math-and-meta

[whole heron]

Yes, it's silly to do it that way. Which is why I'm considering it.

[unreal spire]

You can also smart split packaged water and other dry goods to use a single belt for small things

[woeful shore]

You could send packaged water upside, unpack it upper, doing a pipe right to the ground and a uturn going back to the upside.
Then, you could say, you ve a silly way to push liquid water to the upside without any pomp

[deft lichen]

If it's already on a belt, why go back to a pipe

[astral hornet]

I'm building a Supercomputer and derivatives (ai limiters, high speed connectors and regular computers) factory. I've used the sf tools website and I need about 1200 copper ingots per minute. I've set up my rail station to bring in the copper ingots that are being produced off site, but I'm a little worried the throughput of a single platform with a single train won't be enough.

If it isn't, what's my best recourse? Add another platform and split the load that way or add a second train?

[wind spade]

that depends on a lot of factors

[astral hornet]

I'm assuming length of rail track as well?

[wind spade]

adding a train can help or harm depending on round trip time

[astral hornet]

I guess I'll let it run for a bit and watch the number on the platform

[wind spade]

usually I recommend one belt per platform (with industrial containers in front of each platform connected by two belts to plafrom), that usually handles any reasonable distance

two cars will be a safe option if you go for that

[astral hornet]

Oh, interesting. That would automatically translate to two platforms in my case.

[vapid gorge]

yeah, there's math you can do to figure out if your train can do X parts per min but generally 1 belt per platform is safe.

[wind spade]

unless you do LGIO level of train tracks

[vapid gorge]

1200 on one platform is possible but it'd have to be a fairly short trip

[astral hornet]

And it's not, my factory is built on the west coast where the oil is, but it's a loop that goes to the west of the rocky desert and to the quartz nodes under the pink bamboo forest.

[vapid gorge]

oh it's making multiple stops too?

[astral hornet]

yea

[wind spade]

I personally recommend one train line per trip (trip = between two stops)

[vapid gorge]

oh yeah that's going to add huge times to it. if it's more than a 5 min round trip I'd use the equations on the Wiki to see if it's enough

I generally don't recommend multiple stop trains

[astral hornet]

😅 last time I asked here, other people said "do dedicated trains per factory"

[wind spade]

not sure what that means

but you can still have dedicated trains per factory, just each train only goes between two stops

[vapid gorge]

yeah that sounds like doing a dedicated train per Point A to Point B. Maybe you misunderstood?

You can do multi stop trains - but absolutely do the math to figure out if you have the throughput

[astral hornet]

no no, they meant "have a train that does multiple stops. I think, English isn't my first language, maybe I misunderstood 😅

Dang, don't know if I have the room to add more stations, time to expand I guess.

[wind spade]

well, then they obviously had no idea what they are talking about

there's always room, map is humongous and you can build vertical 😛

[astral hornet]

Currently filling up a manifold of 34 constructors 😅 this is definitely the largest factory I've ever built.

[astral hornet]

I hate building train stations on top of each other though...

Oh boy, yeah... after running for about 30 minutes, looks like the throughput of the single station is only 500 ish copper ingots... Far from the 1200 I need 😅

[vapid gorge]

Like I said , you can do multiple stops on a single train - just needs to be more careful

[vapid gorge]

500 is real bad, do you have buffers going out and in from each platform?

[astral hornet]

No, not yet, that might be part of it as well

[wind spade]

that is definitely part of it

put industrial storage in front of every platform (both for loading and unloading)

connect it with two belts to platform

[vapid gorge]

You might completely fix your throughput and keep your multistop train if you put the buffers on

[vapid gorge]

you don't have to run train stations one in front of the other either - you can have them side by side

[astral hornet]

Oh I know, but I wanted to keep the space free

I'm measuring the round trip time, I'm afraid it's still too long :/

About 3 min 38 seconds. If this post is still accurate, https://www.reddit.com/r/SatisfactoryGame/comments/mm0l3q/analysis_everything_you_need_to_know_about_train/, that would mean about 850/minute. I'm good then.

[vapid gorge]

yeah 3.5 min isn't long

[vapid gorge]

don't rely on Reddit tbh - stick to the wiki https://satisfactory.fandom.com/wiki/Electric_Locomotive

[astral hornet]

Hmmm, would I need to balance my Copper ingot foundry somehow? I have two lines of foundries, each producing 600 ingots per minute, each line going to their own industrial container, then going to a platform. The platforms on the other side consume copper unequally. So I'm ending up with one platform being full and the other empty. Would it be enough to simply use an overflow splitter on the one that's going slower?

[vapid gorge]

yeah over flow manifolds would be fine

Are they in the same location? why are they being used unevenly?

[astral hornet]

The math works out, that would put the limit of my platform to 780 copper/minute, as I'm only hooking up a single belt, I guess but I only need 680...

[astral hornet]

I have one platform going to copper sheet production (678/minute) and one to fused quickwire (487.5/minute)

[placid oyster]

Either overflow manifolds or a load balancer would work fine considering its only 2 lines

[vapid gorge]

well you can do a manifold on either end and do fancy spliter's or over flow - or you can just have the copper feeding into it like this

[placid oyster]

high speed edit 😜

[astral hornet]

appreciated!

[vapid gorge]

assume the 2 inputs are the two station platforms

you have 2 different products you need that consume uneven amounts of raw goods

So just have manifolds coming out of each station as you would normally, but clock your machines to produce the right numbers of each item type. Does the diagram make sense?

[astral hornet]

yeah it does, thanks

[wind spade]

I'd prefer if both platforms used it evenly, then you don't have to solve this 😄

[vapid gorge]

For some reason controlling production and throughput through machine timing and selective merging doesn't occur easily to people.

[wind spade]

so build two sets of machines each needing 600/min

[astral hornet]

Ehh... I ratio'd it out and with some alts, this is fairly nice.

[vapid gorge]

I mean that's basically the diagram - just merging the outputs of the 2 dif manifolds in such a way as to get the numbers they need

[astral hornet]

I like whole numbers on as many machines as possible 😅

[placid oyster]

whole numbers rule

[vapid gorge]

eh if you go by whole numbers per min then it's completely arbitrary what you make the time unit.

[astral hornet]

sshhhhh, let me have this 😛

I have like 500 hours in this game but have never gotten this far 😂 I keep restarting or building myself into a corner

[placid oyster]

not really, making it anything else than mins will overcomplicate your life with conversions

[vapid gorge]

items per minute

[placid oyster]

theoretically you could say "1 item per hour"

[vapid gorge]

items per second

[placid oyster]

but it will be a pain in the ass when machines display per minute

[astral hornet]

this ain't factorio

[vapid gorge]

I'm just saying 'even items per unit of time' is arbitrary 'evenness'

[astral hornet]

of course

arbitrary evenness, best kind of evennens. Just like technically correct

[placid oyster]

theres no real "evenness" tho, it all depends on your number system 😜

[vapid gorge]

right

[snow dove]

i mean basically everything is arbitrary evenness

unless you’re measuring down to the atom, and even then it can be uneven by a small margin

[vapid gorge]

I don't know, I'm making like 12,000 quick wire per min in one factory. Per second is a reasonable unit for them too

[placid oyster]

yeah, but if I made 0.16666666 nuclear pasta per min I would claim its "1 per hour" to make it an even number

[vapid gorge]

again arbitrary 😛

[placid oyster]

and if I made 1.3333 high speed connectors / min I would say "4 every 3 mins"

[snow dove]

well you’re really making .1667 but mkay

[placid oyster]

even if it may sound nicer it would mess up your math

[placid oyster]

arent machines rounded to 3 decimals? so 0.167

[vapid gorge]

if clocked right then the occasional hiccups would bring it down to 1.666 😛

[placid oyster]

or is the internal counter more accurate?

[placid oyster]

it would actually be 10 per hour, I fucked up my math. just goes to show lol

[oblique hollow]

how about we start measuring in Million Units Per Annum?

60 per min? nah
31.563 MUPA? perfection

[wind spade]

murica intensifies

watermelons per football field

[snow dove]

“if clocked right”

[oblique hollow]

i clock stuff to make 1.666/min and it works perfectly fine so imma pretend like it wont matter in a few hundred to thousand hours

[snow dove]

oh yeah i do the same

idrc if there’s a yellow light for one second 700 hours from now

[frosty owl]

The precision limit affects the clock, all the other numbers respect that, even if with more decimals (though the UI show rounded values ofc)

[vapid gorge]

4 decimals. but if you round up the 4th decimal and precisely feeding them then every once in a while the machine will stutter off.

[astral hornet]

Most "annoying" part of this thing: the 270.9 raw quartz required 😒

(I've long decided that this thing isn't gonna be 100% effective 😅

[ember fractal]

Not 100% efficient, not good enough

[snow dove]

like 99.9999

[quartz violet]

Good 'nuf

[prisma kraken]

honestly, sometimes the decimals don't much matter

rates in the game are expressed in per minute quantities, sometimes this leads to a repeating decimal because the game itself does things based on cycles that I'm guessing have millisecond precision

a rate of .33333 repeating per minute just means the cycle time is 180 seconds

i believe that if you have a non-repeating decimal as the cycle time, everything ends up being exact

i'll add that the only set of recipes i've seen that really result in messy numbers that won't resolve nicely as a rational numbered cycle time are with the combo of iron wire & stitched iron plate

[placid oyster]

what are the issues people fear so much about pushing 600m2/min on mk2 pipes? backflow?

and why does nobody advise the same about mk1 pipes? wouldnt there be the same issues?

or is it something precision related?

[quartz violet]

I'd think the amount of pipe derp is higher with higher flow

[snow dove]

backflow is more prevalent and weirder with mk2 pipes

[placid oyster]

isnt backflow solved with a loopback?

or is that just a temporary measure?

[prisma kraken]

with mk2 pipes, if you push them to the max, some floating point rounding errors can occur if you have very large manifolds

[vapid gorge]

why? you can clock the miner to 270.9

[placid oyster]

I wonder what the hell the game is doing to get rounding errors, that should only start happening at numbers a lot larger than 600

maybe it measures fluids in millimeters squared

[vapid gorge]

pretty much yes

[placid oyster]

:xd:

[prisma kraken]

when lots of things are on the same pipe network, the addition and deletion of numbers on odd cycles can just cause a propagating rounding error

it takes a large pipe network to get that affect

[placid oyster]

so, for the record, should I:
a) push a bit less than max
b) add a loopback
c) both

[vapid gorge]

there isn't often you have very large pipe networks, besides turbo fuel and even then you can mitigate that by having small volumes in pipes

[vapid gorge]

just loop back is fine - it's not hard to get full 600

[placid oyster]

I was planning on having some full mk2 manifolds for DF

[prisma kraken]

internally, yeah, the game does represent fluid quantities in 1000th's of a m^3

[placid oyster]

thats why Im asking

[placid oyster]

makes sense considering the rates displayed on tools

[vapid gorge]

as long as you keep the pipe systems simple, don't do anything weird, simple loops work fine

[prisma kraken]

but a lot of the math for the pipes is done using floating point

and it just leads to rounding errors

[vapid gorge]

if you do the fluids from point A to point B and use a loop like this - 600 will work fine 🙂

[prisma kraken]

you can do stuff like pipe 600 water from 2 OC'd extractors into a OC'd nuke without problems

[placid oyster]

allright thanks

[placid oyster]

I wonder why

[prisma kraken]

because there's no inexact fractions in the math

[placid oyster]

given it uses mm3 they should be able to get away with int math

and even then, I think for floating point using bigger units is better? because theres more resolution the closer to 0

[prisma kraken]

truthfully, i think that they should redo the math on pipe calculations to be fixed point

but i've heard that they're planning to just scale the numbers for oil down a bit so 300->250 & 600 -> 500

[placid oyster]

yeah fixed point would be super fine

[prisma kraken]

what happens is 6000 is just a bit too big of a number for all the rounding to be exact

[placid oyster]

maybe use 10000ths of a m3 just in case

[placid oyster]

how would that help? you could still fill up a mk2 pipe, just not with a single oil node

[prisma kraken]

keep in mind as well that a lot of this computation is highly optimized in the game

[placid oyster]

yeah fixed point would be a bit slower

because theres no hardware for it

[prisma kraken]

and they're doing stuff like vectorized math using the smallest variable type they can so that the numbers fit into an MMX/SSE/AVX register, or the equivalent on the GPU

i don't know exactly how the code works, and i'd love to actually see how they're doing it all

(i love those types of programming problems)

but just understanding how to do things quickly and work with large volumes of computations, i'm supposing the problems that the game has are similar to ones i've worked on

technically fixed point can be quite a bit faster, but there's probably a good reason for why things are the way they are

really though, steering back to the subject of pipes, the rule of thumb i follow is keep pipe networks small

there's also a lot of misinformation & allegorical stories out there

[placid oyster]

given modern day fp hardware I think it can match it but its not gonna be a whole lot faster

of course, Id have to benchmark it to prove it

😉

[prisma kraken]

people try pushing 600 fuel into a pipe and hanging 50 generators off a manifold

that just doesn't work

the pipe network is too big

just the length of the pipe to loop around that many generators takes forever to fill the pipes

[placid oyster]

so my idea is busted :(

[prisma kraken]

well, just break it into 2 300 pipes fed by different groups of refineries

[placid oyster]

boring

[prisma kraken]

but really the misinformation that comes from it all

[quartz violet]

If it's all in 1 line it isn't gonna work but I was fine with splitting off like 24 groups of 8 from a 600 pipes for a field of turbofuel generators

[placid oyster]

I like to have 50 generators in a row covering half the map

[prisma kraken]

is that people don't wait for the entire pipe network to fill before starting to online the generators

i made that sort of mistake in my first playthrough in the U4 era

the network just NEVER got full

[quartz violet]

I connected up power lines for generators after everything was built and it was alright

[prisma kraken]

yeah, i mean, atm i have a 300 generator TF plant, its 15 groups of 20 generators, each fed by 2 blenders

(i'm UC'ing the generators if you are trying to math that out)

there's nothing wrong with making large fuel power builds, just be careful to keep the pipe networks small

[quartz violet]

I left mine at normal since it would have been a pain in the but to mess with the clocks peed on all 400 of em

[prisma kraken]

i don't think the problems with mk2 pipes are necessarily pushing them to max, but that if you push them to max, it goes against the rule keep pipe networks small

i'll also say that this is my advice gained from experience, and i haven't scienced it all out completely

[quartz violet]

Less pipe less derp

Makes sense

[whole heron]

What do you define as small in that case?

I think a manifold fully utilizing 600 m/s crude oil would need at least 20 junctions.

[prisma kraken]

I think keeping networks to < 30 producers+consumers is probably about the right rule

i think also there's another variable in it all

[quartz violet]

What if you split off of a main artery into 30 other pipes though

[prisma kraken]

i've gone as high as 36 TF generators off of one pipe

with the total number of machines being 41 irc, but the generators were fed from above

so slosh wasn't as big of an issue

[quartz violet]

If you use valves you don't have to worry about slosh

[prisma kraken]

valves do not work as you think they do

[quartz violet]

I know

[prisma kraken]

i can't even explain exactly how they work, lol

all i can say is that they rarely if ever solve problems

[quartz violet]

They are dumb sometimes but i think I used them to make sure each split off was it's own thing

[prisma kraken]

something else i'll add about pipes

is that headlift is buggy AF

there's some sort of hidden variable in a pipe network that remembers a head lift value

and placing a valve on a pipe can sometimes reset and change the calculation

i was experimenting with this a bit the other night

this should not work

and after tearing it apart and rebuilding it, it didn't... the top pipe segment after rebuilding wouldn't fill, as expected

[quartz violet]

That looks like more than 10m to me

It's 10m of lift off extractors, right?

[prisma kraken]

foundations are color coded so you can count

[quartz violet]

16m it looks like

If those are 4m

[prisma kraken]

its actually 20 meters, the extractor is placed a few meters lower so its pipe is at ground level

[quartz violet]

What is the extractor headlift?

[prisma kraken]

10 meters

that picture should be impossible

[quartz violet]

If the game had perfect physics it would be fine because of siphoning

[prisma kraken]

the vertical segment should be partially full, the top pipe segment should be empty

its a bug with headlift

when i deleted everything and rebuilt it, it actually did result in the expected

what happened is i did a baseline test of having a pump on the line initially

to verify the buffer would fill

afterward, i deleted the buffer, pipe and pump

keeping the extractor in place

and rebuilt it with the result pictured

somehow the extractor remembered the previous headlift value

i fiddled with it a bit more and actually couldn't reproduce what i pictured

conclusion: there's just a bug somewhere in the code where headlift isn't properly calculated

i can tell you roughly when the bug was introduced; it was during update 6's experimental cycle

caused all my coal power plants in my world at the time to run out of water, lol

[quartz violet]

Maybe they will squash the bug in u8

[prisma kraken]

its one of the things that they really should fix sooner vs later

i've caught a few yt streamers run into it on camera as well

[snow dove]

might make a couple bugs they need to squash

[prisma kraken]

usually in the context of headlift not transfering through a floor or wall hole

on this subject, i actually used to run into the bug quite often

at the time, i was building coal power using extractors that fed the generators from below

i'd run into it so often, that i started building the pipes in a different order

if the LAST thing you do with the pipes is place the pumps on, you avoid whatever the problem with headlift being incorrectly calculated is, in my experience

[vapid gorge]

what idea? you can loop 50 generators

[prisma kraken]

you can, but should you?

[vapid gorge]

easily.

just loop it, flood the system, go wild

hell, with turbo fuel you can have dozens of generators on an UNlooped manifold and have it work

[prisma kraken]

filling the pipes before onlining the generators, that's something i can't confirm or deny works

[vapid gorge]

if you just turn off 1 generator you'll be over feeding and it'll flood it

it'll jsut take a while

[prisma kraken]

i tried that back in the U4 days

the manifold never filled, lol

i tried it looped, forked, etc

it was also my first play-through, so god knows what other sins of ignorance i had going on

[vapid gorge]

ages ago there used to be a buffer loss bug where liquids disapeared on load

But if you built it right and let it go for a long while it should have flooded

[prisma kraken]

this was in U4, so that may have been what i was running into

it was my first game, and i have no desire to even open that file and figure out the trainwreck i created, lol

i am curious though about how broken the rifle ammo production line from it actually is now

...that was back when you needed beacons for ammo

i should actually open that save and take a look at that beacon factory though... might be some lessons in there for doing nuclear with the alt chain 🙂

i'm laughing a bit at some old pictures... U4 in all of its glory:

man, that stuff hurts to look at

[whole heron]

Why did things need beacons for ammo? Seemed to be overcomplicating something that's just a casing, primer, gunpowder, and a bullet.

[prisma kraken]

yeah, pre-u6, rifle ammo needed beacons as an ingredient

[whole heron]

No, I know. I'm just wondering the reasoning.

[prisma kraken]

i don't know, i think the devs wanted it to be more of a late-game OP thing at the time

[whole heron]

Funny since they've had to buff rifles multiple times.

[prisma kraken]

i think they envisioned the rebar gun as being projectile mk1 and rifle as projectile mk2

the new crab hatchers sorta made it so the rifle needed a buf

[placid oyster]

just use the magic music box

[whole heron]

You mean bombs?

[placid oyster]

no, the boombox turbobass

[prisma kraken]

it kills the flies

[deft lichen]

Nobelisks used to need beacons, with cartridges being made out of nobelisks. Then they rebalanced nobelisks to not use beacons anymore, but probably wanted to keep them in the cartridge production chain for some reason

[whole heron]

That makes sense. I guess they were supposed to be used for the transmitter and fuse.

[quartz violet]

Wait the boombox does something?

[whole heron]

1 damage and knockback apparently.

[quartz violet]

I had no idea about that

I thought it was just music

[whole heron]

It was probably meant as a joke but turned out to have utlity.

[quartz violet]

Tbh I forgot it existed until now

[placid oyster]

yeah it counters crab flies

[wind spade]

Ficsit was tracking your ammo efficiency and accuracy

[terse stone]

Added here! https://github.com/landonairey/LocationOptimization

[deft lichen]

Great!

[prisma kraken]

iirc pre-u6, rifle ammo was made with beacons directly, and nobelisks were always pipe + black powder

[deft lichen]

what I mentioned is pre-u3

https://satisfactory.greeny.dev/items find the nobelisk here, this is a snapshot of U2 recipes

[prisma kraken]

ahh, i started in U4

i forgot about the 'seismic nobelisk' alt thought, that was still around when i started!

[oblique hollow]

https://questions.satisfactorygame.com/post/644ede8f5298eb579df106ae
reminder that THIS bug exists, in relation to "whats up with headlift" discussion above

[prisma kraken]

interesting

they're probably all related to one another in root cause

[oblique hollow]

Pressure Loss variable is one of them, but i cant tell you the other issues

[deft lichen]

lol what a bug

[oblique hollow]

another bug thats easy to reproduce

[prisma kraken]

file it all in the 'headlift is wonky' bin, lol

[oblique hollow]

make a pipe junction in vertical arrangement, connect a water extractor to the side and then build a pipe roughly 40 m up from the top pipe output

junctions have some ridiculous issues with head lift

[prisma kraken]

did you splice the junctions onto the pipe?

[oblique hollow]

it can be either splicing or pre-placing

doesnt matter

[prisma kraken]

gotcha

i think that there's some variable related to headlift that needs to be re-computed each time a pipe network changes, but isn't

[oblique hollow]

i think the bug related to valves generating infinite lift if you simply slam the slider from left to right also still exists

[prisma kraken]

i wouldn't be surprised

i started doing my experiments the other evening after evil steve pointed me to a recent kibitz video where he was stating that valves remove headlift 🤦

[oblique hollow]

they only block the blue hologram indicator

same thing with buffers

[prisma kraken]

he had a pipe in his build where headlift wasn't transfering

[oblique hollow]

rarely happens

[prisma kraken]

one of those 'rebuild the pipe, jiggle the handle and it works' things

but kibitz asserted that valves negate headlift, and that led to a bit of a lively discussion

but kibitz asserted that valves negate headlift, and that led to a bit of a lively

oops

sorry for the repeat

[oblique hollow]

yeah that was the one thing that really stood out to me from the vid

mid you that valves do specifically have an internal variable so they can transfer head lift

but, as with flow, thats limited to 8 bits

and i do not know what a full 8 bit head lift value is

[prisma kraken]

i'm starting to theorize that a pipe network has a ref-counted shared state for all elements

[oblique hollow]

ref-counted shared state?

[prisma kraken]

like there's a chunk of data shared between all segments of a pipe

that chunk of data has some reference count of the number of things that share it

[oblique hollow]

well all connected pipes do exchange data by nature of pipe networks existing

everything connected together is a pipe network. buffers and valves and pumps included

[prisma kraken]

i'm theorizing that the code's internal structure does something like that

[oblique hollow]

it stops at machines

and the network handles a few different things

like the fluid type for example

that info is pretty much instantly transmitted

[prisma kraken]

i was a bit surprised in my experiment that i got the wonky headlift that i had

[oblique hollow]

its not from pipe to pipe, all pipes register to a network and that network then shares the fluid type with all pipes of that network

[prisma kraken]

it took me a few tries to realize that the extractor was some how 'remembering' the headlift from a previous network it was connected to

[oblique hollow]

i think i remember that pipe networks get the first and last piece of a network and their height

[prisma kraken]

that might be why i was able to trigger the issue

i'm personally not going to stress too much about it all besides knowing it can get wonky and how to resolve my in-game problems, if i don't have access to the code and a task of fixing it, there's no point in much more 🙂

[placid oyster]

Where is this code from?

[oblique hollow]

thats from the modder's resource of game header files

this is a line from the pipe network headers i believe

[vapid nest]

Uhm quick question: If I have 3 mk2 pipes (1800) that need to be split evenly into 4 pipes (each 450) can I just make a loop connecting all 3 pipes and split them off into 4?

[vapid gorge]

I wouldn't - keep pipe systems w/o junctions. What are you using these 4 sections for?

[vapid nest]

4x15 refineries

[vapid gorge]

all doing the same thing?

[vapid nest]

Yup

[vapid gorge]

make it 3 groups of 20

[vapid nest]

Can't

[vapid gorge]

why not?

[vapid nest]

Symetry

And I will not tear down 40 hours of work

[vapid gorge]

still doesn't stop you - you can keep the 3 manifolds of pipes and just snake them around

you can try to merge the pipes but it's going to get messy, harder to stabalise and will be a much bigger pain in the ass if you're trying to diagnose the problem

[vapid nest]

Don't fluids balance themselves?

[vapid gorge]

yeah that's not the point - a lot of junctions means you're making the whole system inter reliant on other parts

it means sloshing can be more likely to happen, for 1 machine to throw out the whole system rather than just it's sections

I'm not saying you can't do it that way - but it's less reliable and can be more difficult to deal with in the future

[visual cove]

and I followed up to point out he realised his current head lift end bands were just before the valves which is why he thought they negated headlift, but he also said that doesn't explain why fluids aren't rising up the 4m and 8m steps.from the wall mounts.

[vapid gorge]

Good rule of thumb with pipes is send X fluid pm from Point A to Point B. Don't have junctions, don't merge them with other systems.

These aren't rules you have to do to make it work, just good rules for making simple reliable systems you can more easily fix if something is wrong

[prisma kraken]

yeah, its bugged in the game, but valves should not remove headlift

[vapid gorge]

yeah I skimmed through that video and saw that bit - that was entirely incorrect right?

[vapid nest]

could i just put valves or pumps at the 4 pipes to prevent backflow?

[vapid gorge]

no valves

and the back flow problems will be between machines

[prisma kraken]

btw, i'm planning on getting to your aluminum wastewater observation and sciencing that a bit myself... just haven't gotten there yet

[vapid gorge]

you ever see The Incredibles?

[visual cove]

yes, see my comment immediately above.

[prisma kraken]

there's bugs with how headlift is calculated

McGalleon and i were commenting on it earlier this afternoon

its like with some sequences of building the values aren't re-caclculated

[vapid gorge]

I didn't realise they also had a headlift calculation issue

[prisma kraken]

without actually sitting in a debugger with the code, i can't tell you what's going on

just that there's a problem with it all

when i run into the issue, i just rebuild the offending pipe, replace the pump on the pipe and it suddenly starts working

totalXclipse ran into it on-stream as well recently when he was building a coal power plant

irc, he stated sometimes pipes don't truly connect through floor holes

and perhaps that's it

[vapid nest]

a "no capes" reference?

[vapid gorge]

If you do the whole system merge you can put powered pumps on spots you’ve calculated properly- if it doesn’t work after you flooded the system and turned it on

But keep the rule of leaving a machine off so that the system is over fed at the start

[visual cove]

or it could be ~~~https://questions.satisfactorygame.com/post/63867a41ca608e08035275da~~https://questions.satisfactorygame.com/post/63862593ca608e0803527554. only time will tell

[vapid nest]

Alright. I've also just found a way to have 4x450 instead of 3x600

[vapid gorge]

splitting it at the start?

[prisma kraken]

without seeing the code, we're blind men feeling the elephant

i don't profess to know how it works at the code level, its just empirical, this shouldn't work, this should work, ya kno?

[vapid nest]

3x pure nodes overclocked to produce 450, then 1 normal node overclocked to 250% (300) and one impure node overclocked to 250% (150)

[vapid gorge]

ah is that using all the oil available?

[vapid nest]

There would be one more normal node, but I have no use for it

Yet

[lusty summit]

dont overcomplicate stuff, if you need 450 per pipe, feed that pipe with 450

underclock and overclock is a thing

for that reason

[delicate chasm]

Messing around with clipping. I don't like this but before I move on, was wondering if anyone could maybe tell me why 3 of the 4 refineries hooked up with this lift-lift-merge configuration work but the 4th one doesn't even after repeated rebuilds?

[vapid gorge]

can't tell you w/o more info, could be math, could be a bad connection shrug

[delicate chasm]

I committed this sin to make it work, but when this is built using the same spacing as the other 3 mergers, the second lift does not connect to the merger.

It does snap to the position but the connection isn't established to permit items through, specifically.

[vapid gorge]

it would clip less if you just put a belt under the pipe/across the pipe

[delicate chasm]

Well I know that, I'm not keeping it. I'm just curious what might cause it to not connect when the other 3 have the same spacing/height and do connect.

[vapid gorge]

where doesn't it flow? does it go into hte first lift but not the next? find out where the items stop and you'll know where the issue is

[delicate chasm]

The items go into both lifts and do not exit the second lift into the merge, like I said: the second lift doesn't actually connect to the merger. I just don't know why.

I rebuild the second lift and merger again and it does work this time, but I did not make an error when building it repeatedly before posting here - it definitely was perfectly lined up with the other mergers and behaved differently.

[vapid gorge]

I mean it's a pretty messed up clip job - did you actually hear the sound effect 'click' when a thing snaps into place?

It's a messed up clip xD not designed for it - not surprised if it doesn't work 100% of the time

[delicate chasm]

🤦‍♂️ The only question I have is "does anyone know what was specifically the problem here?" Not trying to be rude, sorry...

[vapid gorge]

'something wrong with the clipping - probably'

[delicate chasm]

Like is there known to be a bad interaction when clipping pipe + lift or etc. that I should be aware of, or is this an enigma to all who behold? Y'know?

That kind of thing, so that I get the meta knowledge.

Thank you for the insights in any case.

[vapid gorge]

no it'll be completely independent of hte pipe - if you clip so deep into a merger/spliter it may not hit a point where it'll establish a connection. And even if you think you placed everything in the exact same place the ORDER of the parts placed can matter if something will connect or not, and sometimes order of placement will have very minute differences in orienatation or distance relative to other objects

This generally doesn't matter at all - however I can see how it'd mess that tight a clip up

There's a big enough tolerance for things to connect even with the basic slight differences that can crop up like that.

[delicate chasm]

That's my current theory. I know the design itself was replicated and the build order was the same, but there have been some times when snap-to has given me almostbutnotquite. This instance was strange because usually it's immediately obvious which step was done incorrectly.

Using holes for instance. That'll get you a variety of small issues depending on what you do but there's a general flow and logic you'll fall into naturally through experimentation.

I was just certain it had to have something to do with the pipe though, since without the pipe in the equation it's just output -> lift -> lift -> merger all on the same precise z-level due to foundation. 🤷‍♂️

[whole heron]

How do you ensure a splitter -> lift -> merger connection doesn't break?

[snow dove]

you place the splitter and merger, then place the lift

if it works at the beginning it will always work

[whole heron]

I mean when building, to be clear.

When I build too tight on a blueprint, it's hard to tell if the lift actually connected on both sides.

[snow dove]

you can hear a “click” if it properly connects on the output end

[mystic moon]

You can also highlight the lift in dismantle mode and see if the rubber thing is extended

[whole heron]

Thanks.

[whole heron]

Hmmm... pure iron recipe... 780 iron ore -> 770 (22 refineries) + 10 (1 underclocked smelter) -> 1440 iron ingots, which is actually very divisible.

Does this make sense?

[prisma kraken]

pure iron doesn't resolve nicely with 780 ore input

its kinda a cruel joke

[whole heron]

No, I mean does adding the smelter make the next step resolve nicely? Or am I sleepwalking and missed something obvious?

[prisma kraken]

no, it just always proagates the messy numbers

anything with 780 input always is b/c its 60x13

that thirteen as a prime number is sorta killer

what may be helpful in your number noodling, is that 780 = 480 + 300

[whole heron]

That's not what I'm asking.

[prisma kraken]

only thing in my numbers that actually works with pure iron is making screws

everything else ends in a messy fraction or decimal

its been a while since i spent a day trying to resolve that math with pure iron

i can't recall the numbers off hand, but realized it was a blind alley

[wind spade]

you can always underclock to get nicer numbers

[whole heron]

That also wasn't the point.

Just don't mind me for now.

[oblique hollow]

well then you should explain your point better if everyone keeps missing it

[frosty owl]

I did that once. I thought the smelter fit nicely, dividing the 2 rows of Pure Ingot refineries... A bit silly looking, but quite the break from the usual refineries fields that are pure recipes ^^

I don't recall the output numbers matching up particularly well with anything though... I just break up the refineries' outputs in numbers I like (eg: 420, 600 or 720/min are numbers I like) + whatever is left

[fierce cypress]

because of the build order, build the lift connecting to the merger first, and then the one to the ref

[sand epoch]

Or just tap r...

[delicate chasm]

I guess so? I fully disassembled and rebuilt it several times before posting here, and the only difference between the final attempt and any other (I did try merger then lift as well as lift then merger) was that I had made a temporary spaghetti belt to empty the machine while I came here to ask.

I tabbed back over, deleted the spaghetti, built the exact same thing one more time and it functioned "normally." (like the others)

You've reminded me though that I meant to go back over there and fix that. It looks terrible.

[void hollow]

is there a better h. mod frames number than 10? lot of awkward numbers to throw around

needs like..... 3.02 belts of iron ingots for example

[median heath]

11.25

[void hollow]

rgr I'll try it

[median heath]

11.25/2.8125 = 4 Manufacturers.

[void hollow]

yeah it makes sense I think my brain is still locked into round # per min which isn't as important as I think

[median heath]

Unless you're on base recipe.. in which case stop using the base recipe.

[median heath]

45-81 rule.

[void hollow]

45-81?

[median heath]

Book incoming, you were warned 😛

So the premise of the 45-81 rule is getting production in "clean numbers" with respect to clock speed.
As you may or may not know, the game enforces a 4 decimal limit on clock speeds (it does not care about parts per minute, nor does anything in the game actually operate on parts per minute, EVERYTHING is done in individual cycles with respect to clock speed for cycle time)

Example: if you tell a Smelter to do 20/min instead of 30/min, the clock % changes to 66.6666%.
The UI will say 20 ppm, but this is a lie to make you feel better about yourself. The machine truncates anything after xx.xxxx%, so you do not actually get 20/min, you get 66.6666% of 30 (which is 19.99998).
Now, to MOST people, this doesn't matter. But I am a stickler for precision, so not actually getting 20 bothers me (even though you'd literally see a production difference only after a couple hundred thousand cycles).

So the math was done to find out how to make production lines "behave", and the first thing we solved was the 81 part of the rule. With respect to Rubber/Plastic - always do them in multiples of 81 and all machines in the recycled loop automatically fall within the 4 decimal limit. No repeating numbers, no imprecision.
The 45 part of the rule applies to non-oil products. Do your final product amount in multiples of 45 and everything in the production chain automatically behaves without you needing to think about it.
Exceptions: Caterium Circuit Board, Pure Iron, Fine Concrete
(CtCB and PureFe usually outweigh the rule because you'll be doing them in such large quantities that they fine a common multiple somewhere up the chain.
No one should ever use Fine Concrete.)

Note: 45-81 is the lowest whole number multiple, but you can cut them in halves to achieve closer to your actual goals.
I.E. 45/22.5/11.25/5.625/2.8125, or 81/40.5/20.25/10.125 (going lower than 10.125 on oil causes issues)

So if you needed 100 Plastic, I would instead say to make 81+20.25=101.25

[void hollow]

yesssssss this is the good stuff, ty

[median heath]

Now, before Iroh can get to it:

The amount of loss/extra you will have from repeating decimals is something that 99%++ of the playerbase will never even notice.

So it is up to you if you care about the rule.
But it does help with making "how much?" decisions.

@snow dove I gotchu ❤️

[void hollow]

I broke every rule to get to trains but now the real game starts, I'm gonna care as much as I can heh

excited to do it The Right Way heh

[median heath]

Before all my haters come out of the woodwork, I would never say that using the rule is the one "right way" to play the game.

I believe it is a very good way, but there are very few instances where there is a singular "correct" way to do something in this game.

[void hollow]

yeah for sure "The Right Way" for me is to try to get the numbers to all line up

idk for anyone else heh, I've just got a clear vision in my head for what I want

[cinder silo]

I keep forgetting about this to the point I just copy pasted the above in to a text file as something to review when I'm trying to un-stuff numbers.

[median heath]

I just type in the search bar "The premise of the" and it takes me to the last time I posted it so I can copy/paste. 😂

[mystic moon]

I have a large text document full of all sorts of long texts I need often for discord lol

[median heath]

I have 6 things on my phone clipboard.

Throughput Equations: (1560 is used because it is the max of x2 mk5 belts, and all Times are in minutes.)

First you need TimeToFill. Which is reliant on Cargo Stack Size and Car Capacity.

TtF = ((StackSize * Car Capacity) / 1560) + 0.45133...

If TtF >= RtD (Round Trip Duration)
Throughput = ((RtD - 0.45133...) / RtD) * 1560

If TtF < RtD
Throughput = (TtF / RtD) * 1560

^ number 1

50 per Stack
-88.62s RtD
-1083.3 Items/min

100 per Stack
-150.16s RtD
-1278.66 Items/min

200 per Stack
-273.23s RtD
-1405.4 Items/min

500 per Stack
-642.46s RtD
-1494.25 Items/min

^ number 2

300 Oil

5 Rubber Refineries @ 100%
5 Plastic Refineries @ 100%

Byproduct HOR routed to
4 Coke Refineries @ 93.75%
Feeds 18 Coal Gens with no waste.

^ 3

As it stands now the total amount of the base Project Assembly parts you need to complete the game are:

Smart Plating: 6,550
Versatile Framework: 13,000
Automated Wiring: 60,850

^ 4

Non-consumable items worth storing:

Plate, RIP, Beam, EIB
Pipe, Rod, ECR
CSheet, ASheet, Plastic, Rubber
Frame, HMF, FMF
Wire, QW, Cable
Crystal, Osc
CB, HSC, AIL
Rotor, Motor, Turbo
Comp, RCU, Super
Casing, Cooling
Concrete, Silica

32 in total.

[cinder silo]

I'll find out if my conveyor mixer that uses containers will work

[median heath]

^ 5

Mods 🤢
^ 6
Even though I am not allowed to use that one anymore 😂😂😂

[vocal tundra]

This used to help me so much lmao ty

[median heath]

Used to?

[cinder silo]

Mods 🤢 anyway 😄

[vocal tundra]

Yeah used to need to refer to it but now its in my head

[mystic moon]

Mods 🤢 has been said almost 400 times

[median heath]

Fair.

[median heath]

And then Ben struck me with his righteous sword... And cut the word+emote from my tongue.

[snow dove]

ty!

[cinder silo]

@steep rivet So far it looks like the mixer cluster works as intended.

[delicate chasm]

@livid canyon 210/m iron rods from the player would make it 14x the crafting speed of the machine. I didn't do a very controlled experiment but that's close to the other estimate.

[steep rivet]

nice

[livid canyon]

Fair enough

[prisma kraken]

hi guys

[livid canyon]

quick place to post math, pure aluminum alt recipe is 60 scrap to 30 ingots 2:1 ratio
default is 90 to 60 3:2 ratio

[prisma kraken]

someone in pm called my attention to this fact as a fact

and observation

there's a set of triables that have an even nuber of degrees, like a 30/60 90 for ecavple

45/45 is another sucg ecample

the sat of triangles with an even rumber of ecgress

and the sset of triangls with world buildaboles in s are different

and disjont

you can never resolve a prime number that's an integer into a rational number

[true junco]

What? All prime numbers are integers...

[haughty glade]

You sure? Have you checked them all?

[vapid gorge]

by definition they are

[haughty glade]

Seems like a cop out answer to me

Can't be sure without empiric proof

[vapid gorge]

🧌

[fierce cypress]

The definition of a prime number is a natural number that is greater than 1 and cannot be written as the product of two smaller natural numbers. Seeing as natural numbers are a subset of integers, yes, all prime numbers are integers.

[haughty glade]

Spoken like someone who never checked each prime individually for integerness

[fierce cypress]

Spoken like an uneducated troll

[haughty glade]

Hey! I may be a troll, but that was uncalled for

[fierce cypress]

You're the one that said 'integerness' so I beg to differ

[haughty glade]

Fine, but you're the one who forgot to add an exception for 1

[wind spade]

Not to mention that you're trolling in a wrong channel, #off-topic-general please

[haughty glade]

I'll stop

[quartz violet]

May I direct you to the video of Michael from Vsauce reading every prime number for 3 hours

[placid oyster]

Not every prime number

[quartz violet]

Michael is supreme lord and there are no prime numbers other than those uttered my Michael's all-knowing head

[terse stone]

Couple updates to this analysis coming along nicely. I've added nitrogen gas well locations and more importantly it shows results for top 10 locations to build a factory.

This test is looking for best locations in proximity to 300 coal, 300 iron, 300 limestone

[solar rampart]

is there something I don't understand about the per minute math?
I made 4 constructors putting out a total of 160 screws per minute, into two assemblers doing rotors (total requested 200 srews per minute)

both assemblers are accumulating, not draining, screws

what gives?

i should expect 80% efficient from both

[whole heron]

How many Iron Rods are you putting in?

My guess is that either the output is blocked or it's running 80% of the time and stalled 20% of the time.

If you want 80% efficiency and 100% uptime, you should underclock.

Also possible there is something wrong with your belts. But that's always a case.

[snow dove]

belts don’t spawn screws from nothing

[delicate chasm]

Could be their rod belts, guessing that's what they meant.

[snow dove]

ah

[delicate chasm]

What would you all do with 110 spare iron ingots that you don't need for your current plan on-site? Just turn it into something anyway? Sink and be done? Underclock the smelters and miner? Recommendations/favorites?

[wind spade]

I recommend to not have spare ingots 😄

[delicate chasm]

That's the idea of asking yeah.

[whole heron]

Output belt could also be a problem, but that's less likely.

[wind spade]

what I mean is to not plan a factory that has leftover ingots in the first place

[whole heron]

Sometimes you get weird floating point ratios.

[snow dove]

why would i produce more ingots than i use?

[whole heron]

I suggest turning the extra stuff into cable. You can always use cable.

[delicate chasm]

I'm planning the factory to not have 110 spare iron ingots right now... I didn't think that was unclear to begin with, sorry.

[wind spade]

then why do you have 110 extra ingots now?

[delicate chasm]

o_O

Because the math isn't done my guy.

I'm still playing the game, not looking at the finished result.

[wind spade]

I'd need more context then, because this seems like you're planning in some weird way

[delicate chasm]

I dislike the iron wire alt on general principle but that would be kind of small and cool. I could make a little addition wing thing.

I know what I need, the node potentially offers an extra 110 ore and I'm not piping in water. Result is I have 110 extra potential iron ingots on-site that aren't needed.

You don't need the context if you trust that I do in fact have 110 spare iron ingots on-site after taking into consideration my infrastructure and what I'm willing to build. But to not write a 2-paragraph post I gave the salient number: 110 ingots. =/

[wind spade]

the node potentially offers an extra 110 ore
ah this is the context I needed

underclock the miner and keep it for later

[delicate chasm]

🤦‍♂️

[wind spade]

which is exactly what I said in the first place - don't plan the factory to have extra

[delicate chasm]

No. That isn't the situation.

[frail light]

in case you want them for later....

[delicate chasm]

The situation is that the ingots are going to be made because the node is going to be walked away from after this build.

[snow dove]

for what???

[wind spade]

then just underclock and leave it not at 100%, not like you'll ever need the extra iron

and stop planning based on nodes, plan based on what you need

[delicate chasm]

Totally legit. One of the options.

[frail light]

to build an ingot statue or expand or just to sink while laughing manically..

[frail light]

expand production of what your making to use the 110 ingots.

[delicate chasm]

I could even tap a whole other iron node and just slightly more than double the output of that whole place. One of the ideas I have is to come back much later and do that - this is modular frames.

I think I'll go with Not's suggestion for now, with cable, as that's small and easy to dismantle when I do decide to update this facility.

[frail light]

i would double it as modular frames are useful later on and it sucks re expanding sometimes

[solar rampart]

@whole heron it ended up being a false alarm (i guess). The rods were saturated at the time and screws were going up, but since then it evened out and settled at 80%. It must have been a transient condiiton. Faith in math has been restored

[delicate chasm]

That's very true, I'm on a restart and have experienced setting up HMF. Keeping my options open and minimizing rebuild at that location in anticipation of that expansion is one thing on my mind.

[frail light]

i personally plan to make enough for the final item in the chain once i reach a stable power grid so for example how many hmf do i need to make the desired fmf. and in turn how many mf will i need for hmf

[true junco]

Plan whatever way you are comfortable and successful with...

[snow dove]

oh i just go “i want like…………. 6ish FMF’s” then build all new machines for the FMFs

[delicate chasm]

Yeah, that's what I will end up doing - the initial builds are for producing building materials only. I make iron rods and plates in excess of what's needed for the frame chain for example, and when I come back, that will be changed.

[frail light]

that was fun until i ran out of room and had to tear everything down to rebuild from the ground up for that line

[snow dove]

wdym “ran out of room”

[frail light]

i had entrapped my line in other lines and yeah

not my most brilliant moment

[true junco]

I am well aware that the way i think about things is atypical...

But because almost my entire career has been in resource extraction (Mining and Oil and Gas) my paradigm is always to "Pull as much material out of the ground as possible"

[frail light]

okay i didn't have to tear down and rebuild but my desire for neatness said no we are keeping it all together all the rod making for this line all the mf and all that

[true junco]

I figured when anybody says this, they meant they got in their own way, or exceeded the planned design limits that were set by an aesthetic choice or factory elements that were already in place.

[delicate chasm]

Resource efficiency is a perfectly valid school of magic. =)

[frail light]

i got in my own way by expanding other lines into the reserved space for expansion later

[sly belfry]

https://www.youtube.com/watch?v=WIRK_pGdIdA

[wind spade]

#off-topic-media

[wind spade]

@serene rapids

[ember fractal]

Alum casings are like a late-game version of screws. You need a ton of them

[vapid gorge]

eh, you don't need like 12 belts for one fctory

[quartz violet]

If it's a really big one

[vapid gorge]

12xmk5 belts of casings?

[tribal cedar]

is there some spreadsheet for optimal factory ? i know its lot of calculation but the game has few years so maybe some1 did it 👀

[vocal tundra]

You could use satisfactorytools are the calculator on SCIM

[snow dove]

depends on what you want to produce

like Master said, you can try satisfactorytools or satisfactory calculator

[tribal cedar]

everything....map has limited resources and there are only specific recipes so there is limited amount of what can be produced and so there also has to be optimal way of producing as much of it as possible from available resources

[snow dove]

entirely depends what you want to produce, and how much of each

[vocal tundra]

^^^

[snow dove]

and an astronomically small amount of people use all of the map’s resources

there’s 70k iron

highly doubt you can even use just all the iron, let alone everything else

running out of resources is a non-issue for a vast majority of people

[vocal tundra]

^^

[snow dove]

like 99.999% of players won’t get to that point, ever

[tribal cedar]

i guess im not average player i made my own basic calculator but im lazy to finish it to the point where i could calculate optimal path

[vocal tundra]

Yeah some people have saves that have 1000s of hours and havent used everything

I dont wanna sound like a broken record but satisfactorytools and SCIM calculator

[snow dove]

really think you can use 90 completely full mk5 belts worth of iron

let alone the other dozen resources

[vapid gorge]

there's no 1 optimal path - it's entirely depending on your final goal

Also your computer will melt unless you're using a super comp at a university

[tribal cedar]

i mean there could be multiple optimal paths which would yield pretty much same result im not saying there is exactly 1

[snow dove]

12.5 bauxite
14.1 caterium
39.6 coal
37 copper
19.5 oil
90.2 iron
67.7 limestone
20 nitrogen
13.4 quartz
8.7 sulfur
2.6 uranium
MK5 belts/mk2 pipes worth

[snow dove]

there are “best” recipes items, yes

[tribal cedar]

when we were trying it in update 5 we had mods for 2k belts and we got to point of processing like 20k iron in 5 days of building but then we got bored 🙃

[vapid gorge]

but it entirely depends what items you want to build

if you want 1,000,000 screws a minute you'll have less resources for everything else

[vapid gorge]

go to www.satisfactorytools.com and fiddle to your heart's content

[tribal cedar]

i was trying satisfactory calculator in past but from some point it just gets insanely laggy so i just wondered if some1 did it as finished spreadsheet 🥴

[vapid gorge]

calculator sucks
Use tools

[vocal tundra]

PraiseGreeny

[vapid gorge]

and as the person who created Tools said- there's like 10^6 recipe paths you can do for things

so no no spread sheet. Go to tools and make your plan

I'm going to assume you also just bought the game

[vocal tundra]

PraiseCobalt? I apologise i thought it was just greeny lol

[vapid gorge]

missed a word before the edit xD

[vocal tundra]

Ohhh

[vapid gorge]

it makes a big difference for being such a little word

[vocal tundra]

Yeah lol

[snow dove]

the reason it can’t be done, is you have to decide yourself how much of each item you want to make

and what items you want it to make

like if you have 100 iron ore, you could use it to make 100 iron rods, or you could use it to make 66 iron plates

[tribal cedar]

well my thinking was opposite - not to craft specific item but to use all resources such that priority are those items which are final (there is nothing to craft from them like nuclear pasta)

[vapid gorge]

again - in what numbers? here have a production plan for world maximum assembly directors https://www.satisfactorytools.com/production?share=RbgXe416UxZR5zJmheRS

its not hard at all to get a world optimal plan, you just have to decide the end goal

[oblique hollow]

then planning effectively comes down to doing the very very last parts first, and picking a final amount for them

so in the end you yet again have to settle on at least 4 values

[quartz violet]

and screwing with the numbers until it doesn't look like they were beefed up by the cloudflare lavalamp randomizer

[snow dove]

yeah but certain items use the same resources as another item, so producing more of one item, means you produce less of another

[quartz violet]

Some number times the recipe for every item is impressive enough

[lapis geyser]

What's funny too is there's 30,000 buildings in that build. (not counting power of course). Even if you could place 1 building a second it would take 8.5 hours just to place all the buildings. Let alone belts and power infrastructure, foundations etc.

[whole heron]

Blueprints will save work on the redundant parts.

Also, holy hell.

[lapis geyser]

Of course it's also impossible since it uses all the oil and coal on the map so I have no idea how you're going to get the required 530 GW of power.

[snow dove]

nuclear?

setup a power plant, and a bajillion power storages, charge the storages, disconnect the power plant, connect the factory?

[lapis geyser]

Nuclear wouldn't work because you couldn't make the uranium fuel rods when all the resources on the map are consumed.

[whole heron]

Yeah, Nuclear still needs a lot of non-Uranium things.

[median heath]

He's talking about pre-loading the power.

I.E. By charging storages or pre-making a couple dozen ISCs of fuel rods.

[lapis geyser]

I understand, but you get to run for what like 20 minutes then have to stop?

[median heath]

If you pre-load fuel rods into storage you can go a lot longer.

[lapis geyser]

Alright, well I'll amend my statement from impossible to "highly impractical."

[median heath]

Possible but with limited run time.
Yes.

[lapis geyser]

Though I guess it's technically impossible for another reason. It shows 11700 oil on the map and the last I counted in U7 there was only 9900 oil. Don't know about the other numbers.

[median heath]

You counted wrong.

There is 11700.

[frosty owl]

I doubt any of us counted, but the data used on the planner are accurate

[lapis geyser]

I'm curious now. SCIM shows 8 Pure nodes, 12 Normal nodes, and 10 Impure nodes which equals 9900 (4800 + 3600 + 1500). Where's the discrepancy?

[median heath]

In your world do Wells not exist?

[lapis geyser]

Ah, I did skip those. Thanks!

[turbid plinth]

hey guys, I am a bit new to the system, can someone guide me to a 0 waste nuclear power setup?

[vocal tundra]

Do you have plutonium unlocked?

[turbid plinth]

i have about 200 hours in the game, everything is unlocked

but the formula for the 0 waste setup eludes me

[vocal tundra]

Satisfactory tools might help you

Link is in #welcome

[median heath]

It is just making Plutonium Rods and sinking them.

If you're burning Plutonium Rods, it isn't 0 waste.

[turbid plinth]

I am using the tools, but i didnt knew that it shows it for power production

[vocal tundra]

If you input plutonium rods im pretty sure it will automatically add the nuclear powerplants as the waste from them is needed to create them

[turbid plinth]

ahaaa, so plutonium and sinking

thanks both of you

[vocal tundra]

No problem!

[placid oyster]

Only beta can do fuel production I think

[wind spade]

have you told it how much uranium waste you have?

[placid oyster]

Since you have to enable uranium power generation

Unless you input the waste

[wind spade]

it can't make plutonium rods without uranium waste

[turbid plinth]

even more lost now 🙂

[wind spade]

go to Items, Input and put Uranium waste into the input field

[turbid plinth]

nope, still doesnt work

ok, simpler, what do I burn in the plant? Encased U Cells?

[wind spade]

can you share the production line? press the blue share button and paste link

[turbid plinth]

or U fuel Cell?

https://www.satisfactorytools.com/production?share=GHJAPp7kTDb4r8PhcYNn

[deft lichen]

too few waste

[wind spade]

you need more waste: https://www.satisfactorytools.com/production?share=ktKLsqBiJJZ9dHrUdKjp

[turbid plinth]

ok but what do we burn in the plant?

[vocal tundra]

Uranium fuel rods

[placid oyster]

Uranium fuel rods & plutonium fuel rods

[median heath]

Uranium Fuel Rod

[turbid plinth]

[median heath]

You don't burn Plut Rods since your stated goal is waste-free.

[placid oyster]

Ah yeah nvm

[turbid plinth]

yes yes, 0 waste, thank you

[ember fractal]

Burn plut rods in vehicles

[pallid hull]

Yes

[vapid gorge]

I mean I've spent at least 200hrs just on my uranium rod factory xD so doesn't sound like much in that respect

[vague frost]

How do I split a 60 per min line into 5 12 per min lines

[median heath]

I literally just answered that in the other channel...

[vague frost]

That link you linked didnt help one bit

[vapid gorge]

it gives a diagram?

[vague frost]

I had to go onto imgur

[vapid gorge]

that's for an overly complex Load Balancer

the link was for hte simpler Manifold

if you read the link you'd know that a manifold also feeds a system evenly - it just takes time to spin up

[vague frost]

A 2 year old could make the one I just showed i dont even understand this one

[vapid gorge]

You do you and spend 10x the time with belts Load Balancing everything

gl

Manifolds work for everything

You have to build custom Load Balancers every time

[vague frost]

are the boxes storage containers?

[median heath]

No

You put 1 splitter behind each of the 5 machines you're talking about and just run the belt.

[vague frost]

im getting 60 from one machine and I need it split into 5 diffrent ways equally

[median heath]

Yes

Put 1 splitter behind each of the 5 machines you're talking about and just run the belt.

[vague frost]

oh my god my friend is actully stupid

he made it sound overly complicated

[median heath]

The diagram I linked you shows you how simple it is.

[vague frost]

I see now

[vapid gorge]

people's brain's are different

some people automatically just do manifolds w/o thinking about it

others assume you have to manually split things perfectly or it won't work. Shrug

Neither is wrong - it's just what initially makes sense to you

[median heath]

Some people assume you have to manually split things.

Other people assume that until they remember nodes are infinite. 😉

[vapid gorge]

I think it's expectations on organisations and that machines filling up and overflowing isn't explicit?

[wet condor]

is it worth it to use turbo fuel in fuel generator? how more efficient it is ?

[vapid gorge]

depends on you - a lot of people don't see hte point since diluted fuel will get you to nuclear

[median heath]

No.

[vapid gorge]

I think the process is fun personally 😄

[median heath]

If you're doing it just to do it, or for the fun of it -- knock yourself out and enjoy every minute.

[wet condor]

alright

[median heath]

If you're talking about "optimal" -- Turbofuel's only practical use atm is making bullets.

[wet condor]

good to know

ty

[median heath]

That being said

The very, very small amount needed to make bullets means it's entirely justifiable to use 1 sulfur node to make Turbo, diverting the amount needed to make bullets and possibly packaging some just in case they upgrade the jetpack in a future patch, while burning the remainder in generators because "why not?"

[vapid gorge]

do you have a diluted fuel recipe yet?

[wet condor]

Yes

[quartz violet]

enlighten me

[vapid gorge]

caterium CB has a lot of use cases for it

Steel plates is amazing if you have the infrastructure and need a butt ton of them

[quartz violet]

im going to ignore the steel plates

[vapid gorge]

Electrode CB... I feel has very niche value. It's good if you only have oil around and need a small amount of CB but take sa lot of refineries and set up.

Plus the things that use CB need other resources anyway so I just put computer factories elsewhere

[quartz violet]

I picked out caterium boards

I just went hard drive hunting to find some oil alts for a new factory

[median heath]

All 3 are good.

[quartz violet]

yeah, I was stumped

[whole heron]

On the upside, anywhere that has oil has a LOT of oil.

[vapid gorge]

yeah but then you have to ship them anyway right?

[whole heron]

Depends on your setup. If you're shipping everything to Blue Crater, then you can just do whatever since it has nearly everything.

[rocky grotto]

how would i be able to split 60 into 50 and 10?

[whole heron]

What's the context?

[sand epoch]

With a splitter

[whole heron]

What is it you're trying to do?

[median heath]

Single splitter does this just fine.

[mystic moon]

Split into 3 of 20, split one of them again to get 2 of 10, merge all but one

[median heath]

Ok but why would you ever?

[mystic moon]

I wouldn't
You already gave the correct awnser, I gave the awnser they asked for

[rocky grotto]

oh okay im making a 5 computer per minute factory but i also want to make some circuit boards for other stuff.

[median heath]

Use 1 splitter.

[rocky grotto]

ay do you have any idea with what i can do with the heavy oil from the factory?

[median heath]

Look at your Codex to find recipes it is used in.

Learning how to find ways to deal with byproduct is a very important skill you will need later.

[rocky grotto]

just got some ideas on what to do with it.

[fierce ruin]

What are the best alternate recipes?

[vocal tundra]

Best is subjective

[vapid gorge]

there are some that have a lot more opportunities to be good?

[oblique hollow]

Theres only some "really strong combination of recipes"

individual recipes are weak as hell

[vapid gorge]

Early on Cast Screws can be convenient

Encased Steel Pipe is almost a straight up upgrade

[vocal tundra]

There are however objectively worst ones

[vapid gorge]

Oh, Solid Steel ingot is also just about a straight upgrade to base recipe

[fierce ruin]

https://tenor.com/view/nerd-emoji-um-actually-nerd-actually-um-gif-26028792

I'm on tier 6

[vocal tundra]

Good for you!
Still best is subjective, alternate recipes heavily depend on playstyle and use with each other.

[delicate chasm]

The ones that let you switch your base resource out for another one, thus adding the unique ability to create a chain in the total absence of a resource without logistics.

This is my opinion; basis is the structural basis of 0 and 1. Do or do not have access. Other alts just make a location more or less convenient, not possible where previously it was impossible, to make a given part.

[deft lichen]

Your question has no correct answer

[delicate chasm]

We can answer the spirit of questions. It's okay to interpret someone's query when they obviously lack the knowledge that would preclude them asking.

If they knew the question wasn't good, they would ask a better question. Pointing out that the question isn't good is never valid, polite, or productive; it's a waste of your time as well, since you get nothing for your non-answer other than clout from other people giving non-answers.

Math is fine but don't forget about the meta.

[deft lichen]

Oh definitely

[snow dove]

i always take inventory expansions over any other alt, then HOR, then whatever looks good

[deft lichen]

Well the best would probably to prepare a tutorial on how to decide which alts to choose, that could be linked easily

Answering an question and getting a non-answer isn't useful, it's a question how useful it is to give an incorrect/biased answer that does actually answer

[oblique hollow]

Best answer is still "pick one and find out"

[delicate chasm]

See now that's valid - it's easy for someone to maybe miss something that would be important to them if they just get my opinion.

I like the idea of having a guide that is more about how to frame things in this game so that you get good questions to ask (of yourself, of the game, and of the community), rather than a guide that is just answers to common questions.

Might be a very worthy undertaking for the community to engage in.

[deft lichen]

Hmm right

I have a plan to extend the wiki with guides on specific topics

This could definitely be a good addition

I can't really write all of them myself so contributions would be much welcome

[delicate chasm]

One point with my opinion is that I think both cast screws and steel screws are better than one another. It's hard to quantify that in 2 sentences, lol.

[oblique hollow]

i had the silly idea of making the recipe summary thing into its own page, seperate from the excerpts on each page

[deft lichen]

That's still the plan, well, to do both

[oblique hollow]

"Snutty & Juice's Recipe Deal-Broker"

[delicate chasm]

Of course the answer is that it's situational; steel screws are nice for pooping large quantities, cast screws are nice for building fewer constructors in an early game situation.
Both are better depending on which situation you find yourself in, and you may end up using both recipes at the same time depending on where you go. One is not truly better (steel is better 100%) than the other. 😉

[deft lichen]

Base screws are good with steel rods

[delicate chasm]

I can do the actual "how to think" part of the guide; the preface. I'd defer to more experienced players on which recipes to use as the most poignant examples though.

[oblique hollow]

The plan is to simply line out what each recipes ups and downs are

current alternative method to the WP tables are the alternative recipe summaries on select few pages

like on the cable page

those get shortened if needed and extended to all parts to become the new norm

and a dedicated page for analysis of them all with still follow.

ideally, that page doesnt just say "this plus this is good" but instead something like "Heres a few interesting combinations:"

and then you can list things like HOR + diluted fuel + recycled on there

Or the aluminum paths

[true junco]

A bit of a narrowing down on this idea but I did some math on it the other day, determining that hands down the steel beam into screws is the best* way to make screws on a remote location if you need to freight in resources. The comparison was basically "which recipe spits out the most screws per stack." So thats the same ratio per truckload or per traincar and steel beams have the highest per stack density of screws output.

Best meaning, most economical in terms of freight costs. Ie, how many train cars and loadingdocks for input material you need per screw. Which relates to power consumption and labour costs (ie time) to put everything together in game.

[delicate chasm]

Heck yeah, we're both allergic to aspartame and we both like steel screws. 🤝
That's also the reason I prefer them, all other things equal. If there's an iron node I don't even need onsite I might consider the cast screws just because it's easy, but if I have any designs on that node in the future I'll want to leave it open and bring in beams.

[delicate chasm]

'scuse me HWAT?
(vanilla game)

That's the only one. The other refineries are all behaving. 🤔 Do we already know this bug? I used copy/paste to set the machines before connecting and turning them on for the first time so I think it's not the one @cinder silo is documenting (this week, lol).

[cinder silo]

Ahh no the bug I documented uses 0 inputs 🤣

[delicate chasm]

I did observe my machine using inputs, yeah. 👍
It's properly vented now. I left it running while I was building the stuff down the line. I expected it to, y'know, stop at 200.

[cinder silo]

I did notice a funky input one when you ctrl click stuff in to a machine, sometimes all the stacks get piled in to the spot, I should document it when I feel more up to it, looks weird seeing 1600+ items in the input buffer.

[granite nova]

which one

[oblique hollow]

all are good

pick what you need more urgently

[granite nova]

im good on all of them

tbh

[granite nova]

do you think that the iron wire could benifit me beacuse i have stitched iron plate

[oblique hollow]

if you pick iron wire, you get to use iron instead of copper.
if you already have copper, access to water AND refineries, you can use pure copper ingot to reduce the copper cost

either way, its a trade

[granite nova]

hmm

[oblique hollow]

iron wire needs a lot more iron per wire compared to the default wire recipe

[granite nova]

huh

could rotors benefit me?

[oblique hollow]

its pretty efficient

you can open the codex and compare

press ESC and you can close this recipe window and come back later.
or maybe E?

[granite nova]

i have scarce water

and mid iron

ok

i just had a glitch

i got given back the harddrive

when i pressed esc

and the recipes gone

[oblique hollow]

huh

that should not happen

[granite nova]

idk why it did

[oblique hollow]

funny game bug

[granite nova]

im rescanning now

should i report it?

[oblique hollow]

eh, verify game files, save and see if it happens again

[granite nova]

I got a different recipe lol

Free reroll

[oblique hollow]

yeah if you have to wait 10 minutes it rerolls

if you save shortly before it is done, it wont reroll

[granite nova]

Now the recipes are terrible

which one

[oblique hollow]

these are also all quite good

again: pick on what you need more

since you have stitched plates (out of spite for screws i guess) you wont like bolted plate

[granite nova]

Ok

[whole heron]

I never found Iron Alloy Ingot useful. Usually where there is Iron, there is even more iron. So why stretch it with a rarer resource.

[oblique hollow]

because i despise pure iron

and i combo iron alloy with copper alloy

[magic island]

there's spots that have coal+iron+copper nodes close together, and iron alloy > solid steel can take pretty good advantage of that

[whole heron]

I was curious about something else, so I looked at the map. Outside of the northeast part of the map, if an iron node is isolated, it's pure (with one exception).

In the northeast part of the map, there is just Iron scattered everywhere.

You figure that was deliberate or it's just a logical consequence of the devs wanting you to have easy access to iron since it's used constantly?

Oil is never isolated, Quartz mostly comes in pairs*, Uranium is always isolated but there are only 4 nodes of it.

And I'm not counting wells because they're special to begin with.

*There are two non-matching Quartz in the NW and two sets of triplets in the east.

[median heath]

West islands where there are 4 oil nodes and nothing else for a good long distance is "oil is never isolated"?

[granite nova]

How much water does 8 coal power plants need

[median heath]

Entire Spire Coast where the devs have said they are adding more nodes so there is not just oil is "oil is never isolated"?

[median heath]

45 * 8

[whole heron]

I see you're going on a rant to try to show up someone you don't know, but...

I mean isolated from other nodes of the same type.

[median heath]

Ah, that is an important clarifier.

👍

[primal flicker]

I thought that was easy to understand from the context of your other comments.

[placid oyster]

where can I read up on train signals?

Ive heard sev say that he made a guide, is that on reddit or somewhere here?

[oblique hollow]

wiki and the pins here

hm wait...

not the pins in this channel

its the pins in #old-questions-and-help

https://www.reddit.com/r/SatisfactoryGame/comments/qgy9z0/signal_logic_rules/

[vocal tundra]

I miss #old-questions-and-help :(

[placid oyster]

thanks!

[west cedar]

Best recipe for aluminum?

[oblique hollow]

all of them

literally

there is no single best recipe

[median heath]

Scrap or Ingots?

[west cedar]

scraps

[median heath]

2 options

[placid oyster]

electrode + sloppy or instant

pick your poison, they are equally efficient in terms of bauxite

[median heath]

Sloppy + Electrode or Instant

Both have the same conversion, your decision comes down to "which black rock?"

If Coke - Sloppy + Electrode
If Coal - Instant

[oblique hollow]

if for you, best means "most aluminum scrap per bauxite" then its the above stuff

[west cedar]

that was the question. Seems like bauxite is the most scarce resource in the tree. That and maybe sulfur

[placid oyster]

uranium is the most precious resource by far

[oblique hollow]

bauxite, sulfur, quartz, maybe nitrogen, uranium

[placid oyster]

but unless you do max nuclear youre unlikely to hit the limit

[west cedar]

I am referring to the aluminum tree.

I am actually working out a full use uranium power plant

[median heath]

You have enough sulfur to do max nuclear + all Aluminium via Instant and still have left over amounts to make hundreds of batteries per minute.

[oblique hollow]

if you wanna get the most out of bauxite that means you also gotta get a lof of silica

[west cedar]

Thanks everyone!

Helping me plan it out

[ember fractal]

Sulfur is really at a premium

They need to add some sulfur nodes to spire coast

[vocal tundra]

I think they said that they were planning in adding nodes to the spire coast back when update 6 released

[wind spade]

sulfur is hardly needed for anything

[ember fractal]

on the contrary it's needed for a lot of things
turbo fuel, ammo, nobelisks, batteries, nuclear power

[median heath]

One of those is where most people waste their sulfur 😉

[vocal tundra]

Tbf turbo fuel is mainly for ammo

[median heath]

That said they most likely will add 1 more node of it to Spire if I had to wager.
But that opens up more options like Compacted Steel imo, not "hey keep wasting it on turbofuel".

I would sploosh so hard if they added 1 more bauxite node to Spire, because that is truly my limiting resource.

[vapid gorge]

eh too far from coal there, still wouldn't use it

[delicate chasm]

Why would you not do a coke setup there, given a bauxite node? 🤔 There's as much water as you could want and enough crude to do pretty much whatever you want.
Not a challenge, but curious what your reason would be?

[median heath]

You have oil. It's prime for Electrode 😉

[wind spade]

turbofuel is rarely used together with nuclear, usually TF gets skipped for nuclear
nuclear uses very small amount
ammo and nobelisk use even less (since you need very small production of those)
leaves tons of sulfur for batteries

[delicate chasm]

Compacted coal itself is the resource I find myself not really wanting to make much of or having much use for. There are only 2 spots on the entire map I've been to physically that looked like compacted coal would be the most feasible thing to do.

There are needs for it but they are small compared to most other resources.

[wind spade]

maxed nuclear (pretty much never needed) uses 45% of map's sulfur
leaves enough sulfur for 2488/min batteries (or some of it will be used for ammo)

I don't think sulfur is limiting at all

[delicate chasm]

It's mostly just impure when you'd want it to be normal. 🙁

[whole heron]

Now I'm just imagining someone yelling at sulfur nodes like they were an unruly teenager.

[median heath]

I can think of 4 Compacted spots off the top of my head. 🤷‍♂️

[wind spade]

🤔

[median heath]

5

Both Coal Lakes, Crater Lakes, Southeast of GF, East of DD.

6
Northwest of DD

[wind spade]

I didn't even mention compacted 🤔

[median heath]

Discord moved... fuck.

[delicate chasm]

NW DD and Crater Lakes are the locations I had in mind. Compacted coal there seems to make a lot of sense. Anywhere that you can add sulfur to coal theoretically extends your coal limitation so anywhere with both is a 'compacted coal spot' really.

And if push comes to shove there's always logistics.

[median heath]

Coal Lake East is an excellent spot for it.

[delicate chasm]

Which are the Coal Lakes? Up around Titan/Red Bamboo/Maze?

[median heath]

Coal Lake West:

Coal Lake East:

@delicate chasm

[ember fractal]

bauxite is at a premium

[vapid gorge]

Wasn’t Sev talking about sulfur up there?

[median heath]

I mentioned both.

Mainly that I suspect there will be a sulfur added
And what I WANT is a bauxite added.

The 2 Mark specifically mentioned in the last interview on the topic were sulfur and copper.

[vapid gorge]

Still need to coal for compacted recipe though

[median heath]

Ok you're jumping around between conversations and this isn't making sense, lol./

Compacted conversation had nothing to do with the add bauxite to the Spire comment which led into the "you can make Electrode Aluminium Scrap" comments.

[vapid gorge]

Ah see I was replying to someone saying they’d use compacted recipe in the spires

[whole heron]

Ok, my mind is screwed up. I'm starting to imagine oil pouring out of the heart of a dead ancient giant.

And the Satisfactory world is some dark fantasy world.

And for all I know, the planet is alive and it's all aliens.

[delicate chasm]

No, not compacted. Petroleum coke.
My question was why not do a coke setup there, given a bauxite node?

I included the bit about bauxite just in case I was misinterpreting you saying you wouldn't use the sulfur node as you saying you wouldn't use the bauxite node.

[vapid gorge]

That’d be fine, sure?

[livid canyon]

@vapid gorge

[vapid gorge]

Ah yeah, that’s shouldn’t need a valve or pump though

[livid canyon]

fair enough

don't like dealing with liquids largely in this also

[vapid gorge]

Well this is probably the tidiest way for it 😄

[livid canyon]

it also very neatly fits what my bauxite production is, or close enough (had overclocked to 250% because I wasn't sure exactly on this part, but now I know

[west orchid]

guys i have a question. one youtuber that i trust, he says if you try to use all 600m3 oil at %100 efficiency it will bottleneck and you will get problems. Is it fixed with new updates cuz the video recorded 1 year ago.

[median heath]

"says if you try to use all 600m3 oil at %100 efficiency [and fail to build it properly] it will bottleneck and you will get problems"

Yes.

Solution: Build it properly.

[wind spade]

You'd be surprised how many big youtubers have wrong/outdated info. 600m3 is possible if built properly.

Recommendation:

• loop the pipe
• feed from above
• don't use buffers or valves
• simplify the setup as much as possible

[median heath]

-immediately split into mk1s.

[west orchid]

i feed at same level is that problem?

[west orchid]

why

[median heath]

You asked if 600/min is possible.

We are telling you how it is possible.

And you're asking why?
I am confused...

[west orchid]

mk2's max capacity is 600m3 so why am i using it with 2 pipes instead of 1

[median heath]

Do you want it to work?

If yes, do one of the above options.

If no, keep asking why you would do one of the above options.

🤷‍♂️

Because the answer to "why" is "because it makes it work."

[west orchid]

so you dont know why its working with mk1 instead of mk2

[median heath]

🤦‍♂️

I don't know why when I press the thin pedal my car goes faster.
I just know that it does and I use that knowledge accordingly.

[west orchid]

so you dont know why 2x mk1 pipeline is better than 1x mk2 pipeline lol

i want the answer anyway cuz if ill build more complicated things i have to know more details about that

[median heath]

I do.
Because one works and the other does not.

Working = better.
Mystery solved 🤷‍♂️

[west orchid]

im not asking is it working or not bro im asking why its working and the other one is not working do u understand me

[median heath]

Sometimes the answer is "because it does" 🤷‍♂️

Like why does looping make the mk2 work?

Because it does.
And not looping makes it not work.

[west orchid]

its game and its made from codes

[median heath]

The key takeaway for "more complicated things" is to not use just a plain mk2 and expect it to work.
Because it won't.

Pick one of the listed options to make it work.
Why? Because they make it work.

[median heath]

Computers are not infallible.
And there are a host of things in binary that simply don't exist so the computer and compiler "do their best" to provide what is being asked of them.

I.E. The game is telling the mk2 to push 600, and yet it cannot.
Hence the need for the listed workarounds.

Why do they work?
Because they do.
Why does the mk2 not fully work as expected?
Because it doesn't. 🤷‍♂️

Wait until you get into the fact that Valves don't actually restrict to the number you set them to 😉

[west orchid]

if you were a dev of this game and when you find a bug that player's save file is deleted for no reason and the when players ask you why did that happened will u say "because your save file deleted omega lul" when they ask why cant i connect my friend's world and you say "cuz you cant 🤷‍♂️" when players ask why game is not running properly on my computer and u say "cuz its not running on you computer properly🤷‍♂️ 🤷‍♂️ 🤷‍♂️ 🤷‍♂️🤦 🤦 🤦 🤦 🤦 " damn bro what a logical answer good job you answered all of the questions

[median heath]

They are bringing mk2's down to 500 flowrate in a future patch.
Why? Because 600 doesn't work.
Why? Because is doesn't, and 500 does.
So... they are making it 500.

🤷‍♂️

So why would you immediately split a mk2 into x2 mk1s?
Because mk1s work as expected, and the mk2 doesn't.

It really is that simple 🤷‍♂️

But as greeny listed, there are multiple other options available if you do not wish to take that path.

It's funny you bring the dev thing into this because when talking foliage being bugged and someone asked "why not just make all trees X" Jace responded with "Because Unreal Engine doesn't know what a tree is." and the answer was literally that basic 😂

[frosty owl]

We're not Devs, nod do we look at or care for the code; we just observe the game and plan accordinglt. I understand the frustration in not finding the exact info you want, but such info is not quite necessary to be able to predict how the game behaves in (nearly) all scenarios, including complex ones; if you wish for more in-depth info you should probably ask in the modding server, where they actually do look at the code...

[median heath]

||The code on these glasses indicates the Allspark is exactly 230 miles from here.||

[zinc crater]

Are you Sam Witwicky?

[wind spade]

are you ladiesman217?

[zinc crater]

Barricade was really cool in that movie haha

[oblique hollow]

Nobody can answer you that because you literally cant peer into the code deep enough.
and if we could we likely wouldnt understand it.

One of the only responses from the developers we got, on ONE of the issues mk. 2 pipes have, is floating point precision issues

Not even modders can peer into the pipe code deep enough to try and understand

[west orchid]

the guys that i disputed with he said 50 times it is because it is and i asked why? the answer actually "we dont know" that guy spammed it is because it is and just try to find why is it happening but that guys again said it is because it is.

[oblique hollow]

From a community perspective, it is entirely a mystery

We dont REALLY know why they dont work

floating point cant be the only issue

[west orchid]

i just wanted to hear that answer but that guy said "its because it is". When he says this, that means some players knows this but this guys is not

i understood like that

[oblique hollow]

Because the mk 2 issue is so. inconsistent, its hard to even blame floating point at times

ive heard of people who had 0 issues with running mk 2 at max capacity

Hence the "if you build it properly, mk 2 CAN work".

These are all all workarounds to a mysterious issue.

[whole heron]

What's actually the worst that can happen if you don't use a workaround? Some stuff stops working and you get lowered efficiency?

[oblique hollow]

Yes

You simply dont get the number you expect

[vapid gorge]

The mk1 thing might be the ‘safe’ more ‘reliable’ method if you’re less confident with your pipes. Tbh though there’s a few simple rules you can stick to to make mk2 stable

For example this is a max flow stable mk2 fed from below. It looks weird but is actually really straightforward

Besides the basic rules it just means you have to build your factory with your pipe layout in mind. Which honestly is something you probably ought to do anyway

[west orchid]

i built the entire factory with mk2 and there is no problem xd

i used 600m3 of oil

[vapid gorge]

Look - it’s not impossible it could work without looping and special care - but I’m guessing from experience there’s stutters that happen and you just haven’t seen it

[lapis geyser]

When you split pipes you create backflow opportunities. Backflow will make your oil extractors think the line is full and not pump. Don't split the pipes or if you do split them evenly into mk1s.

[oblique hollow]

splitting into 2 mk 2 also works as those are unlikely to run full

[west orchid]

I made with looping

[west orchid]

I split it but still didn't use mk1

[median heath]

Hi. It me. I am "that guy".
I do have a name... 😥

[whole heron]

Is it Robert Paulson?

[delicate chasm]

The precise underlying reason is not known, but the symptom that is observable is a culmination of factors:

Pipes are NOT pressurized; the effects of pressurization are simulated from the height of the pipe and its fill percentage. A vertical, bottom-fed pipe filled 75% will drain into a connected junction at its 50% mark just fine. A connected junction at the 70% level will work fine too, even if the pipe below is draining the vertical pipe at 99% of its potential flow speed.

As long as your vertical pipe is a mark 2 and the two connected horizontal pipes are mark 1, the lower pipe will only ever take 300m3. If your vertical pipe's flow rate is 301m3, your upper pipe will be fed at a theoretical rate of 1m3/m. Keep in mind it's theoretical unless the vertical pipe's fill level remains above the connection at all times.

Now let's say that instead you split those to Mk2. Now the lower pipe segment can take 600m3 at once and this is a problem because that means it can ask for 301m3, receive it, then say "Nah changed my mind" and send the 1m3 back. Then on the next update ask for it again.

Your flow rate in the upper pipe is now approximately 0.25m3 but it's inconsistent and difficult to accurately gauge. It will swell to 50% fill then backwash down to 10%, fill to 20% and then be emptied by the intake + the intake on the pipe below. And none of that seems to make sense and it is absolutely not intended behavior.

But we don't know precisely what the issue is, because they cannot be fully solved with the tools that should solve the issues (valves and pumps) if backflow was the only problem.

You also need to understand that pipes are bidirectional but flow happens one way at a time on any given game update, meaning that a pipe that is drawing from a source pipe can be starved for 1 game tick or be DRAWN FROM by its own source pipe in some instances.