S--S
| | | | |
Like the 3/5th split... How does it know to always send 3 out of the 5 in one direction and 2 out of the other
it doesn't, it sends 50% of what it gets
but since you're looping back 1/6th, it's essentially 3/5 that goes down
It just splits in 2, gets rid of half then splits in 3 (making sixths) and merging back 1/6 to have the two sixths remaining
The merged back 1/6 makes the 2 remaining outputs FIFTHS of the original input
I'm just dumb and don't understand math π
that's where you dump any balancers and you just manifold everything
Fair enough.
It's a bit finniky to get the hang of, but the basic idea is simple: split to the closer multiple of 2 or 3 and merge back any excess
question but asking bout the 3/5 split but asking bout a 1/8 split is there a way to do it evenly or just shove it in the middle and let it fan out?
Eg: want fifths? Split in 6, merge back 1
Or split in 8 and merge back 3... So on
Anyways it works and I'm done questioning it haha
Is building vertically tight like this ok or pure evil?
if it's too tight you can have some problems: at least some visual clipping, and maybe you won't be able to build everywhere on the upper floor
Yeah, smart-mod sure has issues if you use minimal clearance.
Yo. Does splitways work for automated trains now?
automated train β
automated trains β
?
if you add another train on the same track it would eventually stop working
it was never broken for one train
ah, okay. Makes sense. I'm using the blueprint, got tired of my single wonky track around the map :
It has two tracks. But i guess it goes for the same
@fierce ruin if you really want to understand balancer, there is a page in wiki for that. The image you shared is probably uploaded by someone else who wrote 'prime splitter' instead of balancer
Oh interesting! Thanks I'll read up on that.
Ugh, this is a nice example of why I could never trust mixed belts. The internal buffers of splitters make no sense. I accidentally fed my oscillator factory the raw quartz line instead of the processed quartz. Rather than deleting the splitters, I've experimented with allowing them to disgorge the raw quartz, and sometimes they're allowing dozens of processed quartz to pass before going back to spitting out some raw quartz.
This one in particular is a mystery. The raw quartz seems to be stuck in this one indefinitely, but I know if I don't remove it, it may pop out and jam the input of the manufacturer.
I've long suspected that splitters and mergers work on a last-in-first-out basis. Does it not clear if you disconnect the input belt?
I'll return and test
Minimally I saw multiple instances where dozens of crystals would pass, before one raw would inexplicably pop out.
Huh. Sure enough, that cleared it. 2 left in the splitter, and 4 in the input lift. Perhaps they were popping out one at a time before each time there was just enough of a lull in the quartz feed that they were the only thing left in the splitter?
The buffers in splitter and mergers are first in last out.
It is mentioned in wiki, i check it out
little bit of math for resources for my mega
and then the resources needed for the machines
That's quite a number of different types of irons there
You got the rate and material to built, now the remaining problem is where to build
I have a foundation layed around in the desert up north
its 81x81 and its gonna house my mega factory
I prefer to build as I go, attach segments together, etc rather than build a big floor and force fit the machinery
yeah ive done that for the rest of my stuff
and i tried doing a set foundation before
but i learnt that its better to make it wayyy bigger than you think youll need
because my last one was way too small
I LOVE SATFISFCOTRY
I don't like clipping, so I call that evil π
On the flip side, if you give the veiling another 4m of clearance, you'll be able to hang pipes, hypertubes and lights from it
I never had such issues 
I always make the splitters so they "material A" on one line and overflow on another line. That way, even if I select the wrong material, all I have to do is clean up the belt once. If I put the wrong material on the belt instead, nothing happens as it just overflows down the line (towards a sink eventually)
Tldr: I think how convenient mixed belt are to use depend mostly on how you set them up
mixed belts would be easy for me if we had priority mergers...
You can send Questions, Feedback, Suggestions, and Bug Reports at https://questions.satisfactorygame.com @swift robin you can suggest smart mergers on the qa page (probably has been suggested a bunch already though)
<3 @strong whale
lol yeah no point in me posting that when its been done 1000 times
:D
An upvote could help too... Maybe π
Though, I find they work well enough for me until now
Still trying to make sense of whether I need a second drone going back and forth between these two ports. Reported trip time with one drone running is a consistent 3:21. Any trip more than 3 minutes in length should mean less than 3 stacks per minute of bandwidth. Am I wrong? Because this port is reporting a 3.24/stacks minute of max bandwidth.
is it actually getting that many or items are stuck?
Lights? Oh, right, those new U4 things.
I should probably build some already π
Everything is just going into a sink for now, so it's hard to know
does the input station have full inv?
Yeah, input station is continually jammed
speaking of jamps - is there any queue at either station?
The sending port's belt is constantly backed up, so hypothetically enough drones pulling from it should max an MK 5 belt. Though that's not necessary in this instance. The receiving port is just dumping everything into a sink.
What I ultimately need is 270~ aluminum fluid canisters going back and forth.
I still have battery production to spare, so adding a second drone will achieve this easily, I just want to understand how it's calculating that max bandwidth.
Finally completed 1 floor of my mega fuel factory... this floor should produce enough fuel for 300 generators
normal fuel, not turbo, right?
ok yeah seems like it. nice
Dilutted fuel
@fringe pawn the final 1800 a min plastic setup is almost done! Its currently in the middle of warming up the recycled loops and manifolds...takes a good while
Once done, will have a full 3600 plastic and 3600 rubber build
you need that many, or just bragging rights?
Yes actually will be using it
Already half is scheduled for use but decided to just setup it all for future proofing
Using 2400m3 oil from the blue crater. No idea what I will do with the final 150 tho
Pancakes
How are you using all the plastic? @upbeat tide
I remember seeing an outline somewhere I think? But the one I just found only has the plan for the plastic and rubber itself.
I feel like I should time my factory booting up 
Should I include the oil water and fuel prefilling times? (All produced locally)
Bruh... You're missing out! Those with glass foundations make for some belt/pipework that is just 
#screenshots message
-
It will be used for a few things, caterium circuit boards, classic batteries, and standard super computers
-
I stagger oil setup startups. Begin with the first stsges, let machines prime then move on
That's the most convenient, unless you wanna wait for real ages before you can check if it runs well or not 
Agreed but also the safest imo
And you can let a stage prime while your setting up the next
And rubber
Caterium computer
Heat exchanger
Radio control system
Insulated crystal osscilator
And a bit for rifle ammo automation
Lights question - are the sevn color slots on light panel global, or per panel?
@frosty owl full build now completed
just nuts...
looks like you need 4 more fuel per minute
I think its a SCIM rounding error, that said one blender maybe set one or twounit too high
But its so far working as intended
how do i do an even... 12.5-way conveyor split?
as in, 12 conveyors that each output the same amount of items per minute, and a 13th conveyor that outputs half of the amount the rest do
all from the same input
you don't really, might be a way that uses low mk belts as limiters, but generally any balancer outputs an X number of outputs split evenly
you need manifolds for that
!wikisearch manifold
tldr: stuff backs up and then lines will only take what they need
anyone can help me make a setup for this?
@wanton belfry cracks knuckles One to 12.5, here we go.
Start with a 1:5 splitter. Input > Merger > 2-way Splitter
Branch 1: 3-way splitter
Branch 2: 3-way splitter, one branch back to merger
This gives you 5 outputs of equal items/minute.
Each of these 5 outputs goes to another 1:5 splitter of the same design.
This results in a 1:25 splitter.
Merge 24 of the outputs into 12 lines.
The last one will have half the output of the others.
hahaha that's so complicated, pretty sure @frosty owl would be interested in this nonsense
XD
Oh, I'm straight-up recommending manifold for this as well. Each 1:5 splitter is gonna take... what, 2x3 at a decent density, 3x3 to do nicely? And there's six of them so maybe 54 squares of nothing but the 1:5 splitters? But hey, it should work.
i love you
Eh, thank me when it's done.
...also maybe send a pic this sounds absolutely insane.
i'm 16- oh wait you meant the conveyor system nvm yeah i will send a pic when it's done. uh, what does manifold mean btw?
One line in, splitter to every machine.
that would be really uneven
Nah.
it's not, they back up and then only take what's needed
The first one fills up, then the second one fills up, then the third, and so on.
splitters alternate between available outputs, they don't just send stuff into the void if an output is blocked
Mhm.
that sounds awful. sounds much more effective, don't get me wrong, but i really don't wanna for some reason
I did link to the page explaining it earlier
Thinking about it a bit... you might be able to cut it down to 5 machines if you do overclocking, but that'll take a lot of shards.
it is very inefficient until it reaches steady state, yes, but once everything backfills it will operate well
Mhm.
also on the crazy plan: do you need to split each 1/5th branch by 1/5th again? if you're just gonna merge them?
bruh
I thought about that, but the basic ratio I saw is 1:25 then 2:1.
i just realized
You might be able to pull off a 1:5 split then 1:2.5 on each.
I just don't know how.
yeah you're right, I guess that is the path to turn 1/5th into several 1/12ths
so im doing the 12.5 after a 16 output, one of which goes outside of the crazy 12.5 and continues on
i have no idea how to do this, i think ill just do manifold
And this is why we suggest it.
that's basically what ends up happening, balancers are cool but they get complicated to work with arbitrary numbers
and god forbid if you want to expand a little
Mhm. Some balancing is generally a good idea, but yikes.
seems like a challenge
I... I can't, I have to. I'm sorry.
There are challenges. And there are challenges where we never hear from you again. This is the second one.
there's two techniques to ease the warmup time issue of manifolds:
- prefill machines with stuff from storage
- just turn stuff on as you build the whole system, so stuff fills as you're building the rest of the factory
start by building a quick manifold, and before you end up placing all the mergers and splitters you need for your 1:12.5, the manifold will be backstuffed and ready to roll at the same efficiency as you want
oh other good technique whether you're doing balancers or manifolds: don't merge lines that don't need to be merged
like I wouldn't ever merge those three rod lines, I'd just mess with clock speeds and machine counts to keep them completely separate
BUT IM NOT GONNA LIKE IT
That's literally what everyone says. "I'm gonna buck the trend and do it the har- oh god I understand now."
yup.
especially on a recipe that needs screws or wire in the triple digits
im also planning on building an identical factory on top of that one once i get miner mk2, so it'll be easier to replicate
Definitely go with manifold.
Fine, but I'm not gonna like it
someday precise splitters will exist and bring peace to the land
and remember your belts can probably only do 270/min if you haven't gotten mk2 miner yet
or 120 if you haven't done steel
That cliff must've given you a very good time

Staying away from balancing such numbers might be the reason I'm still running my mouth off about balancing 
I just 1:1 that sort of stuff (or clock machines to get multiples, you get what I mean)
... Or manifold it. If items are required at a very low rate, so the belt can fill the machines fast, meaning I can go crazy with sushi belts instead
Gotta have at least one, right? 
Manifacturers' sushi manifolds go brrr
sushi manifolds 
and yeah I'd have a hard time imagining even you doing something like that, it's a lot of effort for very little results
(not the sushi belts, the crazy 12.5 splitter)
Considering there's a very conceient way to get the same result... Not worth it imo ahah
Still... I did manifold the iron ingots for the wire/cable/iron rods/iron plates needed in the beacon factory
I sinned

I mean, it's 1 straight belt Vs 2 sets of... what... 8 foundations worth of balancing?
Nah :disappointed_snutt:
that's always what it comes down to eh? space & time to set up the balancer, even when you know how to do it
I will admit, Smart! Helped alot with the foundations
I need to make a entire separate dedicated factory for heavy modular frames
I'm just lazy, I simply overproduce all items, and syphon off the overflow with a smart splitter in to a container for personal inventory use and then overflow into an Awesome Sink.
Same as above, except i don't build sink, i let them stack up and the machines will eventually stop
Not sinking everything, but when a process has two products, I will sink the overflow of both to avoid clogging the process for either
The only setups I use sinks on to make them not stop is stuff with loops or byproducts.
Recycled loops, aluminum, uranium cells, etc
Uranium cells?
I think residual rubber is the only part of a recycling loop where a sink is needed.
Maybe, but I like to put sinks on the final output belts too
I think the ability of a recycling loop to stop and start would be affected mostly by the ability to disgorge built up solids, which will be limited by belt bandwidth. Fluids probably wouldn't present an issue. I'm curious about sulfuric acid, though. I unplugged the water extractor and have been running it as a closed system. It hasn't had downtime, being power infrastructure, obviously, but I think it would handle it just fine.
Perhaps I take it too conservatively, but I like to use sinks as βpressure valvesβ in a manner of speaking. Allow the system some breathing room just incase.
Also, a good number of my plastic/rubber belts will be trained. So, sinks are a bit useful imo there
The bug with the Mk2 pipes when you fill them to capacity... Is it affected by distance?
If I put a Mk2 pipe coming out of an oil extractor and immediately split it, will that help negate the issue?
No, that wont do much I think. Its the number of junctions that causes issues
Heh, as I'm visiting my nuclear facility for the first time in a while, I'm finding I'd rather use the hoverpack and occasionally use inhalers, versus walking.
Like the actual pipeline junction item?
Yes
I have done a few long distance pipelines without issues, besides headlift as usual
Im planning to have most of my nuclear stuff off-site or nearby. That way the radiation bits have their own spot to fester
Plutonium is the only super dangerous bit.
I allowed 3 containers of encased plutonium cells to fill before turning on the plutonium reactors... yeah, that's a hotspot. π
I built up 15 containers of uranium cells and those aren't so bad.
Come back after your character has grown a third arm π
There's where the plutonium waste ultimately ends up. I calculated enough containers for 1,000 hours of runtime.
Nice
Looks like what I did with U3 setup. But mine was kissing the death wall
My old setup
Also used a train π
Nice! I just kept building foundations until I started taking damage, and then automated a factory cart to run the route. π€‘
Works too
I'm doing a dumb challenge where I complete the game as a balancer. Can someone help me split 2 conveyors into 5.33?
Balancing in thirds is...fun not
Ok, don't help me, i found a solution
i just have to split the miner line into 2 and build 2 seperate factories for the 2 items i wanna produce, instead of making them all in one big factory
I swear this game will make me get a degree or something
well just use manifolds, they are technically balancers as well
then what's the point of the challenge
idk, it's weird challenge anyway π
if you want to make it a little easier on yourself you can use clock speeds so that you don't need to get franctional parts out of a split. Like instead of doing 5 machines at 100% clock speed and 1 machine at 33.3334% clockspeed, you could do 6 at 88.8889% clock speed and just do a 2 to 6 balancer.
hmm, ill take that into account
Yeah, I would way rather balance with machines than some noodlely belt-mageddon.
clock speeds: the duck tape of satisfactory
in infomercial voice : is your ratio wrong? make it right today with clock speeds! only costs 1 green, slimy boi to be sacrificed at the altar of the MAM!
Regarding waste processing as a closed water loop, it seems to be entirely stable without any advanced technique required. You can see my non-fissile uranium blenders on the left, feeding to sulfuric acid refineries, which go right back to the blenders. I've been checking the levels of an intermediary water tank, and it's holding steady.
I was trying to load-balance fluids... I have four inputs that I need to split evenly into five pipes. I fed the four lines into junctions into a perpendicular pipe, and fed that pipe into 5 fluid buffers, hoping they'd sort of fill up evenly and distribute evenly... That's not the case though. I put valves on the outputs, but sometimes the valves are letting more fluid through somehow... What's the point of valves if they can send excess fluid?
More fluid than what?
Than I limited to. I set this valve to 360 cubic meters per minute
You can see the needle closest is way above that line
Hm. I'm on multiplayer and I'm not host...
I didn't check to see if this is a known issue
can the host check if the valve has a limit set for them?
Yeah, I'll ask
how do yall deal with water in aluminum? just curious
I try to separate the refineries using recycled water from the ones using fresh water.
isnt that like realy hard to balance the ratios?
Have you seen the pipe manual/tutorial?
There's this thing called a VIP junction, apparently in a vertical pipe junction the bottom input gets priority, no matter what the head lifts are.
here is the image McGalleon shared for reference
hm i actually havent tried that one but when i used valves it went badly
Can be hard, but they have to be balanced just as much if you mix them.
idk i looked at the numbers and the 105 water from electrode scrap just messed everythign up
you can use overclock/underclock to set the number as you want
For me adding a valve seemed to fix it, putting the refineries producing water higher did nothing.
thats REALY hard to math out for 100% alumina solution efficency
depends on how large you are willing to go - dont have the numbers in my head right now but I'm pretty sure there are a few points/values where it works out alright
I think the problem is that setting numbers only works as long as everything runs 100%, which doesn't always work out.
ye its 7-18 or something like that
I'm def gonna try the VIP jucktion next time.
same
your problem is the byproduced water to reinject in the loop ?
now the problem is that if you scale done the refinery for the water your alumina solution also goes down
resulting in wierd numbers
scale the water for your refinery then
Having the water producing refineries higher may have led to me doing the VIP backwards, it might have worked better had I made sure the junction was flat?
first count how many refinery you need for your bauxite production, that give you how many water total you need, and how many by-produced water you will get. Just set the extractor with underclock for making exactly the difference
@fierce ruinI tried that, it didn't work until I added a valve, but I don't think a valve should work.
with priority junction from McGalleon that should work
That's what I said.
But I didn't know about that then.
i prefer to do that in 2 differents water circuits, but the math for scalling how much you need is still the same
also if you try that keep in mind the the water from aluminum should be at the bottom
@fierce ruin I said that as well.
i know, i was mainly trying to explain how to balance the number easily
||easiest solution is to make water canisters and sink it from the system||
easiest solution is trains <3
@eternal slate Less wasteful would be to try to balance it and have an overflow hump leading to the packager.
also i realy dont like the idea of switching stuff if they cahgne the ratios again
I'd prefer using the water for something else that can be over designed and starved of water a bit without it mattering.
Iike pure iron/copper
wait the source water can just overproduce if you use the VIP right?
I suppose. I haven't tested it yet.
it's supposed to; i never tested it myself
but setting the source water to not overproduced isn't that hard anyway
idk it takes alot of time and math if you make alot
regarding to this: I assume if you input water from bottom side of the junction and attach excess/input water to other sides it should work fine
even without balancing
other way around but ye
#math-and-meta message
one thing you need to add is valve to prevent water flowing to pipe section where excess water is coming from
it's been working like charm so far (I made some fluid storages just in case but they are empty all the time)
idk when i tried it the first time with valves i just sometimes came back to all pipes beeing full with water halting the scrap production
I have just one valve to prevent back flow where recycled water is coming from
ye thats what i had aswell with the correct numbers
but then server lag or unloading or something happened and it stopped so i switched to trains
How do trains help? The water still has to go somewhere, whether by pipe or train?
you do something like this
I might have one pump for recycled water as well, but it's been long time since I checked it out. I cant access the save rn
this way the water from the aluminum is always prioritised
i guess this is the same if you use the VIP i just didnt know about that at that time
Ah, ok.
still not 100% sure if it works like that
cool, it should be fine :)
what ratio would you go for with that assuming base alumina solution and electrode scrap
3x alumina solution 2x scrap 3x extractor?
I have 2 refineries using 500 water/min each, recycled water 600/min coming from the left 400/min from extractors from the right, the refineries producing water are bone dry, all the water disappears right away.
The refineries producing water are quite a bit higher than the extractors.
But it didn't work until I added the valve.
If I were redoing it I'd use VIP.
Another solution would be to overclock the closest refinery so it uses all 600/min recycled, underclock the other and remove the pipe with the valve competely.
That's with Sloppy Alumina and standard scrap recipies.
I do have an oil well nearby, half used for fuel/turbo fuel power, so I'll look into electrode scrap when I expand the aluminum.
Design on the left seems to have an issue keeping up a flow of 600. Design on the right seems to run fine. I forget how either design originally came about, but I need to investigate further, as some of my 250% nuclear plants are having slight water issues. 2 extractors at 250% feeding directly to a nuclear reactor at 250%.
--
I wonder if the reason my setup is working now is that with the valve I'm merging 100 recycled with 400 fresh water and the smaller amount is semi-prioritized? That's what would happen with belts I think.
god pipes are weird
@fringe pawnDid you mistype? It seems like a 250% reactor would need 750 water/min? 200% for 600 water/min?
Nuclear reactors don't scale like other buildings, 250% clockspeed results in 200% production.
Oh, of course, I forgot that generators worked like that.
nuclear generators scale slightly differently than other generators, which scale way differently than other types of buildings
When I get home I'm going to pack plenty of inhalers and fix all the pipes using the design on the right and see if I still get problems.
Still, isn't 250% OC supposed to be 202%
no, read my msg above π
Nuclear power seems to have received some special care from the devs, due to the MK2 pipe limit.
technically each generator and machine can have different exponents for OC scaling. Most machines have the same exponent though
i still wouldnt trust 600 mΒ³ in mk 2 pipes but thats just me
only difference is nuclear gen, instead of using 1.3, it uses 1.321928
straight pipe usually can do 600m3
My uranium plants are safe to fly around it seems, so I'll spend an extended amount of time observing them. My uranium array already has 100% of the functional design, and none of those plants seem to have problems. It's only the plutonium plants that are a mix of the two designs, and some have problems.
Maybe the VIP juction is somehow setting off the 600/min glitch?
I'm not using VIP or anything fancy. Just 2 extractors merged at a single junction, which then flow directly to a reactor. Each also has an MK2 pump. Every nuclear reactor is its own system like this, nothing is linked.
@fringe pawn Yes but the left design has the junction (semi) vertically, that's what the VIP junction is.
But, yeah, it's still a pretty simple design.
Huh. You're right, that is a key element of VIP, and I've been inadvertently including it.
Lesson learned: flat junctions going forward. If the system needs to be run at 600.
Maybe.
damn that was fast
Yup!
I feel like this channel and some others should have a requirement of 20 posts. Or something like that. Just so bots can't immediately spam so many channels and make more work for mods.
well mods just ban and delete all msgs with single command, so that wouldn't help
since bots would still msg #satisfactory or something that doesn't have that requirement
Ah, makes sense
can confirm: angled pipes are the oddest things. and junctions too
angled pipes are in fact so weird, you can set them up in a fun way and they dont need head lift, they just keep climbing infinitely
?
and i dont mean water tower
this shit here.
"Mk 2 Pipes to heaven"
2 mk 2 running up a slope
dont need pumps, they generate infinite head lift in each other
Oh, and even without the VIP junction.
VIP doesnt affect head lift afaik
With max flowrate?
yep, can run max flow
Wow!
some weird feedback
one pipe needs to be slightly offset from the other
then they start oscillating and induce head lift in each other
Due to slosh?
kinda. it has to be very dynamic
mk 1 is too slow for this
doesnt oscillate as easily and as naturally as MK 2
So it might get patched out. But I suppose that should always be coumnted on.
those start oscillating if you even as much as blink
I blink a lot!
omg great
mk2 pipes just keep getting more broken
pipes and belts need to implement graph theory to have optimised game performance - it should also fix all these weird glitchy behaviours ppl still keep finding
but graph theory is hard enough on paper, let alone in a game D:
google graph theory, there is a good short description there
basically turning the whole thing into a mathematical structure. the reason why i say it's good for this game is because you can instantly know exactly what is happening at any point in the graph or very quickly make a route, or find out the "value" of a route
Ah, like for the Travelling Salesman problem and such.
but it's also difficult to implement a system that allows user generated graphs
especially one where the user does not know they are generating graphs
I was wondering how that would help, make the water go the shortes route rather than going around in circles?
you can precalculate the flow
so you have 8 coal gens, each one cycles at a set interval, you have 3 water extreactors, each one cycles at a set interval, you can instantly extrapolate that e.g. in 39.2 seconds from now generator number 3 is gonna shut down for 2 seconds
Might even be more efficient, lagwise, if you're lucky.
Yeah, sounds messy, maybe it would be better ti give it a built in (hidden) buffer to even out the flow.
all the other things are taken into account like the max flow rate of each section of pipe, where the junctions are, the angles of the pipes and junctions etc etc
and the capacity of pipe sections
currently the system works by checking the state of each component on every tick, it does not know what will happen in the next tick let alone a minute from now
and each component does not know anything about any component not directly connected to it
in a graph, all the component are just nodes or data points and there is something that can see the whole graph
Not sure if this is the best channel to send this but I'm planning my first external factory to make computers (Silicon Circuit Board + Caterium Computer recipes), is this a decent plan for location and resources? I was thinking the train line would be good because I can expand it to include new factories when I make them.
anything I could do better or does this just suck
Oh and I already have rubber going to pretty much where it is
Caterium computer and silicon circuit board pair well. Consider also making silicon high speed connectors there.
If you can shift some quartz production, you'll also have all ingredients for oscillators present.
oooooooooooooooooooooooooo
yes yes i will do all those things
we're gonna need more trains, aren't we
now the question remains
should I take the time to put them on the same grid
like the two factories
If you can add aluminum casing into the mix, you can make radio control units there as well.
not there yet
but yes would it be worthwhile to make sure the two factories are lined up on the grid
Power grid? IMO everything on one grid except the machines necessary for creating the coal, fuel, ir whatever.
no like
placement grid
like I build out from my factory all the way over there so they could possibly physically connect
That can end up being useful. I have accidentally expanded in such a way that things did connect as you describe.
@fringe pawn update, is this good so far?
and more importantly is this how it works
will that unload
it should, if the freight platform is set to unload and the station is facing the right direction
holy shorts it's working
Looks good. As with all things, keep a close eye to make sure it works consistently. I'm about to fix a tiny flaw in the water flow of my power plants, which took quite some observation to spot.
I would try to build double train track routes. Sooner or laterβ’οΈ they will be adding in train colision and signaling.
Bidirectional single track setups would be annoying to fix when that comes later this year.
but if it's only 1 train and 2 stations it's fine. you can always add more freight cars
I need help setting up a oil system I've got two pure oil nodes practically next to eachother and I wanna make plastic and rubber with both of them how can I get the max from them?
I've only just unlocked oil processing and also packaging as only just done phase 2
If you're just starting then making rubber/plastic and HOR, turning the HOR into coke and either sinking that or using it in coal generators is probably all you can do.
But there are a lot of very good higher level and alt recipes that will give you much more rubber/plastic so be prepared to redo it.
Look up recycled rubber/plastic in the Wiki for the ultimate.
day 158 of wishing i had heavy oil residue
day 300 of realizing my oil factory is inefficient