#math-and-meta

1 messages · Page 100 of 1

lime wadi
#

Its more than 50% less crude oil

#

Less byproduct too, not sure if you wanted to do anything with it or not.

proven fossil
#

I mean the issue isn't really crude it's just transporting it in pipes

#

I mean how early do you get t3 pipes, bc techincally i don't need that much fuel as I don't have all the fuel gens up yet

#

I can leave it sub optimal till i get the next pipes

lime wadi
#

t3 pipes have not been added and there's no plans from Coffee Stain to add them

proven fossil
#

Oh shoot

lime wadi
#

t2 is the best it gets, though t2 pipes can be buggy at times

#

probably a big reason they dont want to add t3 lol

proven fossil
#

wait what's wrong with t2 pipes?

#

like in what way do they bug out?

lime wadi
#

But in short, there will be fluid loss.

#

You can carry something less than 600 in the pipe safely, but you probably shouldn't use exactly 600 in a pipe. I learned this the hard way. My fuel power plant never runs at 100%.

#

So the 600 fuel output here, I would recommend carrying in 2 pipes, actually.

#

I would suggest working out how much crude oil you want to use for this project. For example if you want to overclock one Normal node, you could get two Mk.2 pipes of 400 each

image.png
proven fossil
#

Hmm that video was interesting

lime wadi
#

Personally I limit myself to 580 in Mk.2's, and I dont seem to run into the bug.

proven fossil
#

Ok I guess I will just scale it down slightly so it ends up being a couple machine less from 600 fuel output

lime wadi
#

Yeah, sadly that bug is probably going to persist for a long while.

proven fossil
#

Yeah I wonder how much more taxing it would be if they ended up using doubles

#

Or maybe a togglable feature

lime wadi
#

Maybe one day we will see a feature— yeah

proven fossil
#

Glad i found out this way

lime wadi
#

But for now, just plan around using less than 600. Maybe 590 max, per pipe.

#

Yep, I built a... very large fuel plant

#

With 600s in my pipes. It doesnt work'

proven fossil
#

Oof

#

By chance is your plant visually appealing

#

Trying to get some inspiration for my build

lime wadi
#

I wouldnt say so, I was going to get to decorating after i fixed it lol

proven fossil
#

lmao

#

Aight thanks for the help tho, appreciate it

#

Btw how many hours are you at?

lime wadi
#

About 600. Built the fuel plant at about 100 lol

#

First major failed project

#

Theoretically balanced, this was the first quarter of the thing.

image.png image.png image.png
#

But thanks to the Mk.2 pipe bug...

image.png
proven fossil
#

Damn that looks better than most of my factory

#

💀

lime wadi
#

If you look in the last screenshot youll see the gens in the back are starved.

#

Thats because the pipe was supposed to carry 600 fuel.

#

Its also not producing 600 fuel because the water for it is being carried in 600 pipes as well

proven fossil
#

damn screwing you over

#

That production line is just

#

painful

lime wadi
#

Yeah, thankfully im playing on a new save for U8 and Im not gonna make that mistake again,

#

Never carry 600 in a pipe

median heath
#

?

lime wadi
#

But do use Mk.2 pipes

median heath
#

600 in a mk2 is fine.

proven fossil
#

Do you think it's an issue if the input crude is 600?

median heath
#

No.

lime wadi
median heath
#

You just have to loop it, which you didn't seem to do in your screenshots.

#

It's about preventing slosh, not floating points.

proven fossil
#

Yeah I'm pretty sure when a community manager says it's an issue I'm gonna believe them

lime wadi
#

Slosh is a good point too, but I did not repair this factory because it carries 600 in its pipes.

median heath
#

Soing something as simple as this guarantees you're getting 600.

Screenshot20230209-20200000000-1.png
lime wadi
#

Otherwise, you are correct in saying I could reduce slosh

proven fossil
#

When would slosh occur?

lime wadi
#

Pipes are bidirectional, so.

median heath
#

Looping eliminates that issue and allows the full 600 flow per mk2.

lime wadi
#

This is true, but machines themselves can also become deadstops should they become full. A loop allows the fluid to move to more than one place in the manifold.

median heath
#

☝️

proven fossil
#

Sorry wdym by deadstop?

lime wadi
median heath
lime wadi
#

Note there are some advanced topics in that PDF

#

But its also great for learning the beginning parts of fluid dynamics

proven fossil
#

i mean does it turning left not count?

median heath
#

Do not think about pipes in terms of belts.

lime wadi
proven fossil
lime wadi
#

It's great for learning how to do fluids like a pro

#

According to the wiki (https://satisfactory.wiki.gg/wiki/Pipelines) loops will significantly reduce the Mk.2 flow rate issues, just note that it may still happen even with best practices.

Satisfactory Wiki
Pipelines

Pipelines are used to transport fluids between structures. The Mk.1 Pipeline can transfer up to 300 m3/min, the Mk.2 twice as much. Head lift has to be applied to transport fluids vertically, which can be increased using Pipeline Pumps.

wind spade
#

good luck (though fair warning: analysis of all recipes in all situations is basically impossible to do)

wind spade
lime wadi
wind spade
#

(and also, any analysis is usually very biased)

vapid gorge
lime wadi
#

I use loops nowadays

#

That factory, even after having put loops into small parts of it to see if it woudl work, was suffering fluid loss.

#

Likely because my game, at the time, would only run at low framerates

vapid gorge
#

Oh I guess, I don’t understand why you were telling someone else to not use mk2s though, a big part of the issue seems to be from longer manifolds

#

But yeah the loop solves that

lime wadi
#

And I still use mk.2's in longer manifolds, just not at the full 600 capacity.

#

I learned a lot of lessons on fluids in that save, lol

median heath
#

You can do the full 600 if you loop 🤷‍♂️

proven fossil
#

Ik this is wrong channel but does anyone have a bunch of blueprints they can share

#

Kibitz had some but i believe they're paywalled

median heath
#

More useful to make your own tbh.

wind spade
#

yeah, even free blueprints are still kinda bad blueprints, since they rarely help your case exactly

proven fossil
#

I mean like cool pillar designs or road pieces, just some generic stuff

median heath
#

Cool is a point of view 🤷‍♂️

wind spade
tepid forge
#

I thought I would ask here .. for as long as I've known running a Mk3 Miner at 250% for 1200 output, has always been a waste (unless running mods) due to the speed of Mk5 belts.

Has there been anything in recent times added to the games OR ways to make use of the 250% playing vanilla without mods or are we still limited to the Mk5 belt speed exiting the miner?

upbeat tide
tepid forge
upbeat tide
#

honestly, planning a factory around a 780 input gets messy. 600 is a magic number for just about everything I have found. Take recycled plastic/rubber as example, it just makes everything clean.

I do wish they would allow direct splitter connection in vanilla to a miner, but maybe one day

tepid forge
#

Yeah I've never understood why we cant slap splitters and the like directly on the front, would make so much sense but then I'm not a game developer 😛

I've never really tried doing a cap at 600 on things, I normally just crank it and let it go, but I might look at that number at some point.

true junco
#

600 works well because anything with factors in common with 60 is going to be easy to manage when calculating rates in "per minute"

That said 780 should also work for a lot of things because it has 60 as a factor (and thus all the factors of 60). Its only screwy because the other factor is 13.

lime wadi
#

Work in progress

image.png
prisma kraken
lime wadi
#

I'll be sharing it once it's done

vapid gorge
lament wraith
#

I’m trying to perfect loop 300 crude oil into packaged fuel at 100% efficiency, all residue recycled back into fuel to be packaged, so the only output should be packaged fuel and the resin from the original fuel recipes but the math just isn’t matching

#

Help

prisma kraken
#

but 300 oil makes 400 hor + 200 resin, the hor can make 800 dilluted fuel and the resin can make 50 plastic or 100 rubber (or some amount of fabric)

lime wadi
#

All after a long day's work.

#

Now to actually take this and run it against alt recipes in different situations... i smell a new spreadsheet

deft lichen
#

before you spend too much time on this, we're planning to remove the WP analysis entirely in favor of text descriptions
because trying to objectively compare alt recipes is like trying to mathematically solve chess

oblique hollow
#

Too many possibilities for a mortal

lime wadi
oblique hollow
#

A few pagea on the wiki have an older earlier concept of those text descriptions

lime wadi
#

This would simply be an analytical tool to help create said text descriptions

deft lichen
#

excellent 😄

oblique hollow
lime wadi
#

But of course, folks are welcome to use what I have already if they want to build in a specific area with specific research.

deft lichen
#

we keep delaying it and at this point I wonder whether to wait until the recipe rebalance

lime wadi
#

Is there a rebalance scheduled?

deft lichen
#

coming at one point before 1.0

#

we don't know when

#

my personal bet is on update 9 (or the 1.0 experimental cycle if we're close enough)

lime wadi
#

Because I could just. Tackle this for the wiki once I feel confident in my research and run it by other folks in #wiki-workers

oblique hollow
#

I wonder if i should continue my spreadsheet analysis of weighing machine volume vs their production yield

lime wadi
#

I do want to take a data analysis approach though, to be sure I'm not missing anything

oblique hollow
lime wadi
#

I don't think it ever hurts to have it all worked out somewhere. It doesn't necessarily need to be in everyone's faces, but I do like listing things like this as a resource or on a separare wiki page.

oblique hollow
#

I wonder how i could justify my scuffed math on how i weigh this xd

lime wadi
# oblique hollow specifically this

There's also considering only the footprint, since y'know. A lot of the time if you put multiple types of buildings on a floor, you're only really concerned with the tallest one. Or if it's outside, you don't care. Context is quite a bit.

oblique hollow
#

it literally just machine volume / (production efficiency + production speed)

oblique hollow
lime wadi
#

But anyway, yeah, to be clear I wouldn't be putting all the raw numbers on the wiki. It's a tool to help figure out what to write and such.

oblique hollow
#

Alt recipe analysis has been a curse since forever

lime wadi
wind spade
prisma kraken
#

someone a while back was working on something of a heat map that allowed you to enter a recipe and it would give you a 'best' location for it in the form of a heat map based on proximity of ore sources

#

interesting idea, but i don't know what became of it

vapid gorge
#

they made it but it was also the type of thing you could eyeball with SCIM. The benefit of the eyeball method was you could pick a better spot than the algorithm because you could take into account terrain

#

it also couldn't decide to say, ignore 1 resource type in finding an optimal location. For example if a spot was super optimal for all but 1 of the resources you needed? it'd pick a mid ish point between the best spot and the missing resource. Instead of picking the best spot and accepting that that 1 resource would have to be moved

wind spade
#

@clever raft why what? 🤔

rustic patio
#

is there any way to make it curve around the piller? 😩

image.png
#

been trying and trying and i think it might just be impossible

wind spade
#

two step construction won't work?

rustic patio
#

not any that ive tried

wind spade
#

or make a straight, connect the S bend to straight, remove straight, connect other S bend?

rustic patio
#

like this?

image.png
wind spade
#

that's the same image? 🤔

rustic patio
#

oops

#

uGH, my screenshot program is being stupid

#

for some reason it hates taking screenshots of satisfactory

deft lichen
#

I can't take screenshots of any fullscreen apps

#

F11 should switch fullscreen quickly

rustic patio
rustic patio
wind spade
#

yeah that was my idea

rustic patio
#

yea, too sharp :/

deft lichen
#

and closer to the pillar?

rustic patio
#

moving it half a foundation back on either side works but id prefer to have it as short as possible

rustic patio
#

i think ill try not doing the pillar instead

#

the stacked interchange is confusing me anyway 😩

#

wouldnt it result in trains hitting eachother?

deft lichen
#

path signal junctions have to be level

#

so not sure how do those giant junctions really work

rustic patio
#

i dont use path signals

#

i mean stacking rails

#

okay nvm ive been doing it all wrong anyway

#

ive had a denkfehler

#

MMMM clipping 🤤

image.png
deft lichen
#

nooo

rustic patio
#

it works 😈

primal flicker
rustic patio
#

exactly

#

but id rather it also dont look atrocious 😭

primal flicker
#

Wdym it's very symmetrical

rustic patio
#

😭

#

any ideas how to make it not look like that without being smited by the clipping gods?

primal flicker
rustic patio
#

sad

#

yea thats the issue, i cant thik of a way to make it not clip vertically without making it so much bigger 😭

primal flicker
#

Turbine interchange?

rustic patio
#

yes

#

cobalt posted one like that

#

but surely there has to be a step iyn between my abomination and the turbine interchange

primal flicker
#

Cloverleaf won't work well unless it's huge (turn radius and all that)

wind spade
#

cloverleaf doesn't have splits before merge 😦

deft lichen
#

got a question about distributing parts
got belts of 600 and 513.75 cables, and need to split them between 101.25, 312.5 and 700

#

is it enough to just 2:2 the two input belts and split one of them on the output?

#

mmm yes

image.png
rustic patio
#

i would just use one of the input belts to feed the first two output belts and then combine the rest for the 700

#

should balance itself out

deft lichen
#

I'm worried that would block the input of the other belt sometimes

rustic patio
#

shouldnt happen if u use smart splitters and overflow

deft lichen
#

hm, true

rustic patio
#

unless theres hiccups in ur production lines down the road

#

but then loadbalacning wouldnt help anyway

deft lichen
#

this wasn't too hard to build though, found an ever more compact 2:2 design

rustic patio
#

👍

#

if it works it works

heady vine
#

is there a possibility to make a sorta enclosed loop of canister-water-fuel, while also distributing a bit of packaged fuel?

#

aka so that i don't build separately fuel production for personal use and sinking, but reuse the same logistics?

#

rn i run into problem that addition of canisters to replace ones possibly used by player ends up clogging outputs and stopping the loop

#

even though there's a smart splitter at the inputof empty canisters

rustic patio
#

u want empty canisters, water canisters, and fuel canisters on the same belt..?

#

im not sure what exactly u mean

heady vine
#

will repackaging fuel solve this?
meaning loop of water-fuel (for dilluted fuel) simply circles canisters, but surplus of produced fuel gets packaged back and used either to fly or to sink

heady vine
rustic patio
#

ah

#

id recommend to siphon off the liquid fuel into a packager

#

thats how id do it

#

u could use a smart splitter to prevent clogs but thats kinda pointless imo

#

would just end up wasting a bunch of resources

wind spade
#

DPF wants to be in 1:1:1 ratio anyway, a nice small loop with like 10-20 canisters put into it, much better than some big system with thousands of canisters that tries to distribute them across everything

rustic patio
#

is this an acceptable amount of clipping?

image.png
static zenith
#

is there an acceptable amount contact a train can make with an overhead bridge ?

rustic patio
#

not sure

#

thats why im asking

static zenith
#

yea its fine ,#

rustic patio
#

if its way too high it could crash into other trains

static zenith
#

its perfect trust me xD

image.png
rustic patio
#

?

#

thats not a train

prisma kraken
#

you'll never have a problem with them clipping into terrain (except aesthetic concerns), just be careful if you have a rail above it that 2 trains can pass on the rails at the same time

rustic patio
rustic patio
prisma kraken
#

park a train on the above rail and drive one below it to test if they collide

#

i've learned that on a long enough time horizon, trains will always end up at the same place more frequently than you'd anticipate

#

the probability of it all is nearly 100%

#

there's probably some math that can prove that involving discretely occurring events that can be quantized vs continuous state events, but i'm not a statistical math wizard

deft lichen
prisma kraken
#

haha no clipping for you!

deft lichen
#

well, depends if visually or by collision logic

#

if visually, no

#

keep in mind the flaps at the top expand

prisma kraken
#

i let a tiny bit of terrain clipping slide for rails that squeeze through terrain features... they're hard enough to run as-is that getting them perfect takes a very long time

rustic patio
#

the best i got so far is this :

Screenshot20231210-20034400000.png
deft lichen
#

Nuh uh

#

It needs like 1 m above

prisma kraken
#

not sure if it needs that much more, but yeah, too close

deft lichen
#

To account for the flaps

#

Realistically it'd need even more

rustic patio
#

😭

#

can u help me?

deft lichen
#

Take a screenshot from further away

rustic patio
#

simple stacked t intersection

golden needle
#

best thing I can do for rotors at mk1 level is 120 iron -> ingots and then 5 constructors for screws (200 screws in total which take 50 rods) and use the remaining 70 rods together with those 200 screws in 2 assemblers for the rotors... right?

#

lacks 10 rods :x

rustic patio
#

there isnt really a "best thing"

graceful egret
#

Just make sure you have enough constructor capacity, and it just evens out after some time

golden needle
#

well if somebody told me to take 1 less constructor because 10 additional rods is a better deal at the cost of 40 screws... that would be more efficient

rustic patio
golden needle
#

efficiency

graceful egret
golden needle
#

hm I see

scarlet path
#

hey guys! just unlocked coal power for the very first time, but i have no idea how to use it efficiently, is there any chart that tells me the best water extractors to coal generator ratio?

golden needle
#

8 generators 3 extractors

scarlet path
#

alright, thanks!

rustic patio
#

u can also just calculate it urself by looking at the numbers

scarlet path
#

oh yeah, just saw that the description has rates and stuff

rustic patio
#

not sure if it has the water or coal rates in game, but u can look at the wiki for that

scarlet path
#

wait, does the stats part say the amount of MW generated or the MW required to start it?

graceful egret
golden needle
#

yep that won't work at 100% efficiency because of the 300m³ limit :<

graceful egret
golden needle
#

but I feel like things get messy if you have to provide extra pipes

rustic patio
graceful egret
rustic patio
#

u just, connect a second pipe

graceful egret
#

Once you put in the coal it does

#

Because you can also have compacted coal

rustic patio
graceful egret
rustic patio
golden needle
#

depends on your layout, I copied some nice looking layout off a youtube video with each extractor... you'd require a 4th extractor with its own pipes then it would work but it would make things more ugly in this case lol

rustic patio
golden needle
#

it's as you say, unless you aim for perfect efficiency it doesn't really matter anyway

graceful egret
graceful egret
rustic patio
#

thats why u use two pipes

golden needle
#

I guess you could split off the 3rd extractor's pipes and route part of them into the main pipe network

#

but everything is kind of symmetrical now

primal flicker
graceful egret
rustic patio
#

this here would work without issue (just imagine theres more coal gens to the left and they're all connected. i was too lazy to build a full coal setup for just one screenshot)

graceful egret
#

Not with mk1 pipes

rustic patio
#

yes

#

try it out

graceful egret
#

because 8 generators take up 360 water per second, and the pipe max is 300

viral ravine
#

What about gens to the right, why not place two more directly in front of the water extractors snuttstach_think

rustic patio
#

oh nvm i forgor something

#

a coal gen only takes 45 water

snow dove
# rustic patio

if you utilize 8 coal gens with this setup you have to add the 2 extras to the right

viral ravine
#

see we are on the same page with Iroh here

rustic patio
#

yee i forgor that a coal gen only takes 45

#

this would work

Screenshot20231210-20240600000.png
viral ravine
#

thats really bothering me visually tired_jace

rustic patio
#

still only one pipe really, definitely not complicated

snow dove
#

you could do something along these lines

image.png
#

as long as you feed from two different points on the manifold, it works

#

just can't feed into the same point

rustic patio
#

technically its feeding in two ways in the fwirst one because junctions have infinite throughput iirc

#

but it would still lack 15 water per minute on the left due to the first coal gen only taking 45 and not 60

snow dove
#

yeah junctions are magical beings capable of infinite throughput

rustic patio
#

my point is u dont need to do complicated stuff with separating the water extractors

#

thats all i was trying to demonstrate

snow dove
#

yeah really truly doesn't matter what two/three points you feed at, there just can't be only 1 point of input into the manifold

wind spade
# 
  G  G  G  G
E-+--+--+--+
E-+
E-+--+--+--+
  G  G  G  G
snow dove
#

and those two input values i have marked there in that diagram don't matter, still equals out regardless

#

could be 300 and 60, still works the same

rustic patio
#

yep

#

as long as its more than 60

snow dove
#

coal generator manifolds aren't picky

#

there's a million different ways to do it

rustic patio
#

tempted to just do this 😩

image.png
snow dove
rustic patio
#

iroh ur good at the game, how would u a stacked t intersection?

snow dove
#

i am not good with trains tho

rustic patio
snow dove
#

i'm assuming each of those rail segments are uni-directional?

rustic patio
#

nope, its the same as normal bidirectional trains

#

just rotated to be vertical

deft lichen
#

Directly vertical switches are cursed

rustic patio
#

i want to do vertical to get a smaller footprint

wind spade
rustic patio
#

😭

snow dove
rustic patio
#

im asking for advice, i would never build that

deft lichen
#

Although real monorails do allow vertical switches

#

Because the switch track moves between end points

#

Rather than the blatant clipping that happens here

rustic patio
#

yea it hurts 😩

#

CoD made a vertical turbine intersection

#

but im choosing vertical to get a smaller footprint so that isnt very appealing

prisma kraken
# rustic patio yea it hurts 😩

i think there's some attractive space savings with a stacked rail design, but i don't think you're going to see much in terms of that with intersections, you just can't cheat the turn radius limitations for rails

rustic patio
#

ye i dont want it to be smaller than horizontal

#

im looking for something comparable to a horizontal intersectino

prisma kraken
#

what exactly are you trying to do?

rustic patio
#

one like this, but vertical

#

if u just connect directly it turns into this abominaton

#

smaller than the horizontal design but a billion times more ugly

image.png
rustic patio
prisma kraken
#

give me a few minutes to noodle that

#

i don't think you need all those connections as you have them

prisma kraken
static zenith
# rustic patio

in this first picture you have two seporate interchanges side by side. each track allows you to turn or go straight through but not allowing for cross over from left to right track.

in the next picture in the vertical setup. you are connecting the top track with the lower track. allowing for cross from left track to right track.

it looks crazy because youve added another crossover in the junction.

rustic patio
rustic patio
static zenith
#

if you want the verical setup to function the same as your horizontal setup yes. you have two tracks, they both interchange but they dont cross over.

if you want to cross over. do it at another point.

rustic patio
#

hm

prisma kraken
#

i think what you want instead is something like

image.png
rustic patio
static zenith
#

or look up full clover interchange

wind spade
#

never clover

static zenith
#

never clover? xD

wind spade
#

turbine or nothing

#

clover is 🤢

static zenith
#

oh, im just thinking theory

prisma kraken
#

the other thing you can try is to only make the turns go in one direction and then after the intersection provide turn-arounds for changing direction

rustic patio
#

hmm

#

would it work to connect top to top and bottom to bottom?

#

i had the feeling it would cause issues and not act the same as a regular horizontal t intersection

static zenith
#

first ... this needs to function properly ?

prisma kraken
#

so just screwing around in Paint for a sec the only tracks that need to change height are the green ones

image.png
static zenith
#

i still say change height at another interchange. . too much in one

prisma kraken
#

now, if you just did this instead , that becomes a lot more straight forward

image.png
wind spade
#

also a lot more limited

prisma kraken
#

yeah, the loop around would be a congestion point

deft lichen
#

imo the best way to handle this would be to move each track aside and just build it level

#

the loop would take up way more space and be a bottleneck

prisma kraken
#

still, this is a rail network in a game where you're not going to have huge amounts of traffic, it isn't the jersey turnpike

deft lichen
#

you CAN have a lot of traffic if you make it so

prisma kraken
#

you can artfully as well look at your rail topology and see that you're only ever going to have 1-2 trains travel on a certain path and realize if you orient an intersection like that correctly, it doesn't matter that there's a bottleneck

rustic patio
prisma kraken
#

from my experience with using a fair number of trains in the game, congestion really doesn't start to matter unless you absolutely do the worst possible designs ever

rustic patio
#

im not really worried aobut congestion

#

i want it to look good and take up relatively little space

prisma kraken
#

i think Casual sort of marked up same idea i was having

rustic patio
#

aah oki

prisma kraken
#

if there's a difference, i can't see the nuance 🙂

#

but ya know, if one spur is going to your nitrogen plant, and the other spur is going to your plastic plant, there's never going to be a train running the pathway between them unless you're self-driving

#

if you build things with that knowledge in mind, sometimes you can cheat 🙂

primal flicker
#

How hard is it to do a 3way over/under with rails?

RDT_20231210_1520487102811823789273781.jpg
scarlet path
#

figured it out

image.png
#

went from this to that

image.png
prisma kraken
rustic patio
#

this wouldnt work, right?

Screenshot20231210-21402700000.png
prisma kraken
rustic patio
#

ye it couldnt work

slim matrix
#

do i need to set the center to overflow?

Satisfactory_10_12_2023_21_42_32.png
rustic patio
#

you'd go out where u come in and come in where u come out

rustic patio
prisma kraken
slim matrix
prisma kraken
#

(you left that part out, had to piece that together by remember there was a copper node there)

#

for that (any, overflow) or (wire, overflow) will work

slim matrix
#

i set left to wire

#

and middle is on "any"

#

not sure if it should be overflow

prisma kraken
#

middle should be overflow

slim matrix
#

otherwise it chokes no?

prisma kraken
#

what any will do is split half to the wire port and half to to the overflow always

#

for you're setup, you're always going to be pushing down the line more than 60 wire/min, so it should accidentally work

#

the correct way of setting it to prioritize the cable constructor's input is (wire, overflow)

#

since you only have one item on the belt (any, overflow) would work and is less clicks to configure in the splitter

slim matrix
#

the thing is that its not only one item

prisma kraken
#

i try to be kind to my future self and make the filtering as specific as i can

#

1000 hrs from now, i may forget, lol

slim matrix
#

it would be 60 sheet and 120 wire, so the idea is last with smart to split the wire into cable

prisma kraken
#

then have it set to (wire, overflow)

#

when in doubt about the setting on a smart splitter, 99% of the time, overflow is the correct setting

#

things like 'any undefined' are really very niche-case things to set for very specifically funny things

wind spade
#

sorting in storage

prisma kraken
#

and some other sushi craziness

slim matrix
#

i see

rustic patio
vapid gorge
ashen stirrup
#

What exactly is the deal with recycled rubber/plastic alts? I see people talking about loops, but I'm not getting it

snow dove
#

to get the best of one, you need both

ashen stirrup
#

I have both, and all the other alts currently available to me

vapid gorge
#

essentially you make all oil into HOR > diluted fuel
make some residual reubber from the polly

fuel + rubber

#

to recycled plastic, plastic + fuel > recycled rubber loop reprocess materials as needed

ashen stirrup
#

Ok, I think it's starting to click conceptually

#

Just gotta work out what's going on with the ratios

vapid gorge
rustic patio
#

agHHhh this stacked rail stuff is frying my brain

#

i color coded it to try to wrap my mind around it

#

it doesnt align

#

on the horizontal one black goes to yellow and green

#

on the vertical one black goes to yellow and red

#

it would align if it always went from top to bottom and bottom to top

#

but thats not how cobalt built it and he has it working in his world

#

Horizontal:

Entry: Left (blue)
Exit: Right (Green and Magenta)

Vertical:

Entry: Top (blue)
Exit: Top (Red. Would be green if it exited at the bottom)
Exit: Bottom (Magenta, lines up with horizontal)
its the same for all other combinations
on the horizontal ones it always enters on one one side (left or right) and exits at the opposite side (left or right)
on vertical it always enters top or bottom and exits split, top and bottom

#

on the horizontal one it comes in on the left side (frim the POV of the middle) and goes out on the right side (from the pov of the middle)

image.png
rustic patio
soft trail
rustic patio
#

is it about the cosmetic aspect of building?

soft trail
#

You're designing a 3 way junction

rustic patio
#

im trying to figure out how to make it work

#

not how to make it pretty

soft trail
#

Still under "design"

rustic patio
#

design and architecture:
This channel is all about the cosmetic aspect of building. Think of this as the artsy version of math-and-meta. This is not another screenshots channel, but for discussion on the design aspects
math and meta:
A channel for sharing and discussing factory builds, material ratios, so on and so forth. Useful sites for making/calculating factory designs and outputs will get pinned.

#

id say it fits better into math and meta

rustic patio
#

holy fuck stacked rail is mindwarping

#

unless im so sleepdrunk that it seems a lot more complicated than it is

#

it feels like knoth theory stuff or somethign

wind spade
#

I mean the problem I can immediately see is where does traffic go, since you can't have left-hand or right-hand traffic

rustic patio
#

yes exactly

#

how does one solve that?

#

well u switch left/right to up/down

wind spade
#

not doing stacked rails

rustic patio
#

u twist it

#

but left and right changes depending on perspective

wind spade
#

you can't really twist it tho 🤔

rustic patio
#

so you need to chang up/down depending on perspective

wind spade
#

yeah

rustic patio
#

fi.nally someone that recognizes how weird this is 😭

#

i think this is the solution

image.png
wind spade
#

well you can twist it but you can't say "upwards traffic goes -> and downwards traffic goes <-", because it depends from which side you're looking at it
this problem doesn't exist in left/right, because when you look on it from other side, you also flip which track is "near/far" or "left/right"

rustic patio
#

u can if u twist it hard enough

wind spade
#

how is in/out defined? 🤔

rustic patio
#

in/out of intersections

wind spade
#

oh like for junctions?

#

ah, I was talking about straight lines

rustic patio
#

for straight lines the answer is obvious

#

its the same as on horizontal rail, its both

#

for a portion of it you are at the top and for the other half u are at the bottom

image.png
wind spade
#

yeah but what if this segment was 2km?

rustic patio
#

you need to switch up and down once within those 2km

wind spade
#

then it'd have a large portion when one track is up and one is down

rustic patio
#

always one switch between junctions

rustic patio
wind spade
#

and you can't come to that segment and know which track is in which direction

#

well obv yeah

rustic patio
#

im not talking about feasibility

#

but it is possible like that, right?

wind spade
#

but my point is that there's no definition there

#

oh yeah it's possible

#

just... the list of issues could make a book

rustic patio
#

but its thinner

#

(this server doesnt allow outside emojis for some re,ason)

wind spade
rustic patio
#

not for me

wind spade
#

(yeah, colored privileges)

rustic patio
#

aaah, fancy!

wind spade
#

it's more that people were abusing it so it got taken from them

rustic patio
#

making something rn

#

okay so

#

left and right is relative

#

so if we want to solve the mystery of how to do vertical rail we need to think relatively

#

drawing rn

#

or as the kids would say, cooking

#

ignore the fact thatg a rail is missing

#

The green arrow is the train. it enters on the left junction (pov u) and exits at the right junction (pov u)

#

if we follow the point of view of the train it stays right the entire time

#

but left and right is relative, and in this case the point of view that matters is the point of view of the junction

#

the junctions (red box) point of view is marked with blue and yellow lines

#

blue is left and yellow is right

#

the train enters the first junction
from the point of view of the junction, the train is on the left
the train goes through the junction and the "junction entity" turns to look at the train
the train exits on the right rail (from pov of the junction)

pov of the second junction
the train enters from the left side
the train goes through the junction
juction looks at train
train leaves on the right side

#

if the train were to loop back to the first junction where it entered the first time it would appear to the first junction to be on the left side again

#

usually we dont think about this because left and right is relative

#

but when you make it absolute (by changing it to up/down) you need to make sure it is always aligned to the POV of the junction

#

you always need to switch up/down inside the junction and between junctions

#

you would think it cancels out and u dont need to switch at all but that doesnt work

#

@wind spade what do u think? has the sleep deprivation rotten away my mind and made me deranged, was this obvious all along, or is it interesting?

vapid gorge
#

Either your brain is fried or mine is from the last few days.

Or both

rustic patio
wind spade
#

you always need to switch up/down inside the junction and between junctions
I think... you don't
obviously it's one way to do so, but...

if you have a junction that has a junction on all sides... then you can remove all the "side switches" around that junction and the junction would basically be used inversely (trains would enter down and leave up)

rustic patio
#

hm

#

im not sure what u mean

#

but i forgot to add an ought

#

if you want all junction "puzzle pieces" to be able to connect in any way you ought to make them always switch like this

#

obvsly u can still make something that works for your purpose without that switching, but if u want it to be "complete" then that switching is necssary afaik

wind spade
# 
  |
--+
  |  |
  *--+--
  |  |
--+
  |

+ and * are junctions (both are T junctions that you designed above)

* doesn't need the side switches around it, because it can just be used inversely (trains enter through bottom rail and exit through top rail) and nothing happens

rustic patio
#

oh yea, that wouldnt need the switching in between

#

but it would still need switching inside of it

wind spade
#

oh yeah

rustic patio
#

otherwise you could end up going in the wrong direction

rustic patio
wind spade
#

obviously this means more planning and make upgrading almost impossible (or painful), but it was just a reaction on "you always need to switch between junctions"

rustic patio
#

yes i forgot an ought clause

#

if you dont want to be could outside in the snow you ought to put on a coat. its not a fact that u need a coat, its just something that u should do to reach certain goals

wind spade
#

I was thinking if you could design a set of blueprints/modules that would connect to each other in a way that you could never build it wrong 🤔

rustic patio
#

good old humes guillotine

#

https://en.wikipedia.org/wiki/Is–ought_problem

Is–ought problem

The is–ought problem, as articulated by the Scottish philosopher and historian David Hume, arises when one makes claims about what ought to be that are based solely on statements about what is. Hume found that there seems to be a significant difference between descriptive or positive statements (about what is) and prescriptive or normative state...

rustic patio
wind spade
#

I'm now thinking more generic, not constrainted to SF

rustic patio
#

aah

#

hm, im not sure

#

because you need to switch up/down once per every two junctions

#

so u cant have one switch per junction

# 
  |
--+---------|
  |  |      |
  *--+--    |
  |  |      |
--+---------|
  |
rustic patio
wind spade
#

I think you can

#

make the modules

#

you have these (it's all side view, just the T junction is kinda hard to do so lol)

blue and red are some markers or something

you're only allowed to connect a module to other module if two same colors don't touch

image.png
rustic patio
#

hm

#

im not sure if the first two would ever work

wind spade
#

building from it indeed forces me to put one (or three, or five, ...) side switches between junctions

image.png
rustic patio
#

oooh, u mean like as just straight pieces of rail

#

i thought u meant as an intersection

rustic patio
#

you could also put another intersection between there if you have one with inversed colors

#

but then the output of that is inverted too and it might make it messy

wind spade
#

I guess inverted intersection could be a thing as well

#

since basically every tile is passed red -> blue, so you could have stations and stuff

#

and it would guarantee proper running of trains

rustic patio
#

i mean, inverted colors intersection would be the same as a normal one but with colors switched since its totally arbitrary

wind spade
#

modules could even have signals in them built in, because there's only one way they can fit

rustic patio
#

yep

wind spade
#

with the colored blueprints/modules, I can see this actually being semi-usable

rustic patio
#

well

#

if u dont care about clipping it is really usable 😈

#

so tiny

wind spade
#

I mean nothing prevents you from having the switch modules and junctions non-clipping

rustic patio
#

it would make it really big tho :/

#

at least the intersections

#

(in comparison to the regular "inefficent" ones)

wind spade
#

well yeah. I was just focusing on "how to make this at least remotely usable so that people don't have to have a PhD to be able to build it"

rustic patio
#

would a turbine interchange built like this be significantly bigger than a regular turbine interchange?

wind spade
#

uh no idea

rustic patio
#

hmm

#

i think it would be very difficult because every connection changes up/down

#

so lots of stuff to avoid

wind spade
#

I mean worst case is "make tracks go next to each other like normal and then build normal turbine"

#

tho I'm pretty sure it can be smaller than that

rustic patio
#

true true

wind spade
#

just don't want to deal with 3D intersections with weird driving directions at 4 am 😄

rustic patio
#

relatable

#

also 4am here in germany

#

sounds like a fun challenge tho

#

would probablyu look cool too. and you would save space on the regular track length

#

i call it.. the ABOMINATRON8000

HighResScreenshot20231211-034833.png
#

and it works 🎉

HighResScreenshot20231211-035153.png
round talon
#

Does longer pipeline effect efficient of water speed per min?

median heath
#

No.

#

Just takes longer to fill.

round talon
#

It does for me, take like one second down. A bug, maybe.

median heath
#

Did you let it fill completely before turning on whatever it is feeding?

round talon
median heath
#

Screenshot of system?

round talon
#

What system? The whole pipeline?

median heath
#

What it is feeding.

round talon
#

Water extractor 240min to nuclear power 100% clock speed.

vapid gorge
#

could be any number of issues - it's why pics help

median heath
#

Pipe is full and Reactor is full prior to providing rods?

round talon
round talon
median heath
#

So that's a "no"?

round talon
#

Rod and waste are fine. Just water

median heath
#

Take the rods out.

#

Let the pipe and the reactor completely fill.

#

Then you won't have issues.

vapid gorge
#

what you're looking for is to always have the water buffer in the machine full, replaced instantly.

round talon
vapid gorge
#

then something else is wrong in the system - the way you built the pipes, the clocking of something Shrug

round talon
vapid gorge
#

a single wrong number somewhere can do it, or a weirdly placed pipe

#

and generally the only way to figure it out is with a lot of pics on here

round talon
#

Maybe issue with pipeline floor hole?

vapid gorge
#

that bug generally kills ALL flow, it'd be very noticeble

round talon
#

Wait. Taking screenshot, the line is long

true fog
#

MK2 pipes with ends at random places comes to mind

round talon
#

Only use mk1

vapid gorge
#

and you're just just feeding it directly into 1 generator right?

round talon
#

Yup

true fog
#

Does the generator use 300 water?

round talon
#

240 only

true fog
#

Overclock the water extractor to get 300 produced.

#

Check if the generator starts filling

round talon
#

Empty the line first?

true fog
#

Nah

#

You’ll be pumping in enough excess to notice

round talon
#

But that going to increase power usage.

true fog
#

Yes, but it’ll mostly be a temporary thing.

#

Worst case you can add in another extractor, assuming you currently only have 2, then underclock them in such a way that they need less power.

round talon
vapid gorge
#

that's the first step

true fog
#

Yeah, those are screwing with things

vapid gorge
#

Maybe - but they aren't helping thats for sure

true fog
#

Also giving you no gai-

#

Mobile typing is slow

round talon
vapid gorge
true fog
#

Are the buffers full?

round talon
#

Yup

vapid gorge
# round talon Yup

you're saying that the nuclear gens are emptying - if they are emptying its unliekly the buffers are all full

#

all the time anyway

true fog
#

Buffers are kinda wonky. If I remember right they only give headlift up to the amount they themselves have water

#

Judging from the pictures you’re having the pipes go even higher further down

round talon
#

Yes, but it always full and give 8m headlift

round talon
true fog
#

It’s likely they’re temporarily falling down at some point when water passes through, with in turn gives too little headlift for a small section of time and cascading down the line.

round talon
#

So, remove the buffer? It is not helping when startup the power system when it down.

true fog
#

Once you have the system up and going it shouldn’t go down. That’s where power switches and priorities come in.

#

But yes, remove the buffers.

#

If you get power issues then do it gradually through the system

round talon
#

removing the buffer are not the answer it seem.

new.png
lime wadi
#

cursed factory planning

image.png
deft lichen
#

Bolted 😭

lime wadi
#

bolting everything 😭

#

i need to know just how annopying screws are because ive always avoided it

deft lichen
#

Don't steel rods+default screws work out better? I don't remember now

#

By resource efficiency, but it might take up way more space

lime wadi
#

this is an intentional learning experience

#

Yeah i went with this for the space

round talon
deft lichen
#

Don't use buffers

round talon
#

before, i use buffer. Now, i don't use buffer, but the issue seem become worse than before.

vapid gorge
#

could send me your save and I could have a look if you like

round talon
#

maybe it take time to stable, i'm not going to watch the switch for hour or more.

#

but i do use mod to test the reactor

#

It seem stable now. Maybe time is the answer.

#

Nope, still have issue

vapid gorge
round talon
#

Water extractor set up, might be

vapid gorge
#

yeah too many missing mods, wouldn't be surprised if something wonky is going on with that. Missing most of the map too? did just auto unlock everything?

round talon
#

Nope, i do use advanced game setting to get item for mam

#

before this factory save, i already make the same factory but different design, short pipeline which is nuclear power above water extractor. that one have not power issue.

vapid gorge
#

yeah sorry I can't diagnose it, but honestly you never know with lots of mods, maybe apparently clocking isn't what the true clocking is? maybe there's a setting that's gone wonky Shrug

round talon
#

So, buffer help?

vapid gorge
#

why would it?
According to you you preflood the systems before turning them on - once flooded yo uturn them on , and then over time they drain.

#

there's eithe ran issue with how much is being consumed, how much is being produced or getting teh fluid to the location

#

and you had buffers before

round talon
#

It drain for like 1-2 sec

round talon
vapid gorge
#

yeah you've got some sort of stuttering happening which shouldnt, especially with single feeding into it

#

It's very very weird is what I'm saying

#

and it's happening with more than 1 of them right?

round talon
#

Yup

vapid gorge
#

maybe try clocking all the water extractors higher? like I said maybe the numbers shown isn't really what is being pumped

round talon
#

I have my old factory save, the one with short pipeline. It have no issue of power. Logical think would mean longer pipeline come with stability issue.

wind spade
#

Buffers never help, they either do nothing or hurt you

round talon
#

I'm already remove the buffer, doesn't fix the issue.

wind spade
#

Well the issue is somewhere else then, but a buffer wouldn't help

vapid gorge
#

lots of mods involved too , maybe the modded extractors are being weird

round talon
#

The shallow mod, even without it. The extractor working fine. im Already test it before.

#

Not yet test on update 8. Let me test first.

proven fossil
#

me and my friend were gonna re make our "iron" facility and we needed a blueprint

#

nvm 4k screws 💀

#

Could someone assist me with this design. I want to only use caterium, water and the polymer input

image.png
#

But it's like too messy if someone has a better one

lime wadi
#

You mean 6,240 screws

lime wadi
#

Is there a specific amount of output you need, or a specific amount of input you want to fully consume?

proven fossil
#

I mean i just want to create a computer farm using the caterium alt recipes but it's just going to being annoying to split stuff

#

So yeah just clean numbers, I mean the thing is I don't want to make more plastic and using the 480 polymer resin I already have from my fuel farm

vapid gorge
proven fossil
#

like the way the polymer resin is being split, along with the quickwire

#

Bc i would need to underclock machine and stuff and am just looking for a cleaner design

vapid gorge
proven fossil
#

I mean i just would rather use a cleaner design

vapid gorge
#

you can split it in two just by having half of the refineries making polymer merged onto one belt and merge the other half

#

There's nothing not clean about it xD

#

OR don't split it at all - put it all on one manifold

proven fossil
#

Aight just wondering if someone has another design i could copy

proven fossil
lime wadi
vapid gorge
#

every single project is individual - the chances of someone making that specific number of computers with that recipe set and remember it and being here are basically nil. Keep tweaking the number of computers if you want numbers to change

proven fossil
#

For future reference how did you make it so not everything has decimals

#

Did you just tweak the computer output?

lime wadi
#

Yes. Check out the WIki and read into the ratios of each recipe. Caterium Computers hits clean numbers every 3.75

#

I'm.. greatly simplifying it but

#

In short, it'll do ya good to look at recipes and know a little math yourself

wind spade
lime wadi
#

Satisfactory Tools gets a lot better when youre willing to help it out and use it as—well, a tool, rather than your sole planner

wind spade
vapid gorge
proven fossil
#

Yeah especially when you have infinite decimals places

wind spade
#

You'll meet more and more decimals the more you play. It's easier to embrace them sooner than later

proven fossil
#

I just tend to find the second image to look nicer

vapid gorge
proven fossil
lime wadi
#

Some folks like to load balance across the whole game, I dont think its a wrong way to play.

vapid gorge
#

you can load balance decimals xD

wind spade
#

Just round up to 4 decimals and once in 100 hours there will be 0.2 second pause

vapid gorge
#

isn't it more like every 10k hours?

proven fossil
#

Nah I don't like load balancing i plan on using manifolds

proven fossil
lime wadi
#

Oh, sorry I got confused since you were concerned about splitting the polymer resin

wind spade
#

Idk, depends on what it is I guess, but it's practically not a problem

proven fossil
#

Also I wasn't even sure if that factory was optimal

lime wadi
#

If you're manifolding it, it shouldnt matter

wind spade
lime wadi
#

Computers are a goofy one. Some people opt for crystal computers

proven fossil
#

I would but then I need like 3 additional resource types which I don't feel like making

lime wadi
#

A lot of it depends on location, how much tech youve got, and how much youre willing to transport.

#

Yep, exactly.

#

Satisfactory Tools can't tell you all the answers, you gotta give it context

#

Just.. maybe dont copy me and use both bolted frames and bolted iron plates in the same factory.

vapid gorge
#

steel screws make it pretty straight forward

lime wadi
#

Its a fun ratio, yeah.

#

3 screw constructors to 2 frame assemblers to 2 plate assemblers

#

And if i had mk.5 on that save, it would all slip onto one belt!

ornate echo
#

anyone has an estimate how far i will get with 2 heavy frames a minute?

wind spade
#

to the end of the game 🤷‍♂️

#

(assuming you only build as fast/slow to use 2 HMFs/min on average)

ornate echo
#

how many should i make per minute then? its the most exepsnive item

wind spade
#

it definitely isn't

#

and make as much as you need now

#

if you'll need more in the future, you build more

ornate echo
#

i dont even have the manufacturer i just noticed

arctic willow
#

i got to endgame with 4hmf/min

true junco
#

For your very first HMFs I would suggest you just set up to run 1 manufacturer at 100% for now. Build more later. 🤷‍♂️

wind spade
#

depends how much you build trains, explorers, fuel gens, trucks, manufactureres, blenders, drone ports or nuclear plants

arctic willow
#

thing about hmfs is the only things they actually get used to build (that can be automated) are fused frames and adaptive control units; most of what you produce you'll end up hiking around on foot to build stuff

#

automating them is mostly about building up a decent supply

wind spade
long lance
#

Okay... so i'm making a station dedicated to several materials and I have a possible 750 Steel I can transport to the site. The whole factory there currently needs 705 Steel Per minute. I've been trying to minimize how many materials I need to transport due to the area having several options for materials. The whole place is going to have 10 Manufacturer's that I need in the building and 5 of them make 10 Modular Frame's Per Minute. While the other 5 are going to be responsible for 10 Computers Per Minute. I do not know how many train cars are going to be needed for Steel.. maybe 1 or 2 would be my best guess.

true junco
true junco
# long lance Okay... so i'm making a station dedicated to several materials and I have a poss...

RCVD FROM: Sevrahn

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

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

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

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

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

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

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

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

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

Fluid Trains
-107.08s RtD
-896.52 Items/min

#

from what i understand. this is the MAX thruput calcs for one train car. so, given that you wont have the exact RtD time, you can assume you will have a lower thruput than what is shown here.

that said. Ingots are stacks of 100. so MAX thruput is 1278.66 ingots/min. you want/need only 750. That's about 58.66% of Max. so you should have a pretty comfortable margin of error in your RtD to achieve that rate.

long lance
#

I'll just use 750 as max input for one. I can't make more than that or transport it all in or out of stations without using 2 Mk4 Belts to keep the throughput the same.

wind spade
#

I personally just do 1 belt = 1 platform

long lance
#

I suppose it's time to make 1500 Steel per minute in one location

wind spade
#

and if it's not enough, add another train

true junco
wind spade
true junco
#

thats what i thought. but i was concerned the original statement was interpretable as "just attach one belt per station", and somebody else might not realize the other steps involved.

incidentally. that's my Railway Freight paradigm aswell.

true junco
# long lance I'll just use 750 as max input for one. I can't make more than that or transport...

no reason to use mk4 if you have mk5. with that rate of material and a single Freight to car to freight railway setup, you will either saturate the factory where the material is dropped off and eventually fill back into the station, and thus the car will only ever be able to drop of what is actually consumed (the full 705 or whatever it was) or your factory will shut down periodically waiting on the steel to arrive. in which case you are either starving the loading station, or the RtD is off enough to prevent you from getting enough thruput.

long lance
#

I can't Mk5.. I have everything Except Tier 7&8

true junco
#

ah. well. idk how the train calcs change for lower belt speeds. shouldnt be hard to figure out tho.

figured it out. its simple enough. the thruput calcs all have 1560 in them. thats the max load and unload rate to/from the station itself with a pair of mk5 belts. so for a pair of mk4s we just do change the thruput calcs to have "960" everywhere they say "1560"

long lance
#

This station is going to be my central location for grabbing materials to make more stations/factory inputs and or outputs since I don't really have a main location for things.

true junco
#

understood. ive done the same occasionally. and it doesnt need to be a huge place to do it if you are gone for long periods. also if this is just making items for storage for buildables, and you dont sink overflow. it wont matter much when since it will all go idle once storage is filled anyways.

wind spade
#

the one belt per one platform works no matter the belt speed

long lance
#

It's made in one of the pits near the Grassfields so then the factory is under the whole platform rather than having it underneath the whole production area. I found a spot where the train network can connect in several locations and it was where my current station is.

true junco
#

if thats the case and they have mk4s. then they should use 2 freight loading docks per rail station, and 2 cars.

true junco
long lance
#

I can cap out my Steel production in this area to 1440 per minute if my math is correct with how much copper is needed and the Coal Nodes. I need to double check my math

true junco
#

because of the lake. i ended up doing Pure Iron Ingots and Solid Steel.

wind spade
#

you should probably figure out how much you need first 😛

true junco
#

you dont need anything tho

wind spade
#

well if your goal is e.g. "build 15 HMF/min", the there's amount of steel you need

true junco
#

yes. but some folks don't know what a reasonable goal is yet. you can only determine that based on how much is available. determining capacity first skips the erroneous step many folks make by deciding to do something silly like "make 1/sec of everything they can possibly make in the game"

long lance
#

I need 400 Copper ore per minute and 400 Iron Ore per minute for Iron Alloy Ingot. That makes 1000 Iron Ingots per minute and only 960 is going to be utilized for Steel. The Nearby coal is limited to the belts so that is also 960 Per Minute so that amounts to 1440 Steel per minute. I'm making it two loading docks so that's 720 Steel per minute for each loading dock.

wind spade
#

finding a place to build it is part of the process of planning, which eliminates those wild cases of 60/min FMF

true junco
true junco
# wind spade finding a place to build it is part of the process of planning, which eliminates...

well yeah. but thats all "after realizing what is reasonable" which is a step not everyone has taken yet.

even then tho. i like to look at all the "local resources" which is certainly an arbitrary selection of nodes based on what i think will make a nice factory area. then calculate various max productions related to what i want to make there. its just an exercise in setting upper bounds when designing a factory to achieve a particular output.

long lance
#

How would it fail to move that much materials if i'm using two belts per loading dock? That's 960 per minute and also given the time to move the materials that should be plenty. Of course storage buffers would need to be setup to make sure nothing is being clogged up at steel production so there is material ready to be loaded.

true junco
#

its impossible to move as much material as can be loaded into the station because of the station loading and unloading cycle.

#

thats why the max rate the 2 belts can load/unload a station is 1560/min for 2 mk5 belts, but the max thruput for stations is from the following options dependent on stack size (with optimum times chosen already)

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

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

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

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

#

also. i dont believe i have the max thruput calculated correctly in the earlier calc tho. so that needs to be redone. i just assumed RTD = TTF as max. and its not. lol

long lance
#

I guess this is going to be a learning curve then because I am only using about half of the steel i'm transporting.

prisma kraken
#

i'm not sure if that would be practical for 200-stackers due to the smaller max roud trip time

true junco
#

i meant it looked impossible to move that volume thru the number of stations with the belts he was using. but i also dont think i am using the calcs right. as im not getting sensible numbers for somethings. so im going thru the definitions of terms with Sevrahn and I am going to run it all thru excell shortly.

prisma kraken
#

the calculation is really derived from how long it takes to fill up 32 slots at 780/min

#

the way that math is presented on the wiki isn't the most helpful for understanding it all

true junco
#

basically. im seeing a mismatch for TTF<RTD where RTD-TTF<0.45133 basically its showing transfer rates that are very nearly the i/o belt capacity. but i believe its because a situation with RTD-TTF<0.45133 doesnt follow the equation because it wouldnt allow for the station to i/o the amount being transfered.

prisma kraken
#

this table is a little less confusing, imho Stack size: 50 Maximum items per trip: 1600 Mk1: 13:20 Mk2: 6:40 Mk3: 2:57 Mk4: 1:40 Mk5: 1:01 Stack size: 100 Maximum items per trip: 3200 Mk1: 26:40 Mk2: 13:20 Mk3: 5:55 Mk4: 3:20 Mk5: 2:03 Stack size: 200 Maximum items per trip: 6400 Mk1: 53:20 Mk2: 26:40 Mk3: 11:51 Mk4: 6:40 Mk5: 4:06 Stack size: 500 Maximum items per trip: 16000 Mk1: 133:20 Mk2: 66:40 Mk3: 29:37 Mk4: 16:40 Mk5: 10:15

#

where the x:y are mm:ss

true junco
#

times for what tho? table doesnt say

prisma kraken
#

they are the max time a round trip can be to sustain the belt throughput

#

same numbers

#

but you can interpolate that if you're trying to move a normal miner's output of 600/min, the round trip time has to be less than about 2:30

#

when i'm guestimating round trip times, my usual starting estimate for a round trip is 4 minutes

#

but that really depends on the distance, a loop around the periphery of dune desert, for example, that's about a 10 minute trip

#

also, those numbers are calculated using a 32-slot inventory for the train cars as theoretical max. In reality, you will never, ever, ever get a full 32 stacks in a train car, so more realistic estimates should use 31 slots

#

i've been meaning to write some code that does a more realistic set of calculations that would simulate multiple trains & lockout periods

median heath
#

If there is even 1 item left in 1 stack slot, you do not have TtF < RtD and therefore use the other equation.

true junco
#

ok. well. i am fairly certain i have figured out the issue. the conditions of the equations are...

  1. if TtF >= RtD

  2. if TtF < RtD

but it fails for all TtF < RtD where RTD-TTF<0.45133

in otherwords. condition 2 should be something like

if Ttf + 0.45133 <=RtD

because the equation that follows the condition assumes enough material has moved into or out of the stations to allow full loading and unloading. but that isnt possible if the RtD < TtF + HandlingTime

ie that equation only covers RTD >= Ttf + HandlingTime

median heath
#

You're misrepresenting RtD and that's why you're adding a third lockout time for some reason.

prisma kraken
#

changing the > to >=

true junco
#

well. the equation gave me the thruput of the 2 full belts. so it cannot be right

median heath
#

Give me the scenario.

true junco
#

RTD = TTF + 0.0001seconds

median heath
#

What is RtD?
What are the belt mks?
What item is being moved?

prisma kraken
#

its all kind of moot to use the second set of equations for when RTT (to use the proper terminology for round trip time), is greater than it takes to fill a train load

median heath
#

If you don't have a Roundtrip Duration you can't run the equations.

prisma kraken
#

the point is that you don't need the equations, just the max round trip time you can tolerate

true junco
#

thats what im getting at. the conditions allow you to enter numbers that do not work into the equations. its an issue of mathematical rigour.

prisma kraken
#

if the round trip time is greater, you have to double up your trains or use multiple cars

median heath
#

sigh.

You give me actual an actual, in-game scenario where it fails and I'll be open to it.
But you keep dodging that specific question.

true junco
#

condition 2 should start at an RTD that allows the TTF to actually laps. so RTD has to be at least TTF + handling time for the equation to be valid.

if RTD is any value between TTF and TTF+handling time. then the station doesnt have enough time to i/o a full car.

median heath
#

No.

true junco
#

im not dodging anything. im looking at an equations conditions that do not work.

#

where is the math is math gif when i need it. lol

median heath
#

If RtD is more than TtF "is any value between" then you use the other equation.

prisma kraken
#

i actually think the discrepancy is covered in something i've already said, the numbers should be calculated with a 31-stack car capacity, not a 32 slot capacity

true junco
#

the second equation is RTD>TTF the equation allows for RTD = TTF+.000000001 sec. this equates to very nearly 1560/min this isnt possible

median heath
#

I'm done. Enjoy whatever you believe you've discovered 🤷‍♂️ I don't care anymore.
Every time I ask for specifics you give vague replies and declare they are specific.

true junco
#

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

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

these equations do not line up at TTF=RTD and TTF<RTD for RTD-TTF<0.45133...

median heath
#

As I said, they have been vetted and graphed without any anomalies.

prisma kraken
#

have they been empirically verified? I've not seen that data?

#

graphing a theoretical equation != verification

true junco
#

Well. plugging in the equations with the given conditions plotted over Duration of round trip scaled to the Time to fill (ie Drt=0.01Ttf, Drt=0.02Ttf etc which means that the point wher Ttf=RtD = 1 on the X axis) gives me this graph.

note that it also goes crazy when Drt>Handling time because its not possible to have a Drt that is less than the 2 station lockout times. so i cut that data off as it was all negative thruputs.

image.png
median heath
#

Error found. Confirming data and will repost.

#

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

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

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

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

If TtF < RtD
Throughput = (Stack Size * 32) / RtD

#

@true junco new second equation.

#

Given the conditional is "station completely empty before new train", fill time and even belt speed are irrelevant.

#

So you just need Stack Size * Capacity (32) / Roundtrip Duration.

#

Let me know if you find something that breaks that.

#

Already let Ondar know to update the wiki page.

wind spade
#

you know you can update it yourself

median heath
#

No hehe

true junco
#

alright. that fixed it.

image.png
median heath
#

We dids a smort thing.

#

👍

true junco
#

yay! (i hope we didnt miss something... lol)

something doesnt feel right to me still tho. because now the max thruput is always when the Rtd = Ttf

is that right?(note i have Drt for Rtd in my graph... lol

median heath
#

now the max thruput is always when the Rtd = Ttf

Yes.

#

Max Throughput is when you get a frame-perfect drop-off.

#

Like the animation begins the exact frame the last item is loaded/unloaded.

prisma kraken
#

the question i have is when would you ever actually wish to use the calculations to calculate throughput for a delivery system that falls on the other side of the cusp?

true junco
#

idk. i only want to use freight exclusively for things that are far away. in the case that was graphed the inflection point is at a round trip time of only ~150 seconds.

true junco
#

alright. i sorted myself out. lol. i had forgotten that Time to Fill included the station lockout time. its why the graph starts at t>0 because any Rtd shorter than the lock out times gives a negative thruput... because it makes no sense to have an Rtd shorter than the minimum time for the stations to do their loading/unloading.

prisma kraken
#

my thinking on that, and perhaps i'm incorrect, is that if the rtt dictates that the train car fills up, you're going to add another train, not settle for less throughput

true junco
#

so... besides correcting the second equation. the "conditions" should state that Rtd must be greater than the lockout time. its obvious in hind sight, but not overtly stated. lol

prisma kraken
#

yeah, you're not going to get an RTT of less than the blackout time EVER, if the train stops at another station

median heath
prisma kraken
#

more difficult to calculate, and i've not seen the problem tackled, is how many trains can a given RTT support

true junco
#

lol i understand it. the results of the equations made sense there. ie. negative thruput for Rtds that arent possible. equation fails when impossible input is provided. this is acceptable and anticipated behavior.

the hitch from having the wrong Eq for the other half being resolved now is great. and the last bit of confusion i had was resolved by remembering that Ttf included the lockout time. lol

prisma kraken
#

i'm not sure the math on that is very useful, i suspect that for really any reasonably sized load you could construct in the game, 2 trains to sustain a rate of 780/min is enough for anything but a 50-stack item

true junco
prisma kraken
#

i ran into some real issues with modular frames, btw... i'm shipping about 550/min of them around on a train

true junco
#

the question really is, is there a risk of lowering your thruput with more trains. and i think the answer is yes. because you cause more lockout time.

median heath
prisma kraken
#

yes and no, of course you can gimp the transfer rate with too many lockouts, but just kind of thnking about it, as long as stations don't have a train docked for more time than they have a train docked, you'll be able to sustain 780

median heath
#

Answer is almost ALWAYS yes

true junco
# median heath When is this a question??

it is the question about "if and when" doing multiple trains on the same route is good or bad. my suspicion is that it is always bad for thruput except when your Rtd is very high. like, Rtd > 2xTtf or something like that.

prisma kraken
#

it depends on whether your transfer rate is above or below 780

median heath
true junco
#

well. yeah, i was just trying to be more specific about it.

prisma kraken
#

and i really don't think that these are problems that looking at the math on paper gives you the answer, you need to experiment with it

median heath
prisma kraken
#

like whatever you calculate isn't going to take into account the changes in travel time that will occur with your railways growing and becoming more congested

#

you just kind of have to get a feel for it, and say 'i can do 600/min from a single train car of a 100-stack item for a 4-5 min rtt'

true junco
prisma kraken
#

that's about what they can handle, 2x that for 200-stack items, don't even try pushing 600 /min of a 50-stack

#

if you're doing a long haul from deep dune desert to north forest, yeah, that's going to take 2-3 trains

true junco
prisma kraken
#

well, mean, of course you can create longer routes, but typically you're going to be shipping stuff more locally because, well, its cheaper by every metric

#

that was probably about the same distance as the dune desert crate to north forest

true junco
#

if its local, why use trains tho? seams like a lot of trouble to use trains over routes with Rtds shorter than the Ttf anyways.

prisma kraken
#

it is really easy to underestimate the size of DD

true junco
#

yeah. thats true. DD is pretty big. kind of feels like it should have been considered several biomes. it def has parts that feel very different. like the 3 big oases areas, and the rivers in the canyons that lead to the spire coast sea. they dont feel like Dune Deserts. nor does that very green river canyons area feel like its part of the Desert mountain plateaus either...

#

anyways. this was vaguely that old railway route...

image.png
rancid topaz
#

ten points to anyone who can guess what this is

image.png
median heath
#

Needlessly complicated.

#

Boom, 10 points.

prisma kraken
#

about twice as much storage as you're ever going to need 😄

#

in actuality, i'd guess some sort of item sorter that sits on the recieving end of a remote delivery system

rancid topaz
#

Well sort of, more of an ore sorter that will act as a buffer for the iron factory

median heath
#

Cool I was right 😮

rancid topaz
#

Ore comes in and is evenly split into 30's

prisma kraken
#

yeah, i have to reluctantly agree with Sev on that

#

does you no good to ever cache ore, pull it out of the ground at the rate you can process it

#

well, i lie, if you're speed running, there is a case to fill a container with ore before you have the power build the smelters for it

lime wadi
#

Thankfully satisfactory is not factorio

median heath
lime wadi
#

You only really need buffers if it serves a purpose other than stockpiling

median heath
#

Trains are the sole place you need buffers.

prisma kraken
#

there's a few things i'd consider caching for speed reasons like autowire

lime wadi
#

Train stations start and stop, yeah.

prisma kraken
#

atm i have a few buffers sitting on ore lines to measure mk5 belt drift

sacred orbit
true fog
#

It's hard to understand exactly what is going on in the picture. But if they provide 30 and you use two exits from a splitter, both sides will be getting 15. If they provide 45 then you can use three exits.

sacred orbit
#

the manifold belts are going to be feeding 240-270. there's three separate belts each manifolding into one merger

#

@true fog is this a better showcase?

true fog
#

Far as I understand, you're doing two exits from each splitter and combining them into a merger.

sacred orbit
true fog
#

If all input-belts have equal amounts, one really only needs to focus on one. For that machine to get exactly 15 limestone you would need each belt to have an input of 10.

#

Are you feeding them from three separate limestone-nodes?

sacred orbit
#

yes

true fog
#

I'd much rather recommend feeding the maximum amount of machines possible from one, for each node.

#

The cap being based on your belts, of course.

sacred orbit
#

that doesn't really work out well with concrete, since it's 45 limestone input.

true fog
#

How much is each node outputting, and what's your max belt?

sacred orbit
#

max belt is 270

true fog
#

Consider the nodes with max overclocking. :)

sacred orbit
#

the idea here is to divy that up into units of 15 per constructor, and feed one unit of 15 per belt into each machine, totaling 45 input per constructor/

true fog
#

But yes, I do believe your system will work

#

It won't work in the fashion you're thinking though.

sacred orbit
#

the long line of these in the back should cover 240 limestone per input belt

image.png
vapid gorge
#

are you just feeding items into constructors? this seems overly complicated even for load balancing

true fog
#

Say your nodes output 120, and you have 3 of them for belt A, B and C.
Splitter A, B and C will act identical to each other, so we ignore B and C while focusing on A.
A will get an input of 120 and distribute this equally to both exits (as two are being used). That is an output of 60.

The merger for machine 1 will then be getting 180 limestone per minute.

Moving to machine 2, the splitter A2 will be getting an input of the other 60. This will get split into the two exits yet again, each getting 30. Machine 2 will be getting 90 limestone per minute.

Machine 3 will get 15 from A3, 45 in total per minute.

But, you'll be filling these machines up down the line over time, rather quickly, and it'll even out.

vapid gorge
sacred orbit
#

it's sort of like this spaghetti setup i made before, except using a manifold setup into the mergers instead of a load balancing setup

Screenshot_2023-11-24_221201.png
vapid gorge
sacred orbit
true fog
#

If you aren't having issues with space, I'd much more recommend doing as Cobalt said, underclock to an even number.

sacred orbit
vapid gorge
# sacred orbit C O N C R E T E.

feed one belt along multiple machines - change the clocking of one or more of them if the number of machine's item use doesn't line up with what's on the belt

true fog
#

Our issue is, you seem to be overcomplicating. :p
But yes, your method WILL work ^^

vapid gorge
#

if all yo ucare about concrete this is a lot of work for nothing

#

if you do just like making wildly complicated things?

go hog wild

sacred orbit
#

and aesthetics

vapid gorge
#

eh ok, have a ton of belts everywhere I guess

true fog
#

Remember, some people like that. Tons down in #screenshots 🤭

vapid gorge
#

as long as the goal isn't purely 'make concrete' whatever really

true fog
#

You've seen the BP's I'm making for machinery. It's overly complicated for little return. But we each enjoy our own kind of aesthetics :)

vapid gorge
#

possibly? bp stuff tends to float past me for the most part

sacred orbit
true fog
#

But yeah, if all your belts and elevators are Mk3 and the average output of the three extractors are 240/min your design will work perfectly, Major. 👍

vapid gorge
#

Just overclock things if you want less space taken up imo

true fog
#

I tend to do some decent underclocking instead.

median heath
vapid gorge
median heath
#

Well then fuck you.
Respectfully.

#

🙃

sacred orbit
vapid gorge
#

but you sounded like part of your goal was to make something elaborate? which is also fine

sacred orbit
#

a little bit.

vapid gorge
#

my suggestion is purely based around only 'making the thing'

#

yeah if you want to do complicated stuff have fun. Everyone has their goals in the game

#

mostly I didn't want you to go to huge effort if all you wanted was a working system

true fog
#

That would have been my suggestion too. Underclock machines to fit with one node each, then expand later on as one gets better extractors and belts. But that's entirely moot when one wants the design. ^^

sacred orbit
#

part of the design that i wanted was to not need to mess with the speed of the constructors.

#

i find it a bit fun to work in ratios like that

vapid gorge
#

if your personal challenge is 'build stuff w/o clocking' ? sure. Sounds like a TON of work but if you like doing it you like doing it right?

true fog
#

You only need to mess with one, really. Set the speed and recipe you want, then Ctrl+C and Ctrl+V the rest.

It works out for me personally, because I'll be Ctrl+C/V'ing them all anyways, so adding in the speed at the start is like, sure.

vapid gorge