math-and-meta

[plush gulch]

Yes. I don't want a load balancer cuz I always want belt 1 having full amount cuz that's my belt where I take everything from

[deep jacinth]

Do a manifold and preload the input of machines at the end of the line

[worthy karma]

i was exploring my world and found an entire factory that i never built

[deep jacinth]

it will take time but they will balance out

[glossy wagon]

Is it worth it to have a fluid buffer in your setup? sorry for the stupid questin

[unborn dome]

Why not just have a storage container to take stuff from?

[glossy wagon]

This is my example.. Left oil extractor is set to 240 right one is set to 60 and every machine is taking in 30 for 300

[deep jacinth]

In some instances yes. If the fluids on the output side back up, everything stops. A fluid buffer keeps things moving and will give you time to sort out excess

[worthy karma]

how did this happen?

[plush gulch]

Would be too much of a footprint

[unborn dome]

Standing there picking items off a belt isn't exactly efficient 😳

[deep jacinth]

I might not be understanding the problem, why are you picking items off the belt?

[plush gulch]

You understood me wrong.
With taking everything from I mean a splitter from this belt going to the machines

Multiple blueprints of different machines

[deep jacinth]

ok then yes I think a manifold is the best answer. Load balancers are better for a game like Factorio where inputs are limited; with unlimited inputs in Satisfactory a manifold will always eventually balance itself if the input matches the machine output

S is a splitter, M is a merer

[plush gulch]

I already have a manifold at my machines.
That's not the problem.

But I have a main line of screws going to both factories, consisting of 2 belts due to me producing more than 480 screws and only having MK4 belts. Now I want to merge those lanes into 1 lane, however the result is one belt is now overflowing back to the machines

thats roughly my setup
line 1 completely overflowed back to machines and the second set of machines didnt get enough ressources to produce everything
i just want to have belt 1 at max capacity and then belt 2 resupplying belt 1 when i run dries after 2 or 3 sets of machines taking from it

[deep jacinth]

You are describing an injected manifold, where the overloaded belt should have a splitter and merges into the undersupplied belt

If you labeled the belts with throughput and machines with output/consumption you could identify where the overflow is

a splitter divides a belt throughput by 2, so belt 1 will get half of whatever 2 is moving, and only the top row of machines will get items from belt 2 because the of direction of the belt 1

if you put the splitter on belt two before both rows on belt 1 each row of machines will get the parts being split from belt 2

[plush gulch]

Belt 1 is at max capacity though til machines row 1 though

But doesn't have enough ressources for machines row 2

[deep jacinth]

That is what the splitter would do

[plush gulch]

Guess I have to tinker around, find a solution by trying things out. Thanks for you trying to help.
Maybe I find a solution tomorrow when I slept enough and I am bit clearer in my head

[deep jacinth]

sure; ideally you should merge all inputs into one belt then do the split; that is a true manifold.

[plush gulch]

Yeah. But there is no belt for the amounts of items I need to transfer later

[lunar ingot]

a quick bandage fix like this could work for now,

but i think a long term fix would be "don't throw a ton of screws onto a belt, build screws right next to whatever needs them"

[plush gulch]

And machines themselves have manifolds.

[plush gulch]

Would only partially work. Cuz then I need to transport either iron or steel to these machines

And going above MK6 limits then late game

But I will go to sleep now. And hopefully find a solution tomorrow

[deep jacinth]

good luck friend, sleep well.

[prisma plover]

My plastic/rubber/turbofuel factory at the crater right now seems to have a problem with keeping the pipes full, seems that eventually one of the refineries goes idle with no oil. It looks something like this rn (yes, it's 120m height and it has 4 pumps decently spaced, I decided to make 4 levels to organize it better). I tried a simple manifold, with all oil coming from one of the sides, looping it at the end while keeping it level and right now a loop under the foundation.

Any tips? (Yes, I tried flushing the whole pipeline, redoing all connections with supports/floor holes/pumps/junctions)

[plush gulch]

Which is hopefully aesthetically pleasing as well and easy to implement

[deep jacinth]

pumps dont stack

[plush gulch]

Thx. Already 6:30 AM💀

[prisma plover]

which is why they're spaced, I put 3 to lift everything and 1 before the manifold

[deep jacinth]

fair; let me digest that diagram

Did you check each level of the pumps to see if header pressure fell off somewhere between them?

[prisma plover]

pipe is full, at 600m3 right before the last pump which is before the start of the manifold

[deep jacinth]

so the last pump is the problem section; is there a pipeline junction downstream and close to that pump? I've had issues with those junctions blackholing all flow before

ive had to delete the junction and rebuild it before to get the flow right

[prisma kraken]

it is a pretty common build. what i do for the 600 from extractor is split it into 2 mk1 pipes that feed 2 refineries each at 250%

[deep jacinth]

ooohhh split into 2 mk 1s would help the flow

pipes fill from low to high

[prisma kraken]

anything i can do to reduce the number of machines on a full pipe helps with the problem

[prisma plover]

seems like an option, will try to cut down to 4 at 250

[prisma kraken]

also avoiding full pipes helps too, but that isn't an option off of a pure extractor

after reconfiguring, make sure the crude pipe is completely full

a pipe is only completely full when you see its flow rate drop to zero

it takes a while for them to completely fill, be patient on that

i think there's some hidden overfull pipe state that the simulation uses where there's actually more in the pipe than is reported. oft times getting the pipe to that point helps with slosh

i suspect though that splitting the pipe into 2 mk1's will be the thing that solves the problem though. don't ask me why besides experience with solving such issues

[prisma plover]

would you split right after the extractor or pump it all the way then split at the manifold?

[prisma kraken]

i'd pump it to close to the destination then split and stick a powered pump on each side of the split

the pumps are probably overkill

[deep jacinth]

uhh yeah the pumps are only needed if you have more verticality to overcome

they dont increase flow

[prisma plover]

I see it more as a way to prevent backflow in that case, but im not sure if it might help or hurt

[prisma kraken]

when dealing with such things, it ends up being a lot of trial and error and goat sacrifice while standing on one foot naked under a blood moon 😄

[deep jacinth]

the point of splitting a 600/m pipe into a Mk 1 is to increase flow pressure and match input of the refinery

[prisma kraken]

i think more it just takes opportunity away from liquid to flow the wrong way

[deep jacinth]

agreed - I think that is the same I was trying to say

the way liquid works is kind of nebulous; thats just my way of thinking about it that tends to work for me

[prisma kraken]

the game's simulation of liquids tries to be as complicated as it can be to be challenging but also allow for fast simple computation

sometimes it just does some odd things

i think under the hood, what the simulation is probably doing is for each pipe segment, see if a neighboring pipe segment has less and transfer as much as the pipe's capacity supports. sometimes it pathologically just picks the wrong way

that's kind of my working theory after 1000's of hours with the game - take it or leave it

[deep jacinth]

It is so interesting to see each person's interpretation of the black box that is fluid dynamics in this game

[prisma kraken]

whatever it is doing, it has to be very quick to compute

[prisma plover]

i think it feels extra frustrating seeing it fail in such weird ways because im an engineer, so i've done this stuff IRL lol

[deep jacinth]

That is exactly why I thought each pipe has an input/output and flow decay rate and that rate is modified by height and distance of the neighboring pipes

[prisma kraken]

i think as well neighboring segments are prioritized bottom-to-top

[deep jacinth]

yes, bottom to top is definitely a confirmed mechanic

[prisma kraken]

i can build things that just will not ever work based on that observation

[deep jacinth]

big pipe squeezes flow into small pipe and pressue go up?

[prisma kraken]

there's no real concept of pressure in the simulation, that's part of the reason why it is so whacky

[prisma plover]

yea, thats exactly why it feels so weird

irl, pressure is king

[deep jacinth]

right! You should be able to extract from a huge pipeline and reduce diameter or flow to increase pressure and push fluid where you want it to go instead of worrying about backflow

[prisma plover]

so flow rate itself isnt exactly determined by the pipe (simplifying a lot here, but you get the point), so you get points of different pressure, but flow is constant

[prisma kraken]

yeah, it really doesn't work that way in the game

[prisma plover]

also, it seems to have stabilized right now

[deep jacinth]

and pumps shouldnt be constrained by vertical flow, a pump is a pump. it increases pressure to increase flow regardless of direction

but eh, its a quick simulation

[prisma plover]

did a bit of everything, removed the loop, simplified the manifold by removing redundant junctions (which seems to bug out sometimes, at the cost of looking ugly af) and changed pipes, kept mk2 up to the 2nd refinery and mk1 from there and at every connection to them

[deep jacinth]

Ive had an entire coal plant die because a single pipe junction decided not to work and just blackholed all fluid to the buildings

[deep jacinth]

deleted the junction, rebuilt it exactly as it was, and suddenly everything fine

oh wow

I do mine like thisw

[prisma plover]

yea, i dont like the feeding from below thing

[deep jacinth]

I used to put junctions right on top of splitters and feed water up top

[prisma kraken]

nor do i; floorholes have about a 1/50 chance of not connecting correctly and i just minimize the number of chances by bringing a pipe through the floor once

its a pain to look at every machine feed to find that stuff

[deep jacinth]

ive read that the floor holes have problems; I havent encountered it yet

[prisma kraken]

a lot of what people say of 'zomg floorholes are broken' is incorrect

sometimes they fail to connect the 2 sides. its rare but does happen

you just rebuild the pipe and it works. simple as that

[deep jacinth]

same fix for the junctions in my experience

hey you people are cool

[prisma kraken]

what i can't figure out is this... no matter how i rebuild those pipes, the left one's indicator is always different, lol

[deep jacinth]

oh that is annoying.... clean pipes?

[prisma kraken]

i could, but i want the flow indicator

[prisma plover]

and it's the best part of pipes, IMO

[deep jacinth]

hmmm.. dont the pumps act as a flow indicator

[prisma kraken]

yeah, but not from across the factory

[deep jacinth]

clean pipes below the foundation, switch to indicated pipes above?

[prisma kraken]

i could, or i could just not care

[deep jacinth]

hey you posted it

[prisma kraken]

i mean...

[deep jacinth]

You are a monster.

I cant with whatever you got going on there; but dang that one pipe indicator above the others is super noticeable

[prisma kraken]

yeah, it'll be solved with a wall at some point,lol

next floor up is fine

[deep jacinth]

always interesting to see people's builds; Im a fan of "tiny things" so I like to expose the belts with their parts and see the little boxes of screws and bars and rotors and such reach their buildings

but hey the pipe indicators match up on that floor!

[prisma plover]

im a bit of both, right angles and keep overlaps to a minimum

[prisma kraken]

yeah, that one indicator is the weirdest thing

[deep jacinth]

I really wish vanilla would display the distance and height of belts/buildings relative to a known 0

it would be so much easier to make right angles and verticality work with the world grid and not have to delete and rebuild multiple times

[prisma plover]

i just hope they can do the straight pipes soon, would make so much stuff look better while being easier to build

[normal merlin]

i have a question about oil to fuel.... if I have 1 Pure Oil, how many refineries could that supply without loss of production?

[prisma plover]

1 pure oil with overclock is 10 refineries (considering the oil -> fuel and polymer recipe)

[normal merlin]

okay, then i think my pipe setup is a little messed. but got 2 prue, 3 normal, 36 refineries... still good?

[deep jacinth]

wait what

[prisma plover]

at the crater with the big lake, right?

[normal merlin]

up north area

[prisma plover]

ok

[normal merlin]

there is more oil if i need it but its lower than normal nodes

[prisma plover]

1 pure = 10, 1 normal = 5, 20+15 = 35 (quick mathing here)

that is considering he has the alternate recipe for HOR

[normal merlin]

im working on unlock a lot of recipies... waiting 10min each takes a lot of time lol

but thank you, this was helpful

[prisma kraken]

yeah, drop a mam whenever you gather an new drive and start the research while you're still hunting

[prisma plover]

you can dismantle it too, it keeps track server-wide

[normal merlin]

i know, i watch a video and someone did that, i was like thats brilliant on the go

[deep jacinth]

I must be silly or something

1 pure oil is like 4 fuel refineries

[prisma plover]

4 with 250% oc, 10 without

[deep jacinth]

oh with OC

[normal merlin]

37500 MW

just once it was fully operational i started loosing the center section(first built)

[deep jacinth]

240 oil > 4 refinery > 160 fuel / 120 polymer resin?

[prisma plover]

im used to OCing mostly extractors, since it multiplies power consumption over 2,5x and there was a limited number of shards before slooping and t9, so I guess you can oc everything

no, 400 fuel, 300 resyn

oh, 240 oil

yea, that maths out

[deep jacinth]

1 pure oil node is 240 without OC

[prisma kraken]

anything oil tends to be an easier build if you use 150/300/600 as input quanta

[normal merlin]

glad i popped into this section... specially when its getting late and brain dosent want to brain anymore haha

[deep jacinth]

google satisfactory-tools

it does the math for you friendo

[prisma kraken]

except for blended tf, that kind of wants 450 or 900

[normal merlin]

been using both calc and tools depending, but sometimes from real players helps more

[deep jacinth]

for sure; things like @prisma kraken said about input quanta is real

wont get that from the tool

[prisma kraken]

lotta stuff on youtube as well, some good, some not so good

[prisma plover]

yea, go for easier math, it compounds on itself later

[deep jacinth]

but it does help ease the mental load on input/output maths

[prisma kraken]

building in this game is tricky enough without having to deal with fractions of prime numbers and repeating decimals 🙂

[deep jacinth]

I'd rather underclock and run away to forget about it or use a buffer or pick a thing rather than deal with that minutae

[prisma kraken]

figuring out how to sandbox nasty numbers that are multiples of 7, 11 or 13 or use them exactly is kind of something i really try to do

good example: steel screw. awful numbers, but direct belting it into copper rotor makes for nice numbers

[prisma plover]

mine, at the crater, about 30GW, 400 rubber, 675 plastic, 337.5 fabric

[deep jacinth]

I prefer to focus alt recipes on eliminating production runs entirely to divert those resources into other factories like heavy encased frame

[prisma kraken]

yeah, there's no getting away from using most parts if you're building past a certain size

[deep jacinth]

@prisma plover My brain says "make concrete supports for those belts to make gravity happy"

[normal merlin]

so far mines just for power, prepping for space elevator items factory...

[pure crow]

You should make what I call the N-Hub then... merge N inputs and split to N outputs by making O(N^2) Industrial Storages and chaining them together. (cough Ridiculously large in most cases) -- but yeah, in most cases, parallel belts'll do the trick.

[prisma kraken]

pfft, play more minecraft 😉

[pure crow]

My designs were prepared for everything to be like screws and then... most things turned out to not be like screws... kind of overkill.

[prisma plover]

nah, F support, the moment you start designing bridges, the thing you wonder the most is "why is there such a thing as gravity?"

[pure crow]

Nevermind bridges, I just tried to make roads and there's like 1 billion meter dropoffs at certain points in the map

[prisma plover]

i meant as IRL bridges

[pure crow]

Wait, you design IRL bridges?

[deep jacinth]

The idea of floating foundations stretching out across the landscape for miles and miles with no respect for the landscape or the laws of physics gives me conniptions

[prisma plover]

yea, civil engineer

[normal merlin]

thats pretty cool

[deep jacinth]

Im gunna give you more plates and concrete to keep going, but im also going to come behind you a build supports.

[prisma plover]

the moron here works with spreadsheets all day, then comes home to play with spreadsheets

[normal merlin]

dont mess with them spreadsheeters...

[pure crow]

I play Satisfactory without spreadsheets, I only do the math in the blueprints mostly unless it's something critical.

[deep jacinth]

All of the math can be done with the codex, really.

[prisma plover]

i started planning the aluminum stuff this week, i did like 5 pages of math, so I knew how many of each mat i had and all the stuff i needed

it ended up on like 700 sheets and casings each

[deep jacinth]

sloppy alumina is your friend

nice round numbers there

[prisma plover]

sure, but then i would need more water, which means more pipes, so.... yea, i prefer silica lol

and extractors.... man I hate their encroaching area, so damn big

[deep jacinth]

sloppy is less water

12 x 18 vs 10 x 10

unless you need silica

[prisma plover]

overall, maybe, but byproduct i think i can handle with wet concrete, the extractors that are most of the pain

[deep jacinth]

thats fair

I usually go straight for a quartz input instead of pulling silica as a byproduct but I get your take

[indigo wren]

guys, is it possible to re-cycle water that comes out of refineries from creating scrap aluminum? as in, routing that water back to where i'm feeding the refienries making alumina solution and limiting the previous input to match with the re-cycled water

[prisma plover]

oh, sure, producing silica is kinda annoying because of the numbers, but it evened out well for the math i had

[prisma plover]

it's preferred to dedicate some of the refineries (so that the fluid doesnt bug in the pipes) for that water or create wet concrete (alt recipe for concrete)

dont try to mix input from extractors with that, it might end up being a pain to solve

[loud trellis]

What makes underclocking so inaccurate?

Just tried to split 300 limestone across 7 machines, the closest I can get reasonably is 299.999 and it kinda irks me

[deep jacinth]

The water from scrap aluminum is equal to a single water extractor

if you have a use in your build for a single extractor worth's of water, then yeah for sure its possible to recycle it

[indigo wren]

we need water sinks

[deep jacinth]

but its dependent on your build

[loud trellis]

✨ There's a mod for that! ✨

[prisma plover]

if youre using constructors to create concrete, try 3 machines, set 2 of them to 33,33 and 1 to 33,34

or 10, at exactly 10 per min, only underclocking

[deep jacinth]

im not sure how that relates... but im pretty drunk. can you explain?

[loud trellis]

44.999

lol

UE really hates decimals

[prisma plover]

i think i replied to the wrong one, instead to the one right before you

[deep jacinth]

oh you meant to reply to the underclocking math

[prisma plover]

yea

[deep jacinth]

yeah UE.... is UE.

that decimal is offset by an autosave

[unborn ermine]

So! I have a fun little discovery to share.
I am planning to use the rocky desert's quartz and 600/min nitrogen to do purification, all up on the cliff by the caterium + copper nodes, using that pond for water.
(you can fit 5 extractors in that pond under the uranium node, and im using 2 already)

its so CLEAN

[wind spade]

No, that would make things too easy

[deep jacinth]

Did you know that packaged nitrogen has a x4 compression ratio?

just sayin

[unborn ermine]

Yeah im delivering it there via drone

[deep jacinth]

hell yeah you rock

[unborn ermine]

that one well near there, splitting it 3 ways

Train would be a nightmare.

Pipes a mess

[deep jacinth]

I saw someone on reddit today that was doing nitrogen by train

I felt so bad for them

[wind spade]

Eh, it's fine

[deep jacinth]

I'll pray for you

[fringe seal]

wait, what the fuck

[wind spade]

What's wrong with moving it via train?

[deep jacinth]

it would be x4 less efficient than packaging it

[unborn ermine]

Something something throughput and gasses being finnicky

[deep jacinth]

and fluid trains are... difficult

[fringe seal]

@prisma plover no, the thing is that it works but it should not

[wind spade]

Efficient how? Packaging uses extra power

[unborn ermine]

with trains you can have wonky throughput, gasses also being able to be compressed, make something like (for me) 337.5/min trivial for a train.

[prisma plover]

I mean, it's clipping, and that pipe isnt exactly lined up, so sometimes it just doesnt want to connect right/fail, but if it connected, i think it should work (as long as there's head lift, i think)

[fringe seal]

that is the thing
the headlift is insufficient

but the pipe is entirely filled

[deep jacinth]

power is rarely ever a bottleneck. two trains being equal, one moves 4 times more product than its fluid counterpart

[wind spade]

Wonky throughput means you built it wrong

[wind spade]

But if I don't need that much throughput?

[deep jacinth]

then efficieny isnt a factor and it doesnt matter

[wind spade]

Then it's better to not package

[deep jacinth]

the cool thing about Satisfactory is that input is unlimited from the source so if efficiency isnt the goal you dont lose anything material.

[unborn ermine]

Well think of it like your packaged fuel greeny, you can unpack at the machines and save piping.

[fringe seal]

the headlift of the bottom pump is maxed out, and there's still ~5m to climb
but the pipe beyond the blue ring is filled

[deep jacinth]

Its objectively better to package; but subjectively it doesnt matter.

[prisma plover]

yea, i see it, wtf

[wind spade]

The thing is that "efficiency" as a word has no meaning unless you define what kind of efficiency are you talking about

[fringe seal]

now I can actually claim it is broken

[wind spade]

Why save piping at the cost of power?

[wind spade]

It's not.

[deep jacinth]

I define efficiency as obtaining maximum output from the minimum input in the shortest time possible. With that definition, packaging is the most efficient. If it doesnt meet your need or you don't subscribe to that definition, thats ok. you do you

[fringe seal]

it's better for the throughput to package lol
esp. bc the gases have problems in buffers as I heard?

[wind spade]

I hate generalisations people make based on their subjective preferences

[unborn ermine]

This metric is getting all of the material from point a->b without dealing with any potential throughput loss, and only using one means of transport.

[fringe seal]

yours is not objective either

[unborn ermine]

@wind spade Dosent mean there WILL be loss its just cleaner and smaller numbers to deal with

[wind spade]

I'm not saying mine is objective. I'm also not saying it's better to not package.

I'm saying both are fine and there's no reason to package if you don't need the 4x throughput

[deep jacinth]

You are in the "math and meta" channel; objective is kind of the goal here.

if you dont "need" a throughout, that is subjective.

[unborn ermine]

Need a "philosophy" channel

[wind spade]

If you need it, that's subjective too

[deep jacinth]

mathematically, a 4x compression ratio is one of the best ratios in the game. if you dont need the extra, you can sink it. It is still the optimal answer for storage on trains.

[wind spade]

Objective would be "assuming you want to minimise train/car amount and don't care about power usage and unpackaged gas doesn't fit into one car, packaging is better"

[deep jacinth]

An objective solution does not start with an assumption and then go on to disregard other inputs. that is not objective in any sense,.

its fine that you choose to do that; there is nothing wrong with your choice

but it doesnt make it efficient or objective.

[wind spade]

You just don't understand what "objective" means

[unborn ermine]

tbh if you are debating packing or not for a remote nitrogen node, you probably arent close to hurting for power (yet)

power for late game is always a moot point

[wind spade]

Sure, but again that needs to be included before calling it "better"

[unborn ermine]

and it always seems like a strawman's argument

[deep jacinth]

We could even remove "objective" and go to "practical" and your solution would still fall short.

[prisma plover]

tbh, power is mostly for early game, when youre starting coal, and when youre building your first oil plant

after that, just go ham and get like 400GW in a single power plant

[unborn ermine]

Yeah like im already 200k MW and I went minor ham

[wind spade]

"Objective" means "not affected by opinions", not "doesn't include restrictions"

[wind spade]

*GW

[deep jacinth]

200K GW?

wat

[prisma plover]

MW, should be 200gw

[deep jacinth]

Unaffected by opinion: Nitrogen Gas has the highest compression ratio of all liquids and gases, with the 4× compression making packaging the more optimal choice for storage and transport on a Freight Car.

[wind spade]

kMW -> GW

[unborn ermine]

Greeny you have to remember not everyone debates the meaning of words because its the internet, people speak different languages and have different levels of understanding

Also you dont have to be so snippy

[prisma plover]

dude is being pedantic for the sake of being pedantic. The game literally displays in MW, doesnt have a GW, so yeah, if people want to use MW, more power to them (see what I did there?)

[deep jacinth]

That isn't an opinion; it is widely accepted fact. Its meta.

[wind spade]

*unless the throughput is low enough for one car or power is a concern

[wind spade]

Point is that SI system works how it works so lets use it

[unborn ermine]

#math-and-meta message not reading eh

[wind spade]

?

[unborn ermine]

Everyone does there own thing and has their own way of speaking, pointing out things like that in a discussion is just crass and poor form.

[wind spade]

k means kilo (10^3)
M means mega (10^6)
If you want 10^9, you use G (giga), not kM (kilomega??)

[unborn ermine]

[wind spade]

You don't say a file is 30 kilomegabytes, you say it's 30 gigabytes

[unborn ermine]

https://tenor.com/view/missthepoint-the-point-gif-7984004

[prisma plover]

you should send coffee stain a message, I think they dont know about that

[wind spade]

The point is that SI is literally designed so that people across the world can use it

[unborn ermine]

Yeah lets throw everything under the bus because greeny wants to be pedantic

[wind spade]

That's fine tho?

My point is "kMW" is not a thing, not that we need to use GW

[unborn ermine]

One of my pet peeves is people doing this shit, and never listening, just going on and on because they hijacked the flow.

[prisma plover]

no no, that's a kilomegawatt, coffee stain are doing it WRONG, they dont know what theyre doing ||/s||

[deep jacinth]

yeah we are feeding the troll at this point.

[wind spade]

Eks dee

[prisma plover]

i mean, it's not that im mad, im laughing here, gotta wait for the buffers to fill anyway, already watched my series backlog lol

[deep jacinth]

lol im manually crafting my way through phase 2

but its not a productive conversation if the other party wont at least consider a counterpoint

[wind spade]

30000 MW ✅
30 GW ✅
30 kMW

[unborn ermine]

Like I said, we need a philosophy channel for greeny to go be pedantic in his own way

[deep jacinth]

well what about MWh?

[leaden cosmos]

2TW is my own goal,
2000000MW feels like inflating the number to something humans can understand 😄

[wind spade]

I literally did but ok

[wind spade]

Unit of energy, totally fine

[deep jacinth]

unit of energy over time*

[leaden cosmos]

that is fine, as its MW * Hours
e.g. 1MWH is 1MW for 1 hour

[wind spade]

No, energy

[deep jacinth]

I dont recall, but how is it expressed in the batteries?

[wind spade]

Power is energy over time
Energy is not over time

[deep jacinth]

I think its MwH

[wind spade]

Correctly, yeah

[deep jacinth]

yeah, that is energy over time?

1mw hour

im sorry I see what your saying; there is a difference between energy vs power

im a cyber engineer; energy is the same a power to me. is there a nuance there I dont understand?

[wind spade]

MW is power
MWh is energy

[deep jacinth]

oh yeah that makes total sense

[leaden cosmos]

its like a 3000mAh powerbank
the runtime is 3000 / current
so 300mA is 10 Hours

(Similar enough, but mWh and mAh are just voltage apart) 😄

[deep jacinth]

amount of power through a conduit vs energy expenditure over time. cool im there

similar expect that amps are what kill people

lol

[oblique hollow]

Its not energy over time because there is no fraction here, only multiplication

Its more like "Power times time"

Divide power x time by time again and you get power back

Speed would be distance over time as that is m/s or km/h

[timid zenith]

is it good to have 6 hours worth of solid biofuel

no wait I have 6 biomass burners uhh

4,4 hours of solid biofuel for 6 biomass burners

[oblique hollow]

It doesnt matter much just get to coal power ASAP

[timid zenith]

ok

[oblique hollow]

You can stockpile as much as you need to get there

[timid zenith]

I remember seeing some good coal place near the sea

to the northwest

[pulsar notch]

yeah, that area has a bunch of coal nodes.

it also has small demons, and a cave with more demons, plus quartz and SAM, but it's got coal, which is nice.

[mint lichen]

best radio control unit recipe?

[amber umbra]

Per earlier topic, 30 kMW is displeasing.

@mint lichen They’re all reasonable recipes with tradeoffs. I used default.

RCUs next to crystal oscillators is kinda nice imo if you have alts like crystal computer.

[spare jolt]

the only problem is that the quartz is in a cave

[amber umbra]

Definitely a situation for long distance transport of quartz.

[vast jungle]

What is this with kMW... Has the giga prefix been discontinued for SF? 😉

[vast jungle]

You could put a drone port into the cave... (I just belted it out towards the coal)

[spare jolt]

But clipping tho :(

[vast jungle]

no clipping for the drone port 😉

[wind spade]

Lack of basic education 😦

[amber umbra]

Even like 100,000 MW is fine.

Guess the thread is called math-and-meta so is on topic.

[broken kiln]

Is crude to HOR alt to Turbo Blend still the most efficient route? Trying to make myself a big ol' turbo factory that I can then augment/upgrade to rocket when I get nitrogen

[broken kiln]

So speaketh

[spare jolt]

Wouldn't the drone clip through the terrain to find its way?

[vast jungle]

yes... but no clipping on the building necessary 😉

[spare jolt]

Well yea, I'm talking about the drone, not the building

[vast jungle]

stupid SCIM 😦

[wind zinc]

Lucky power core 7 more like UNlucky power core 7

[wispy jay]

I recently passed screenshots of a Satisfactory Tools Visualization to ChatGPT to help me organize a plan. It worked quite well except for when the visualization was too complex it made a lot of errors. I hoped to be able to get more to a text-based version of the graph to use instead but couldn’t find one. Anyone have any tips or ideas for this?

[rancid fossil]

if I have this problem with fortnite what can i do. I reinstall all the drivers and download the new updates and i have the same problems

[spare jolt]

First tip - do not ever use chatgpt or any large language model for math-related calculations or planning. It cannot calculate by itself, it just hallucinates based on data it feeds on.

[wispy jay]

What I ended up doing was turning the plan into a json representation, then passed it to another prompt where it generated and executed a python algorithm to follow a plan I laid out. Then it could print the results from code. Worked really great for my needs except for when I noticed errors in the original image-to-json parse.

[wind zinc]

it gave me that too and i didnt see any problems no stuttering no artifacts(the game ran perfectly fine) so i dont know why fortnite gives that error

[wispy jay]

But yeah my actual problem to solve is how to get a text-based plan description that’s accurate from Satisfactory Tools

[spare jolt]

What do you mean by "text-based plan"? Like, turn the graph into the text, or to plan where to start the building from?

[wispy jay]

Turn the graph into text

[bitter grail]

What kind of text? Sentences? Bullet points? More of a graph/chart with words/numbers? What exactly do you want?

[bitter grail]

If you want to look at different calculators cause you don’t like the format of satisfactory tools, on the wiki there is a page with a list of various tools (it also states if the tool is compatible with 1.0 which many seem to be)

https://satisfactory.wiki.gg/wiki/Online_tools

[flint rapids]

When placing Water Extractors for Nuclear in one of the lakes and not in the ocean do yous overclock them straight away?

[wispy jay]

Perhaps a different calculator is the answer. Honestly any text form is sufficient. It doesn’t matter because the LLM can convert it to what I need as long as I have something to start with. I might just try to start over and rebuild my plan in Satisfactory Calculator

[bitter grail]

In the link I sent there are a good amount of different options to take a peek at. I use one of the different ones cause I don’t use SCIM or Satisfactory tools personally because the layout doesn’t work well for how I think

I don’t think using ChatGPT will work in the end cause of how complicated Satisfactory planning can get, but then again I am not familiar with such things

[wispy jay]

Like I mentioned above, I got it to generate me an executable algorithm to make my plan work so I don’t have to worry about that. It’s just that my input data was inaccurate because I was using the LLM for parsing the images which is really tough.

So now that I’m putting my plan into action and trying to manually adjust I’m saying “yikes if I have to make too many manual adjustments I’m going to run out of wiggle room and my plan will break, I need to rerun this with more accurate recipe data”

[wind spade]

what more you need than number of machines? 🤔

[wispy jay]

@wind spade I think Nilaus’s efficiency bus system is really neat bus also super ugly. I also really like some vertical modular stackable designs I’ve been seeing. So my system is to have 4 buses in concentric squares. Outer square is external inputs like ingots and plastic. Center is finally products and sinks+elevator. Between these squares are squares of buildings with these modular stacks. Buildings can do inputs and outputs only to their adjacent buses. Figuring out which buildings go where is actually really hard to figure out to make this work so I needed an algorithm to generate an arrangement that works. I’m really excited about the results and I’m loving how it looks so far.

It looks like a big city

[wind spade]

I heavily recommend not using a bus

[wispy jay]

Yeah I’m fully aware of the downsides, so this design makes it much easier to have multiple belts of the same product and still have room. Also since the buses are in big squares, it’s also much easier to have extra buildings that can refill the buses at various points in the bus to keep it full enough for my plan.

[wind spade]

my recommendation is https://satisfactory.wiki.gg/wiki/Tutorial:Independency

[wispy jay]

And pretty much once you get past the first bus, the parts/min of the next 3 buses are all low enough to fit a single belt.

[wispy jay]

Yep this is totally valid and much more how I approached my last playthrough. Just not what I want to do here 😊

[wind spade]

that's fair, but expect the game throw stick under your feet 😛

[ember fractal]

Don't play fortnite, problem solved

[prisma kraken]

more to the point, why is he asking here?

[ancient halo]

I think my brain broke, 15 nuclear powerplants make 150 and not 750 Nuclear waste /minute right?

[prisma kraken]

iirc, it is 10/min for ufr's

and 5/min for pfr's

10/min ufr; 1/min pfr, sorry

[amber umbra]

On the bus mentions above, a bus is just an organization system/tool. The term doesn’t indicate scale. A huge, singular bus as your only factory has significant issues in Satisfactory. Smaller busses to organize a factory is a good idea and what many people already do without calling it by that name.

[velvet venture]

is the ratio of the product items okay? gonna do a planned 2nd playthrough, trying to plan a good iron factory to take in only iron and power and will have depots at the output.

Wanted to use a max of 480 ore, but either I have to skip out the Motors or Have extremely low and unusable quantity of other items (like 15 plates/min)

[fringe pawn]

There's no ongoing need for extra screws or miners in there. Once you fill a container of each, you could repurpose the underlying ingredients if you want.

[velvet venture]

wanna do a build and forget type of factory

[fringe pawn]

Keeping it alive won't hurt anything, but you're not going to use more than 1200 miners.

[sand epoch]

Why 1200?

[fringe pawn]

That's one container.

[amber umbra]

Keep in mind, 15 plates/minute for 20 hours is 18,000 plates. Really don’t need much if it’s for personal building use.

[sand epoch]

Since when do miners stack..o0

[spare jolt]

also you can't set up more than 1 miner per node, so eventually after filling everything with mk3 miners you don't need them anymore

[spare jolt]

1.0

[sand epoch]

Oof

[amber umbra]

Drones do use a portable miner afaik.

[spare jolt]

they do, but are you going to use only drones as your means of resource transportation?

[fringe pawn]

If you do an industrial container that'd be 48 stacks and 2400 miners. I'm not about to add up the amount of nodes on the planet, but it wouldn't surprise me if it's impossible to need that many barring an insane amount of drones.

Same with screws. Unless you have some bonkers plan to decorate your base with AWESOME shops.

[amber umbra]

Idk drones are just iron ore I thought. I added it to my mall items since it’s trivial to do and saves manually making miners for miners, drones.

Any non-zero item/s is good.

[velvet venture]

ik but in my first playthrough I think I had like 10/min and I was constantly running out while making very long belts

[amber umbra]

That’s due to t1 belts being plates. The demand drops greatly.

[spare jolt]

you can always run around and gather stuff like HDDs and mercer spheres while they're getting stockpiled

[velvet venture]

I wish there was a guide of some sorts which gave you a rough idea of the ideal /min resource production for an average playthrough

Then again, since playthroughs can vary extremely player to player, maybe an improtance wise tierlist of every single resource, listed by how much of it u need throughout the game. For example, the list starts with concrete at #1 and so on...

[amber umbra]

I play kinda slow, but I have 1 constructor making concrete that has built all the concrete for my playthrough. Mall items just need to be automated to any extent is the rule.

[spare jolt]

for me as a solo player, i use 1 constructor or about 5 items/min, whichever is less

e.g. 1 contstructor produces 20 plates, but 2 manufacturers produce 5 computers

then it gets stockpiled AND uploaded to dimensional depot, so in case i'll need a big amount of one thing like concrete, it'll quickly replenish the depot

[amber umbra]

The big distinction is items for construction (mall items) and items for elevator and/or higher tier items. Mall items can be very low item/s. The other category can be very high especially for the intermediate items.

[viral patio]

how much would i need to overclock my coal gen to allow it to use 33.75 coal/min

and in return how much water would each gen need

[deft lichen]

if the default is 15/min, then (33.75 / 15) * 100%

water usage will be 45 * clock speed

[copper seal]

is this a good spot for aluminion production?

might have to use some pumps for the water to get it to build height but there's bauxite, coal, quartz and water all in relativley close proximity

2 pure bauxite and 1 normal

could either use the water well or the rivers maybe

or maybe the coast 🤔

[stiff rune]

whats the perfect place to build a turbomotor factory

[deft lichen]

anywhere that has most of the required ingredients nearby

[true pier]

what is this? I only have a water pump needing 50 MW and 2 coal generator on this line and it says cons. 150MW when max cons. is 50MW

[unborn ermine]

the oil nearby too if you do electro scrap

[spare jolt]

you may want to bring the bauxite down left to mix it with coke for an electrode scrap

[unborn ermine]

I brought the coke up

[spare jolt]

me hard read new messages

[copper seal]

idk what that means

but it seems like the perfect place for baux

[unborn ermine]

[copper seal]

bro its my first time in phase 4

my 5th world but lets not talk about that.

[spare jolt]

electrode aluminum scrap alt recipe

[copper seal]

oh

whats the difference?

[spare jolt]

gives 1:1 of bauxite:ingot if combined with sloppy alumina alt recipe and pure aluminum ingot alt recipe

[unborn ermine]

sloppy and scrap are really nice yeah

[copper seal]

so i need to get both sloppy and electro alt recipes first?

[spare jolt]

takes petrocoke instead of coal

[spare jolt]

sloppy alumina, electrode scrap and pure aluminum ingot, yea. they'll eliminate the silica from bauxite chain

[unborn ermine]

and if you DO end up doing the silica you get more for regular recipe(but its nutty to do)

[copper seal]

now i need to find out why thats significant

[spare jolt]

i just told ya

removes the silica from bauxite -> one less ingredient to deal with

[unborn ermine]

also makes for a really clean setup

[copper seal]

hmm

[spare jolt]

it's not the most efficient combo, but the simplest one indeed

[copper seal]

should i use the pressurized oil well on the west coast for that?

i'll make the petroleum coke there and then... train or belt it to the bauxite setup?

[toxic hemlock]

How is electrode simpler than regular?

[unborn ermine]

Ratios and no silica

[toxic hemlock]

Silica isn't involved in regular either

The silica comes in when you're making ingots

[unborn ermine]

[toxic hemlock]

You can do pure ingots with regular just as easily as with electrode

[unborn ermine]

sloppy + electro

[toxic hemlock]

That's more complex than sloppy + regular because you need to make coke

[copper seal]

isn't there a better spot then?

because the only reason i thought that might be a good spot was because it had quartz

[toxic hemlock]

It's not a huge hurdle but it's additional complexity

[unborn ermine]

theres nitrogen nearby too, dangle spire area

[copper seal]

i dont know why that matters

[spare jolt]

electrode by itself isn't simpler at all, it just uses less alumina solution

[toxic hemlock]

Yeah

[spare jolt]

0.6667 for default vs 0.6 for electrode

[toxic hemlock]

That's accurate; saying it's simpler isn't

I wasn't sure if I was missing something

[unborn ermine]

well you are planing aluminum production

[copper seal]

i've never touched nitrogen

don't even know what its used for

[toxic hemlock]

You will

[copper seal]

and its not in the aluminium production tree either, at least until aluminium ingots

[spare jolt]

though sloppy + pure definitely simplify the overall setup, that's what I was referring to, not to electrode in particular

[copper seal]

though with the alt recipes i can also get plastic from refineries while making heavy oil residue

[toxic hemlock]

Nitrogen is the last resource you can gain access to

[copper seal]

and i could use that plastic for alclad sheets

[toxic hemlock]

It has byproducts but they're all made from other things you find in the world

[spare jolt]

alclad sheets take copper, not plastic

[toxic hemlock]

Actually uranium might be one milestone after it

[copper seal]

wait which ones take plastic then

wasn't there a aluminium product that uses plastic?

[unborn ermine]

its the combo of things, one 300/min oil node = exactly what I needed to do mine, and having a nice clean setup were I built it that was done frighteningly quick.

[toxic hemlock]

No, that's an iron plate alt recipe

[copper seal]

ah

[spare jolt]

classic battery alt and hazmat suit. that's it

[toxic hemlock]

Classic battery is nice

[spare jolt]

both use aluminum product and plastic

[copper seal]

i mean batteries would be nice so i could start using drones

[unborn ermine]

forgot to mention I did 2940/min aluminum ingots

[copper seal]

but i'd want them to be produced somewhere more convenient

[copper seal]

that spot that i was eyeing allows 2160 for me right now

more if i get mk6 belts

[unborn ermine]

I pulled that extra pure node near the middle

[toxic hemlock]

Drones take any fuel now

[copper seal]

ye but i don't have a automated packaged fuel setup

[spare jolt]

since 1.0 drones can be fueled by packaged fuel(s) and radioactive rods as well (except fisconium), not just batteries

[spare jolt]

not any. they don't take e.g. coal

[copper seal]

i have a hand fed packager for my jetpack and thats about it

[toxic hemlock]

Coal isn't a fuel

I guess cars can use it

[spare jolt]

can be used in a tractor -> fuel 🤓

that's why i said "packaged fuel(s)"

@copper seal either way, my additional advice - get yourself a hoverpack ASAP

[copper seal]

i have a hoverpack

[spare jolt]

good

[copper seal]

i cheesed a bunch of the phase 4 milestones

[spare jolt]

such a hard drug can't be allowed to just pass by 👃

[copper seal]

also got mk5 belts too

but

okay, i've been thinking about aluminium the most but

[spare jolt]

mk5 belts can be unlocked later\

hoverpack can't

[copper seal]

noooo

belts must go

miners too

always my priority to get highest tier belts and miners asap lol

[unborn ermine]

Belts are a MUST for scrap

let me keep my building/belt count nice for my build

[copper seal]

its always what i do first ;p just gotta

[unborn ermine]

I will say, I did a hand feeding setup at my base before building aluminum using the collected random bauxite ore, let me have so many belts before I was done

[copper seal]

but anyways, i asked before but i didn't get an answer, is that still a good location to build bauxite if im gonna try to get all the alt recipes or is there a better one then

the only reason i picked it was because of the quartz node

but if im not gonna be needing quartz

[flint rapids]

When it comes to Gases - Valves do the same thing as Pumps without using power, is that right?
Also do they geenerally help to fix flowrate problems on gas lines?

[unborn ermine]

Its nice, but you might want to build lower than the cliff if you do.
Quartz is used for some of the later stuff like the Radio Controlled Units aka crystal oscillators/crystals

[copper seal]

why lower than the cliff?

not messing with pumps or some other reason?

[unborn ermine]

its a pain to get stuff UP the cliff

[copper seal]

mweh im totally fine with building tendrils of foundations

and i'd like to make it easy to get trains to it

but I guess i could just put the train station ontop

but that just means i'm sending the ingots up later xd

[spare jolt]

if you're gonna bringing liquids from below - i'd still advice to build down the cliff

[unborn ermine]

I have the logistical challenge of getting 1560/min copper up to that large arch, and some other things to send over.

[copper seal]

no i was thinking about building it in this

[unborn ermine]

Good spot, nitrogen is RIGHT next to it as well.

[copper seal]

roughly same height as the surrounding mountains though

[unborn ermine]

oh and if you do have a train, you can have one going into the red forest from that spot too.

[copper seal]

I could also grab the SAM there and just chuck it on a train to be processed somewhere else

later, i don't really need SAM anymore

[unborn ermine]

For the moment*

[copper seal]

ye

[unborn ermine]

[copper seal]

wait where does the copper come in?

[unborn ermine]

Alclad + lots of alts

[copper seal]

oh ye alclad

[unborn ermine]

like making alclad casings/heatsinks

[copper seal]

which i need for more mk5 belts

[unborn ermine]

and im planning the heatsink alt that wants rubber

[copper seal]

ooooh

[solemn gorge]

if I have 660 aluminum ingots/min how much of everything should i create with that?

[copper seal]

copper situation there isn't particularly great though

about 900 with 3 fully overlocked mk3 miners

[unborn ermine]

I cant help with that, still figuring that out myself

with 2940 ingots

[spare jolt]

"guys i turned the entire map into ingots what should i do with them"

[solemn gorge]

okay, im just not sure how much i'll need later

[solemn gorge]

uhh, will you ever need that much? I mean you probably will but in that case for what?

[unborn ermine]

.

lots of stuff needs it, just sorting it out

[solemn gorge]

okay

[unborn ermine]

example

[copper seal]

wait so there are only, really, 2 viable aluminion locations? lol

the one i already found and the one in the forest??

[unborn ermine]

Yeah the two main spots, or you just transport it somewhere else entirely

[spare jolt]

any location is viable if you use automated transport

[copper seal]

i mean ye ofcourse xD

i mean i could just bring everything to my base

😂

[flint rapids]

Is there a better way to find throughput of my Fluid Train than manually timing it with stopwatch and dividing volume of cars by time?
Also, is it me or throughput of fluid cars kinda sucks even on medium distance? I need two cistern-wagons to max out one mk2 pipe.

[copper seal]

doesn't it say so on the platform?

at least with regular cargo container carriages it will just tell you the items per minute

[copper seal]

yes, they're bad

you're better off packaging it and then transporting it with regular cargo carriage

like, MUCH better off

[flint rapids]

Yes I foudn the transfer rate, thank you. Was looking for it in wrong places

[copper seal]

ye thought so ;p would be weird if they did it for solid cargo but not for liquids

but yeah, if you care about efficiency its better to just package it instead

that being said, i do like the aesthetic of having more than just box trains driving around

[flint rapids]

No I think I'll stick with cisterns for a while, at least until I ever need more than 4 of them to get throughput I want. I want to have variety on trains moving around

[unborn ermine]

yeah you either need reaaaally long trains or packing it to have higher throughput.

I heard 4 is doable

(dunno about hard slopes though)

[copper seal]

you can just add more locomotives though?

i do that anyways regardless of train length, i put one on the front and one on the back

but you can make a train more powerful by just adding more locomotives

and you can always just use more trains @flint rapids

as long as you make sure they don't crash into each other and make it a 2 way line you would be able to just run multiple trains for the same route to get a higher throughput

aside from just adding more carts and then adding more locomotives when the carts get too heavy

[unborn ermine]

I remember someone mentioning a falloff for adding trains, but when you get to THAT many trains its better to package.
(dont ask me for details I remember nothing )

[copper seal]

ye i can imagine there is some, i don't know the limit either

but its nice knowing that even though liquid trains are shit you can still do them

they really should just make liquid trains better than packaging it, it makes no sense that you're able to fit that much more fuelm3 in a roughly same size container even with all the added air and packaging

[quiet wraith]

speaking of trains i just unlocked them for the first time im keen but not keen to play with them

[copper seal]

they're fun!

and they're also a conductor of energy so you don't need to create a massive line of power poles

[quiet wraith]

yeah i realised that last night makes running power easier

[copper seal]

well yea only if you place down a station ;p

there are also some really good train blueprints available, i'm not so anal that i would decorate every single thing but some people have so you can get it to look good

[quiet wraith]

yeah i found a couple that i like the look of im about to jump on and start building some stations

[copper seal]

enjoy ;p

and then when you're done you can plop down a train and have some fun driving around yourself

[quiet wraith]

haha thank you

[restive timber]

Hi, I tried to come up with a 4 to 3 loadbalacer. To you think this can work?

1. Split the inputs into 6 (2 x 2 to 3)
2. Load balance them (3 x 2 to 2)
3. Combine them
   #screenshots message

[prisma kraken]

it is easier to build a 2:3 balancer twice and then merge output pairs

[restive timber]

O, that would reduce the complexity quite a bit, thanks, didn't see that

[prisma kraken]

if you're looking for a compact design, here's something i came up with

[restive timber]

Good to know that screenshots are allowed in here. I don't really need a compact design thankfully in this case

[prisma kraken]

something i'd recommend is to think twice about why you need a balancer, oft times there's some pivot to a problem you can make that removes the need for such beasts

[restive timber]

The numbers are just like that. 2x 720 + 2x 400. I don't have mk6 yet, so I can't combine the 2 400 lanes

[prisma kraken]

idk, sometimes an injected manifold is the way to go, sometimes you can make due with merging things in an ISC and relying on back pressure from outputs to get things where needed - load balancers are rarely needed in this game

the only time i really think of using them is when i need to ensure train car drop-offs are always consumed at an even rate

usually some creative belting solves such problems

[restive timber]

I already made it tighter than it has to be

[prisma kraken]

gotta give yourself lots of space in this game 😉

[restive timber]

I hope I'll never need more than 1680 QWire

[desert oxide]

I think im making over 20k Quickwire 🙂

[restive timber]

I like being in this state where 3 freight stations are enough to handle my wire

[prisma kraken]

don't move wire, move ingots

they're faster than quick/wire in & out of stations and especially for qw, most recipes that use it take some multiple or fraction of the 90/min output from a fused qw assembler

[restive timber]

I'm moving it to multiple factories though, each factory pics it up as needed

[turbid patio]

10 hmf per min,
i am free o7

[amber umbra]

Nice looking factory.

[agile osprey]

Turbomotor factory

[spare jolt]

phase 3 in 40 hours. the time is cumulative, and I'm surprised it took me less to finish phase 3 than phase 2

[fringe pawn]

Well that's a first. I threw a nuke nobelisk at a cliff hog, and it hit the rock they kicked up. Normally that detonates it right there, but this time it stuck to the rock, and didn't go off until I pressed the detonator.

[prisma kraken]

haha

[lean fable]

hey! Just started needing oil! Quick question. It's pretty far away from my base so, will i need a lot of it?

[deft lichen]

that depends on what you consider a "lot", but multiple nodes of it probably

because pipes require head lift, it's better to process the oil on location and deliver the solid products

you'll get access to trains soon, which you can use for this purpose

[lean fable]

For now do i setup a truck route or should i pick it up from time to time from a storage

[deft lichen]

you can set up a truck route

[lean fable]

Awesome thanks

[timid zenith]

this is the time when I need like a lot of rotors

I only get 4/min and I need amounts like 50 or 25 for basic stuff like coal or yellow slugs

[lean fable]

Oh yeah, got my rotor production setup dont worry

[timid zenith]

and I already know where im gonna build a rotor factory thing

2 pure nodes of iron 💪

[lean fable]

Also just got the jetpack, i see that rubber (required for logi mk4) has heavy oil residue as a bi-product, so i can just route it towards another refinery to make residual fuel. Would that make sense? i figure if the bi-product gets full the machine will simply stop

[deft lichen]

correct

there's another thing you can make from heavy oil residue that you can deal with before fuel generators

spoiler: ||petroleum coke, can be burned in coal generators or sunk||

[lean fable]

oh yeah i see that now

i am workign around here tho, there are 3 coal nodes south of crude oil wells, if needed i can just make another coal generator plant. Maybe ill sink the coke

😮‍💨 got a lot of work to do! Thanks for ur help Ondar

[spare jolt]

depends. for the start, 60/min of each of plastic and rubber will be fine to kickstart the milestones. HOR can be refined into Fuel or Petroleum coke, the latter can be burned in coal generator, and the former can be packaged and used in jetpack or to power vehicles.

[timid zenith]

I need plastic before I can make explosive rebar 🙁

by the time I get plastic I wont need a rebar gun

[spare jolt]

yea that's a bummer, i don't even bother with it

especially why would you need it, if you can throw some nobelisks around and detonate them at the same time

[timid zenith]

I wanted to explode stuff with my rebar gun in earlygame

I have 210 steel pipes somehow before coal so ill get nobelisks

[lean fable]

When you say packaged will i need another machine to "package" it?

[spare jolt]

yes, the fluid packaging milestone unlocks a packager which turns canisters (made from plastic) and fluids into packaged fluids, and vice versa.

if unpackaged, the canisters are refunded, but if burned as fuel or in the jetpack - they are not

jetpack can only use six types of fuel, three of which are (according to your progress) solid biofuel, packaged fuel and packaged liquid biofuel

[lean fable]

so packaging gives me the opportunity to recycle ressources?? If i understood wel

[spare jolt]

not sure what you mean by recycle. sinking fluids into Awesome sink? yes. Turning liquids into other products? you don't necessarily need a packager for that, just a corresponding machine and a recipe

[lean fable]

So packaging is optional? what is the advantage of packaging

[spare jolt]

yes, packaging is optional. packaging allows liquids to be transported as solids (by conveyers), be used as fuel in the vehicles and carried by a pioneer

[lean fable]

ohhhh i get it now

makes it possible to transport on conveyors and no longer in pipelines

[spare jolt]

exactly

[lean fable]

As such i can store it in my inventory

[unborn ermine]

and when you get to dealing with gasses? They can be compressed when packed so its FAR easier to transport.

[lean fable]

ohhh alrigh it i get it now

[spare jolt]

but to get them back as liquids, you need to unpackage them (using the same Packager building), which will also refund the used canisters and can be a pain in the butt to deal with 🚬

[lean fable]

SO, if i put the unpackaged fuel into my jetpack, theres no way for me to take it out. But if it it packaged i can do what i please with it (drink it?)

[brisk smelt]

turbo slug racing

[lean fable]

Refunds the used canisters as a bi-product

i would have to get into it i dont want to sworm you with questions

havent even placed my first oil extractor yet

[unborn ermine]

Yeah loops for packing/unplacking are nice for canisters, but logistics can be a nightmare if done poorly.

[timid zenith]

coal power lets gooooooooooooo

[lean fable]

First playthrough actually not doing spagetthi and actually using manifolds and balancers so ill have to do some trial and error with that one i think LOL

[spare jolt]

that's fine, relax, i was the same way 🚬
but yea, you'll have to find a way to deal with empty canisters if you choose to not transport liquids by pipes.

[timid zenith]

good place for coal generators?

[spare jolt]

one of the best probably

[timid zenith]

ok now time to grab some resources and turn this into a battery charger

[spare jolt]

if you don't stay on phase 2 for too long, two of those can even be used for steel, there's some iron nearby

by too long i mean for so long that you have to build a lot of coal gens because of sheer size of your factory

[timid zenith]

the lower for power and the upper for steel becuase lower are closer to water and I dont want any fuses blown

2 impure...

and 2 normal
yeowch

[spare jolt]

i had 2 impures for steel and 2 normals for coal gens

solid steel alt greatly cuts on the amount of coal you need

also, I usually don't sink my excess items at this point yet, simply to save on power

[timid zenith]

no copper on me time to go back

[spare jolt]

there's also sulfur nearby btw, you can set up a tractor for nobelisk too

[timid zenith]

i already have a conveyor full of sulfur

and a storage full of sulfur, which has 2400 sulfur iirc

[primal shadow]

Should put better belts on the straight ways to make it better

[brisk smelt]

i added em, mk3s on the straights

[primal shadow]

Niice

[prisma kraken]

decided to take a little break and do something a bit different... wanted to see what it would take to supply an ammo factory with wood and sam:

...it truly isn't eating nearly as much wood as i thought it wood to fill containers with the different ore types

[golden ridge]

guys, is this type of intesection doable with the ingame trains?

[spare jolt]

Technically this wood be possible with any biomass source, not just wood

I wood assume

[prisma kraken]

yeah, i have biocoal cooking too, but really everything is coming from the wood

[prisma kraken]

you can build it, sure, but trains always take shortest path so such things aren't the most useful where you're trying to add another rail for throughput

if you're building 2 loops and need the traffic to cross over between the loops, that'll work find and is pretty easy to build

[golden ridge]

basically what im trying to make is a 2 way track, where a train terminates and reverses. and im going to have two terminus stations, ie one to drop off coal and sulfur, and another for petrochemical loading, and i want it to always take the right side track, how would i do that?

or am i going to have to basically put both stations in one straight line

i can but its inefficient lol

[prisma kraken]

not quite the same as what is in your photo, but this works

note that that is 2 unidirectional tracks, i just have the stations connected to a single one of the rails and handle turning around elsewhere

when you have bidirectional tracks, you really need to either a) keep it extremely simple or b) plan it out extremely well before you start building

what i would suggest is set up a testbed somewhere with a minimal railway on a big platform where you can prototype your idea w/o all the other terrain and distance between the stations and just run some trains through model designs

(making a big foundation slab over one of the map edges is a great place for such things)

i'm not sure exactly what you have in mind from your description

i count 3 stations you named, and there's 4 'ends' to a rail like that

[broken kiln]

anyone got a design for a low-power but far-launching cannon? best i'm getting with anywhere between 2-4 entrances is most of the way across the rocky desert biome. i want to yeet myself back and forth between the blue crater without consuming over 100MW per cannon

[prisma kraken]

do you know the pulse nobelisk trick?

[broken kiln]

saw that earlier, actually

haven't automated COs yet though

[prisma kraken]

for the power consumed, that probably is your best bet

CO's can be cheesed by making them with the base recipe, iron wire and stitched plate

[broken kiln]

well sure, 0MW is hard to beat lol

[broken kiln]

that is a good point....

[prisma kraken]

it isn't super efficient, but for what you need to make nobelisks a 4x4 bp of that solves your oscillator problem

some fugly numbers with it all, but not that bad

[fringe seal]

ah, ok. train forcibly stops at a red signal even if it can't brake in time

just like how it stops at the physical end of line

[prisma kraken]

yeah, pretty odd when you witness it

something like what this mod does should really be available as a late/after winning unlock

[fringe seal]

Output multiplier
hmmmmmmmmm

[fringe pawn]

Calculating maximum tickets would become hell. That's a difficult non infinite number to work with.

You could choose to forego dimensional storage entirely for the sake of more sloops.

[prisma kraken]

or just sloop the conversion and make more of both 😄

(kind of the point of the mod)

[fringe pawn]

Unlimited sloops becomes less interesting because you can use them on underclocked things freely

[amber jacinth]

At that point just spawn in extra artifacts

[prisma kraken]

yeah, there really should be some endgame prize like 100 sloops or something

[amber jacinth]

Converting from one to another is an interesting mechanic, but infinite is just boring

[fringe pawn]

Infinite sloops also gives you unlimited resource-free power.

[prisma kraken]

agreed, needs some noodling, but i feel like there should be something cool

[unborn dome]

Would a train with 1 locomotive and 5 cars be too many cars for a single loco?

[prisma kraken]

yeah, as it stands that is kind of broken as balance, but there should be something like that, i think

5 carraiges are the limit for one engine on a slope

[amber jacinth]

Imagine using the packager for those recipes 😂

Hmm, actually…

[prisma kraken]

i think 5 full cars make the train really untolerably slow in climbing

ymmv, i've used 5 cars before without issue on pretty gentle slopes, try it and see

[unborn dome]

For the sake of having an even number of platforms, I might just go with a 1+4 train, but just have every platform end with a blank segment...

[prisma kraken]

that's what i've standardized on unless i'm doing something funny

[unborn dome]

Like this?

Or you mean standardized on 1+4

[prisma kraken]

4 cars makes loads easy to deal with

loco + 4 freight

[unborn dome]

Literally sole reason is because if I do five platforms total, pillars legs aren't evenly-spaced anymore, since each is 3-foundations-long

[prisma kraken]

the capacity of the cars supports 600/min of ore with a 4.x minute round trip, that often keeps you from needing to double up on trains, and it seems often what i'm moving is 2400/min or less

[unborn dome]

Oh wow, each platform costs 50 MW. That's gonna add up quick.

[prisma kraken]

um, platforms are 2 foundations long. exactly.

[unborn dome]

Yeah, but my pillars are 3-long

So two pillars per three platforms

[prisma kraken]

gotcha

[unborn dome]

Maybe I just need to redesign the pillars, since every station is probably going to end up looking like this...

[velvet venture]

the male urge to start satisfactory from scratch has got the best of me lads

[dusky dust]

Okay. Time to actually learn a thing, I think. How does one actually come up with equations for how to split up Recycled Plastic + Recycled Rubber in the "optimal" petrochemical loop?

I've been searching around and playing with numbers and such myself, and I know for instance that if the loop is only producing Rubber + Plastic (ie: no Fuel as an "output") you can know how much fuel to send to Recycled Plastic with (P*17/27) + (R*8/27) (P + R being the amount of plastic + rubber you want at the end, respectively)

And likewise the amount of fuel to send to Recycled Rubber can be gotten with (P*7/27) + (R*16/27)

[vapid gorge]

I just make recycled plastic and rubber in groups of 1200 output and keep reprocessing until I have my desired amount of >.>

[dusky dust]

And those numerators are obviously related to how 810 plastic/min would involve 7 Recycled Rubber + 17 Recycled Plastic, and 810 rubber/min would invovle 8 Recycled Plastic + 16 Recycled Rubber

But how do you get to those numbers, etc?

[vapid gorge]

… tools? xD

[dusky dust]

Heh, yeah, I've always just used tools and trusted the results. But I hate relying on external tools; like to understand stuff myself

[vapid gorge]

Or know my final output of plastic or rubber and keep dividing in half

If I need 1500 recycled plastic? That’ll be my last group, fed by a group of 750 recycled rubber

Not that clear cut of course, I may break it down into smaller groups depending

[dusky dust]

Oh, and the ratios above work so long as the only outputs are rubber and/or plastic. Throw in Fuel as an output as well and those ratios don't work anymore, since you're generating more seed rubber than you would otherwise

Really I just want to understand how exactly those ratios come about. Going through the loop iteratively would be enough in practice, of course, but I'd love to know how to arrive at an equation

(Younger me is screaming at me that I was once Actually Not Shit At Math. :D)

[vapid gorge]

Does it? You make as much diluted fuel as you want plastic or rubber, use all the resin as residual rubber to start it off

You can treat the residual rubber as a chemical catalyst that just turns your fuel into proper products

[dusky dust]

Right, I understand the loop well enough from a practical consideration, but I don't understand how to figure out the exact ratio of Recycled Rubber to Recycled Plastic

At least without iterating through the loop.

Like how do you get to (P*17/27) + (R*8/27) for the Recycled Plastic fuel requirement?

[vapid gorge]

Are you wanting a sort of formula to just plug numbers in? Hmm

I suppose it depends what steps you doing it. Do you have just a bunch of buffered plastic and rubber to do it in 1 step?

[vapid gorge]

I’m not sure what that formula is or where it’s from?

[dusky dust]

Yeah, knowing how to derive a formula for this stuff is my main goal

Heh, that particular one was gleaned from comments @ https://www.reddit.com/r/SatisfactoryGame/comments/x94t7i/could_use_help_with_the_math_behind_the_oil_loop/

And it works great so long as there's no Fuel output

a la: ```
$ satisfactory-petrochem.py -r 405 -p 405
Crude Oil Input: 270
Water Input: 900
-> 720 to Diluted Fuel
-> 180 to Residual Rubber

Number of HOR refineries: 9
-> HOR to Diluted Fuel: 360
-> Resin to Residual Rubber: 180

Number of Residual Rubber refineries: 4.5
-> Rubber to Recycled Plastic: 90

Number of Diluted Fuel blenders: 7.2
(Or, number of Diluted Packaged Fuel loops: 12)
-> Total fuel to Recycling loop: 720
-> Fuel to Recycled Rubber: 345
-> Fuel to Recycled Plastic: 375

Number of Recycled Rubber refineries: 11.5
-> Total rubber: 690
-> Rubber looped to Recycled Plastic: 285
-> RUBBER OUTPUT: 405

Number of Recycled Plastic refineries: 12.5
-> Total plastic: 750
-> Plastic looped to Recycled Rubber: 345
-> PLASTIC OUTPUT: 405

But I don't understand how it was derived

And, as I say, it doesn't work if there are fuel outputs, 'cause it doesn't account for the added seed rubber you get, since you're producing more fuel than you need for the loop

[vapid gorge]

This person is really overthinking it. For example you never need more than the initial residual rubber to start the loop process

So their initial premise is just weird

[dusky dust]

I think the initial post was just iterating through the Recycled loop, not adding in external rubber sources

But it's more the comments which provide some basis for formulae

As I say, with only plastic and/or rubber outputs, those two equations for the Recycled Plastic/Rubber fuel requirements work great. But I don't really get how they're arrived at

The numerators seem to be based on the refinery counts for all-rubber and all-plastic loops, but I don't get how they gel together

(Or why exactly 27's the denominator, though that's clearly related to the 81-divisibility thing)

[vapid gorge]

Maybe if I get home and can have a proper look? But it looks like their trying to over complicate things

[dusky dust]

I think they basically just had the same goal that I had: arrive at a formula (or a set of formulas, anyway) which, when you plug in your desired rubber+plastic outputs, gives you exactly the ratio of Recycled Plastic to Recycled Rubber.

My own additional goal is throwing in a Fuel output, but I figure I'm starting with trying to understand the simpler case. :)

Presumably once my brain clicks into place with that, it'd be easy enough to add Fuel output to the mix

[vapid gorge]

Maybe I’m not understanding specifically what you’re trying to do? Why can’t you treat each individually?

[dusky dust]

(And those formulae are already there in the comments -- I just want to understand how to arrive at the formula. Just plugging it in and knowing that it works isn't enough for me)

[vapid gorge]

Hmm ok, I’ll take your word for it, I didn’t get to sleep till 6am last night xD

[dusky dust]

Heh, no worries

As I say, if there's only plastic and/or rubber outputs, those formulae work great. The text block I pasted above uses them, and all the numbers seem totally accurate

But I don't understand how they were constructed; how to figure it out myself. If I didn't have the formulae just given to me in a comment there, my little util wouldn't be able to just do simple math to know how to assign refineries.

[sick oracle]

new meta right here

[dusky dust]

(I've never seen 0 UNDEF N/A being produced so quickly!)

[sick oracle]

ikr

[sturdy solar]

I used the loop (plastic + rubber) to only make plastic at a 2-to-1 ratio. (6 refinery make plastic, 3 refinery make rubber)

with the Resin + water, I make rubber

300 crude makes 100 rubber and 270 plastic

TLDR:
10 refineries: crude > heavy oil
7 refineries: heavy oil > fuel
3 refineries: plastic > rubber
6 refineries: rubber > plastic
5 refineries: resin + water > rubber

that, and i dont have blenders / diluted fuel yet

[turbid oar]

hi all, i got a problem some ratios :/ what is the best way to describe it that you understand it easily?

[vapid gorge]

Make a plan in Satisfactorytools.com, share the plan

Unless it’s just a one step part of a system?

[turbid oar]

i set this part up without the satisfactorytools, everything is correct according the numbers but at the end its still not running as it should.

tried to remake it in satisfactorytools and even there i cant figure it out, the numbers there are incorrect

[past reef]

At least screenshot or describe the part that is having issue

We can't magically know which recipe you use and how much you input and whatnot

[pure crow]

Brainstorm, model, test, iterate.

Also, it probably shouldn't nearly be that complicated, when a building eats 6 and produces 12, a clean answer is going to be related to some clean multiple of 12/6 = 2.

[past reef]

For plastic rubber run 2 equation using variables machine count of recycled plastic-rubber

[turbid oar]

want to make alu sheets and alu casing but running dry on ingots

[past reef]

How many ingots are you making and how many machines are on sheets/casings/alclad casings

[fringe seal]

I am feeding 300 in a pipe that requires 250
it's been an hour
blender efficiency is still 100%
how the fuck is it still going

ah ok

finally

[past reef]

The machines feeding water will be minus efficiency

Also in aluminum setup its easy to run into inefficient scrap/solution machines because of attempt to merge wastewater

[fringe seal]

(it's rocket fuel in my case)

[vapid gorge]

ok follow the issue backwards, are the ingot producers clogged or starving?
if clogged? you have flow issue from there
if starving? follow it to the srap producers

repeat

my wild guess based on no info is that your solution/scrap makers are janky but curious to see

[turbid oar]

producing 160 alu ingots,

3x assembler on alu sheet @ 77.77 requires 70 ingots
1x constructor on alu casing @ 100 requires 90 ingots

[heavy gust]

anyone else getting framedrops through the roof after reaching tier 9?

its still playable since it drops from 100 to 60, but doesnt feel good.

[vapid gorge]

that doesn't answer the first step I asked you to do

are the ingot producers starving or clogged?

[turbid oar]

yes and no

start from the beginning then, moment

[vapid gorge]

so they aren't getting enough scrap?

go to the scrap producers. are they clogged with water or scrap? or starving of solution or coal/coke?

[turbid oar]

480 bauxite - 240 coal - 30 copper ore - 760 water (4 extractors, 2 pipes mk 2)

[vapid gorge]

unless you made some math errors and didn't follow the tools plan that doesn't matter right now

you're machines are starved for scrap
so check out yoru scrap producers

[turbid oar]

3 foundry full with 500 scrap, the rest to sink
240 needed, making 720

silica running dry for ingot foundry

producing 200 silica with 4 refinery, looks like the refinerys are idling..

they got enough resources

[vapid gorge]

so you're running out of both scrap and silica?

[turbid oar]

and not full

[vapid gorge]

you didn't answer my other question about scrap

[turbid oar]

got enough

[vapid gorge]

so the scrap is CLOGGED or STARVED

[turbid oar]

still not getting enough silica

[vapid gorge]

untill you tell me how the scrap refineries are going I can't help you

[turbid oar]

2 scrap refinery full with input, output immediatly empty after constructing

producing 720 scrap but needing only 240

[vapid gorge]

so they are running fine but your ingot machines are starving

[turbid oar]

got smart splitters on overflow to sinks

ye the silica

[vapid gorge]

but you said you weren't getting enough scrap in yoru foundries

[turbid oar]

no was running dry on ingots, but its not the scrap, its the silica

[vapid gorge]

right ok follow the silica - are the silica producers clogged or starving

[turbid oar]

producing 200 silica, 4 refinery input full with water and bauxite

after construcing resources leave immideatly

[vapid gorge]

so you're not producing enoug silica

[turbid oar]

alumina solution not full

[vapid gorge]

they aren't stuttering then? show your tools web link

[eager solar]

can you send a screenshot of the problem?

[turbid oar]

i think i fixed the problem,,

[past reef]

For normal aluminum ingot the silica from refineries wont be enough

Did you check the place making the extra silica

[turbid oar]

got 3 foundry, had the first on 40pm and second and third on 60pm.

changed the first two to 60pm and last one to 40pm

[past reef]

Clocking on the ingot foundry issue then?

[vapid gorge]

that shouldhnt' do anything

[turbid oar]

first and last foundry full of input and running good but need to wait a little for filling the middle

but the middle is still running dry sometimes

producing 200 and requiring 200

[vapid gorge]

well you keep refusing to answer what is being asked, it takes like 5 minutes to pull any one question from you, so I'll wish you good luck

[eager solar]

send us a screenshot of the problem, it will help a lot

[timid zenith]

we outta modular frames

[turbid oar]

ye srry i cant do 5 things at once..but thanks for the help. it was the smart splitter at the end of the line, while the last foundry was full and the middle not it was sending half to the overflow

[full thunder]

Hello there !
If some ppl are kind enough to check if everything is alright in my maths 🥹

I want to produce 9.6 uranium fuel rods / min:
https://www.satisfactorytools.com/1.0/production?share=Cp0vQ3IVmr8Yv20LgTX1

Then, 480 /min uranium wastes into 4.266 plutonium fuel rods / min
https://www.satisfactorytools.com/1.0/production?share=5SeiXPtxotK29N4ZvHCg

Then, 42.66 / min plutonium waste into 21.34 ficsonium fuel rods
https://www.satisfactorytools.com/1.0/production?share=7buUSRIV0nUyGEHubG2y

[deft lichen]

That's not 480

https://www.satisfactorytools.com/1.0/production?share=hc4gDpyqRfIBmNJtd4OU

Use the maximize option

The plutonium alt is messing up the numbers, it has a clean 2:1 uranium:plutonium ratio otherwise I think

[full thunder]

oh I missed this option entirely, thx a lot !

[full thunder]

you mean the numbers are hard to work with because of my use of the alt recipe?
if yes, i know, but without the recipe I drop to 3.675 plutonium fuel rods min, you would choose less rods?

[deft lichen]

I have no opinion on this, I would not burn plutonium or produce ficsonium in the first place

Uranium rods and sinking plutonium, or just a large RF setup

RF is so insanely overpowered

[full thunder]

I was recycling uranium wastes into plutonium rods then sink them as well in my previous save.
This time I want to go for ficsonium, part for the beauty of it, part because I need a ton of power for my ongoing project 🙂

Since i'll need enough to feed 40 ish portals as well as continuously making all project parts allowing me to feed all phases the fastest way possible, I'm trying to minimize the number of machines in each of my factories, and end game production machines eats your power grid away when overclocked.

[opal pivot]

is there a way to copy and paste trains?

Like, Ive made 6 of the same train so far and I need about 20, do I really just have to keep manually putting in the same schedule for each? :/

[fickle roost]

Is there an easy way to see what machines have power shards inside?

[past reef]

if you're looking at them just go dismantle mode and hover over them

[vapid gorge]

... you do not want 20 trains stopping at the same station

[vapid gorge]

their power lights are dif colour

[fickle roost]

Can you send an example?

[vapid gorge]

no sorry, but if you have it overclocked it'll be blue instead of green. A lightish blue. could be mistaken for white

[vapid gorge]

red is for no power or no recipe, yellow for standby, green for on, blue overclocked and on

[fickle roost]

Gotcha cheers

[vapid gorge]

I guess you should also check yellow light ones cause I think if overclocked machines are on stand by they are also yellow

[fickle roost]

Geysers are just free energy right?

[vapid gorge]

pretty much? but there aren't many of them and they are scattered so their uses are niche.

I've used them to slowly charge up a power storage facility to kick start a big fuel station though?

[spare jolt]

geothermal is legit the only reason to use power storages

[opal pivot]

What? Yes I do? Otherwise I wont have enough throughput...

[deft lichen]

not the only, there are fluctuating power consumers too

geothermal generator, particle accelerator, quantum encoder, converter and technically trains (locomotives and freight platforms)

[vapid gorge]

every train you have stop at a particular station reduces the throughput of those platforms. You'll have almost zero throughput

you know platforms have very limited throughput right?

at most any platform might be able to do 1.5 belt of throughput.

morelikely it'll be more like 1 belt's worth if not less. That's with 1 trains stopping

[past reef]

I'm starting automated radio control unit, already have 100s of crystal oscillator pm to spare, I should go for radio control system?

[eager solar]

depends on what else you have, for true optimization of parts you may even want to use the base recipe on top of the alt or something

[bleak ivy]

yeah, fertile uranium also preserves the ratio https://www.satisfactorytools.com/1.0/production?share=Yl16qvDzACqVrP0aOdPm

[opal pivot]

How else do I get 1200 throughput if I only have 1 train every 30 minutes? xD

[past reef]

how do you even have the train RTT being 30 minutes

[bleak ivy]

split your payload into more freight cars

[deft lichen]

is that train doing a trip around the whole map?

[opal pivot]

Its a LONG track

and a slight exageration

[bleak ivy]

1200 throughput you can easily just do with like 2 freight cars doing 600 each

[opal pivot]

pretty much, yeah

[deft lichen]

that's why you use multiple smaller trains and not 1 big train

[opal pivot]

exactly what I'm saying

[bleak ivy]

im pretty sure like the maximum possible throughput you could get from a 30 minute rtt is ~106 items/min

per freight car

cause they carry 3200 items / 30 (for 100 stack size)

[opal pivot]

Right, so more trains means more throughput

[past reef]

as long as you double load the freights and a train arrive every ~5 minutes you can still do 1 belt throughput

[opal pivot]

exactly

[deft lichen]

the round trip durations are solved

https://satisfactory.wiki.gg/wiki/Tutorial:Train_throughput#Solved_maximum_throughput

for a stack size of 500 and mk6 belt, the round trip has to be exactly 427.08s (7.118 min)

[opal pivot]

how does more trains not increase throughput?

[deft lichen]

any more or less, and the throughput will be less (having a shorter round trip time is much less diminishing than longer)

[deft lichen]

it does, but only as long as they don't stop at the same station

all of this math exists because a docking train blocks the platform I/O ports

multiple trains stopping means more blocking

[opal pivot]

I have 2,700 resources going through a train with 3 wagons, Im not 100% sure of the RTD but could time it in a second

[opal pivot]

only if they try and stop at the same time...

[deft lichen]

no, in general

[opal pivot]

I have 6 trains and they have never once stopped

[deft lichen]

you can't ensure unbunching

[opal pivot]

what? how?

[opal pivot]

they have never once had to wait for the next train

and even if they did, as long as there is a waiting area there is no issue with that...

[deft lichen]

I recommend reading the linked wiki page

(I don't have all the explanations on hand, sorry)

[opal pivot]

the train stations have 2 inputs and 2 outputs, as long as you transfer that into one using a double storage warehouse then its fine

Ill read it but I have no idea how more trains is supposed to reduce throughput when one isn't enough

Surely I just have to make sure that I have enough trains that even when they wait it never causes the belts to stop?

[turbid sapphire]

is this unreasonable? lol

would go in the desert

[opal pivot]

Looks similar to mine

but why so many Encased?

[outer vale]

consder the extreme case where as soon as one train finishes the docking sequence, another one's there almost immediately to dock again. The station would spend almost all its time locked down by the docking sequence, not actually letting anything out on the belt outputs

[opal pivot]

But that's why there's a buffer there...

thats why you turn the 2 outputs of the train into 1...

so that the lows and highs are smooth

[outer vale]

buffer doesn't help if it never gets a chance to use it

[opal pivot]

you may get 2400 coming out of the station for 2 seconds, then 0 for 1 second, but thats still 1200 per second

[turbid sapphire]

i just hit maximize to use all the coal in the desert

[past reef]

do you need 1000 singularity cell per minute or what, why that many encased beam

[outer vale]

indeed. So if you get 2400 coming out for 2 seconds and nothing for 58, that average drops right down

[opal pivot]

I dont see any need for that much encased, so maybe less encased and more iron

[outer vale]

like I said, extreme case, but the point is that eventually you stop being limited by what the trains can carry, and instead by what the station can physically unload in the time it's not locked down

[opal pivot]

Sure, but I'm getting more 2400 inputs than I am 0...

Surely as long as the train station never completely empties I'm fine...

which shouldn't happen as long as I only take 1200 from a single wagon...

[brisk smelt]

just make the entire big PCC chain factory at that point

[turbid sapphire]

PCC?

[brisk smelt]

pressure conversion cube

-> FMF -> HMF -> MF

[turbid sapphire]

iron? you mean more steel beams/pipes

[opal pivot]

steel sorry