math-and-meta

[amber jacinth]

Other calculators? They definitely exist, yeah. I have a testflight one developed by a friend.
There's also the SCIM calculator as well

[gusty edge]

Is this working?

[minor hawk]

Well I guess this new discovery will make Satisfactory Tools a bit more satisfactory to use

[forest solstice]

when do you unlock bauxite? i just found a node but it didnt trigger any research for it

[calm mauve]

after oil

[forest solstice]

im onto fuel already, maybe a bug?

[calm mauve]

its in the next tier

[amber jacinth]

Not at all a bug

[forest solstice]

gotcha

i assume 7

[amber jacinth]

yep

[forest solstice]

thankyou!

[dawn lark]

Yeah it's not MAM related so there's no "you could learn more about this" message

Likewise with uranium

[forest solstice]

ah ok, i thought there might be MAM research line for it

[amber jacinth]

(do not the uranium)

[forest solstice]

thats been killing me lately, "oops i landed on a uranium pillar and am taking way to much rad, oh hey im ded"

[amber jacinth]

[forest solstice]

when do you get a suit, i havent ever made it to nuclear, i assume in that same tier

[amber jacinth]

It's a tier 7 unlock, yeah

[forest solstice]

nice, i should have enough modular engines in like a couple min

[urban kite]

ahhh i need 300 more coal

[edgy leaf]

any idea how to vertically align pipe junctions?

like, without having them next to eachother

[ashen girder]

Looks like it. When a train enters the block before it, it will try to reserve a path through it.

Path signals only turn green when a train has a reservation through them.

[grizzled ember]

is it possible to distribute 1 conveyor belt, into 5 destinations using spliters/mergers and having 20% on each destination?

[edgy leaf]

you can use a manifold for it

[tender niche]

Yes. One 1:3 splitter, feeding three 1:2 splitters. Six outputs total. One output is fed back to the input line using a merger.

[edgy leaf]

wouldn't that cause issues if you have a full line?

[ashen girder]

Mind doing it like this requires you to be able to fit 120% through the belts. So you can't split 60/min into 5 this way on a Mk1 belt.

[lone igloo]

What would be your recommendation guys on the height that a logistics hidden floor should be?

[ashen girder]

There's an alternative way that involves another splitter and merger that lets you do it though.

[tender niche]

@ashen girder is correct that you need extra capacity on the belt with this method.

[edgy leaf]

in case you dont know manifolds, you just put splitters in a row. at first the first one will get more than the others, but it will fill help, then the next one and the next one, and if you wait a little bit itll be split perfectly

[ashen girder]

Here's the diagrams for both ways.

[grizzled ember]

yeah manifolds arebad

[ashen girder]

And also, yeah. Manifolds work a treat. 😄

[edgy leaf]

why are they bad?

[tender niche]

Ahh, the bottom left solution is clever!

[grizzled ember]

cuz each time the conveyor goes thru a spliter it halves the income

so the first always get more than the last

[tender niche]

Manifolds aren't bad where you have enough material to saturate the line. They're only bad when material is low.

[weary sable]

manifold takes just a little to ramp up but when it does it works just as well. easier on my smooth brain

[lone igloo]

Whats so clever about it? you make 6 equal splits and recycle one xD

[edgy leaf]

okay lets say you get 50 as input and each machine consumes 10

[ashen girder]

You split one in half and remerge it downstream.

[edgy leaf]

at first the first machine will get 25

but it only consumes 10 so eventually it will fill up

the splitter will still "try" to push 25 to the first one, but it will only take 10

the rest will get transferred to the next splitter

[ashen girder]

Splitters only send as much down a line as they can, then they send the rest down the other lines. They equalize naturally.

[edgy leaf]

same happens with the second one, it will get 20 instead of 10

but eventually it will be filled and will only consume 10

[tender niche]

The solution I proposed was to create six outputs and recycle one, but this recycles to two earlier inputs, allowing a lower quality belt.

[edgy leaf]

im not saying belt balancing is bad, it can be a fun challenge and not having aall the machines filled can be very useful for nuclear, but i think you might not be fully informed aabout how manifolds work

theres nothing wrong with using belt balancing, but saying htat manifolds are bad and inefficient is incorrect

[tender niche]

In general, you want to use a manifold unless you have a reason to use a belt balancer. They're simple, and cover most applications.

[lone igloo]

Yeah but it makes sense to do that with lower quality or am I just being stupid I got those BPs since update 7 ?

[grizzled ember]

lol dont be so defensive about mainfolds, do whatever u want on your save, its not the awnser i wanted, and i think i was pretty clear on my question

but jay already helped

[edgy leaf]

im not being defensive..?

[grizzled ember]

so thanks

[edgy leaf]

i just wanted to explain to you how manifolds work, im not trying to impose my playstyle on you

[ashen girder]

@tame frigate This one scanned fine for me. I was standing beside it when my save loaded, so your theory might have merit? 🤷‍♂️

[tender niche]

If you were using a Tier 6 conveyer, 1200/min, with a full 1200 input, the top left will not consume the full throughput. The bottom left will consume the full 1200.

[lone igloo]

I've not tested with T6 as I'm not there yet.

[ashen girder]

To be clear, balancers are largely a waste of space and effort and will become increasingly frustrating and annoying to implement compared to manifolds.

[lone igloo]

Yes but they make your factory run clean.

[tender niche]

Same, but the application is when you can't create a better belt that what you need to flow.

[edgy leaf]

with manifolds the factory also runs clean

[ashen girder]

Manifolds run just as clean after they saturate.

[tame frigate]

Thanks. That'll be quite the trek for me, but it's one I can test.

[lone igloo]

Disagree on big part there, if you want clean look with a factory and numbers are easy you never manifold because you get stacks of resources in machine instead the perfect number which never goes up and stays perfect.

[edgy leaf]

i think manifolds arguably give a cleaner look

[ashen girder]

You're in Rocky Desert yeah? I've got some in the ravine north of NF that I still need. I'll let you know if I get one closer to you.

[urban kite]

oh manifolds def look cleaner from the outside, than balancers? maybe not in the machines or items on belts but yeah

[tame frigate]

I've briefly penetrated into the Spire Coast north of the Northern Forest, so those may or may not be ones I've grabbed. Last time I was there though, zero drives scanned.

[lone igloo]

I agree they might look better but having to constantly just sink overflow and having to find a place to place AWESOME sink or have it in a build all tiem is annoying

[edgy leaf]

you dont need to sink overflow when using manifolds

[ashen girder]

This has nothing to do with manifolds or balancers.

60 in, 60 out, both will run nonstop at 100%.

[amber jacinth]

If you math right with manifolds, sinks are not used in the lines

[edgy leaf]

and if you math wrong, even a balancer wont help you

[ashen girder]

70 in, 60 out, your balancer's gonna back up too.

50 in, 60 out, your machines won't hit 100%.

[lone igloo]

Yeah its that easy numbers but manifold will always prioritize 1st building meaning eventually you will stack to max resources.

[ashen girder]

It'll be less obvious with a balancer when you have a deficit, in fact.

[ashen girder]

Yeah? And?

[lone igloo]

overclocking and underclocking is a thing?

[edgy leaf]

why is it bad when the first building is full?

[ashen girder]

That has literally nothing to do with this either.

If you underclock you machines to consume 50, that's 50 in, 50 out.

[amber jacinth]

Both belting methods have their uses, for sure. From what I've seen more folks prefer manifolds as they're simpler to set up overall. I do prefer balancing for endgame products due to their low throughput.

[lone igloo]

if your manifold is longer than certain input and consumption rate your last machine will get starved or you need to pre-saturate it.

[ashen girder]

No, it just takes longer to saturate.

[lone igloo]

I work with good numbers and if numbers dont work well I make them work well.

[ashen girder]

If your inputs and outputs are balanced, all manifolds will saturate eventually.

[tame frigate]

Just to jump into the fracas, if I want to add machines to a manifold because I'm getting more resources, that seems simple enough. Never tried adding machines to a balancer.

[potent plinth]

how would i get 7.5 items into a 2, 3 belts and 1 1.5 belt?

[edgy leaf]

except in some rare edge cases. if you have 10 machines in a row with a manifold and you're under producing by 50% and the last 2 machines are critical for power production and the first 8 arent then you might get a grid failure with maanifolds but not with balancing. but thats bad factory design

[tender niche]

Low-quantity products are exactly where you want to use a balancer as opposed to a manifold. Late-game is when this becomes relevent. Or nuclear, where you don't want radioactive stuff sitting and accumulating.

[urban kite]

no? if you dont have enough to saturate, balancers will starve too

[amber jacinth]

If your manifold is longer than the input... even a balancer wouldn't help 😛

[weary sable]

How do you all decide on how much of a resource to make when progressing? im at the endish of phase 2 so just trying to deliver stuff to the elevator. just hard to know how much of a resource to try to make. Should i aim for the highest tier resources to be made with one assembler at a 100% rate? IE: 5 versatile framework crafts (so i think 10 total?) a minute? Then i know to aim for 25 modular frames, 60 steel to craft that and have it castcade down to ingot, ore, etc? obviously other things as well. For some of the items used for building i just split it off to a storage so the machines are a little inefficient for a bit but it catches up when full

[lone igloo]

No. go make a 8 power coal gen with 120 input and run it from start and wait for last to become online in 4 hours...

[ashen girder]

Yeah, that's literally exactly what I do every time.

[amber jacinth]

Yep, aiming for a certain end product is the way to go

[lone igloo]

Sadge

Anyways good convo going back to make 200 fuel gen setup

[urban kite]

only 200?

[edgy leaf]

does anyone here know how to vertically align pipe junctions?

[calm mauve]

the 5x5 BP machine really helps with making fuel gen power tower BPs

[lone igloo]

Still on phase 4 I dont need no more that 50k power atm going to mega nucleor power plant after

[edgy leaf]

im too autistic to just eyeball it and too lazy to make a second blueprint..

[lone igloo]

I dont do power towers I make awesome builds dude. Thanks

[ashen girder]

Pipe stands and snapping to outputs/inputs?

[amber jacinth]

Do you know the two-snap rule (or whatever it's formal name is)?

[edgy leaf]

i have a vertical pipe and i want a reproducable way to put the pipe junction at the same height in independent locations

[amber jacinth]

ah, I see

[ashen girder]

Pipe stands is how I'd do it.

[calm mauve]

is there a foundation below?

[edgy leaf]

it odes not snap to pipeline supports

[amber jacinth]

Build a pipe that intersects horizontally with the pipes at the desired height, and then place the junctions on that

[ashen girder]

Build supports to the correct height, build a cross pipe, then build the vertical pipe after.

[edgy leaf]

it snaps to it when its horizontal or diagnoal but it unsnaps as soon as i make it vertical

so either its not aligned horizontally or not aligned vertically 🤔

[ashen girder]

Ah, that sucks.

Will it snap to a machine vertically?

[tender niche]

You can place + junctions vertically on walls. Put a wall on your foundation, nudge it to your pipe, and place the junction on the wall.

[edgy leaf]

OHHH

i didnt know that, thanks !

[fossil cypress]

hey quick question how much urainum fuel rods do nuclear power plants take per min? i forgot

[amber jacinth]

0.2

[fossil cypress]

thx

[edgy leaf]

do you think this here can still be considered bottomfeeding?

[vapid gorge]

yes because you're pushing it up into the feed pipe

[edgy leaf]

epic

[vapid gorge]

why is there a connection at the bottom of each feed line?

[edgy leaf]

the horizontal upper connecting pipe fixes the wonkiness of it all

[edgy leaf]

wdym?

i hadnt added the bottom pipes yet in that screenshot, one sec

[edgy leaf]

the fuel comes from below

[vapid gorge]

you want to avoid a lot of merges within a manifold loop

[sand epoch]

Yea.. just go direct from the lower into the refineries

[vapid gorge]

adds a lot of wonk

[edgy leaf]

addint that upper horizontal line made my refineries 100% efficient

it removed all the wonk. im guessing it helped because it connected all the pipes on a level above where the refinery input is i think

[edgy leaf]

doing that made my refineries stall

[calm mauve]

do those pipe floor holes work when you place the BP?

[vapid gorge]

look if it's actually working fine I'm happy for you, just run into a lot of people misdiagnosing pipes and there's no reason for tons of branches to help

[edgy leaf]

yep !

[sand epoch]

You would need pumps...

[raven bolt]

I dont feel like doing a lot of math rn so here I am asking, is turbofuel still worth it?

[edgy leaf]

why would i need pumps?

[sand epoch]

How else you getting the headlift to get up?

[edgy leaf]

pumps

heres with the upper horizontal connections

[vapid gorge]

if the yhave enough base headlift before the manifold it's not an issue

[amber jacinth]

Don't think TF numbers have changed much, tbh, but rocket fuel is a step up for sure.

[raven bolt]

ah alright thank you

[tame frigate]

Huh. After crossing about 2/3rds of the distance there, a hard drive popped up on my scanner. 🤔

[edgy leaf]

and now its time to wait

[sand epoch]

Why are there still junctions?

[edgy leaf]

efficiency has already dropped

@vapid gorge removing the upper horizontal line made the extractor stall within seconds

[calm mauve]

it must help with the slosh

[edgy leaf]

becuase i was too lazy to remove them and leaving them in allows me to easily add the horizontal pipe agian

[sand epoch]

You were getting fake headlift from a feedback loop.. im still not seeing any pumps :/

[edgy leaf]

how does one get fake headlift from a feedback loop? also, i have a pump

yep, the refineries are starting to stall. the 100.7mw drop is the exact number of a refinery dropping ou

the refinery efficiency is going down too

[sand epoch]

If you think it works, then do it. I wouldn't be connecting any of it like that, but thats me.

[edgy leaf]

i know it works

i just demonstrated that if i were to do it the way you told me to that it wouldnt work.

i added the horizontal pipes back in, i didnt flood the same i just added empty pipes to it and now the extractor is back at 100% efficiency

[tame frigate]

Three scanning drives in a row, and I haven't even reached yours yet. WEIRD.

[vapid gorge]

you mean vertical pipes?

[edgy leaf]

.

[muted crypt]

if I genuinely don't know what is a good amount, I set up for one machine's output per minute, assuming 100% clock, as an end product. This number can change based on what recipe I use, too.

[edgy leaf]

these pipes

[ashen girder]

Yeah, I really think it's about which ones they've touched recently.

[sand epoch]

I would have changed more than you did in your "test" so dont go using that as your proof.

All that really matters is that you like it.

[vapid gorge]

ah, no, these ones

one long loop that only connects at teh end

[edgy leaf]

well the oil is coming from there, i cant just remove them

[edgy leaf]

it works flawlessly went built like this

and thats using full 600 pipes too

each one of those refineries uses 150 oil, theres two 600 pipes feeding them

after adding the horizontal pipe back in and waiting a bit my powergrid is perfectly flat again, no flooding necessary

[tender niche]

I'm digging into trains a bit and have some questions/assertions I'd like to confirm.

1. Train platforms can either load or unload, not both at the same time.
2. Train stations can filter items to prevent accidental load/unloads. I'm not sure if this has any clever uses.
3. Train platforms always have two load and two unload conveyers. I'm not sure if there's a clever way to use the load conveyers for an unloading platform, but they seem functionally unnecessary.

[muted crypt]

server name checks out

[cedar ivy]

all correct.

[tender niche]

Yay, thanks for the sanity check!

[edgy leaf]

yea, outside of building new factories i havent had the power graph be not perfectly flat even once. if it did happen id check to see hwihc machine it was and spend days trying to fix it until its either fixed or i have burnout

[amber jacinth]

The double input/output is useful for when trains are at the station, and the resource output/input from/to the station stops- can be used to buffer items easily.

[muted crypt]

How do you tackle machines with fluctuating power consumption?

[cedar ivy]

Platforms can only be set to do one function at a time, but it's not like a building can change layout by the flip of a switch, so the in/out ports are always there

[edgy leaf]

i havent gotten to that yet, but i plan to put all machines with fluctuating power consumption on a secondary grid

[tender niche]

I think I understand. Have two belts feeding an industrial storage, with one belt leaving for contant output.

[amber jacinth]

exactly 👍

[edgy leaf]

i plan to put all my nuclear power plants in a long line with priority power switches between them so i can remotely toggle how much power to allocate to each grid

[muted crypt]

👀 interesting

[tender niche]

It makes sense they only wanted one model for the platform. But, I was wondering if the platform literally switched two inventories. I was thinking terrible thoughts...

[cedar ivy]

item filtering is probably most useful to mix items in a single platform without mixing in the car, or visa versa, because it's a per-station option. Still generally inadvisable to mix in a single car.

Nope, same inventory

[tender niche]

Agreed. It did occur to me that station filtering could be used instead of smart splitters at the output for sanitizing inputs to the factory.

[vapid gorge]

are you planning on having multiple trains stopping at a particular station?

[cedar ivy]

yeah but at that point it's already in the station or in the car

[amber jacinth]

sushi trains

[tender niche]

I hadn't really got that far. I was thinking each consumer could have their own train, perhaps going to a single source.

[tender niche]

Agreed, but in the car is one place, all over the factory is multiple places.

[vapid gorge]

kk, careful if you do do that, it significantly lowers the throuhput of each platform having extra trains stopping

[tender niche]

Noted. We'll see what terrible, cursed things I dream up...

[vapid gorge]

and look up how to buffer train stations

[tender niche]

I think I understand buffering, but I'll look it up just in case...

[tame frigate]

Your drive is NOT scanning.

...but you said you were standing next to it when you loaded the game. Hrm.

[edgy leaf]

heres a chart that shows maximum throughput of a train station based on belt speed, item stack size and how often a train stops there

[urban kite]

how does an industrial container act with two outputs. Is it basically a splitter?

[edgy leaf]

lets say you're transporting 500 stack size items with two mk5 belts, if the train comes every 50 seconds you'll only be able to transport 715 items per minute. thats two mk5 belts which can together only transport 715/min total

[calm mauve]

it helps to feed in with two belts to get it working properly though

[edgy leaf]

seems like it

[ashen girder]

Ooh, we're both wrong. 😂

[urban kite]

the one i just set up is not splitting

[ashen girder]

Weird bug. Hopefully they fix it soon.

[muted crypt]

iirc industrial containers will pick an output and prioritize it. I don't believe it's ever been consistently the top or bottom output, either

[calm mauve]

use both inputs too

[tame frigate]

A known workaround is to reload the game next to it. Hence why my theory about 'regions close to where you load'.

[ashen girder]

Ah, that makes sense. I didn't realize that was a thing.

[tender niche]

Interesting... can you send me a link to that calculator?

[edgy leaf]

desmos.com/calculator/djzytmsw4h

R=27.08 is the amount of seconds that the trains takes to unload, during which it blocks the inputs

u dont ever really have a reason to change it but i put it as a variable just to see what would happen if it was different

[wicked ravine]

How do i split 800 fuel between 36.56 Refineries

[tender niche]

Nice. Thank you!

[edgy leaf]

id just put 37 refineries, clock one to 56%, feed 400 in at the front and 400 in the middle

or if u want to be extra safe maybe 400 front and 400 back i guess. its just superstition but i dont trust the fluids, they're always scheming against me trying to overthrow my kingdom

[wicked ravine]

hmmm i overclocked one to 156%

[edgy leaf]

same thing

[wicked ravine]

but then technically one half needs a little more tho

maybe i just use valves

and make sure 400 goes one way and the rest another way?

[edgy leaf]

no, connect both halves

[tender niche]

Valves are complicated. Set your buildings to consume the right amount and make sure your pipes can handle the throughput. Having exactly 400 two ways won't help you.

[edgy leaf]

one long line, all connected. feed one half in at one side and the other half at the other side. dont use valves unless u really really know that you need valves

[wicked ravine]

ah okay cool, i will do that then

[edgy leaf]

the pipes balance themselves out because fluid flows both direction, so as long as no one pipe needs to carry more than its maximum, itll be fine ™️

[tender niche]

Also, I think you need MkII pipes, in case no one else mentioned that.

[wicked ravine]

Oh yeah i am using mk2

[tender niche]

Good.

[wicked ravine]

the next fun part will be supplying 200 Generators with Rocket fuel

[cursive heron]

does scim have a way to let you know where your somersloops are, I kinda forgot which machines I put them in

[vapid gorge]

you can click on containers and look at their inventory

[tame harbor]

As someone who speaks from experience

Be ready

[ashen girder]

"The Mathematics of Wonton Burrito Meals"

[tame frigate]

Doesn't help you right now, but I've started painting mine bright pink/purple when I boost them with a Somersloop.

[wicked ravine]

ready with patience or

i have no idea where im gonna place them

[edgy leaf]

maybe im just way too tired, but how do you divide 500 water needed per minute into water extractors without repeating decimals?

[neat lantern]

I literally came this thread because my rocket fuel setup doesn't work. Currently making 2400 rocket fuel, I can make another 2400 but can't get the initial rocket fuel to fill my generators. Looking through here to see if it's a bug or just me. The blenders just stop making rocket fuel, that's my issue. It's like the coal issue with generators. It just generating and need to set the recipe again.

[edgy leaf]

do they stop because they're full or because they dont have enough input?

[cursive heron]

for 200 gens the footprint needed is actually quite small and just needs a few buildings

[wicked ravine]

yeah its just to figure out where i want them

[edgy leaf]

if its because they're full then your pipework is wrong somewhere. maybe not enough headlift

[wicked ravine]

but still got a little to do before i can start placing

[neat lantern]

I have 4 blenders going to a buffer, so that's 600 rocket fuel. Then that splits up to the 2nd and 3rd floor with 135 generators. It starts filling up the first 10 but then blenders just stop and won't make any more rocket fuel.

[cursive heron]

900 rocket/fuel min to fully overclocked gens

[wicked ravine]

ohhh yeah i should overclock the gens

[edgy leaf]

any ideas? im starting to feel like its impossible 😭

[wicked ravine]

how much does a fully overclocked gen use with rocket fuel?

[neat lantern]

I have head lifts in the beginning of every entry just to make sure. It's the blenders that stop working.

[cursive heron]

just type 250 on two water extractors

[vapid gorge]

You make more and use the extra to sink product

[edgy leaf]

oof

im already using it for wet concrete tho, thats usually the thing u use to sink lol

so its just mathematically impossible to get rid of a repeating decimal?

[wicked ravine]

@cursive heron how much does on overclocked gen use?

[edgy leaf]

!wikisearch fuel generator

[brisk shore]

https://satisfactory.wiki.gg/index.php?search=fuel

[edgy leaf]

noo, im too stupid to use the command

[cursive heron]

its 4.16666 rf per gen just multiply it by 2.5

[wicked ravine]

Ah okay, thank you!

Thats a very annoying number

[edgy leaf]

if you build 6 its exactly 25

[cursive heron]

its 72 gens per 300 rocket fuel

[edgy leaf]

if you build 6 overclocked once its exactly 62.5

[wicked ravine]

Oh sweet! Thank you

[gusty edge]

333,333 Turbofuel/min factory complete

[wicked ravine]

And im gonna be producing something like 1k a minute

Pain

[vapid gorge]

Make a bit more and sink it?

[edgy leaf]

yea i probably should

it just feels kinda like.. cheating? idk

but you're right htats the most straight forward solution

[granite axle]

how can i optimize my AIO iron factory? this is early game so i don't have overclocking. that's all from 2 normal iron nodes, i have 1 extra. also note it's not completely done, i just did it quick and ignored the large rock and debris and didn't care to put a storage container behind the reinforced plates and rotors.

Iron Plates - 20/m
Iron Rods - 22.5/m
Screws - 20/m
Reinforced Plates - 5/m
Rotors - 4/m

[sweet blade]

How long does it take for trees and bushes to grow back?

[neat lantern]

Omg I am so stupid. Nitro rocket fuel has a by product of compact coal which I have not going anywhere

[tame harbor]

Are you removing the compacted coal byproduct?

Lmao

Feed it back to the turbo plants!

[granite axle]

id also like to make it more compact and neater looking if any1 has advice on that

[amber jacinth]

All I’d recommend really is getting overclocking so you can ratio out your buildings better 👍

Looks very neat- compactness isn’t awful, anything past what you have would be weird, really 😛

[vapid gorge]

Or you could change everything down the line so you don’t need repetition :p

[amber jacinth]

Space is not a premium for the game, so expand as much as you wish!

[granite axle]

yeah no absolutely, i just have ocd

[cursive heron]

sink it or turn it into turbo heavy fuel by just scaling up your heavy oil residue

[edgy leaf]

thats really hard when u have long recipe chains, to find a number that you want that doesnt have repitition anywhere

at least i found it hard when i tried it

[vapid gorge]

Yes, yes it is xD

[amber jacinth]

What I’ve noticed a lot is you get weird ratios for the base ores- which is fine, but it also leaves the player with a morsel of ore left over for personal use 😛
The machines never notice the taxes, dw

[tame frigate]

I don't think they do? I could be wrong.

[neat lantern]

Gonna sink compact. It's crazy that 600 rocket fuel can fill that many generators. I don't know what to do with the other 4,200 rocket fuel. That's way too many fuel generators.

[edgy leaf]

ah yes, my favorite activity. using the taxes (iron ore) i collect from my servants (machines) to create heavy modular frames (heavy modular frames)

[calm mauve]

if you have the power shards you can overlock the gens reducing the number you need by 2.5

[tame harbor]

There comes a point where you just build X fuel generators and make regular fuel to fill it out

Like, my base has over 3k crude near it, half tempted to just turn 600 of it into rubber/plastic and the rest into diamonds

2550 Crude/minute is 25.5 diamonds/minute!

[cursive heron]

isnt that 63.75 diam/min

[tame harbor]

Fark did I math bad

[calm mauve]

i get 510pm

[tame harbor]

Oh man
A. I really fucked up my math somewhere
B. Hot dayum that's way better than coal

[cursive heron]

oh yeah 510

[cursive heron]

oil based diamond is the better recipe
for some reason its 10 less oil per 10 diam compared to using petroleum coke

[tame harbor]

Smells like an oversight that

[ashen girder]

Because coke has other byproducts.

So the crude isn't creating only diamonds.

[tame harbor]

Oh no Polymer Resin! What ever shall we do?

[ashen girder]

...account for the extra plastic/rubber in the comparison?

[tame harbor]

Sorry. it's late, thought you meant more about the added logistical burden

[ashen girder]

No, more that 2550 worth of crude into HOR -> Coke -> Diamonds generates 425 Diamonds and 566 plastic.

[tame harbor]

Yeye

But 600 crude can be made into 900 rubber and 900 plastic with the right setup

[ashen girder]

Okay?

The question isn't "how else can we get plastic".

The question is "why are coke diamonds so bad?"

[weary sable]

using satisfactory calculator - looks like it rounds up to the nearest 10%. Is there a reason for this or is it just a way for them to simplify the site?

[tame harbor]

So going for the extra plastic from coke diamonds isn't really necessary

[ashen girder]

I mean, if you need plastic and diamonds, it might be a better choice depending on how much oil you're using.

Or rubber and diamonds, for that matter. 🤷‍♂️

You can also turn runoff HOR into it instead of fuel that way. The point is you have options to fit whatever scenario you're building your factory in.

[opal pewter]

I'm not the cleaniest or most artistic but some good advice I learned in the early versions was verticality...use those lifts as much as possible and butt everything up as far as possible and avoid over-stretching things

[calm mauve]

This is how I think I would do it as a BP. Each of this layout is one floor. 8 floors is 600 oil. Bring the diamonds and resin down to do with as you please.

[tame harbor]

Damnit now I am wondering if going for the Diluted fuel/recycled chain is worth it for the scale I am building towards

Or just going for polymer resin -> residual rubber+plastic and then converting the HOR into diamonds would be enough

[ashen girder]

Tradeoffs are the name of the game. 😉

[digital kelp]

Why are my smelters only running 25% when I have 4 iron smelters off 1 pure node?

[edgy leaf]

is ur belt speed high enough?

[digital kelp]

Should only need 1 mk 2 belt from miner to the first splitter right?

Then that goes 2 seperate ways so only need mk 1 belts running at 60/min?

[edgy leaf]

yep

are ur smelters full or empty

[ashen girder]

And is your miner full or empty.

[digital kelp]

nvm i am feeding the first belt is mk 1 lmfao

[edgy leaf]

lol

ive had that happen more times than i can count, dw

[digital kelp]

Im upset with my build tho only using 115 ignots per minute instead of the full 120 -.-

[tame harbor]

satisfactory tools is telling me to use the basic-ass plastic and rubber recipies

[ashen girder]

Have you found a slug yet? You can underclock your smelters to only output 115 total.

[cursive heron]

way too many buildings and too inefficient, oil based is straight oil into a particle accelerator
2400 oil turning it into hor coke diamond and resin for residual plastic is 187 refineries and 14 particle accelerators for 400 diamonds and 534 plastic

2000 oil turning into diamond is 10 accelerators for 400 diamonds and the remaining 400 oil can be used in the recycled rubber/plastic loop to make 1200 plastic

[digital kelp]

No i haven't done much exploring yet lol

[calm mauve]

the alts are unticked by default

[tame harbor]

i have alts selected, i just purposfully unticked the recycled plastic/rubber recipies to see what it said (after telling it to do petroleum coke -> diamonds)

[ashen girder]

If anything, that just points out how completely OP that loop still is. 🤣

[tame harbor]

and at the scale i am going for (this is before slooping, except for the Pasta where i will be slooping the two inputs instead) and the alts i selected, it only calls for 399 Rubber and 112 Plastic

rubber for Turbo motors, modualr engines, and caterium computers
plastic for super computers

that's it

[cursive heron]

much less buildings, and way more output

[ashen girder]

I am absolutely shocked at how different that is. 😂

[calm mauve]

less water too

[ashen girder]

Eh, water's free. 😉

Even skipping the recycled loop, it's a shit load more refineries and accelerators.

[tame harbor]

fused wire

[ashen girder]

I don't fuck with either of those personally.

[cursive heron]

fused, copper sheet can be removed from the production chain completely

about time too I was sick of making a bajillion refineries for steamed sheets

[tame harbor]

I unironically love fused wire/quickwire

[vapid gorge]

? all you had to use them for was AI limiters, and you never needed a ton of them

[ashen girder]

Circuit boards and heat sinks by default.

Oh, right, he has me blocked. 🤣

[empty maple]

90 a min

[tame harbor]

looks at my love for silicon circuit boards

[wicked ravine]

man i wish we would get a indication for how high elevators are when placing them

[ashen girder]

..the thing that goes to space?

[wicked ravine]

no, conveyor elevators

[ashen girder]

Ooooh. That makes way more sense. Lifts. 😂

Yeah, I 100% agree with that. Trying to eyeball two connections 20m in the air is annoying.

[wicked ravine]

even just a few meters can be annoying xd

[vapid gorge]

you do, in how much it costs to build

[wicked ravine]

ohhhh

didnt even think of that!

[ashen girder]

My brain's far too smooth for that.

[wicked ravine]

Amazing tip

[wicked ravine]

I feel that often in this game

[sweet blade]

How much is too much bio-power... Alternatively how much fuel does 30 biofuel generators even use per hour?

[wicked ravine]

i cant even fathom wanting to setup 30 xd

[sweet blade]

It fits nicely in a 7x7

[cursive heron]

2 constructors worth

[sweet blade]

Good. I got 3.

[wicked ravine]

i would rush coal

[cursive heron]

or just stick a somersloop

[sweet blade]

Nah. Too much pollution lol

[cursive heron]

in the solid bio

[wicked ravine]

pollution??

[cursive heron]

you can skip coal and basic fuel using geothermal

[sweet blade]

Lol yes, no smog in my world /s

[sweet blade]

Smart

[wicked ravine]

is this like a self imposed rule?

[sweet blade]

It's more of a joke

[wicked ravine]

gonna be tough with smelter and literally everything else producing smoke xd

[sweet blade]

I'm actually just over engineering everything and 30 was a nice round number

[wicked ravine]

thats fair

it does probably look cool

[sweet blade]

I'll send/post a screenshot when I'm back on my pc

[cursive heron]

Making rocket fuel is alot simpler and uses much less buildings compared to having to build diluted packaged fuel loops or turbofuel at phase 3

[wicked ravine]

oh yeah go from bio burners to rocket fuel xd

[ashen girder]

Doesn't rocket fuel use turbofuel?

[wicked ravine]

all these weird ratios are melting my brain

[sweet blade]

@wicked ravine the real reason it's that many is because on the next layer above I had to fit enough constructors to be able to make every type of biofuel. Each mob gets its own constructor, even though 90% of the time they're not going to be used and I'll just be feeding wood/grass

12 total

[cursive heron]

very possible now, connecting the vents in the map provides 10650MW, enough to get you to phase 4

[ashen girder]

7.1GW on average..

[wicked ravine]

oh yeah it probably is, but what a jump

[cursive heron]

only thing is you're spending majority of your early game exploring cause you have to pick up the loots needed to hand craft HSCs, upside is you're also getting mercers/sloops/hard drives

[wicked ravine]

love elevators

[calm mauve]

new ficsit meta job. elevator service tech

[wicked ravine]

dont think i would want that job in my world

[jolly sparrow]

what is powering this

[proud totem]

I am figuring out how best I should split my polymer resin between making plastic and rubber. My initial thought is to just do the 60/40 split, so I get 20 plastic 20 rubber per 100 resin. Or am I better off with relatively more plastic/rubber than the other?

[wicked ravine]

Im gonna be annoying and say: it depends on your needs i guess

[ashen girder]

Do you have trains? Do the tracks have any hills?

[vapid gorge]

if you don't have a production line that needs either just go half half, can always buffer each item so you have a bunch backed up for use

[proud totem]

That's the thing, I don't really have any immediate needs for either, but I want to anticipate the future as much as I can. I've only really ever gotten to computers in my previous playthroughs, and so I don't know if plastic or rubber is more needed as the game progresses

[proud totem]

Yeah, that's kinda my immediate thought

[edgy leaf]

is diluted to recycled plastic/rubber still the most efficent way to make plastic/rubber?

[proud totem]

I am going to have 337.5 resin/min though, so I'll probably do 300 to make 60 rubber and 60 plastic, and then send the leftover to plastic so I have slightly more

[edgy leaf]

epic, as soon as i get blenders im bulldozing the spire coast again

this time with the power of hindsight

[proud totem]

Also dang, my turbo-fuel power plant is going to require about 1400-1500MW to run (granted that also includes the plastic and rubber I am making). At least I will be producing 12500MW

[wicked ravine]

man that new steam tool for satisfactory is cool

[edgy leaf]

which tool?

[wicked ravine]

its called satisfactory modeler

its pretty basic right now but its pretty cool

[vapid gorge]

eh, it's faster to design the layouts in game. And then it's built

[vapid gorge]

and tools already does that much

[wicked ravine]

i just like this more

[edgy leaf]

pipes are so weird

[wicked ravine]

i like them

[edgy leaf]

if i connect it the way it is right now in the image i have perfect 600 flow rate

if i connect it where the pipe junction hologram is i have sub 600 flow rate

(yes i placed it and then connected the pipes, i didnt place it directly on the pipe)

i know i know, its probably because the way im doing it makes the system fragile

still strange

[wicked ravine]

wouldnt that be because it runs two different ways?

[vapid gorge]

feeding from one point instead of in the middle is often more stable

[wicked ravine]

where as now it only runs one way

[edgy leaf]

maybe

[vapid gorge]

and you've got fluid flowing in so many directions in that set up who knows what's going on

[edgy leaf]

more connections == more better

anyway i finally have encased industrial beams automated

[wicked ravine]

is it right that if i need 209 generators, but i overclock them 2.5x and then only need 83?

[edgy leaf]

84, because its 83.6

[wicked ravine]

ah yeah also the result i got

[edgy leaf]

yep htats true. generators overclock linearly

[wicked ravine]

thats gonna be way easier to build

[ashen girder]

Might want to make sure you've got 252 shards.

[wicked ravine]

now i just gotta figure out splitting 84 machines nicely

easy

[edgy leaf]

im still so sad that i made 250 powershards without slooping htem...

[wicked ravine]

ohhh big oof

luckily theres so many slugs

[edgy leaf]

yea, thats what u get for trying to avoid spoilers i guess

[wicked ravine]

how is your power so clean

[unborn ermine]

I just finished my exploration trip... gunna have 687 total shards when I process them, 👀

[edgy leaf]

i dont start the next build unless the one im working on right now is 100% flat

[ashen girder]

Do you use trains?

[edgy leaf]

no, not yet. i plan to create a secondary power grid when i get wonky consumers or producers

[unborn ermine]

Im not even in space elevator's 3rd delivery yet too

[wicked ravine]

that sounds insane but i like it

[edgy leaf]

my plan is to use priority power switches in series with sets of nuclear generators to remotely allocate power to the two grids

[ashen girder]

Yeah, keep in mind that trains have regenerative braking. 😂 That other guy's screenshot made me think of you.

[edgy leaf]

and variable consumption. i want my generation AND consumpiton to be perfectly flat

[worn trout]

is there like a "spot" thats near a bunch of pure nodes for copper and iron and coal

[edgy leaf]

and my max cons and cons to be equal

[edgy leaf]

generally its recommended to not look for a spot like that but instead to branch out and learn to make outposts and stuff

i think if u want copper and iron and coal in one "spot" the northern forest cliff is still the best tho

this area

[edgy leaf]

i wouldnt recommend it tho

[worn trout]

i got this spot here wich has a good amount of copper and coal nodes but only one pure iron node with the rest being impure

[unborn ermine]

The rocky desert spot near there is where I set up shop, just off that image to the left

[edgy leaf]

why do u need them all on one spot?

[worn trout]

so i dont have to make belt lines that stretch on for ages

and my one pure iron node over where i am is barley keeping my steel refineries going

i could probably cut down on the iron plate and rod makers but i like having alot of those

[unborn ermine]

Copper is a bit of a miss, but this is the area I setup shop in.

[edgy leaf]

you'll have to figure out how to deal with long distance logistics sooner or later

[worn trout]

im just gonna wait till i get drones to bother with anything long distance

[wicked ravine]

good luck xd

thats gonna suck

[edgy leaf]

if thats what you want to do, sure. but the game is designed to force you to do long distance logistics before you get drones, so you will have to fight the game

[wicked ravine]

if i overclock machines in a blueprint, does it save?

[edgy leaf]

yep

[wicked ravine]

oh sick!

This game likes to annoy you. Oh you want coal and iron? sure, pure iron, impure coal

[edgy leaf]

choo choo mother******

its kinda mean that the game forces you to wait 7 minutes between unlocking trains and signals

[wicked ravine]

huh

[pallid spoke]

ok slug duping is just rediculous x)

[lime ice]

i wish i’d learned of it before making 200+ shards from slugs 🙃

[wicked ravine]

wuup almost 64 generators down

[lime ice]

i’ve reached tier 9 so synthetic slugs are in sight but wow dark matter residue balancing looks like a pain without the pipe equivalent of overflow splitters

any advice on this?

[wicked ravine]

whats the problem?

[calm mauve]

do you mean fluid recycling? like this

fluid lower down at a junction gets used before the fluid above

[lime ice]

topping up sounds like priority merge but dark matter residue is usually in excess so it’d need priority split

[wicked ravine]

i wish we could zoop squares

[vapid gorge]

blueprints

[lime ice]

i find that blueprint of a square of foundation (whatever size you have BP designer available) is faster to place the zooping

[wicked ravine]

i cant get them to line up well

they end up clipping into each segment

[lime ice]

place them in "blueprint mode"?

[wicked ravine]

omg

i didnt know that was a thing

man it would have been so smart to put power poles and connect my generators in the blueprint

[ashen girder]

You can make overflow junctions in pipes.

They're the only special junctions that reliably work. 😂

[lime ice]

how does that work?

[ashen girder]

Oh, wait. It's not a liquid is it.

Nevermind. 😂

[calm mauve]

are you trying to use up the excess so the process doesnt jam?

[lime ice]

yes

[calm mauve]

how about adding the dark matter crystalization alt to the end and clock it so it uses just the right amount to stop the jam

[lime ice]

yeah that was my first idea: clock everything just right so it balances out, keep everything running at 100% (of clock) all the time, and hope slight inaccuracies don’t add up to a jam

but i’ll have variable use of power shards for building, and the excess can’t be sunk

[wicked ravine]

Since i need to move about 871 m3/min, how exactly do i do that with a setup like this?

[lime ice]

run two pipes

[wicked ravine]

ah i guess i can connect in both ends at the top and then just a connection downwards

[gentle jacinth]

guys

[lime ice]

trying for achievements?

[gentle jacinth]

nope

[spiral wigeon]

Whats the best place for a fix it sink mid game?

[gentle jacinth]

https://tenor.com/view/elmo-shrug-idk-i-dont-know-who-knows-gif-17348455

[vast jungle]

What do you mean, best place? You will most likely have multiple of them soon 😉

[cinder pecan]

I am lazy and attach them as overflow on nearly every major line 😁

[edgy leaf]

thats not laziness thats efficiency!

[cinder pecan]

Can't be bothered to find a common spot for them. slap em down and move on lol

[edgy leaf]

ice gecko gave me an idea, im considering building my base totally flat aorund the entire island, mostly on the water. like a giant circuitboard. what do yall think?

bad idea, good idea?

[cinder pecan]

Should be fun 🤣

[edgy leaf]

i loved the way my old oil factory looked like, it gave liminal space vibes

[cinder pecan]

I built mine wrapped around the spider rock formation. Was fun finding build lines that worked.

[edgy leaf]

https://youtu.be/O7MdMEExDoA?si=CAGfgJaBcw-w3lV9 heres a video of it

im thinking of having everything as two layers, one for buildings and one for logistics

or maybe two logistics layers. the buildings far apart from eachother and just expand over the entire ocean

[cinder pecan]

Spider taking over lol

[shrewd mason]

Can you have trucks doing different routes but using the same station? Or is it 100% gonna crash

[digital kelp]

It's 3 water pumps for 8 coal generators right?

[calm mauve]

yup

[digital kelp]

How much coal is needed for those 8?

[calm mauve]

gotta spread the extractors out along the gens though

120 coal

[cursive heron]

my brain is smoothing, if you're looking at machines with full overclock how do you configure the decimal ie 3.6 machines, how much is 0.6 of a fully overclocked machine.

[wicked ravine]

wait can you not clean out fuel generators?!?!

[edgy leaf]

3.6 machines total or 3.6 2.5 overclocked machines?

[cursive heron]

ie 9 iron smelters, divide by 2.5 is 3.6
thats 3 full overclock smelters and how do you configure the 4th one

[edgy leaf]

0.6*2.5

[cursive heron]

nvm had a brain blast, just multiply 0.6 by 2.5 to get 1.5 so 3 250% and 1 150%

[edgy leaf]

it needs to be 1.5x overclocked

[cursive heron]

yea

[edgy leaf]

yep

[cursive heron]

or 4 running at 225%

[wicked ravine]

cant wait to get back on later and find out my gens arent getting enough fuel

[harsh mason]

Woot built my first augmentor today

[pallid spoke]

omg..... I only need 1 node for life 😭

[cursive heron]

whats the prefered recipe for ficsite, just iron?

[amber jacinth]

wdym?

Like using all the SAM for iron conversions, or?

[cursive heron]

for making ficsite ingots

[amber jacinth]

ah gotcha

It really depends on if you're looking to conserve SAM or not- in which case aluminum would be best

[spice depot]

G G G G G G G G
E-+--+--+--+--+--+--+--+
E-+
E-+
E-+
E-+
E-+--+--+--+--+--+--+--+
G G G G G G G G
G - Coal Generator
E - Water Extractor
Would this work? if I'm using mk2 pipelines

[deep citrus]

I mean use whar you find the best considering what is going to be your factory location and how much you want to set up for them

[vapid gorge]

Yes? But there’s no reason to connect them

[spice depot]

should I just keep it to 3 for 8 gens?

[drowsy ferry]

What are you guys doing with all the oil you can get? the spire coast gives 2700/min and I can't figure out how to use it all, as in, other than plastic/rubber and a little bit for aluminum production I can't figure out a decent ratio of things to make.

How much of that should I devote to fuel generators, plastic/rubber, aluminum, coke, etc, ?

It ends up using like 300 refineries if I maximize plastic/rubber which keep me from using it for other projects

[cursive heron]

are you using the recycled rubber/plastic loop, a single pure node is enough to make 900 plastic/rubber already more than enough for most applications

most oil now just gets turned into diamonds

[drowsy ferry]

yeah that's what I mean, a single normal node can give me so much the rest is what I'm thinking about.

I'm super early tier 8 so not at that point yet, should I be saving the nodes for diamonds or something?

[cursive heron]

there isn't really anything that requires that much oil when using proper alts

[drowsy ferry]

:/
I wanna be efficient lol!
That's what I'm seeing and I kinda don't wanna spam like 200 fuel generators with the rest

[cursive heron]

oil is just that abundant, 900 rocket fuel or 216 gens only needs 225 crude oil

[drowsy ferry]

You think its worth it to upgrade above normal fuel for fuel generators?
Getting coal/sulfur for turbo seems a bit cheeks

with oil being so abundant that is

[cursive heron]

I skipped normal fuel and went straight to rocket just because its less buildings overall

[calm mauve]

the heavy oil residue alt into diluted fuel is king

[drowsy ferry]

tbh had forgotten that mam research so I'm gonna look into that

[vapid gorge]

yeah keep fluids as simple as yo ucan

[cursive heron]

with the introduction of drones being able to use fuel there's no real use for sulfur outside nuclear and rocket/ionized

you can cut out batteries from the chain

also you use nitro recipe not turbofuel

[drowsy ferry]

yeah I saw that, excited to try it out but need some actual factories for the higher tier components still

[vast jungle]

Sometimes the MAM wants to troll you...

[cursive heron]

automated miner lock that in

literally just needs iron now

and it stacks

[vast jungle]

automatic miners combined with dimensional depot sounds like a good QOL thing

[cursive heron]

not having to hand craft them in the equipment workshop and having it ready in your dimensional depo is goat

[vast jungle]

but at the moment I am just collecting a few HDs while my Phase-2 parts get finished

[cursive heron]

just find some random impure iron node and stick a mini factory with a depo and you're set for the entire game

[past stirrup]

im so confused in what world would being able to automate the construction of portable miners be useful?

[vast jungle]

easier placing/upgrading of Miners

[drowsy ferry]

single constructor somewhere > into depot > never worry about hand crafting them for miners ever

It's just a QOL thing

[past stirrup]

fair i guess

[past stirrup]

i guess i never really saw hand crafting them as a big issue since you place miners so infrequently

[cursive heron]

the annoyance is more noticeable when you hit mk2/mk3 and are upgrading or putting down more, setting this up early is just a huge qol

[restive mantle]

what the actual hell is a fiscite ingot

[drowsy ferry]

spoilers heads up

[restive mantle]

nvm

dont spoil it

must be something tier 7/8 im too tier 5/6 to understand

[past dragon]

I just read coal gen setup steam guide and saw the infinitely expandable setup from Jeslis. Question is: Is this infinitely expandable in a straight line? (surely it isn't saying modularly, as in repeating the same design and connecting to network. Otherwise, why would they mention it separately from the others.)

I'm having a hard time wrapping my head around what they mean. Can't a network only push so much fluid based on the limits of pipelines.

link to post: https://steamcommunity.com/sharedfiles/filedetails/?id=2124946046

infinite scalable design is in images a few scrolls down.

[drowsy ferry]

Which part are you asking is infinitely expandable? the coal can just manifold forever and as long as you regulate water it is too, is it the water part?

[past dragon]

yeah water

would you just multiply water extractors horizontally?

[vast jungle]

there are three solutions to coal water...

1. keep it simple: 1 water extractor (75% UC) for two generators
2. keep it efficient: 3 extractors for 8 generators (piping is a bit more difficult)
3. keep it tight: 1 water extracter (150% OC) for four generators

I normally go with option 1

[past dragon]

well, yeah, sure i could keep it simple

but i'm genuinely curious

[drowsy ferry]

These two parts can work if you keep going on but water has physics in the pipes so you should either use valves, unpowered pumps, or 1-way fluid gates to keep it from sloshing around.
If you instead just stick another pipe on top with water on it works perfectly fine

[past dragon]

This claims to be infinitely scalable. And seems rather simple at base value. is this infinitely scalable? and if so in what fashion.

My question is not alternatives.

[vapid gorge]

you don't need pumps or valves to get this working ??

[drowsy ferry]

should not need, for efficiency!

[cursive heron]

I'm pretty sure when it says infinitely expandable it just means copy paste that setup for however many times you need

[vapid gorge]

does not need

in fact you don't need valves anywhere and shoudl avoid them

[drowsy ferry]

If you pile the water pipes like this you can use multiple and go for however long you need

[vapid gorge]

do you have mk1 pipes in this image or mk2?

should be good either way just curious

[past dragon]

well that goes back to my original question. shouldn't pipes limit scalability

[drowsy ferry]

Yes everytime its the pipes limiting

[cursive heron]

reject coal embrace geothermal

[past dragon]

So then why do they mention infinitely scalable?

If I take any design restricted by laws that force a size limit, and therefore make copies, and connect them to an overall network. Isn't that the same as any?

[dusky aspen]

Infinity scalable as in spacing the producers out makes it so you're not capped by the pipeflow limit

[drowsy ferry]

Pretty much yeah I'd agree, probably something akin to a typo cause the only way I see that image scaling infinitely is if you keep adding more pipes down the way but if you do that you should just cut off and restart a new one

[past dragon]

okay so now we're circumventing pipeflow limit

[past dragon]

Can you elaborate, because this seems like a lead on an understanding of the multiplicity of this design.

[amber jacinth]

If all it is is tileable 3:8 coal setups, then sure, I guess that is infinitely expandable, barring map resources of course.

[vapid gorge]

they probably mean you can build in modules

[past dragon]

Absolutely true, to both previous messages

that's why I'm wondering why they said it, as that doesn't separate it from any other design

Resource limits is not necessarily a factor in my question. Really it would be the limits of the mode of transport. I find the design in the photo I sent really appealing and very simple. But for some reason I'm caught up on their comment or claim that separates it from the rest. And was wondering if anyone could explain or draw out what they mean, beyond modular design. (because that doesn't separate it). Or am I missing something fundamental as far as terminology and definition?

The wording next to the image is: This design is a expansion on the 3:8 method shown above. Best approach is to build the top half, then add in the bottom half. This type of approach is infinitely expandable.

Infinitely expandable to me reads as such: growing in a straight line. Maybe I should just let it go 🥺

[hollow wave]

i'm sure it's intentional now that i've noticed it but has anyone already noticed that every single upgrade to the dimensional depot uses exclusively prime-numbered item counts? and it's the only place in the whole game where that's the case?

[blissful tide]

So all of the alt recipe analysis based on WP has been deleted from the official wiki... Anyone know what the plan is to replace it?

[hollow wave]

anyway i just put this on the wiki because it's a really fun fact i think

[amber jacinth]

Yeah they probably just meant that it’s fully tileable this way

[drowsy ferry]

unsure what you're talking about but wrigh516 has update 1.0 alt recipe analysis on reddit

[white bloom]

not the one who deleted it, but very glad it's gone. This whole WP thing was completely arbitrary and giving people false ideas.

the worth of a recipe depends on your personal weighting of resources against each other, depending on what your goal is. There is no way getting around that fundamental fact and any attempt to assign a single "objective" score number to it and order them in a linear preference order has to involve some completely arbitrary choices

[past dragon]

Alright, thanks for that answer. Going to look into definition of tileable and attempt to use that to understand the claim.

[hollow wave]

i liked the pre-WP analysis of "resource efficiency: -20%, power efficiency: +50%, space efficiency: -5%"

[white bloom]

but you cant compute resource efficiency, that's the point. and it's linked to power efficiency.

[hollow wave]

"input resource efficiency" would be a better term

[white bloom]

I know what you mean. I'm not being pedantic about words. It is in principle not possible

in general some recipe will cost more of one resource and an alt recipe more of another

you cannot break that down into a single number

[drowsy ferry]

With limited resources on the map it is 100% possible to calculate the best recipes as well as which are more power efficient

[white bloom]

yes, and I did that. but that's always ever going to be best with regard to a specific objective

[drowsy ferry]

^ in reference to this statement which would be false

[hollow wave]

yeah, that's why there's three lol

solid steel makes 150% of the amount of steel, but takes slightly more power; those ultimately are separate measurements to almost anyone, and the only time they're not is a full-map-utilization scenario

[white bloom]

No, you're missing the point. There is not one, the resource efficiency. There is resource efficiency only in regard to a specific goal, not in general

[white bloom]

but you're still trying to keep the resource efficiency in a single number of these, which doesn't work. you have efficiency with regard to every single of the 100+ items in the game and no universal way to combine it

[hollow wave]

precisely; if my goal is to use less iron but i don't care about power, i'd look at those three numbers and think solid steel is good
if it was to use less power or space and i didn't care about output, i'd make regular steel

[past dragon]

okay so you mean an easily placeable blueprint that plays well with its copies?

[vast jungle]

interesting... the MAM Library does not scroll?

[white bloom]

it's not just iron though. it's iron efficiency vs coal efficiency vs compacted coal efficiency vs whatever else you can use to make steel

[drowsy ferry]

correct you are limited to a certain amount of held recipes

[vast jungle]

not sure you are limited, but you cannot see the rest

[white bloom]

some recipe might be more coal efficiency while another is more compacted coal efficient. and there is no way to combine these that holds for any objective or any valuation of the individual resources

[amber jacinth]

Something like that, yeah. Just an easily repeatable design overall.

[white bloom]

WP was an attempt to combine the efficiencies towards all items into one number and it's flawed from the start

[white bloom]

this would work if there was 1 resource, power, and space. but there are 100+ resources. 100+ different efficiencies

[past dragon]

I'm sorry xD, I do feel like I'm being annoying at this point. But aren't they all easily repeatable in the context of blueprints? Or even without, based on perception. I'm just gonna let it go T_T
back to building
@amber jacinth Thanks!

[fading coyote]

is there any problem splitting/merging this way in order to get a 30/70 split on something?

[hollow wave]

sure; i'm not saying it's not still somewhat flawed, but it's infinitely better than just "wp" on a global scale

[white bloom]

Ultimately the best you can do to judge the value of an alt recipe to you is clearly state your objective, then use an optimizer to get the set of efficient recipes for that objective, then collect those. Or alternatively, keep track of your unlocked recipes and accessed resources, and use an optimizer to compute by how much your objective value increases if you unlock recipe x,y, or z, and choose the one with the biggest increase

multi-resource optimization is inherently multidimensional and there is no way to reduce that to a one-dimensional problem

[wind spade]

It was misleading at very best. Will be replaced with rext-based descriptions of what the alt does and abount interesting combinations

[wind spade]

Heavily recommended to not follow blindly

[wind spade]

Yeah that wasn't really usable anyway

[hollow wave]

it was rather helpful to me in early game early-access

[white bloom]

see, this is a lot more healthy way to go about it

[wind spade]

It's not possible to calculate "best" recipe

[wind spade]

You're not. You can use the search box

[wind spade]

It was very flawed

[hollow wave]

honestly the general consensus should be "pick whichever is better or neither, and don't worry about it because by the time it actually matters you'll probably have collected almost all of them anyway"

[hollow wave]

@wind spade while you're here what's your opinion on this?

[dusky aspen]

The smelting recipes are strong
The one that uses water to get much more iron out of ore
The one that adds a bit of gold to copper to get a ton more copper is also strong

[hollow wave]

the alloys in general seem pretty good, especially when they're synergetic with each other (iron copper and copper iron)

[dusky aspen]

And combination recipes like steel rotors + iron wire make it so you're able to just throw down a blueprint at a single iron ore location and instantly start printing Rotors Stators and Engines with 1 node

So subjective

[amber jacinth]

Iron pipe

[hollow wave]

iron pipe seems great but i haven't got it yet

[white bloom]

One could conceive an Alt Recipe Picker where you have to enter all your progression information. all recipes and milestones and MAM unlocked, how many nodes of what resources you have accessed, and what goal you're currently working towards. Then you enter the 3 offered alts and it tells you which gives the greatest increase towards the goal.

But that approach would miss strong synergies. Some recipes are not good until you got another recipe, but then together they are very good. With the above you never catch that. Also you wouldn't always only want to collect for whatever milestone you're currently working on, also future goals. And you'd be willing to access new resource nodes if it's beneficial, that changes the calculation again.

I don't think it's feasible. People want a brains-off way of being told what to pick but it really depends, they need to understand the trade-offs involved. So I agree some textual descriptions should be best, and for anyone who knows exactly what they want they can use any optimizer tool out there to make a shopping list of exactly which alts they need and don't need for the optimal production chain for their objective.

[amber jacinth]

Very inefficient, but nice for basic steel products without having to bring in coal/alternatives for steel

[fading coyote]

anyone haev a comment on this? 😔

[amber jacinth]

Manifolds!

[tiny fog]

Thats what i really like there are so many possible solutions, everyone will do it a bit different

[fading coyote]

dont work for low volume items tho

[woven minnow]

any changes to pipes behaviour for 1.0?

[dusky aspen]

If you wanted to define best alternative recipes for the expected increase to awesome sink value when everything is being sunk as high value as it can, but even that is subjective, then a tiering could be quantified

[white bloom]

didn't mean to create a thread. yes it works but beware the bottleneck. you could re-insert the two backwards-belts after the first splitter to avoid that issue entirely.

[amber jacinth]

Manifold would work regardless- just a longer startup time.

[hollow wave]

oops i thought we were speaking in threads now

[fading coyote]

ok chat here haha

Yeah so you'd want to split, do the feedback merge, and then split again. Because then you'd never overload a belt if it's a concern

[hollow wave]

what tier is the input belt

[vast jungle]

there are certainly recipes that seem to be useful than others... but depending what your resource situation is the priorities might be different.

[white bloom]

you see this is a clear objective, yes you get definitive answers for that with explicit constraint assumptions. I agree personally that seems to be the best long-term goal conceivable in this game, the only number that keeps tracking as you go up. But some people have different opinions on this and rather want to max out power production or what not

[amber jacinth]

Biocoal & charcoal 😔

[fading coyote]

Doesnt matter on specifics, in this specific case im building rn the initial input is fully saturated

[amber jacinth]

Yes, I know folks have found uses for them in remote ammunition setups… but I just can’t bring myself to do it

[vast jungle]

ever were on a HD trip and hit the one damned droppod that need COAL... and you don't have coal? Then you will wish you had this recipe 😉

[fading coyote]

well i dont THINK specifics matter

[white bloom]

like when you say all resources on the map, all mam research and all recipes available and you demand power self-sustained and no waste buildup, you get a definitive best production chain. and the recipes it uses are efficient (in that scenario) and the ones it doesn't are not. simple yes/no

[fading coyote]

i wanna know if the theory works in diferent scenarios as well

[woven minnow]

does headlift measured from the centre of the pipe cross section to another centre of the pipe cross section upward?

[hollow wave]

what does matter is "do you have the next tier of belt unlocked also" in which case all your problems go away if the merged belt is a tier higher than the input belt

[fading coyote]

Right right i gotcha

[hollow wave]

so if your input is a mk6 i'm sorry lol but at that point... you'd saturate a manifold fast enough anyway.

[wind spade]

sure, but that's really specific case that imo doesn't apply to pretty much anybody here

[white bloom]

this, yes. in a 1:n balancer, you always want to merge after the first splitter or so for exactly this reason, unless you know for sure you won't need the full belt capacity as throughput

[vast jungle]

if you have all resources on the map I would assume you also have all alt-recipes...

[wind spade]

given that:

• nobody used the whole map (from what I know), given how PC-intensive would such a factory be
• most people have at least some production to storage for personal use

[white bloom]

yes, I agree! this is exactly my point I've been trying to make here. We shouldn't enforce a specific goal

people must simply accept that there is no universal, one dimensional scale of efficiency that they can map all recipes to

[fading coyote]

are there any other interesting ways to do a balanced for non 2x+3y sort of splits or is feeding back the way to go

[hollow wave]

a multi-dimensional heuristic too complex for human understanding? this sounds like a problem that can be solved with neural networks! /s

[vast jungle]

neural networks are only as good as their training/incentive function

[hollow wave]

feeding back is the way to go

[hollow wave]

incentive: consume

[vast jungle]

its not "2x + 3y"... its any number thats a multiple of the prime factors 2 and 3

[hollow wave]

2^x * 3^y

[wind spade]

which is why I heavily despise any "alt recipe analysis" spreadsheets or "best alt recipes" videos

while people making them usually understand that it's subjective (and often mention it in the content), people following it usually just blindly do "oh, X is a S tier, let's use it everywhere", without thinking whether or not the recipe is actually useful to them or if they combined it with other recipes that would make it "good" for them

[white bloom]

when you have general m:n balancers you get all sorts of very interesting belt circuits as optimal (i.e. minimal node count) solutions. When they don't factor into 2s and 3s nicely, I think there will always be cycles involved somewhere, but it isn't always a simple "merge all, then split all to the next larger power 2^x * 3^y, then feed back the excess". Sometimes there are more elegant solutions.

[hollow wave]

i actually really enjoy that about alt recipes, because so many games stick to linear upgrade paths that make every upgrade useless as soon as the next one's acquired

[fading coyote]

Oh shoot this totally doesnt work when i stop abstracting the input as a single belt lol

[vast jungle]

and we got a whole lot of good new alts with 1.0

[fading coyote]

if i want to do 4 incoming belts and split it into 30% 70% this gets a lot more weird

i guess in genreal thats a 4:10 balancer

[white bloom]

yep.

[hollow wave]

2 to 5, really

[fading coyote]

oh yeah

[wind spade]

well, most of the time it's usually met with complexity increase, which is a valid balancing point. But not always

[white bloom]

I ade a script a while back that searches for the optimal balancers when entering m:n and giving an internal bottleneck limit. optimal as in, uses as few splitters and mergers as possible

[hollow wave]

actually that's a single 2:5 on only half of the inputs, if all of the inputs are the same

[fading coyote]

Looks like calculator website has a bunch of balancers pre computed

[white bloom]

unfortunately it will run for a very long time for any slightly larger numbers 🙂

[hollow wave]

i wonder if my algorithms professor would think this is a cool application of graph theory for bonus credit

[fading coyote]

Gonna go on a limb and say its gonna be some stupid NP complete type shit

[white bloom]

the thing is pretty much the only way of knowing whether it's truly the minimal balancer is by brute forcing all possible smaller splitter/merger networks

so my script generates all of them in increasing order of nodes, throwing out isomorphic duplicates

[fading coyote]

I think that makes it np hard

[white bloom]

but even the number of non-isomorphic networks grows pretty fast

[hollow wave]

on the plus side we don't ever have to worrry about factorio-style lane balancing; phew

[white bloom]

need to find a way to detect networks that will surely not be able to achieve the desired balancing evenly early during construction, to avoid this sort of blowup

[white bloom]

absolutely, it's a very interesting graph theory problem. tough nut though

idk if it's np hard because I haven't shown how this can be used to solve another np-complete problem by reducing it to finding a minimal splitter/balancer network, which would be necessary to prove its np hardness. 😛

but at least that way I currently go about it has runtime growing exponentially with the input numbers

[fading coyote]

I cant quite remember classifications

[hollow wave]

i don't have the time to spend on taking a crack at the algorithm so unfortunately i won't be 🤷‍♂️ but it sounds fun nonetheless

[fading coyote]

But if validating an answer is nonpolynomial time then you are not in luck hah

[white bloom]

tbh that part I never really understood about the classification of optimization problems, rather than decision problems, anyways

[fading coyote]

Oh yeah it might be diffrent idek

only really covered decision problems at my daycare ass school

ANYWAYS definitely difficult to solve

[hollow wave]

tough beryl nut to crack, yea

[white bloom]

yeah me neither right now I'm already procrastinating heavily, need to finish writing my thesis. Maybe some time next week, though polishing the sloop optimization is more urgent for satisfactory priorities 😛

[fading coyote]

well tomorrow i need to go and fix this balancer bc it sure is messed up

[hollow wave]

definitely off topic but what's the thesis topic?

[white bloom]

also graph theory actually. have you guys also been playing Sanctum, CSS's tower defense game before they made Satisfactory?

[fading coyote]

never erad of it

[hollow wave]

nope

[wind spade]

would be funny if it just spit out manifold always

[fading coyote]

I assme its a rule that it has to perfectly output m:n lol

[white bloom]

okay the gist is it's a tower defense game, but you can shape the path enemies walk by building your towers in their path

usually maps are rectangular grids

[fading coyote]

i played some games like that before yeh

[white bloom]

my thesis is about how to find the LONGEST possible path you can make enemies walk on a given map

that's my master thesis

[fading coyote]

is it not just a snake going up n down and up n down

[white bloom]

of course it's formulated abstractly, but Sanctum / tower defense games were my personal motivation on the topic

it's called the longest induced path problem in fancy

[wind spade]

that's useful 🤔

[white bloom]

now given, I worked on an algorithm that can do it on any kind of network, not just on rectangular grids

[wind spade]

does/will it support:

• already existing blockers
• points that the enemies must pass in some order?

[fading coyote]

ah that makes it more interesting

[white bloom]

on hypercubes, it's called the "snake in the box problem", probably the most famous instance of it in acedemia because it's relevant to error-correcting coding

[fading coyote]

goodness gracious i am just remembering i really hated taking algos

[hollow wave]

that's actually really interesting; i'm technically a junior (currently in algos and architecture etc.) but i've ended up being responsible for quite a lot of research of various topics so it's always fun to hear cool things others have been working on

[white bloom]

existing blockers yes, that's easily modeled. however, unblockable tiles not so easily. though you can model it still somewhat by just all-to-all-connecting that one's neighbors

[white bloom]

points passed in order: nope

[fading coyote]

id assume in the game and thesis ur trying to maximize their shortest past i think thats what u said earlier

[white bloom]

yeah that's what it comes down to, in some sense

[fading coyote]

ill stick to software development 😓

[wind spade]

well that is software dev 😄

[fading coyote]

noooo it aint

[white bloom]

path needs to be induced, i.e. there are no connections between non-immediate successors along the path, so called chords.

[wind spade]

I mean if you think about e.g. Satisfactory Tools I've made - what do you think it powers it? 😄 linear optimisation algorithm 🙂

[vast jungle]

I really like the "keep in the library" feature of the new MAM... being able to block two recipes not showing up in new HDs is really good 😄
I have now FOUR of these "do not want both" recipes

[fading coyote]

yeh but the math is the hard part

[white bloom]

you'd be amazed with how quickly and brutally that problem scales in runtime as there are more nodes in the graph you're running this on

[fading coyote]

im willing to bet any interesting satisfactory calculator made to date is gonna be less than 10k lines

uness its like synthesizing some ingame layout

[wind spade]

just the dataset is 60k+ lines JSON

[fading coyote]

dosnt count

[vast jungle]

I think the hard part of the SF tools is all the UI programming to make is USEFUL... its comparatively easy to do this by hand on an Excel-Sheet for one instance of the problem, but backing it into a useable web-app takes time

[fading coyote]

yeah thats prob true

[hollow wave]

i think my desired research will ultimately come down to pathfinding optimization as well, albeit much more physical --- "what's the optimal path, subject to constraints of the control system, a high altitude balloon can take in the current and predicted atmospheric conditions and wind layers to get within a certain radius of a landing/recovery point"

but that's a little too far off topic for the channel so i'll leave it there

[fading coyote]

altho if u wanted to do a quick and dirty copy of the tools u could just use graphviz and yeah have the problem solved in no time

[wind spade]

also, "lines of code" are very hard to compare, what do you count as lines of code?

• third party libraries?
• config files for build?
• comments?
• empty lines?
• datasets?

[white bloom]

if you subtract recipe data, yeah probably. mine is 1.2k lines including recipes and sink values in their own lines each, so probably around 900 lines

no UI though

[hollow wave]

words of machine code 😈

[fading coyote]

im just being very very vague just comparing how difficult the math and algorithm part is vs implementation

its generally true

[white bloom]

can you do linear programming in excel now? you mean with advanced scripts probably, because with default functions, surely not

[fading coyote]

GPTs are not a lot of code at all, but god damn try teaching someone all the math of how they work
Fortunes algorithm i cant understand for the life of me but a python script of it is like 200 lines
etc etc

[vast jungle]

Excel is turning complete I think

[hollow wave]

it is, i've made a functional 16 bit cpu with nothing but =NOT(AND()) and iterative cell resolution

[fading coyote]

lmao

[white bloom]

theoretically yes, practically no it's not going to be feasible to make a Linear Programming Solver in excel

[wind spade]

now make it run doom

[fading coyote]

excel nand to tetris

[hollow wave]

this is literally what i did; i got bored during covid pandemic

[wind spade]

and play bad apple

[fading coyote]

we basically did nand 2 tetris in some shitty hdl simulator thingy when i was in school

[hollow wave]

i stopped at Fibonacci purely out of tedium

[fading coyote]

was cool

i wuz never motivated enough to actually make smth cool and intersting and challenging rofl

[wind spade]

actually now I'm interested in this - if you answer my question about lines of code, I'd like to see how much code there is for SFTools 🙂

[hollow wave]

i think my problem tends to be "i have all the motivation in the world and too much adhd to force myself to follow through with it"
hence stopping after fibbonacci

[fading coyote]

my judgement would be exclusively source code. Definitely no imported libraries. Definitely not comments, empty lines, or datasets.
build configs and stuff, maybe, idk

And I do NOT mean all calculators combined, I mean any single interesting calculator algorithm

[hollow wave]

how about lines that just have a closing bracket

[fading coyote]

getting a bit too pedantic, that probably counts

[wind spade]

oh you mean just the algorithm and not things around that like UI and stuff?

[fading coyote]

Eh maybe include UI, although the spirit of my original statement was basically just the algorithm. UI makes it more interesting bcs thats more of the software developer experience

But yeah u cant go like, my 10,000 npm packages count as source code! haha

[lime ice]

so yesterday i was worried about how to balance dark matter residue without overflow valves

i’ve found a satisfactory solution, and it turned out Snutt kinda gave it in a 1.0 teaser video: consume all the residue to make more dark matter crystals than needed, then use a smart splitter to sink the excess crystals

[fading coyote]

nice pun

[wind spade]

ah, that kinda doesn't play in my favor, as the "algorithm" is me generating a model for external linear solver 😄

[fading coyote]

i barely know what that means

[wind spade]

basically I just tell it "hey solve this", and majority of the code doing the actual calculations is in external third-party software

[hollow wave]

i presume solver is a library and most of the code is asking the library to solve it

[fading coyote]

Yeah but thats my point now aint it

development aint hard in the same way that coming up and proving algorithms are

NOT trying to say its stupid easy or anything, btw

Like do u think writing this display thingy (before chat gpt...), or coming up with the tileset to generate a sierpinski took me longer?

(the code did take longer but not by THAT much)

[hollow wave]

depends on if you're writing to get it done or writing for maintainability

[fading coyote]

started making that for a professor but got bored and graduated before making it real nice

[white bloom]

the algorithm for linear programming is actually fairly simple. look up Simplex Method if you're curious. It's just extremely crucial that you have a hyper efficient implementation of it to solve large problems. And it's so extremely important for practical applications that people have spent a lot of time on perfecting the implementation.

[wind spade]

why is it 3D xD

[wind spade]

hence why I didn't want to reinvent a wheel and used existing thing to do the math

[fading coyote]

It is meant to do things in 3d, the tileset just forces it to 2D by not giving the 3rd axis any connection sockets

[wind spade]

oh I see

[fading coyote]

Basically u start with a single cube, each face of the cube has colors. And you have a bank of cubes with specific face colors. cubes connect when their faces have the same color. And the point is to come up with tilesets that deterministically do what u want

some of my professors did papers n shit on that

[white bloom]

of course, so did I. just have to get a little inventive about the nonlinear aspects of the game. power cost with clock speed and now somersloops.

[hollow wave]

can you make a bad apple tileset

[white bloom]

but the largest part of it is just a linear program and yes, absolutely no reason not to use a pre-existing solver

[wind spade]

my approach is: "just ignore it"

but no, for new Tools, I've been already doing research on things like MILP solvers

[fading coyote]

With sufficient tiles you could definitely make a gigantic thing that renders whatever u want

[fading coyote]

Yeah i guess that can be an actual hard problem for developers to solve

Like thank the lord im not building a database management system or smth those are complicated

I just type away in my baby high level languages

I really do get paid a decent living just to make fucking CRUD apps all day

at least I know how to use the supreme satisfactory calculator though

[hollow wave]

if it makes you feel any better i work at nasa and program almost exclusively in python

[fading coyote]

KEK

[shadow kite]

i tried to get into python but i got burned by the py2/py3 switchover

[white bloom]

I guess the key part is getting the large matrix into reduced row echelon form as quickly as possible with different pivot columns each time. it's a very general lin alg problem so people have gone crazy eeking out every bit of efficiency possible when doing that

[white bloom]

py3 is nice tho, give it another go, you'll get used to it

[fading coyote]

Right now the minority of my days are "how in the world do i even code this" and rather most of it is "bang on your keyboard implementing an ubsurdly simple thing that just takes 4 hours because our system is bad"

although I do get to put some thought every day into figuring out how to unfuck our awful system for sure

[shadow kite]

i mean, now it'll be safe to do so, and if i have a usecase for it, i might.
though, my strong suit is C++ 🤔 (been looking into switching to Zig because I prefer it over Rust though)

[wind spade]

yeah that's like 95% of my day job as well... write a code you've written 10000 times already, just with very slight modifications big enough to justify not making it into a library

[fading coyote]

Yep

Its way too off topic to chat about and I get to complain to my friends all the time, so I'll spare yall, but my company is so screwed LMAO. Little startup gig im doing. it's really ass. I want to work on something that challenging sometimes. Thats only really happened like once in the past year

[frosty owl]

Yes, that is correct.
The only way it can stop working is if you "desync" the Rods and screws production so that only one of them keeps going for a while (or, similarly, if there is an issue in clocking or belting that causes any lack/excess of items reaching the Assembler)

[stray nest]

@oblique hollow I wanted to run something by you to see if you as the resident pipe expert see the same behavior. Hurts my brain to think about and I can't come up with a foolproof testing procedure for this.

I feel like the following linked post covers it.

If this is true, its a very common problem I see with peoples builds, but I do not feel confident enough advising them on doing the same unless I can confirm it elsewhere. Would appreciate the reply, sorry for ping lol
#1286233308170162176 message

[fading coyote]

using a belt for 2 different items seems like black magic evil to me

[royal yacht]

Yeah I don’t think that should have happened at any point. It’s overflow sunk anyways so should correct itself

[shadow kite]

i haven't looked into it, but is micropython heading in py3 direction? (because it's a nice option for some stuff)

[royal yacht]

But anyways I feel like it’s also a good showcase of how sushi increases throughput because conventional methods would require a mk2

[fading coyote]

maybe i need to go back into aerospace because there i really had no god damn clue how to code anything in the way they liked it haha

[hollow wave]

can confirm, aerospace programming gets crazy

[frosty owl]

Ahah, yeah, the rotor production is a funny example of how one can run an otherwise-impossible production at 100% without handfeeding or needing to unlock better belts 😆

[shadow kite]

industrial environments can be like that too

[fading coyote]

I did an internship back when for a company doing that and I lucked out with them liking me enough to put me on an actual dev team instead of verification and stuff. It was crazy what they had me doing

[oblique hollow]

A quick little interjection:
Its possible they do "force the main pipe to fill first" , but after that you sort of deal with upfeeding and fighting gravity.

" Rocking motion" is not an issue as the same happens on belts.
The only time you dont want rocking motion is when you are at the pipe's limit

Seeing as you used 4 pipes for 16 generators, you are well below any danger limits

Buut i will run a few tests sometime in the future again

[white bloom]

if you're used to C++ what do you need any other language for anyways? lol. why are you walking among us mere mortals

[icy kindle]

Hi Folks,

Question about overclocking and powe consumption, is it a lineair thing, or is there a formula for it?

[shadow kite]

some microcontrollers are only micropython or assembly 😦

[fading coyote]

Doing c++ when you don't absolutely need to sounds like hell

[white bloom]

formula, see wiki: https://satisfactory.wiki.gg/wiki/Clock_speed

[fading coyote]

Sorry, I only use programming languages with built in map reduce filter functions 😡 👺 🙂

[vapid gorge]

apart from charcoal and biocoal, which are just very niche, all these recipes are very usable

[hollow wave]

it's a power function!

[white bloom]

power = base * clock^log_2(2.5)

[oblique hollow]

(quite literally)

[shadow kite]

totally depends on what you do. I don't have to deal with multithreading, other than thread pools thankfully.

also libs matter

[white bloom]

... * (1 + slotted sloops / max sloops)^2

[hollow wave]

lmao it is a literal power function; didn't realize that

[white bloom]

exception is power buildings, there it's linear

[frosty owl]

Interestingly, similar sushi setups can easily be made so that they are extremely "robust" (built so they never stop working no matter what, pretty much).
When the source item is just one (in this case Iron Ore or Ingots), as long as all machines in the setup are on the same power grid and there are no "long belts and short belts" involved (them buffering a lot Vs very few items may lead to issues during power outages), one just has to make sure that that single input is running fine and the whole system won't ever clog
(Ofc, the system's output aka rotors is assumed to be sunk/consumed~)

[stray nest]

Would that rocking motion mentioned cause fluid to reverse flow temporarily and cause issues with extractors filling/keeping the gens full?

Guess what I am really asking is if backflow acts as I am imagining it

[fading coyote]

I don't even know c++ my old job was pure C99 or something. I did really adore that though. Reading directly from memory to detect real world events and stuff is very cool

[hollow wave]

this part is quite funny when it makes certain buildings peak at 26,862 MW

[oblique hollow]

temporary but it wont cause a permanent stutter
If the extractors fill up its likely head lift

It may also be that you oversupplied water, cant say

usually, for 16 generators, i use 3 pipes, each filled by 2 extractors

[shadow kite]

C++ and C often go hand-in-hand tbh. C++ is like 10 steps forward, but 9 back 😄

[fading coyote]

I just know modern c++ has a lottttt of extra stuff

[stray nest]

I personally am not having any issues, but I see so many others have issues where machines dont stay full long term, and I cant tell if my little loop over with the pipes is whats preventing me from experiencing that.

I use the same technique with Pure oil nodes, 99.9% efficiency with no issues.

That setup is using 5 extractors [3 normal, 2 OCd on the closest 8 gens] with no pumps and 1 crossover between pipes to [again another assumption of mine] help maintain even pressure across the whole system

[fading coyote]

Well ok it's bed time I need to rip up my faulty balancer tomorrow 😔

[lapis flume]

guys 1 or 2?

[vapid gorge]

are either of them going to be useful for your next project?

[lapis flume]

nope

[river night]

i dont think i ever used either

but by the time i build large i aim to have all of them anyway

[frosty owl]

@stray nest I think that many underestimate the positive effects of building pipelines "properly" (90 degree angles, no weird/confusing clipping or extreme placing of junctions, no chaos) on overall fluid flow.
I've seen many setups start working "just" by remaking some sections into a less messy thing (keeping the original pipe design)

[lapis flume]

and i cant recan it, the first one was drills or + 6 spaces for inv

[river night]

should've taken the +6 😄

[hollow wave]

if you're not using either yet i think the meta now is leave it until you plan on using one?

[fading coyote]

Oh shoot that reminds me. Do water towers work? Can I just pump up high at the extractor and just freely build with infinite pressure?

[river night]

can always leave it, makes future harddrives have different choices

[hollow wave]

yeah but like, that has the same power cost as putting pumps where needed

[fading coyote]

It makes it simpler by avoiding putting pumps on each separate incline and just doing it all at once and being done forever tho

[shadow kite]

imagine no pumps, only fluid trains 🤣

[oblique hollow]

i assume fully overclocked pure oil nodes?

[vast jungle]

I never had much use for the two Fused recipes in earlier runs... never went "big" enough to run out of Caterium... and why the hell should I need Caterium to make normal Wire? 😉

[hollow wave]

you wouldn't have to put them on each incline; only the first!

[fading coyote]

Hell yeah. Train my water up into the water tower to get pressure lol

[fading coyote]

Huh what, if you have a fork in your road and each forked path goes up, they both need pumps right? (Assuming they don't reconnect up high)

[hollow wave]

i mean yeah it applies separately to individual pipes after a split but is anyone splitting pipes before taking them long distance in opposite directions?

[fading coyote]

I don't understand

For me, right now, no

[oblique hollow]

No they don't

[fading coyote]

But sure why wouldn't you have a split pipeline going in 2 directions at some point?

[oblique hollow]

Pump pressure doesnt get divided by junctions

[shadow kite]

i did for asthethics, on the outside of a building once but they weren't that long.

[oblique hollow]

It stays the same all the way

[hollow wave]

if you have to have a pump after the junction though you;d need two

[river night]

because pipes don't contain that much volume really that a long-distance pipe that then also splits is a common occurance

[stray nest]

Nope. Not OCd... yet. Waiting to use all the oil nodes before i start doing that

[oblique hollow]

If you have a pump before the junction you only need that one, as long as you dont exceed the height limit

[river night]

not that I would ever really recommend long-distance pipes

move the factory, move solid products, not liquids 😄

[oblique hollow]

Overclocked pure nodes are only a problem due to thr mk 2 pipe limit

[fading coyote]

If you would view the amazing diagram

[oblique hollow]

Non overclocked pure nodes are not a problem at all

[fading coyote]

The right fork has a pump, the left doesnt.

[shadow kite]

fluid choo choo...

[fading coyote]

But the left fork does need a pump, no?

[river night]

I would prefer that over a long pipeline, in fact I love the look of fluid trains, so might just do that for this reason alone

[fading coyote]

If a water tower works my boys r gonna build on the server for sure

[river night]

of course you can also build a tower if you want, a big idea is to just build how you like

[vast jungle]

trains are always a nice solution

[lapis flume]

will 1 be ever usefull

[fading coyote]

Well, it requires that the tower actually provides head height. Which it sounds like it does

[white bloom]

left

[river night]

HOR is always a good recipe to have, so get that anyway

[white bloom]

right

[vast jungle]

only in combination with other recipes... but then its godlike

[river night]

gotta start getting recipes somewhere 😄

[hollow wave]

it does; if you pump it to (x) meters altitude then you wouldn't need a pump on any hill less than (x + 20) meters

[fading coyote]

Awesome

[shadow kite]

i'll be ignoring the existence of the standard AI Limiter recipe, am tired of it lol