math-and-meta

[median heath]

I'm sure Marv will have it clipped when stream ends.

[frosty owl]

Ah, it's in today's

[median heath]

He explains that the took all they "it's bugged" feedback and genuinely looked into it.
Only to have the guy who made it explain the system is working as it is supposed to. And if you build hard-stop manifolds like that in real life, you get water hammer.

Which is why you don't get the full 600.
And why if you build other ways (such as looping, but he did not specifically say looping) you can, in fact, get the full 600.

He acknowledged the genuine bugs for things like floor holes.

But the 600 throughput on the mk2 is one of those "just build right" things.

[frosty owl]

That would imply that a non-looped system could work fine, so long as it's constantly flowing without fluid hammering, right?
Eg: system where the consumer never stops and fluid has no chance to completely fill up pipes

[median heath]

Uh... sure?

[frosty owl]

Pipe load-balancing FTW

[median heath]

Looping is the simplest solution 🤷‍♂️

People just refuse to not think about pipes in terms of belts.

[oblique hollow]

Other solutions include: switching to mk 1 pipes in places where high flow velocity isnt needed

And maaaaybe even valves if i can figure out how they really interact with all this

[frosty owl]

I'm not sure how that would help

[oblique hollow]

Reduces flow velocity

It is an actually valid strategy

[frosty owl]

Just trying to avoid "flow spikes", so to speak?

[oblique hollow]

Yep

Which means the old "the interruption happens at the first junction" is exactly the right explaination

Mk 2 pipe accelerates up to 600 in the first split and gets suddenly interrupted

[frosty owl]

I (unironically) think load balanced pipes might truly lead to the "safest" solution in that regard.
Not that other methods can't work, it's just that I think that's guaranteed to have to smallest flow spikes possibles

[teal tiger]

can i deactivate this fog somehow?

[oblique hollow]

Our pipes dont move and our pressure isnt quite the same as real pressure. So all that results in a sudden flow direction reversal

[frosty owl]

!wikisearch consolle

[brisk shore]

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

[frosty owl]

Using console commands. Check them out on the wiki.gg

[oblique hollow]

Note the "sudden" . Mk 1 pipes are inherently less dynamic and slower

[teal tiger]

everytime im entering my world?

[frosty owl]

I don't know if those settings grt saved or not, but it takes ~5 seconds to do the trick once you know how

[teal tiger]

cant open up the console for some reason

doing the damn ctrl shift L thingy

but the tilde button isnt doing anything

[oblique hollow]

Tilde depends on your keyboars layout

If its not a UK keyboard its lilely a different key than you think

[turbid cedar]

Because recipies will be rebalanced and possibly even changed

[wind spade]

that doesn't change most math done here

[turbid cedar]

The base equations, no. The hard numbers, yes.

An extra iron plate can throw off an entire production line

[wind spade]

yeah but most math here isn't "how to produce X", we have production tools for that

[turbid cedar]

Why are people so damn lazy (me when I just make zig zag rails cause I don't care to make them straight)

Do the math yourself

[median heath]

Unless we have to help with basic arithmetic

That's always fun

[wind spade]

oh yeah let me solve a linear problem with 150 constraints and 200 variables by hand, sure

[turbid cedar]

I just use a spreadsheet

It is sooooooo helpful

[wind spade]

spreadsheets can't really do solvers like that

[turbid cedar]

Well, what do you need solved?

[wind spade]

given X base resources and set of recipes, find a way to make most of multiple product, assuming I want equal amount of both products. Also choose recipes that lead to this max possible amount and show the production line

[turbid cedar]

Well then how the heck does the program do that?

[wind spade]

linear solver

you make matrix and solve that matrix

[turbid cedar]

Interesting. I don't know Matrix math yet.

[frosty owl]

What, you can't?!
Pfft, git gud
True pioneers make world-spanning factory plans all in their meaty calculators

[turbid cedar]

That is a cool project I should do

[frosty owl]

It's just math^2...

[turbid cedar]

Create a python script that I can use to calculate my factories, without having to do the excel stuff

Or I could make it also generate the excel equations . . .

[frosty owl]

Ever heard of SFTools? ^^

[wind spade]

that's literally SFTools 🙂

[teal tiger]

am i doing something wrong?

why is the first junction with a full machine but in the other junction its barely empty with 0.6 fluid inside the machine

[deft lichen]

the mere thought of this algorithm makes my brain hurt

I'll stick to turning JSONs to HTML...

[frosty owl]

From the little you're showing of the pipework, it looks like the horizontal pipes are full (first Pic) while the parts going higher aren't (second Pic, the pipe goes up).

• Might be a headlift issue (in which case the fluid should be piling up in the producers)
• Might be a waiting issue where the lower parts of the pipework (and connected machinery) need to fill up before fluid reaches the higer parts (fluid should not be piling up in the producers)

[north monolith]

will that kind of setup cause a water hammer or some backflow issues, or is it fine?

[wind spade]

what are the arrows?

[north monolith]

direction of the flow

[wind spade]

but... flow is bidirectional

also depends on amounts and verticality and stuff

basically you want to have any pipe manifold looped, which makes backflow irrelevant and water hammer non-existent (at least in the main pipe)

[north monolith]

that's a side view, so the water comes from the sides from WE, generally same level, then the main pipe goes down

Normally I wouldn't care, but I want to try to utilize full 600m3 with 5WE

[wind spade]

in general liquids want to go down first, so this should be fine-ish. But no guarantee, it depends on setup obviously

[north monolith]

yeah the question here is mostly if the direction of the flow is taken into account in junctions in "water collision" calculations, so not when they are reaching the end of the pipe (I know that causes water hammer), but if 2 flows that head into each other can "collide"

or if they are enough space to flow into, they will just go there without interacting with each other

[wind spade]

I don't think the simulation goes to such depths

[ebon pilot]

Is it a good idea to turn Blue Crater into a GPU factory?

(joke elaborated: I'm thinking of foregoing the other methods of creating Circuit Boards, and make crude oil the sole resource in making them)

[vapid gorge]

well things to consider

1. do yo uthink you'd prefer to use the oil for something else later?
2. do you only need CBs fro mthat oil?
3. is it convenient to whereever you're going to use the CBs?

in general I tend to avoid electrod CBs just because it's very expensive in oil per CB

[median heath]

Electrode pairs exceptionally well with Caterium Comp.

[vapid gorge]

cat board and cat comps also pair well

[ebon pilot]

After my nuclear power plant, I'm brainstorming to turn Blue Crater into a general Circuit Board supplier by directly feeding truck stations with locally produced packaged fuel.

[vapid gorge]

I mean you do you - I tend to just pick a spot that has the resources locally

[ebon pilot]

If I want pure output though, I guess I can fall back on Silicone Boards which can also be produced locally.

[vapid gorge]

Have you put through the numbers for how you'll be using them? working forwards is generally more difficult than targetting a higher tier component and working backwards

[ebon pilot]

I haven't put through the numbers. I'm still working on my plans.

[vapid gorge]

yeah I'd do a plan for probably at least the next step to figure out how much you'll need

[ebon pilot]

Is having 6 nuclear plants a good idea?

[vapid gorge]

if that's what you want? I can't tell yo uhow much power to aim for 🙂

though I would probably aim to only do a nuclear build once on a map. It's complex enough that I'd find annoying to do a second time and I don't feel it's the sort of thing you can just expand easily

[ebon pilot]

Well, I don't look to do a max power build. I wanna have power for automating the final space elevator parts, which I can foresee will cost a lot of megawatts.

[vapid gorge]

use sftools and set up a vague space elevator plan you might want to do to get an idea of the power requirements

[north monolith]

does headlift somehow transfer between the liquids through the machines? Because I can't explain what happened otherwise: I've got dil fuel blender tower, 240m height. Water gets the headlift from built in water loop going over the roof, HOR from the refinery tower. No pumps for the fuel obviously. Only couple of bottom floors are working because I am in process of connecting the resources. Top machines had no water, no HOR. And yet the fuel produced in the bottom machines filled all the available pipes, right to the top.

[oblique hollow]

Nope, it does not

Machines reset it to 10m on their output side

I can only assume you have a bunch of junctions in vertical alignment, right?

[north monolith]

yes

I will now test it without mods, because I disconnected the pump part of water lines and yet it went up to 240m straight from the machines as well... Maybe some mod screws up the machine lift

Lol, no, undmodded and water still happily reaches 240m

Yeah, even after flushing the whole network, it's already over 100m high

[oblique hollow]

in rare circumstances, junctions laid out like that can just forget what headlift is and give you basically infinite headlift

[north monolith]

lol

ok then, well it seems using junctions in place of floor holes gave some unexpected benefits

[oblique hollow]

Its not like you need it

Its a tower where everything flows down right?

all the fuel i mean

[north monolith]

no, waters comes from below

[oblique hollow]

I only care about the fuel right now

[north monolith]

yeah, it flows down

but the water goes up too with the infinite lift

[oblique hollow]

yeah so its not like theres a benefit from it being able to climb up the tower when the top blenders will apply pressure later on anyway

... oh i didnt realize it also happened to the water

[north monolith]

yeah

[oblique hollow]

That wasn't easy to read out from before

Where exactly is this build at

West coast? Northern?

[north monolith]

I forgot the name, west islands with 2 pure, 2 normal oil nodes

[oblique hollow]

east?

[north monolith]

the "floor holes" are all junctions, on each floor it's twice like that

sorry, west

[oblique hollow]

yeah thats the west coast which is quite a bit below 0m world height

going from above 0 to below 0 sometimes causes issues for junctions

[north monolith]

Oh, this is very interesting

But I'm guessing it will all be fixed in 1.0

[oblique hollow]

We dont know, we can only hope

[north monolith]

I just wonder about the headlift, if they will change that one high pipe connected to a network provides the headlift for any other pipe in the network

I started using that in my towers, so there's always one loop going to the roof and then back down and connecting to all the others to provide the headlift

If they change something that will require each pipe have its own pumps, ugh.. I will have a lot of rebuilding

[turbid cedar]

But it would be cool to bang my head over and over to figure it out

[fallow vector]

it's an ~8 minute segment

[median heath]

[fallow vector]

would note that water hammer would occur if pipe segment is ever full when intake stops.

(well, full + flowing)

[oblique hollow]

From wikipedia's water hammer article

[fallow vector]

I'm assuming annotations in this context mean "McGalleon has ran tests" ?

[oblique hollow]

if i cross it out it doesnt work ingame or has no ingame equivalent

i crossed out "this doesnt help with water hammer but reduces damage" because we cannot damage pipes

and manifolds with short pipes experience this too so its been crossed out

[fallow vector]

what would a flywheel on a pump be?

[mystic moon]

No real analogue that I can think of

[oblique hollow]

water tower might be able to help by keeping head lift up to a really high level

as for flywheel - no equivalent

hydropneumatic device, for all i can think of, would be literally a buffer in front of every machine

[fallow vector]

is part of the issue that there's no indicator when water hammer is happening, vs. audio/visual feedback in real world

[oblique hollow]

or at least inbetween main pipe and a few side branches

[oblique hollow]

time to head to the QA site i guess

though detecting water hammer via code can only be a nightmare i assume

[fallow vector]

1.0 patch notes:

• Glass walls now go thonk again
• Mk.2 pipe issue addressed by having pipes also go thonk when water hammer occurs

[oblique hollow]

Anyway -
Certified (™️) list of things that help™️ with Satisfactory fluid hammer:

• Looping (reduces the flow rate through the pipes and creates a bypass, as well as allowing for a shock wave to dissipate in lines with spare capacity) or splitting the pipe
• Using mk 1 pipes where mk 2 isnt needed (lower flow velocity and more stable flow and acceleration)
• Apply more pressure via Mk 2 Pump (no confirmation from devs that head lift actually directly affects flow but tests have shown more head lift = gooder)
• Water tower maybe (same as above)

Unverified:

• Using valves or buffers to either reduce absolute flow speed or to allow for a pressure wave to be absorbed into a buffer

[fallow vector]

does water fluid hammer make water fluid exceed headlift?

like if you put a stand pipe in that fed a package/sink setup or a coalgen to syphon off the hammered excess?

[median heath]

Valves cause issues in pretty much everything I have tested.

[oblique hollow]

it doesnt actually cause a headlift surge

its a surge in whatever the heck pressure is used for flow rates

[fallow vector]

should it cause a surge?

[oblique hollow]

if it did its not readable from pumps

so we just assume it doesnt exist

[fallow vector]

one assumes that if one set up a standpipe that a non-zero value in a segment past the headlift would mean "it caused a surge"

[oblique hollow]

please rephrase

the above statement is nonsense

[fallow vector]

?

[oblique hollow]

not the same kind of surge

head lift is not confirmed to directly affect flow

so we must rule this one out

[fallow vector]

that's testable tho?

[mystic moon]

My best understanding is that there's a working headlift affected by pumps/buffers and a "flow pressure" affected by the fluid itself

[oblique hollow]

thats my current understanding too

i had an assumption that maybe headlift had 2 regions of effect: 0-1.5 m and 1.5 m to infinity
but Snutt replied back with a "nope, head lift doesnt directly affect flow"

so we can only assume that all it does is indirectly affect it - by making it stop when exceeding

[mystic moon]

Which is very interesting, because it makes me question how they calculate flow velocity

[oblique hollow]

there is no velocity, its just flow per second

pipes all have a fixed volume (well, a fixed area. area times length = volume)

[mystic moon]

Yes, was using that as the difference between a full pipe flowing 100 and 200 m^3

I assume that they work backwards against the current net flow direction, but I'm really not sure

[north monolith]

Have they ever said what exactly causes the technical limit of pipes to be 600m3?

[oblique hollow]

their code cant handle more in a stable manner

600/min is already at the limit

[turbid cedar]

Same story for belts, which is why there is no Mk6

A 250% overclocked MK3 miner on a pure node is such a waste

[oblique hollow]

It serves as mental torment

[north monolith]

It serves either as torment, or as an excercise to fight off your OCD

[turbid cedar]

They should just make it have 2 outputs

I mean, the model already kinda supports it

[oblique hollow]

tbh nah

I want them to double down on it

Just a big old "You know what, no, fuck you, more isnt better"

[turbid cedar]

Turbocharge shard: allows you to overclock your overclocking to 300%

[north monolith]

Yeah, I kinda cringe when I see these mods with 2000/min belts and so on

[turbid cedar]

Their new belt optimization might make it even more possible

[oblique hollow]

It might but honestly they already dont like things getting too big

[north monolith]

There's a mod that lets you put a splitter directly onto machines without crashing the game, allowing you for the full 1200 output

[oblique hollow]

You cant even use the current maximum of all iron already

[turbid cedar]

Like, How Can Mod Do It But Dev's Can't?

[north monolith]

Can't != won't

[oblique hollow]

Before a certain fix by the devs, 2000/min worked but was unstable

just because its possible doesnt mean they will do it

[turbid cedar]

BELTNADO LAG

[north monolith]

I wouldn't like that kind of shit in the vanilla game, feels too much

[frosty owl]

I notice a distinct lack of load-balancing there 😒

[oblique hollow]

first point:
loop or split

[mystic moon]

Because it doesn't actually use the same system as the rest of the belts, is super performance heavy, and isn't actually that reliable

[oblique hollow]

extrapolate the rest of the data

[north monolith]

my mother was a manifold, my father was a manifold, no way in hell I will balance anything ever

[turbid cedar]

You forgot one. Just don't have any shutoff. If it always flows, there is no hammer.

[oblique hollow]

Refer to balancers

[frosty owl]

You SOM!

[turbid cedar]

I couldn't build a working 4 belt balancer, so I just combined the 4 belts into 3 and used a 3 belt balancer.

[oblique hollow]

4 belt balancer is just...
split in 2, then split in 2 again

well, load balancer

[north monolith]

I've been building manifolds before I knew they are called that way, most natural way to feed hungry machines

[turbid cedar]

It was a 4:3 balancer but I was going about it completly wrong

[oblique hollow]

split in 4, merge back one

[turbid cedar]

My friend thinks that manifolds suck, are inneficient, and that you have to put in more material than you actually need

[oblique hollow]

or something idk lol

[mystic moon]

Your friend is wrong 🤷‍♂️

[turbid cedar]

I know

[mystic moon]

People love to completely fail to understand how they work

[oblique hollow]

tell your friend to try using overflow smart splitters, maybe then they'll understand

[north monolith]

Yes, because manifolds make some stuff disappear in the Abyss®, that's a legit concern ||actually ocd||

[turbid cedar]

I used absolute splitting in my coal plant cause I didn't want to wait for the generators to fill up, but I use manifolds in everything else

[oblique hollow]

unless you actually suffer from OCD, dont refer to your irks on perfectionism as OCD
If you do have OCD - i apologize, but too many people just throw that word around without thought

[turbid cedar]

For manifold input-output blueprints, you only need 6 for the assembler, not 8

[arctic willow]

horses for courses; if it's a 4 or 6 (or even 8) i might split it but i'm not going to mess around balancing 21 machines

[north monolith]

isn't already too late for that? it seems the word went into the common use long time ago and doesn't describe only the actual mental issue anymore

[oblique hollow]

its never too late to correct one's own ways

small adjustments like that, if taken seriously and also passed on to others, can have at least a local effect

[north monolith]

I can correct my own, but I don't feel I can have any influence on how the language evolves

I don't remember the exact term for the process, but the original words losing their severity with time is completely normal

[oblique hollow]

Language adheres to how people use it.
If you avoid oversuing a word for cases where it may not be best, you can at least locally preserve some of the intended use

and if we can lessen the trivialization of a condition people can actually suffer from - why not do it

words can lose meaning but also that meaning can be restored

its all up to how people use it

[north monolith]

Is that more an expression of hope and belief or are there examples of that? Because I can't think of any word that was restored to its original meaning

[oblique hollow]

It was more out of my understanding that "people use and make words, so people make the definitions"

Dictionary disagrees? Force the meaning in through repetition and popularization
Dont like a word? phase it out

Should you just give up if you see that a word is being used for something else? no

[north monolith]

I don't believe a conscious effort of even a group of people can beat the natural mass processes of language evolution

[oblique hollow]

But using it is part of that process

By not using a word in a way anymore and passing that onto others, you literally shape language

[north monolith]

Yes, but it's a drop vs an ocean

[oblique hollow]

So?

[north monolith]

So the ocean will assimilate it without care

[oblique hollow]

Thats about the same mentality as "Why should I vote?"

[north monolith]

Yeah, I don't

[oblique hollow]

Which is utter crap

Letting a mentality such as "what does it matter" spread is what lets things deteriorate

[north monolith]

For me voting is just theatrics to keep people in believing they have any freedom or influence left

[oblique hollow]

And for me refusing to accept that is motivation to also encourage others to not let up.
Let everyone else do the decision making for you and you've resigned all control.

[north monolith]

I do believe some things matter, but these two things you mentioned, voting and shaping language by individual effort, I consider impossible

[oblique hollow]

You dont have to save the world. Neither your country
It is- in fact- the small effects that add up.

Small communities can in fact do stuff. And so can they also use language and shape it

Manifolds? Balancers? Utter nonsense to the normal person

perfectly normal to the factory nerd

[eager smelt]

So I am a little confused on how to progress as far as factories go. I'm starting to get to the point where my non foundationed starter factory is getting too big and too complicated to not be planned more extensively. I'm nowhere near endgame, but from what I understand if I don't start expanding now it'll hurt later. But I'm not sure how to transition into larger factories. Everyone says build up for more space, and I understand that and can handle where to build for the most part, but what are the factories supposed to be? Do I collect every resource from everywhere and bring it to one place? Do I have a smelting factory, and a constructing factory? Do I have a dedicated factory for iron, steel, etc.? And how much rescources do I need to mine if not all of it? Do I use only trains or do I transport across the map with belts? A lot of people say just do whatever and experiment, which I can do, but I'm getting really overwhealmed because I don't want to start wrong and not be able to come back from it. I know this is a long paragraph and might not have a simple answer, so if there is anyone who has had this problem I'd love to talk about it. (sorry if this is the wrong channel to ask this is)

[north monolith]

Yes, in case of choosing products which encourage producents to make them more and give up the alternatives, sure. But voting... The politics in my country and in the others that I observe are just theatrics for money grabbing, no one keeps their word, and even if they did, most of their promises seem to be clueless and disconnected from what people actually need. There is no one to vote for. And when someone tries to go against it, he's later found "commiting suicide" or "dying in an accident".

[mystic moon]

Give me a bit to type out a good response

[eager smelt]

cheers

[median heath]

Do you have a picture of your current?

[eager smelt]

Not unless you want to have a heart attack

but yeah, give me a minute lol

[north monolith]

Ah, ok, only now the key word fell into place for me - "local changes". Yes, so shaping the language inside small communities, no matter what the general evolution brings. Yes, in that case I agree with your point.

[oblique hollow]

If you feel that way thats up to you but i wont resign like that.
If you wanna continue then its in #off-topic-general, i dont quite care anymore.
One last question I do have tho: Will you keep going with "the flow" using OCD in not-quite-applicable situations or will you try to actually preserve the meaning, if only you by yourself?

[median heath]

Bullet replies while you go take one:

what are the factories supposed to be?
Whatever you feel like.

Do I collect every resource from everywhere and bring it to one place?
You can, but it isn't recommended.

Do I have a smelting factory, and a constructing factory?
Entirely up to you. There is no wrong answer.

Do I have a dedicated factory for iron, steel, etc.?
Entirely up to you. There is no wrong answer.

And how much rescources do I need to mine if not all of it?
You are the sole person who can answer this question.

Do I use only trains or do I transport across the map with belts?
Entirely up to you. There is no wrong answer. Keep in mind Trucks exist, and Trains are not better.

overwhealmed because I don't want to start wrong
There is no wrong answer.

[eager smelt]

this is all my iron/copper and steal

[mystic moon]

So, the most important thing is to avoid upgrading a factory. Decide on a path of progression, like a hub milestone, then pick an item(computers, for example). Then, go find a location that's near one or more of the resources that you'll need, or just in a convenient location. Then, transport the raw resources needed for that product to the location, and build production from start to finish. If you need a product farther down the line, don't pull from your existing computer production, just build more.

[median heath]

This is a map.
This is not a screenshot of your actual build.

[eager smelt]

I'm not sure how to get a good picture with it all in frame

[north monolith]

If you are getting overwhelmed, keep everything at minimum. Don't even think of larger factories. My advice is: before any bigger project, try to find and refine your own logic and process of building. Because building large requires a lot of repetetive steps and if you have them as a process that you've trained on smaller models, it becomes much easier.

I was building various stuff until I figured out the best way for me is to build small modules for individual items in a designer that can easily be put onto each other as towers and connected, so that's what I'm doing and that's the process that allows me not to feel overwhelmed during big projects.

[median heath]

I disagree with your "most important" assertion. Upgrading a current build is completely fine.

[median heath]

Then take more than one? 😉

[mystic moon]

Maybe I should rephrase - rebuilding a factory for more production is fine. Just trying to cram more production onto the side or top of it usually isn't

[north monolith]

I will try to keep the original meaning, because I understood the point you were making and I respect it enough to put my effort into it.

[mystic moon]

Don't just upgrade the miner and then try to overclock everything

[median heath]

Yeah, this has a completely different meaning when you word it this way 😄

[eager smelt]

so in yalls opinion, what's the best way to get rescores from a-b long distance?

[median heath]

There is no best.

Unless you clarify a factor that matters most to you.

i.e. if the sole thing you care about is throughput, and nothing else matters -> Belts and Pipes beat everything and you will never touch any other form of logistics.

[eager smelt]

so why do people like trains?

[median heath]

Why do people like the color red?

[eager smelt]

This game is making me realize how un-creative I am😅

all just preference?

[median heath]

Yes.

[mystic moon]

Way simpler to scale and you can re-use a route once it's built

[median heath]

99% of the questions you have about "what is?" or "how do?" are completely subjective to personal preference.

That is one of the more brilliant things the devs have done with this game.

Ensuring there is no single, universally optimal answer to most everything you could wonder about.

[north monolith]

Creativeness comes only after you get your process and think less of what to do and have more space to think how to do it

[eager smelt]

Ineresting

all iron

copper and some steel

steel

[median heath]

Theft is wrong.

[eager smelt]

oops

[median heath]

Yeah, Foundations will change your life.

[eager smelt]

yeah

[median heath]

Example of my T1-2 setup:

[eager smelt]

what I want to do is build some factories is the sky, and put minecraft ingots as the front so they look like giant ingots

[eager smelt]

puts mine to shame😔

[median heath]

Skybuilder... I'm out. @mystic moon this one is all you now 💛

[north monolith]

Shame isn't a productive emotion, accept that you are just beginning your journey and will eventually reach the creativity levels you aim for. Get inspired by others, not ashamed.

[eager smelt]

Very good advice

[tulip fiber]

im coocked

[median heath]

Use Tools.

[eager smelt]

that look like and American setup to me🦅

[tulip fiber]

i need 2,635 MW to pack it :'(

[eager smelt]

yikes

[median heath]

Why are you packing it?

[eager smelt]

Thanks for the help/advice @median heath @north monolith @mystic moon, it has been very helpful and inspiring!

[tulip fiber]

i want to transtport it to my base i thought it would look cool

to have oil plant here

[north monolith]

All the best on your journey!

[median heath]

If you're solely going for looks, that is one thing.

Else, Fluid Trains are better than Freight Trains.

[tulip fiber]

but now when i see numbers i think i will use 50% for energy and 50% for production maybe less

[oblique hollow]

For anyone:
reminder that it isnt stealing if you borrow design ideas from others, try them out, adapt them and thus incorporate it into your own style.

[median heath]

I was more referring to how they spelled steel.. 😦

[oblique hollow]

oh right

couldnt know, the magic of editing

[tulip fiber]

but they have less space yes?

last time when i played they did

[oblique hollow]

Equal capacity compared with packaging

if you package, you need 2 cars: one for packaged fluids, one for empty canisters

if you use fluid directly, you can just use 2 fluid cars

[median heath]

Hi.
I wrote the equations on train throughput and the wiki page that explains train throughput.

So when I tell you that Fluid Trains are better than Freight Trains with the sole exception of Nitrogen Gas... you can either believe me, or not. I do not care.

[deft lichen]

@oblique hollow on the topic of language shaping above, a wiki page that's a rundown of terms would be quite useful if we could put one together cohesively

I mean the acronyms page kinda is that

[tulip fiber]

oh i forgot about empty canisters

[tulip fiber]

thx for help <3

[oblique hollow]

sure, we could.
And then link to respective pages if they exist
A Factory Glossary

[deft lichen]

glossary, exactly

[lone lily]

TIL you can get higher precision with underclocking using the percentages rather than item values

Also anyone else have issues sometimes with output manifolds when their rates match the output belt rates?

[oblique hollow]

shouldnt cause issues

unless your output belt isnt used up completely

then the machines fill up with items, but that also shouldnt cause issues

[lone lily]

What seems to happen in certain cases is the output belt would be full, and the first or second machine in the manifold will have none in inventory, but the couple furthest from final output belt idle with a full inventory as the mergers seem to prioritize the ones closer to the output

The two things I used to fix were to remove the stuff in the machine inventories and increase the output belt tier, but thinking about the math I didn't think it should cause issues

[oblique hollow]

usually the mergers prioritize slower belts - its just how the math for slow belt vs fast belt works out

machines in the back will have to fight more of these priorities of every merger

so you need to make sure that your output belt is actually used up

[lone lily]

Would increasing the tier of the belts from the machines into the mergers also work to fix that then since it would no longer favour the slower belt?

oh I was just wandering and found this exact thing happening here

left two are at 94% efficiency with a mostly full inv, while the right two are empty, elevator is mk 2. but otherwise everything is mk 1.

[oblique hollow]

i think you should check your production math

does the lift stop from time to time?

[lone lily]

It does not, each constructor outputting 30/m wire

[oblique hollow]

then there is not really a problem in that regard

if the lift is mk 2 and the upper floor uses 120/min, its all fine

[lone lily]

well it was a bit of an issue as the left two were idling occasionally, but you were right, I noticed a gap in the elevator (once a minute maybe?) so I increased the middle belt to mk 2 to ensure it would be fully saturated

and now the efficiencies are at 100%

I think it has to do with timings of the constructor outputs occurring at the same time, and that's maybe causing one of the belts to have to queue for a gap

[oblique hollow]

the mk 1 belt from the 2 constructors should be no issue

as they both make 30, so 60/min total

queue times are not a real problem

[lone lily]

yeah but that's the one that was stopping to let the other two merge 🤷‍♂️

[oblique hollow]

yeah but if belt capacity is sufficient timings dont matter much

[lone lily]

well this is why I was originally confused!

[median heath]

🤔

[oblique hollow]

*usually, the merger prioritizes the belt with less throughput compared to all other connected belts

[vapid gorge]

don't rely on the efficiency meters, they are made of lies. Look at the lights and make sure they don't go yellow is easier

[lone lily]

made of lies? 😨

[vapid gorge]

and it looks flooded, so it seems like you may not be consuming everything they are being produced and that would cause them to be idle

yeah the 'efficiency %' on the machines are not to be trusted.

[lone lily]

Definitely consuming 100% of the outputs

[vapid gorge]

they used to be wildly out of accuracy, no long clear how accurate they are now after fixes

[lone lily]

They seem to be accurate from what I've seen, I mostly just check is something is at 100% or it isn't

[vapid gorge]

when you click on the machines that go idle - is their output full or is their input empty?

[lone lily]

the constructors making the wires that idle are full, but the mk2 elevator never backs up and neither does the rest of the belt up top

[vapid gorge]

wouldn't trust that number, like I said, look at the lights for a bit when testing a system, but you do you

[lone lily]

can I trust my power graph?

[vapid gorge]

tiny blips on a power graph could easily not show up when a machine stops for 1 second

[vapid gorge]

are the machiens consuming wire starving ever?

[lone lily]

Nope!

well not for this case, other times i've seen it sometimes a little

[vapid gorge]

well w/o jumping into the game myself I'd guess you either

a) flooded the system and it hasn't run long enough to self correct with enough stutters
b) making too much wire for the process it's feeding

or
c) haven't stared at the row of lights on the consumers to properly trouble shoot your problem

[lone lily]

probably true for higher scale power graphs but I can see 2MW diffs at 2400 MW production

[vapid gorge]

from what you've said those are my top 3 guesses at the issue at hand

[lone lily]

the system's run for about 7 hours so probably not a) and I could show pictures to prove b) but its 120/m out and 120/m consumed (or at least thats what the machines say)

[vapid gorge]

it's very crowded and hard to see what exactly is going on if its like that other image. More trouble shooting I guess!

open up the consoles of the consumers. See if any are starving or close to starving

[lone lily]

Well from what I can tell, the consumers are either just barely starved or barely not, but the question is why they would be when I'm producing the 120/m and the final output belt is mk 2

[vapid gorge]

well at the end of a manifold the machines will just have enough of a buffer to never stutter when built right, and set ups that need high stack items like screws and wire tend to take forever to fill up.

but if they are nearly starved and you said your producers are stuttering back? sounds like a flow issue

want an easy trouble shooting step?

[lone lily]

Shoot

[vapid gorge]

take the fullish stacks of wire from the producers so they are no longer instsantly cloged
cram them manually into the consumers
come back in 20 minutes after doing something else.

if the consumers are once again starving you definitely know it's a flow or number issue and not just needing time to self correct

[lone lily]

yeah I did this originally when I was diagnosing the problem yesterday, but I also upgraded the belts at the same time, give me a bit to test to see if this alone fixes it without changing the belts

but also i'm drawing a diagram to better show the situation

[vapid gorge]

you didn't build the mergers and splitters directly on the belts did you?

[lone lily]

No, that would be a much more obvious thing

[vapid gorge]

well if the consumers are set up like the producers in your first pic it looks straight forward

[lone lily]

the consumers aren't quite the same, they're assemblers and because of ratios and stuff the splitter up top isn't even ratios

besides I like drawing diagrams ;D

[vapid gorge]

that's fine, manifolds handle anything but go for it 😛

[lone lily]

behold the diagram

also I was thinking of the building next door which has the weirder assembler stuff

so belts 1. 2. and 3. will occasionally stop, but the others on the bottom floor never do

I made a test setup to see if just removing the stacks will fix it/maybe help with solving the puzzle

OK so after about 2 minutes, the issue is already starting to occur, and while the left constructors started empty, they are filling up (albeit very slowly) and the right belt occasionally has a gap

[oblique hollow]

seems fishy

[glacial saffron]

I never understood how to use drones

are there some ressources detailing it ?

[mystic moon]

The wiki has pretty solid explanation I think

[vapid gorge]

so you emptied the wire makers and they started filling up again? are the products of the assemblers asll getting used? or sunk?

[lone lily]

you can see on the right all the wire is going directly to a sink in the new simulated test I made

I made it brand new, they all started with no wire

[vapid gorge]

all mk2 belts? cause this is weird

[lone lily]

everything going into all the mergers are mk 1. the only mk 2. is going into the sink

[vapid gorge]

make it all mk2

[lone lily]

well yeah there would be no issues then

if I make the one between the mergers mk 2 everything's fine

[vapid gorge]

xD

[lone lily]

but one wouldn't think it would have to be

It wasn't a question on how to fix it, it was a question on why it doesn't work xD

[vapid gorge]

I can't recall why but there seems to be an issue with fast enough delivery with mergers/splitters and things backing up. I dont' think it's an actual bug but you generally want your fastest belts especiall if the manifold is running at max throughput

[lone lily]

Yeah, what I was doing to fix it was just increasing the one belt by a tier, I was just curious as to if anyone knew what caused it since ethe belt rates/production rates should all be good

[vapid gorge]

could just be a way things get accepted round robin from different inputs and if you have slower input than output on the merger it could be weird?

[lone lily]

Its interesting because it doesn't always run into this issue in some setups, Ie. 4 smelters producing 120/m ingots

I thought it might be due to the double input for the mergers but I had a one side case like this for screws

[vapid gorge]

might also have something to do with burst output of items and wire I think bursts more?

[lone lily]

that could be it too, caused by things having a larger than 1 output

the wire makes 2 at a time, screws make 4?

[vapid gorge]

could be! and ingots are x1

like I said - it's not a common issue so I imagine a lot of people never run into it. Largely because everyone will generally slap down their fastest belts

combined with (I think) needing to run the manifold at tis max throughput and probaly burst items, it be rare

[lone lily]

what probably happens is while the rate is 30/m, that's an average, so you could dump all 30 at once and nothing for the rest or 1 per 2 seconds, in this case since its multiple at once when they get to the merger at that instant its more than it can hold and it causes a brief clog

[vapid gorge]

But I don't think it's a bug.

I think it's just an obscure mechanical issue

[lone lily]

me trying to somewhat optimize material costs for my blueprints 😔

well i'm glad we figured it out ;D

[vapid gorge]

remember - materials are infinite 😛 also keeping the same belts may mean you need fewer material types to cary around

[vapid gorge]

yeah sorry I must have missed you were using some mk1s as outputs

[lone lily]

all good, you weren't here at the start of the convo

[lone lily]

true but its nice to be able to visually tell belt rates/capacities based on type

useful for sushi things

[vapid gorge]

wdym? dif mk belts look diff

[lone lily]

I mean like, I can see that the total wire outputs >=120/m wire because it looks like a mk 2

vs everything being mk 3 or 4 and then I have to just know or check the machines

[vapid gorge]

ah as in how full the belt looks/

[lone lily]

and the max that the belt could have

cause when i'm trying to merge building outputs, if I know one building has a mk 1 output line it has at max 60/m items on it

[vapid gorge]

if it makes you feel better it's gonna be really rare even with sushi builds for you to use the very max throughput all the time

plus the items on the belts are all virtual. they aren't a good indicator of what's going on with them

[lone lily]

Yes, I like to try to get close to max throughput on my sushi belts though because it means less sinks and less belts to try to maneuver

[lone lily]

yes but its not the items on the belt i'm looking at, just the belt type because that caps the throughput

[vapid gorge]

close is fine and easy enough, you'll have a few less belts, but as for 'less sinks' you can just have the over flow all head to 1 sink

[lone lily]

unless the combined outputs i'm trying to put on the sushi are higher than one belt can handle (I only have mk 4 rn)

[vapid gorge]

well this is assuming you're using the items, at the start of the sushi you shouldn't be overflowing much at all and you're probably not turning on more than 1 sushi section at a time testing things

which means you'll only ever be using a very small amount of the sink belt throughput right?

and once things are balanced out it'll be very rare you get ANY going to sink

like there's a rare dupe bug that *might eventually mean a sushi belt sends something to a sink after balanced?

[lone lily]

basically how my setups are working is all building outputs go onto a sushi which leads to storage sorter, and once all the storage is full it gets sunk

not all the building outputs are used for further processing/parts

[vapid gorge]

ah well that's different then if it's not just hte final output

[lone lily]

for example, my heavy mod frames building adds the heavy mod frames onto one of the sushi's that goes to storage, and then once that container fills it will go to sink

but if I had something that needed heavy mod frames, I would split them off the sushi

[vapid gorge]

yeah if ou have multiple steps that overproduce from storage you would need a bunch of sinks if you want it running all the time

[lone lily]

The way i'm doing my stuff rn is rather than produce a specific amount for a single need, I produce items based on raw resource quantity, and then distribute whatever is needed and sink excess

that way I can get some for storage from the excess without needing specific production lines just for personal resources

[vapid gorge]

yeah I do similar going up tiers, I don't bother with sushi for it though especially since belt changes are a thing

[lone lily]

well I wanted to try them out since I was starting over, and besides then I can do this!

[vapid gorge]

yeah I just do sushi when it's very final and permanent stuff 🙂

[lone lily]

which is fair, they can be tricky to deal with

[median heath]

Sushi is not tricky.

[vapid gorge]

well more that I upgrade belts and lines as I go and sushi doesn't let itself to that

[lone lily]

if you're not careful about sinking excess and item rates going onto the sushi belts you can run into single item backlogs, which is more than you have to worry about without sushi

[median heath]

?

Item rates are irrelevant.
Sushi manifolds are incapable of stalling or breaking.

You can send the wrong item down the line and they will not care at all.

[lone lily]

only if you have an overflow line

[median heath]

"Only if you build it correctly."

Ok yeah? Not sure the point of that reply.

"Buildings only work if you connect them to power."

https://tenor.com/view/segundo-sol-novela-remy-vladimir-brichta-chocado-gif-12304506

[lone lily]

Whoa that got passive agressive quickly, not trying to argue anything here I love me some sushi belts

I didn't realize that overflows were automatically assumed when dealing with sushi manifolds

[median heath]

That's the fundamental principle on which they are built.

[median heath]

Balancer and single-input sushi are the complicated ones.

Sushi manifolds are the old Geico commercial of "So easy, a caveman could do it." Which is their beauty. Reliable, simple, unbreakable.

[lone lily]

What is single-input sushi? Like if you take a sushi belt and try and have it go in only one of 2 assembler inputs for example?

[median heath]

Or 1 of the 4 manufacturer inputs.

[lone lily]

Gotcha, is it necessary to use a programmable splitter for those or can you use a smart splitter in a clever way?

[median heath]

Poggers are key.

[lone lily]

If the sushi belt had only the items that the manufacturer needed, could you use a smart splitter with any or any undefined?

[median heath]

Ven does stuff like this ☝️

I will never, lol.

I think in that one it's all normal splitters too.

[lone lily]

Is that sushi load balancing with normal splitters lmao

[median heath]

No smarts or poggers.

[lone lily]

I was gonna say

[median heath]

Single-input sushi relies on balancing, yes.

[lone lily]

oh I guess because if you didn't load balance, you could get one or two extra of something blocking the rest even with overflow

[median heath]

Yes.

So THAT is complicated, tricky sushi.

Manifolds simple.

[lone lily]

yeah ok I agree there, but I think you could still say more complicated than a single item belt manifold

[median heath]

Doesn't mean you can't have fun with manifold configs though 🙂

[lone lily]

ok now that's pretty

[median heath]

Manufacturer version:

[lone lily]

reminds me of ||factorio|| a little bit

[median heath]

No need to be dropping f words.

[lone lily]

lol my bad

I appreciate the right angled conveyor turns ;D

[median heath]

Just because you said that:

[lone lily]

45 degrees is still ok

I like how you just have these images at the ready

[median heath]

Given my name, are you surprised I have an entire portfolio of sushi configurations?

[lone lily]

Of course not, also those last ones seem annoying to set up with the raised splitters (until you blueprint it that is)

[median heath]

Yes, you would want to BP the Manufacturer one so you only built it the one time 🙂

[lone lily]

Except when I have to constantly go back and fix all my blueprints because I forget a belt or something

😔

[median heath]

Skill issue

[lone lily]

Never said it wasn't 😉

[median heath]

I will say the arguable most powerful use of sushi is Steel. But we're getting into #design-and-architecture so I will hop down there to post that.

[frosty owl]

That's where a little known thing with programmable splitters come into play.
As you imagined, if all input items are provided at the correct ratios, a Sushi belt can be fed to a single input of a machine (this takes away overflow control though, with consequences on how fragile the system is!).
But what if one has enough items to feed MORE than one machine on one belt...? (Eg: merging outputs from one factory, one has enough to feed X manufacturers)
This is where programmable splitters shine as they can split said sushi belt evenly along all outputs (like having one normal splitter for each individual item). Smart splitters trying to do the same with "any" on multiple outputs don't achieve an even split

Feel free to @ me for more details 😁

[median heath]

*poggers

[median heath]

Also @frosty owl does this specific screenshot have poggers operating off-screen, or did you do this entirely with normal splitters?

[frosty owl]

That's just with normal splitters.
Saturated sushi relies on balancing and full belts unlike all other kinds of sushi (most rely on machine-clocking and/or overflow management)

[fallow vector]

May 7th, 2024 Livestream
Q&A: Mk.2 Pipe bug?
https://youtu.be/COJV4Ubn1Ac

[turbid cedar]

Are those SMART splitters?

[wild radish]

Yes, you can tell from the roof

[ashen stirrup]

I'm very curious about this. Would you be able to outline an example of what this looks like feeding multiple manufacturers?

[frosty owl]

Of course! There's a few screenshots in the server too, but the most clear example I can give right now might be in this video after 4:10 (there's a few example setups shown)
https://youtu.be/EiQuFIx9Rlg?si=9AleLra5UNyfvUzr

(Note: in some cases I'm showing a mix of normal feeding or sushi manifolding and single-input. Eg: the Manufacturers making Fuel Rods get the Cells on one input from the Manufacturers directly behind (1:1 connections) and the other 3 items through another input; the Plutonium Rods Manufacturers input 3 items in one slot and Control Rods in another, so the Control Rods can overflow (still from one splitter))

[ashen stirrup]

Thanks! I'll have to check this out again later when my head is a bit clearer, I think.

[frosty owl]

Feel free to throw @s for clarifications :D

[primal flicker]

Nicely done examples there

Overall, how much time/space would you say this Psplitter behavior has saved (or could optimally save)?

I've never messed with them outside of a couple of isolated cases.

[frosty owl]

It really depends on how one likes to place logistics...
Since I like to give them quite some space and prefer not to have stacks of belts, this cuts down the logistic footprint a lot (4x) compared to single-item belts and a decent amount even compared to sushi manifolds (~2x)
Also no overflow management

Technically speaking, each programmable splitter can equal (smart splitter for filtering+normal splitter to split item) times (number of different items involved)

[primal flicker]

The no-overflow fragility is what keeps me from diving into it tbh

[frosty owl]

Fair enough.
There's no way to have overflow control and use less machine-inputs than input items ^^

[wary tulip]

Yeah, that’s probably why I don’t sushi…. I rely on overflow.

[median heath]

Sushi manifolds.

[primal flicker]

That can still be sushi. Just not single-input sushi

[vapid gorge]

sushi is great with manifolds?

[swift vale]

Any link for what best alternative recipes to aim for?

[median heath]

Yes -> Best is subjective.

Best for you can only be determined by what you, personally, care about.

There is no objective "this recipe is better than that recipe." because every alternate is a tradeoff from one category to another.
So your personal preferences on said categories determine what is best for you, personally.

Anyone who has made a list of "Objectively Ranked Alt Recipes" is lying their ass off, as it cannot be done.

Making sure all recipes are different and not better/worse is one of the greatest design choices the developers have made in this game.

[true junco]

Sevr's summary is entirely accurate. I would add the following...

Objective analysis aimed at determining the "best" choice of recipes requires parameters. You can objectively determine the best recipe for a given set of parameters.

Note that subjectivity comes right back in when you decide which parameters matter to you.

Examples:

• ratio of a particular output per a particular input.

• output per weighted value of all inputs.

• thruput per machine.

No recipe optimizes everything. You have to choose what matters.

Its like the quality vs cost vs speed triangle.

You can get it cheap and fast, but it wont be good.
You can get it good and cheap but it wont be fast.
You can get it good and fast but it wont be cheap

[median heath]

You can objectively determine the best recipe for a given set of parameters.

You kind of said it, but just the be clearer than any crystal could hope to achieve:

The fact you have to set parameters makes it subjective, not objective.

[swift vale]

Thanks guys

[thorn bane]

dont listen to them
its subjective as in most persons like to progress where others want to do weird shit that they like
so if you are in the group of 90% of people that aim to actually progress and finish the game, then 90% of recipe choices will be the same
you can check the wiki for a good overview of a recipes breakdown, there is some reddit lists but they are not great
the best option imo is to ask in discord what recipe to us (for example should i use heavy encased frames? with 99% of "subjective" opinions being Yes)

[vapid gorge]

That is also nonsense

There are some recipes that are less niche than others true but why would you lock someone out of potential build styles based on what other people do?

reckless and poor behaviour

[vapid gorge]

In the end basically all recipes have their places and situations where they shine. There are recipes which often end up shining more easily than others. But only a couple that outstrip other options.

The rest you should keep an open mind about and keep in mind recipe options when looking at factory locations when you're planning things out

[median heath]

This is completely false.

[oblique hollow]

Reasons why alt recipe talk has and will always be pointless N° 6874

[median heath]

👈 👆 👉 👇

I just noticed that 3 of those are right hands, and exclusively the point left is a left hand.

[vapid gorge]

can't have the thumb being down

[median heath]

No, but you could have point_up/down be a left hand as well, so you'd have 2 of each.

[thorn bane]

idk man
trying to help > throwing your imaginary hands in the air going "mimi there is no right answer"

[vapid gorge]

'there is no right answer for all situations'

now if you want to talk about a specific situations and desire how to set things up? yeah you can make good arguments for which recipes may be better in those times and places

[oblique hollow]

Telling people to use a recipe for whatever reason is just whatever you think works - based on who knows what. Thats the same for everyone
If people were a little more specific about their constraints then maybe yes we could all single in on a possible recipe choice (or a set of recipes)

But thats never the case

[median heath]

Also no one threw their hands up.
We pointed out that parameters have to be clarified, as there is no universally correct answer.

[thorn bane]

wait you actually think you helped that person?

[median heath]

I know that I did.

[vapid gorge]

If someone said they needed a factory where stators are a big part but they wanted to reduce steel use?

Then you could recommend something like QW stators. but yeah you need specifics

[vapid gorge]

yes. Getting people to think critically about what they may need in different situations is far far better than trying to tunnel vision them to options

[oblique hollow]

That topic is a fruitless tree. Always has been.
This topic should never be "listen to me, i know what im talking about"

[vapid gorge]

unless maybe they give you a specific situation ofc, then you can break it down.

[oblique hollow]

You can discuss it with a person for like 30 minutes and maybe then they will have some clue what to do.
But honestly? Telling them to just try stuff out would help just as much.

[vapid gorge]

Eh, if they come up with an end goal and a few locations they're looking at you could help brain storm what recipes might suit them fairly well

[thorn bane]

i feel like yall forgot what it feels like to get 3 recipe choices that you have no fucking idea about
saying "its subjective" is just gonna lead to them choosing one at random
like throw a dice fucking random
and at that point, having a resource that puts those into tiers on how usefullthey can be is a huge help

[oblique hollow]

Yes. Literally throw a fuckin dice

Alt recipe choice / analysis as a beginner is pointless

[vapid gorge]

or you pick a recipe for an item you may want to work on now or soon.

[oblique hollow]

Just pick one and try it

[median heath]

Any such resource is just "this was useful to me, personally" and does not necessarily apply to the context or situation of the person currently choosing the recipe.

[oblique hollow]

The best thing you can do is tell someone "these recipes produce the same item they depict, just in a different way. Got a recipe thats really hard? maybe one of these helps"

And fear of missing out is just as much of a reason people get paralyzed

so thats also why we tell em "you can get them all if you really want"

[median heath]

Tell me what you want. What you really, really want. 🎶

[oblique hollow]

heck theres a reason we tried this page:
https://satisfactory.wiki.gg/wiki/Tutorial:Picking_an_alternate_recipe

speaking of which...
@swift vale do you think the page linked above could be useful to you?
We are always looking for feedback for that article

[swift vale]

As soon as i have a little time ill see the page atm working

[frosty owl]

Most sushi (by players' use) actually does rely on overflow

[teal tiger]

is that a blueprint or an actual thing?

never seen a 8m foundation

[sand epoch]

...8m is the normal size..?

[oblique hollow]

every normal foundation is 8m wide

[teal tiger]

oh its about the wide

[snow maple]

all the analyses that i've seen say that turbo pressure motor is the worst alt for turbo motors, but i use it anyway bc nice clean numbers and i have pressure cubes anyway for nuclear pasta

[median heath]

Imagine thinking Turbo Pressure is bad....

[oblique hollow]

if you like an alt then dont let anyone stop you from using it

they all offer something

[snow maple]

right? maybe it is from a minmax standpoint but i like it

[median heath]

It isn't bad. Like at all.

[snow maple]

also caterium circuit, everyone hypes up silicone circuit but i hate the ratios on it

[median heath]

All 3 CB alts have merit.

[oblique hollow]

Minmaxers be seething whenever i mention Instant scrap because "oh no my sulfur" but then the same guys will guzzle 50000 m³ / min of Turbofuel

[median heath]

Instant Scrap would be in S tier if tiers mattered.

[snow maple]

im actually planning to use that alt for the first time on my current save bc im sick of refineries

[oblique hollow]

It has the nicest water loop in the game

1:1 water output from blenders to sulfuric acid refineries

[idle stirrup]

Here's an example of using 2 of them in different contexts: 1) I'm making a Computer factory and using the caterium computer alt, the caterium circuit is great to pair with it. 2) I'm making a High Speed Connector factory using the silicone speed connector alt, I'm going to pair that with the silicon circuit board alt (Could use the Caterium circuit here instead, but my preference would be to avoid importing plastic when I can use copper instead)

There are some alts that feel very niche or weak compared to others for some products. Like I haven't found a solid use-case for the Electrode Circuit Board recipe simply because of how much it takes in oil resources

[oblique hollow]

the greatest bonus of electrode is it needs no other resources.
Sure, it needs oil, but it simplifies some part of production greatly

the question is if that trade is worth it to you

[idle stirrup]

Yeah, fair. I do use the recipe - but its usually 1 machine at the start of the oil production tier, just to get circuit board production up for building materials

but in general, I believe that alt recipe was ranked really low in surveys, so players don't prefer to use it. To me it feels worse than even the base recipe

the trade-off doesn't seem worth it at scale

[oblique hollow]

best numbers for electrode i found was:
HOR + Poly alt = 150/min Resin and 60/min HOR
which is 75 Rubber and 180/min coke
enough for a fully overclocked ECB assembler
plus some more coke to spare (67.5/min)
so 90 Crude to 12.5/min CB

[idle stirrup]

from an energy standpoint, thats pricey

[oblique hollow]

burn the coke

202.5 MW regained

its not like you would get away with much less energy for normal circuit boards

[idle stirrup]

I don't know why you'd go through all that effort instead of just building a normal fuel plant instead, and just build logistics. The one consideration a player won't see in the wiki when comparing alts is the "time aspect" of playing the game. Presumably you don't have infinite time to put into playing, so some of these recipes feel a little like a diversion down a niche path that simply takes more time playing the game

[oblique hollow]

i mean sure.
Its 16.875 22.5/min HOR instead of 67.5 Coke that you could turn into fuel

[idle stirrup]

I'd bet this recipe is one that gets retuned - my guess is that it likely can be made more appealing if its per min production of circuit boards was increased

out of all of the 4 choices, its the slowest

[oblique hollow]

and the simplest

oil only

you could run half of a turbo blend fuel blender with the leftover HOR 😂

750 MW

[idle stirrup]

another consideration is more aesthetic - for me I'm trying to avoid the "Refiner" spam of the game which you can fall into in the mid/late game depending on recipe choices

[oblique hollow]

refinery spam is why i never use the pure recipes

and why i also use instant scrap

[idle stirrup]

I use some pure ones, only because I like them better than the default. Like the Pure Caterium and Pure Quartz, simply because yu have no other choice

[oblique hollow]

Refinery spam is what you get for minmaxing resource efficiency

[idle stirrup]

but for iron and copper? Huge fan of alloy recipes

Whats interesting and what a lot of new players struggle with is even asking - what am I optimizing for? Is it resource efficiency, power efficiency, the efficiency of your time, space considerations, aesthetic considerations, etc. Thats why min/maxing for this game is complicated. However, you can tell how some recipes are more popular than others, because I imagine the player-base tends to optimize similarly (on average)

[turbid cedar]

Progress on the Python script: I have no idea what "return" is used for in functions

[oblique hollow]

it gives you back a value

if you have a function that can accept a value, then does something with it, return will give back that value when you do a function call

else you do void if you dont want a return value

[stark iris]

Just tested pyomo optimizations on satisfactory alternate recipes and it’s working great. It immediately used the 1:3 oil to rubber/plastic route when I set it to optimize for resources.

I might use this from now on

[median heath]

Electrode CB + Caterium Computer.

[vapid gorge]

pyomo?

did you manually set up a spread sheet with sets to do calcs or something?

[stark iris]

It’s a python library to use a standardized syntax to test different optimization models.

I ran the glpk model for this and it works fast

[vapid gorge]

oh, I guess? Is it more user friendly than sf tools? https://www.satisfactorytools.com/production

[stark iris]

I don’t have a gui yet

[vapid gorge]

Ah fair, well if you're curious sftools does that and lets you modify recipes on the fly, save plans, copy tabs.

[stark iris]

I’ve used it but I needed the results in code and wanted to map it differently. I found it easier to build a quick version myself than figure out the code in his git file

[jovial wyvern]

What is the calculation update rate in Satisfactory (the "tick")? Is it 1/30 second?

[vapid gorge]

@violet turtle share the link you're using?

[violet turtle]

https://www.satisfactorytools.com/production?share=Sc03W7td9kOpC24j9cnd

[vapid gorge]

so 1 heavy modular frame?

[violet turtle]

maybe 5

[vapid gorge]

what is your fastest belt speed?

[violet turtle]

mk3

[vapid gorge]

ok you know how manifolds work by now right?

[violet turtle]

tho i have stuff for mk 4

whats a manifold again? ive prolly accidently made one

[vapid gorge]

you make a belt in a line and feed a series of machines?

[violet turtle]

o yeah ive made those

i just didnt know what they were called

[vapid gorge]

right so you don't need to split off those exact numbers right?

so for hte first step

145 ore to smelter
102.5 to foundries

just have 1 belt feeding those smelters and foundries and it'll self balance

you can do that for every step if you want.
But after the ingots you could clock sections of smelters or foundries into groups and manually split it that way

[wind spade]

Could have just asked 🙂

[oblique hollow]

The factory simulation doesnt have a time limit on the lower end, so it can be as fast as your pc can handle. However, it is clamped to 30ms. So if your game runs below 30 fps, they clamp factory simulation to 30ms and start doing sub-steps.
That means that game time and factory time process slightly differently there, where the factory tries to keep up on time and your gameplay itself might be more laggy.

[grizzled wigeon]

from which location mining bauxite would be good?

I have just unlocked aluminium and omg there is a lot of thing to do

[oblique hollow]

Anywhere there is bauxite is usually good?
Processing bauxite always needs water, so you should move the bauxite ore to water

[unborn schooner]

im at the same stage as you im making my site by a large lake next to 2 normal crystal nodes

at 51,99

[cinder silo]

If you don't mind epic causeways with belts or multiple trains, mine the bauxite anywhere and just ship it.

The whole point really is logistics, there is no way you would be expected to have like seven separate bauxite processing facilities.

[primal flicker]

The way the bauxite is in a belt across the middle of the map makes it the easiest resource to centralize, imo.

[cinder silo]

In my case, that is literal, here's six bauxite mk5 belts running down a tunnel right across the map to the refinery on the east coast.

[stark iris]

Oh hi lol

Did you use a solver package? If so, which one are you using?

Since this is just a LP, I’m looking at trying HiGHS

[wind spade]

lp_solve

There's a thread on Tools discord, if you want to join, you cab ping me there and I'll link you the thread with all the info on how to use Tools API

[magic island]

by default bauxite needs to combine with water/coal/quartz, and there aren't any locations where you can get ALL of those right next to each other. you'll have to decide on a location that's convenient to bring it all together

I usually bring the bauxite to the location of a coal plant. alt recipes can also change/simplify the resources involved

[median heath]

Any location that has bauxite is a good location.

[true junco]

Most of the bauxite is near water and coal regardless. And more oil than you need to max out bauxite refinement is also very close to the middle of all the bauxite.

And quartz is entirely optional imo. Only needed if you think youll need more than 9780 aluminum ingots/min. Or if you want a lot of production of aluminum without tapping all the bauxite i guess. I dont think it makes for a smaller factory since the production rate is 2x from a building 2x the size, but then requires the silica production to feed it.

[arctic vine]

i got a fun math question unrelated to satisfactory! (not sure if it's allowed).

what is, reasonably, the mass of a sphere with radius r of a big, multi-million-inhabitant metropolis?

let the center of the sphere be at ground level (visual attached)

mass must include the ground as well

buildings, people, cars, asphalt and the whole substrate beneath

[true junco]

Your question is off topic. And doesnt make sense. Is it a sherical metropolis as the premis starts out, or is it a spherical carve out of a metropolis as you have drawn. These are not the same thing. One is the deathstar with billions of people crammed into it. The other once of sufficient radius becomes 50% rock a sliver of atmosphere and nearly 50% empty space. Where the human construction on the surface becomes increasingly negligible. This holds as reasonably accurate until the radius becomes large enough that the curvature of the planet starts to make the rocky proportion fall noticibly away from 50% until the entire earth is within the radius of the sphere and from then the overall density will trend towards the average density of the universe which is nearly 0.

[deft lichen]

For a sphere like on the image, that is with a reasonable radius, you'd first need an estimate for the mass of a skyscraper given its footprint and height and then estimate the amount of skyscrapers within. The mass of concrete, asphalt, soil and air in the remaining volume could be calculated easily afterwards

(and yes this is offtopic)

[arctic vine]

the second option, apologies.
it is a carve of a city
to account for the other issues you mentioned, let the center be at distance r perpendicular to the ground and let r = 200m

[true junco]

So the center of the carve out is 200m above the surface of the planet?

[deft lichen]

I understood it as the centerpoint touching the ground, so 200 m below and 200 m into the atmosphere

[arctic vine]

you were correct, i changed the conditions to make the question more coherent

[true junco]

So if the center of the sphere is at r= 200m above the ground and the sphere has radius r=200m then it contains no ground.

Now you need to define the city. And where it is in the city.

Mexico city has a huge population but most of it is 1 story buildings. Such a sphere would contain very little besides air. Vs newyork or tokyo where the centerof such sphere could be inside a building.

[arctic vine]

then let's say the center is in the middle of Shibuya City in Tokyo, Japan.
35°40'04.1"N 139°41'16.5"E

(picked a city at random)

[stoic gorge]

why is the calculator doing that?

[oblique hollow]

doing what?
not rounding?

[stoic gorge]

like why 563.333 coal and then 563.332

[oblique hollow]

floating point error

[stoic gorge]

so i should go just with + .001

and make 30 steel ingot for steel pipe?

[oblique hollow]

you should use what the foundries need

[stoic gorge]

ik it doesnt make almost any difference

but its kinda annoying

[oblique hollow]

hover over one of the orange blocks to see the actual clock speed

[oblique hollow]

youre gonna have to deal with that personally somehow

thats just how computers handle numbers depending on how accurate they need them

[stoic gorge]

ok thanks for explaining 🙂

[feral valve]

hello, diluted fuel seems to be the most efficient way of producing turbofuel for a limited amount of crude oil, is that true?

[wind spade]

yeah. However usually the recommendation is to not produce turbofuel for power

[feral valve]

why?

[median heath]

Because you don't need more than Diluted.

[cinder silo]

Turbofuels complexity and the consumption of sulphur, basic or dilute fuel can run the entire show easily up to nuclear if you want to build that.

[median heath]

Simple. Easy. Far more than you need to get to nuclear.

[cinder silo]

Back when the trade off was storing masses of nuclear waste, turbofuel was attractive (update 3), when they made it so you can reprocess and sink the waste, nuclear shot WAY ahead.

[feral valve]

yes, i want nuclear, but need to up the power productiong a bit

because of this

[wind spade]

then do just diluted -> power

[feral valve]

but i will consider the diluted power

[wind spade]

no reason to bring in sulphur and coal

[feral valve]

thanks for the help

[wind spade]

and you can repurpose the fuel for plastic+rubber later if you decommission the plant

[cinder silo]

Don't forget even if you built a rats nest plant as a temp measure, you don't lose out by decommisioning it and reusing the parts elsewhere.

[weak plinth]

might be worth keeping some fuel power around either way though since it has much less dependencies than nuclear to run.

[cinder silo]

True but not a rats nest plant 😄 , build a purpose built backup station in the blue crater for dealing with black start woes, and even chuck it on its own grid.

[median heath]

When using Diluted, for ease of planning:

Oil * 2 / 9 = Generators

[cinder silo]

Oh wow my save faux pas actually corrupted two autosaves and the manual one, what the hell.

[median heath]

Time for a Hub0 restart.

[cinder silo]

After 1,0 now, and I still want my existing world fixed for it.

[regal star]

Alright nerds: I’ve a problem for you! Perhaps this has already been asked before, perhaps, perhaps not in this fashion, nonetheless I’ve been cracking my head on this problem for over 6 hours and asked three (lazy, I believe) mathematicians to help me find the equation with no success.
So, we all know that if you want to maximize the efficiency of Plutonium Pellet production using only Uranium Waste as starting ingredient, we have to process the 75% of our Uranium Waste production into Non-fissile Uranium and then combine the latter with the remaining 25% of Uranium Waste. This is easy, the proportions are easy (thanks @knotty venture, or whoever at Coffee Stain decided for this ratio!).
However, I want to know (purely for thirst of knowledge) what is the equation by which we can calculate the percentage of Uranium Waste to convert into Non-fissile Uranium, so that we can process 100% of the Uranium Waste amount we produce every minute, to craft Plutonium Pellets.
I want to know the equation in the algebraic form, regardless of the conversion ratio from Uranium Waste to Non-fissile Uranium or their ration to produce Plutonium Pellets.

[median heath]

Huh?

[oblique hollow]

Can you please restate your question.
I dont get it

You want to know why its 75% and 25%? Or did i understand that wrong

[wicked tinsel]

20 non-fissile is equivalent to 15 waste, so bottom recipe needs 100 / 20 = 5 results from processing, which in turn needs 15 * 5 = 75 waste input

[regal star]

Uhm... basically yes. Why, algrebraically, is 75% and 25%?

[oblique hollow]

(Non-fissle needed by Plutonium Pellets / Non-Fissile made by Non-Fissile Recipe) x Uranium Waste needed by Non-Fissile Recipe = Waste neeeded by Plutonium Pellets (in the form of non fissile)

if we add that to the Waste itself needed by pellets, thats 100 Waste total

If we divide that fraction of the waste needed for Non fissile by the total number of waste needed, we get 75%

Because 100/min non fissile needs 75/min Waste (37.5 x 2)

and we need 25/min Waste + those 75/min for 30/min Pellets

so 100/min total
75/min comes from Non Fissile
75 / 100 = 75%

[regal star]

I've either misunderstood something or the solution is wrong 🤐
Been trying to solve it since you wrote the answer @oblique hollow

[oblique hollow]

0.6051 was already the answer

60.5% ish

[regal star]

oh

[oblique hollow]

Fraction of waste for non-fissile = 56.695

total waste = 93.695

56.695 / 93.695 = 0.6051...

x 100 is 60.51%

[regal star]

omfg it works!

Thank you!

https://tenor.com/view/frankenstein-its-alive-happy-excited-gif-5625959

[weak plinth]

you know that to make the most plutonium fuel rods from waste you dont make any pellets though right? you use the alt recipe that skips pellets instead. Asking just in case, if thats not the goal, then nevermind.

[regal star]

Yes, thank you, I read that on the wiki, but my objective is to sink the plutonium fuel rods to run a waste-free nuclear setup, so I'm maximizing the production of uranium rods and treating the waste so that I can sink it

[oblique hollow]

so you wanna minimize the amount of rods per waste then

[weak plinth]

ok then 👍

[regal star]

👀 Uhm... Do I? I mean... More rods, more points to sink... But overall I'm trying to pick recipes to balance the expense of rare materials to craft the plutonium rods and the energy needed to run the rod's production line

[oblique hollow]

well if you want to max out the points sure, more rods = gooder

default recipes are a decent choice

[weak plinth]

the question is basically whether you want to make the pressure conversion cubes or not.

of course you can sink the plutonium rods either way in the end

[regal star]

IIRC I want to use

• Encased Plutonium Cell
• Plutonium Fuel Unit

[regal star]

Yes. I think they will be a pain to make but I don't mind that. Efficiency at all costs!

[weak plinth]

Ah no sorry I got it mixed up. You can skip the pellets with "Instant Plutonium Cell" which allows you to also make more plutonium rods in the end. thats what I meant.

[regal star]

And the standard Plutonium Fuel Rod recipe is incredibly taxing on the energy grid, compared with the Fuel Unit one, likewise the Instant Plutonium Cell is way more expensive when it comes to the materials used (and the energy as well)

[weak plinth]

yeah it takes some aluminium, thats the difference basically

[regal star]

yes, I got that even if you mixed up the recipes but yeah: waste of aluminum not using the rods for energy.

[weak plinth]

its really not that much though even if you max out nuclear

[deft lichen]

just noticing that you're on the unmaintained Fandom wiki, the new wiki is https://satisfactory.wiki.gg

[oblique hollow]

Just know you trade manufacturers making fuel rods for assemblers making fuel rods, but you also trade simple parts mostly made in assemblers (except the plutonium parts) for pressure conversion cubes, which need blenders and manufacturers for their parts

[regal star]

oh, thank you!

[onyx jolt]

how do i get better beating things

[onyx jolt]

now ive got a refined plate problem now fuck

[regal star]

"The barrels stack to 500, which allows for 12,000 (200 hours) to be contained in a Storage Container or 24,000 (400 hours) in an Industrial Storage Container."
I'm pondering about perhaps using the more expensive recipe in order to use these rods to feed nuclear plants in the future... With an 8x4x3 blueprint of a total of 96 Industrial Storage Container there's enough room to store 38,400 hours of Waste production, which maximizing Plutonium Rod production should last for slightly over 1700 hours, if I'm not wrong

[cinder silo]

I wouldn't want to store the waste coming out of my reactor hall, that place produces 1050 waste per minute, as shown here:

[regal star]

but this is a Uranium-fed reactor.... right?

[cinder silo]

True, I reprocess the waste and sink the 5.25 resulting plutonium rods.

[regal star]

They would produce 5.2 Plutonium Waste/min

[cinder silo]

For now at least, with my over production of power I don't need to burn the plutonium, though should I need it, the rods would be enough to drive a decent amount of reactors.

[regal star]

That's what I was thinking of doing: sinking the rods until I need more power. After all seems that the plutonium waste amount is reasonable enough to be managed

Because building another 533 fuel generator setup is a chore

[cinder silo]

Tell me about it, I have one facility with 888 fuel generators, that was a drag to rebuild.

[viral ravine]

is that not how manifold works?

Each input belt gets saturated one by one 🤔

[violet turtle]

it shouldnt be backing up to this degree

the 3rd and 4th one are now too

[vapid gorge]

the output are clogged right? cause this is the input for something else and seems fine

[violet turtle]

i think i shall let it sit

[violet turtle]

No

[viral ravine]

yeah as far as input side goes, this looks like a normal manifold

[vapid gorge]

ah is the resin using machine clogged in the output?

[viral ravine]

AS i said, each input get saturated one by one, save for the last one in the sequence

[vapid gorge]

starving from water?

[violet turtle]

i think ill let it sit for now, since i dont have any actual fuel generators yet

[viral ravine]

[violet turtle]

Nope

[violet turtle]

Nope

[full dawn]

Wait, so the fuel gets clogged when being produced?

So are you starving your plastic refineries of resin?

[violet turtle]

No theyre being overfed

[viral ravine]

but they keep producing?

[violet turtle]

yes

[viral ravine]

that's a manifold

[full dawn]

Yea, where is your fuel going right now if you have no fuel generators burning it?

[vapid gorge]

Well it looks like you're trying to do some weird injection manifold in the image? you have two belts merging to feed 1 machine

[coarse copper]

Your belts are well organized

[violet turtle]

they go into a buffer, with a valve so they dont slosh n stuff

[viral ravine]

it literally is a properly operational manifold

First refinery got saturated on input, so did the second

The third will too and it will continue on, untill all the input is consumed

Talking about the resin here

[violet turtle]

oh im a dumbass

[full dawn]

Oh so it is just normal manifold operation is what Burntcatto was scared of

[viral ravine]

The need to saturate one by one is the reason why manifolds are slow to start up

[violet turtle]

im so dumb, my brain was thinking of all 5 plastic refineries as one refinery and was wondering why one or two were getting backed up

i thought the first one and two had the combined consumption of all of them

[full dawn]

https://tenor.com/view/there-cheer-up-buttercup-it-gif-21613666

[violet turtle]

i am too tired for this

[viscid swallow]

Manifolds take longer to get to 100%

[viral ravine]

If you have enough refneries to consume all of the provided input, you are fine. Just wait 😁

[coarse copper]

Can turn off all machines and wait until everything full

[violet turtle]

its 3 am for me

[viral ravine]

Not needed

[viscid swallow]

2am for me

[viral ravine]

It will eventually saturate itself

[violet turtle]

i made this

specifically for testing

eventually i willl have another one on the left for the fuel generators going into the main grid

[viral ravine]

I used to have power switched but i've since deconstructed them since I had no reason to power anything off

[violet turtle]

I have one on my ammo factory

its off currently

[viral ravine]

whys that?

[vapid gorge]

could jsut use priority switches

[violet turtle]

idrk

BUT it is time for the main event

the 34 fuel generators

[viscid swallow]

U can use the priority power switch to

[obsidian edge]

Ive got a real difficult math question:

Whats 1+2?

[tacit merlin]

4

[obsidian edge]

thank you

[wind spade]

that's #off-topic-general

[obsidian edge]

No, it's math

[wind spade]

it's not related to Satisfactory

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.

[viscid swallow]

With 624 copper ore that the pure alt makes 1560 copper ingots and 780 caterium ore makes 390 ingots pure alt to make 11,700 wire

And with just 390 caterium ingots makes 3,120 wire

@ebon pilot I did some math

[stark iris]

Hey @wind spade what is your solver's objective minimizing? Resources and? Do you have the weights for those somewhere I can see?

[wind spade]

only resources

[stark iris]

I want to test mine with your site to see that everything is working.

It seems to be right, but your weight for water is a tad lower

[wind spade]

https://github.com/greeny/SatisfactoryTools/blob/dev/src/Data/Data.ts#L33-L46

[stark iris]

Adaptive Control Units used Residual Rubber to save a tiny bit of oil, while mine chose Residual Plastic to save water

[wind spade]

my weight for water is 0 😄

[stark iris]

oh that's why

[wind spade]

my weights are based purely on resource availability, which obviously isn't for everyone, but it gives decent results as a default weight system and for future I plan to make weights adjustable

[stark iris]

Thanks for the link, that should help me test this

I did the inverse of global availability and then gave water 1/100,000

[wind spade]

yeah I'm thinking of having some small weight for water as well by default, but again, that's for future tools 😄

[oblique hollow]

Tiny weight for water would basically indicate "yes its free but there is some tedium associated with it"

[wind spade]

I wonder if that even changes something 🤔

[stark iris]

I does

My solver was trading saving like 3 oil to save like 100 water or something on that test until I matched your weights

[wind spade]

if water is 0.00001 then I doubt that

[stark iris]

I had oil at 1/11700

The inverse of the global availability

[wind spade]

oh

[stark iris]

Must have been 300 water

[wind spade]

I did [total amount of resources] / [amount of X resource] which should basically be the same, but I've normalised the weights so that iron is 1

[stark iris]

I see that

[wind spade]

and I was thinking of having like 10^-6 water as well

which wouldn't change a thing imo

[stark iris]

no

I do think I would rather save 100 water than 1 oil though. Maybe

[wind spade]

I personally wouldn't

that's why I want weights to be adjustable 😄

[stark iris]

I'm going to use power in the objective as well. Some percent of power saved is resources saved.

A checkbox for "Fuel type for Power" would let you solve for that

Set power needed to be produced by that type of fuel, and have the resources for that be needed in the background. Then use that to decide the recipes to show the user.

[regal star]

Between High-Speed Connector and Silicon High-Speed Connector it's always preferable to go with the Default recipe unless you want to save energy, right? I know it's an obvious question but the alternative recipe is so expensive that I'm wondering if I'm missing anything.

[wind spade]

saves on quickwire and replaces cable with silica

depends on if you like that or not

[regal star]

No, nevermind, I got it looking at the two recipes again: the alternative consumes Quartz which is less scarce than Caterium

[wind spade]

"preferable" depends on player preferences, there's no "best" or "better" in terms of recipes

[stark iris]

I tested it against used with no other alternates and with other alternates, and I always get better resources with the silicon high speed connector

[wind spade]

with all alts it defaults to HSC

[stark iris]

hmmm

[regal star]

I always try to maximize efficiency, and my primary concern is to make the best use of the scarce resources, so it always looks to me that -overall- the alternative recipes consumes a lower amount of rare resources

[regal star]

what do you mean?

[wind spade]

"efficiency"? what kind of efficiency?

also, while some resources are rarer than others, you most likely won't ever run out of any resource, unless you do some huge project

[wind spade]

that was reply to Scott

[stark iris]

oh, ran it on my solver and it did HSC too

I was using an old tool

[regal star]

Efficiency as in to maximixe the amount of products I can craft with the available resources

[stark iris]

Must be your weights

[wind spade]

well the whole map is your available resources and you can't run out of resources on the map

(since your PC will burn before that happens)

[regal star]

will it? 👀

[stark iris]

https://www.reddit.com/r/SatisfactoryGame/comments/xutbq3/phase_4_alternate_recipes_ranking_w_spreadsheet/
This chooses SHSC

[wind spade]

this is very subjective and irrelevant 😄

[stark iris]

It is all math

[wind spade]

based on subjective parameters 🤷‍♂️

[stark iris]

The weights are subjective on your model

[wind spade]

yeah

[stark iris]

The same weights are used there

just water has weight

[wind spade]

but the reddit post assumes you want to make phase 4 items only

and I'm not claiming Tools are objective either 🤷‍♂️
there's no answer to "which recipe is better", it's all subjective and the person playing must decide themselves

[wind spade]

yes. Even best consumer PCs can't handle saves that would mine the full map

[stark iris]

Almost better off minimizing UObjects

(which that post also tries to do with buildings scaled)

[wind spade]

yeah, so don't decorate and build on the ground 🙂

in the end what the player optimises for is their own choice. Pretty much only relevant variable is time, everything other is practically infinite

[stark iris]

Alright, I'm going back to work on my program. Thanks for the weights to compare your results to, greeny.

I'm on a laptop on a trip and have very limited screen realestate.

[regal star]

I like that this analysis says that Pure Iron Ingot and Pure Copper Ingot are not worth because I've just finished building a 475-Refineries production line for those two items ☺️

[stark iris]

It gives a weight for water and power use

I absolutely dread building those refineries for pure recipes and finding water to connect, so I think it's a fair ranking for "time".

[stark iris]

I had a small bug in one recipe but now it matches satisfactory tools site results on everything I tested. 🙂

It makes all of the final parts using the exact same recipes and resource counts

[median heath]

Oh, another subjective as fuck tierlist to eviscerate. Fun!

[median heath]

Let's look at what they have in F tier-

Both of the highest conversion rates for Iron Ingots are not recommended. 🙃
The best ways to make Cable are not recommended 🙃
The highest conversion rate for Copper Ingot is not recommended 🙃
The most-recommended Battery alt is not recommended 🙃
The tied-for best Aluminium alt is not recommended 🙃
The only way Caterium Ingot should ever be made is not recommended 🙃
The highest conversion rate Rubber recipe is not recommended 🙃
Residual Rubber, which enables Recycled Rubber/Plastic, is not recommended 🙃
The only way Silica should ever be made is not recommended 🙃

[stark iris]

Now you're using subjective measures

The post is using a spreadsheet with the parameters given

[mystic moon]

I have to disagree with that ranking system I think

Really, it's impossible to rank alts

[stark iris]

I think what you disagree with the parameters for the ranking

[mystic moon]

Even if you have an objective system to rank them, use cases are subjective and entirely dependant on each individual save

[median heath]

The fact you have to give it parameters is what makes the entire thing subjective.

[mystic moon]

No
You cannot tell someone what the relative value of two recipies is to them without knowing the exact details of their save, build style, and plan going forward

[oblique hollow]

Truth ia you should just test and see what fits for you

[mystic moon]

All this list does is scares new players away from recipes that could potentially be very beneficial to them.

[oblique hollow]

You can tell someone the differenc between recipes but the value gained from that depends on the use case

[stark iris]

I think you're really confused. It's mathematical after the parameters are set.

So you disgree with the parameters

[median heath]

Again, the fact you have to set parameters is what makes it subjective.

[oblique hollow]

Well then yes they are bogus

[mystic moon]

F Tier (Not Recommended **Unless Combining Residual/Recycled/Heavy Oil)
Pure recipies

Does not compute

You can't just tell someone "don't use these" when there are very good reasons to use them sometimes

[stark iris]

It's relative to other recipe use

[oblique hollow]

Do not use X recipe for (whatever reason) is imo never a good thing to do

[mystic moon]

No
I disagree with the entire foolish idea of trying to rank them in a TIER LIST of all things

[oblique hollow]

Its better to list possible use cases instead

[stark iris]

It allows you to change the parameters in the link

[median heath]

"I personally chose the parameters that I cared about when making this list."

You have to see how that defines the list as subjective.
You have to...

[stark iris]

And I think the relative changes to resource use, power use, items, buildings is useful to know

[mystic moon]

The sheet is fine
The list is not

[stark iris]

You're arguing my point to the response to why he doesn't agree with it

[median heath]

I am making our point about why your list is harmful.

It is subjective as fuck, yet presented as objective.

[stark iris]

It is clearly labeled

[median heath]

If you think that works, you don't understand people.

[stark iris]

I think you're getting upset because you don't agree with the ranking, but all of the assumptions are clearly written out, and it allows you to change them.

[median heath]

🤦‍♂️

[oblique hollow]

I dont think those parameters are enough for any sort of general ranking

The only thing we can say regarding that tier list is : ranked based on... parameters for sure based on 4 arbitrary goals

[stark iris]

I can't find a place in that post where it says it was objective ranking

Which recipe is making you upset? lol

[oblique hollow]

We generally see people bring that list up for general alt recipe suggesstions or recommendations.

Its kinda the classic "tier list effect"

[stark iris]

It's a result on an analysis. You can't get upset with the result, only the process.

[mystic moon]

"Somebody ranked them, so I'll just apply that ranking to everything"

[oblique hollow]

Sure, lets just say then that the premise itself really odd

[mystic moon]

The process is fine
But presenting any recipie as generally better than another is risky at best

[stark iris]

I don't think it does that. It has a headline saying "for this case"

just above the ranking

[oblique hollow]

We saw what it did

Its not like this is a new post

[stark iris]

It's got to be years old at this point

Yeah, 2 years

[oblique hollow]

I mean why'd you bring it up in the first place

[stark iris]

I was using it to see where it put SHSC

[oblique hollow]

Its not like its metrics are applicable outside of its own premise

[stark iris]

It will give resources used compared to HSC and it isn't subjective on that

[mystic moon]

And you have thus fallen into the fallacy of the tier list
If you applied it to yourself in any way, it has become exactly the problem we said it would.

[stark iris]

It also gives the resources used similar to the greeny site, except gives water a small weight

[oblique hollow]

My issue is that people treated that very specific list as being generally applicable

[stark iris]

Resources*

[oblique hollow]

Greeny's also recommends you to use Cooling Device for Cooling systems

just cause it uses very little nitrogen

On that note:
Do you personally think that minimizing nitrogen is the best thing to do for that recipe?
Just looking for how people rank nitrogen

[median heath]

Nitrogen is like water.
It's impossible to run out.

[stark iris]

I get that there cannot be a general rule for all situations, but let's say about 80% of players just want a general list of what recipes are better to go for. Say they never want to look at another list or do any analysis. What parameters would you use to make that list more general?

[mystic moon]

I would not

I would tell them to just look at the reciepies they have avaliable and pick the one that best fits their situation

[oblique hollow]

"Better to go for"
I hate that phrase
Its not specific at all

[stark iris]

Well, if they want to save effort and time, there is a "better"

[median heath]

None. Because no such list truly exists.
Because better is subjective.
And any suggestion otherwise makes them worse at the game.

[stark iris]

Better to have a reference than no reference at all

[median heath]

False.

[mystic moon]

The goal in this game is not to do every project with the least effort/time invested

[stark iris]

You can I won't agree then. Some players just need a quick answer with little thought.

[oblique hollow]

Thats arbitrary decision making and wiki people like me have tried to cut down on that.

[median heath]

And giving them that makes them worse at the game.

[stark iris]

I don't thin kso

[mystic moon]

I have to disagree with that
If that's the case, they'd be better off just picking defaults for everything

[stark iris]

We won't agree

[median heath]

Ah yes. The world is a better place when people think less...

[oblique hollow]

Yes there will be no consensus