math-and-meta

[opal pivot]

I meant just hold the steel ingots until you need them in a recipe

Only build what you need for now basically

but remember that you have a potential for that amount

[past reef]

you don't need a bunch of steel beams and pipes though

[opal pivot]

Like if you need them in the future for Versatile Framework for example

[opal pivot]

could also even use them for more Iron plates

[turbid sapphire]

Something like that? or even less encased beam? like just a mk3 belt?

[opal pivot]

also, id wait until you have a better iron ingot recipe TBH

[turbid sapphire]

oh wtf i have all of them, i didnt notice it wasnt using pure at least

[past reef]

You won't have enough power for all that buildings seeing that you arent planning for heavy frame in tier 6

[opal pivot]

Yeah, Pure is a straight upgrade if you have water 😅

[turbid sapphire]

yeah this is compltely wrong one minute

[opal pivot]

I mean, they could. They could be doing it one step at a time

[past reef]

Man's in the desert all those water should probably already go into pure copper

[opal pivot]

There's a whole ocean to the east

[past reef]

I'd rather not drag the entire ocean up with 10 freight car for not even 2x iron yield

[honest wyvern]

ficsit would be proud

[opal pivot]

I mean, you can build down onto the ocean and take the items there, that's what I like to do.

[past reef]

That's reversing the direction does it make the process better?

[opal pivot]

Well, you aren't transfering liquids then 😅

[brisk smelt]

please never transport liquids un-packaged

its hell

[past reef]

I have a train for nitric acid for the well on the desert all the way down to north forest

[opal pivot]

I have trains bringing raw Bauxite all the way to the West ocean 😅

[past reef]

I just don't transport water for iron, bauxite or fuel is fine

[opal pivot]

If you package it you'll get a LOT better throughput btw

[turbid sapphire]

there we go, i cant believe i didnt notice that...

[opal pivot]

Ahh Im building a huge pure Iron facility atm 😅

[brisk smelt]

why would u even transport nitric, N2 has a 1:4 compression and water & iron are literally everywhre in the world

[brisk smelt]

remove leached ... if u need that many pipes u need that much sulfur as well

[past reef]

Cause a pure sulfur node is on the way and stopping there doesn't reach max allowed RTT?

[opal pivot]

I mean, I have a nitrogen gas facility where I transfer Nutrogen and Nitric out by drone.

[turbid sapphire]

fair enough that makes it a bit simpler for sure

[opal pivot]

Nitric isn't used in high quantities usually

[past reef]

For steel please at least do molded beam and molded pipe for large factories like that

[opal pivot]

and yeah, I agree, leached is almost never wirth the Sulfur it takes

[turbid sapphire]

thats looking WAY better guys thanks sm

[opal pivot]

which concrete recipe are you using?

My steel factory has loads of Silica nearby so Im using fine

but wet is still better than default

[past reef]

Rubber or wet are nicest depends on where you are

[opal pivot]

not fine?

[turbid sapphire]

wet concrete in its own factory

using all 9 pure nodes from rocky desert

current prod 2880, havent upgraded belts to mk5

[past reef]

Fine has same ratio as wet, I'm never in a spot for fine I suppose

[opal pivot]

Ahh fair

Although I wasn't aware they were the same

I have water there so should definitely swap over to wet then...

[past reef]

Idk I just rmb rubber for mega and wet for simple depot

Havent started in a nice quartz spot to make use those yet

[opal pivot]

I really need to set up some plastic and rubber trains, they're used in so many top tier alternates...

[past reef]

Think fine is still better than wet I hate refineries

[dusky dust]

Heh, yeah, I have of course been doing that as well. :D After being stuck for awhile it seemed prudent to ask in a channel where it's likely someone already understands it and could nudge me in the right direction, though. :)

[spare jolt]

what do you need so much steel for?

[turbid sapphire]

idek i just build mega factories for certain material types and then train/drone them around where i need

eg wet concrete factory

[edgy leaf]

why so many steel beams tho?

[opal pivot]

I love them. Bigger machines are less fiddly xD

[edgy leaf]

they're only used in plut rods, versatile Frameworks, and screws (if you use the alt)

they're also used in EIB but you're already making them with pipes

[opal pivot]

Please censor ||screws||

[edgy leaf]

I love screws

[spare jolt]

screws are the juice of a factory

[edgy leaf]

I want to make 5 million screws per minute some day

[opal pivot]

omg did they change discord censoring?

[past reef]

Use with bolted frame and bolted plate and heavy flexible frame for very small HMF factory

[edgy leaf]

it's always been ||this||

[opal pivot]

D: But think of all the belts!

[spare jolt]

who cares

the screws must grow

[opal pivot]

But that isnt working either xD

[edgy leaf]

put the screw constructors inline with the machines, no belt

[opal pivot]

oh ignore me xD

[edgy leaf]

because I wrote it like this: \|\|this\|\|

[opal pivot]

But then I can't have nice straight rows

[opal pivot]

I thought I tried || originally but I must have been wrong xD

[edgy leaf]

you did |this|

not ||this||

[opal pivot]

Yeah I thought I edited to that haha

[edgy leaf]

you can.

[opal pivot]

but the gaps wouldn't be uniform, my OCD could never 😭

[edgy leaf]

not sure what you mean, you can make the gaps uniform if you want

[turbid sapphire]

honestly just planned to fill two mk5 belts but i realize thats way too much now

[edgy leaf]

I love excess, but I prefer excess with a purpose

[opal pivot]

I mean, if you need 2 screw machines per next machine then you need 2x the amount making it not uniform unless you have really long belts

[turbid sapphire]

No you are absolutely right, i just dont really know the endgame requirements and recipe interactions very well at all

I let my friend deal with that but hes sleepy

[opal pivot]

I mean, you could sink it... But why? xD

[edgy leaf]

or you just ... leave space

or you overclock

[opal pivot]

sure but "space" is what I hate xD

[edgy leaf]

or you underclock

or you choose a different screw recipe that makes more screws

or you build vertically

there's infinite solutions, just takes some creativity

[opal pivot]

I DO need to get better at this tbh

[turbid sapphire]

unpopular opinion: screws are my favourite item

[opal pivot]

I usually just have a single resource per "floor"

Which screws mess up big time

[edgy leaf]

so true

my next playthrough will be screws only. you can make over 5 million screws per minute theoretically, I want to do that

[bronze kestrel]

Is there a knows bug with drone ports, once you selected the wrong port and cancel what ever you selected. It just randomly picks ports and goes wherever he wants? If not and its user error, i'm in need of some sleep

And those "relevant" ports makes no sense at all

[spare jolt]

Do you have several ports with the same name?

[bronze kestrel]

Nope, somehow an excisting port who was doing computer parts and was working flawless is now delivering where bauxide should be delivered.

i mean i have set up drones in the past. This is the first time its really screwing me over

Hmmm

When you have build a drone port, and delete the port while a drone was still flying, what happens with the drone and its "set route" ?

Cause i have drones on my map without a name

Think i fixed it.
Deleted the whole thing and started over.
Note to myself: Never ever, click the wrong port twice in a row.

[copper seal]

Anyone know why statisfactory tools doesn't accept the stitched plate alt recipe?

[bronze kestrel]

Currently using it

[outer vale]

turn off the default RIP recipe

[copper seal]

Satisfactory tools isn't dewfaulting to it

ah

[outer vale]

because it's judging it as less resource-efficient

[bronze kestrel]

Oh on the website

my bad.

[outer vale]

IIRC it's weighted by how much of each resource there is on the map. There's (very rough math here) about a third as much copper as there is iron, so it counts each copper ore as worth 3x as much as one iron ore. So for 10 a minute, the default recipe's 120 iron vs the alt's 50 iron + 100 "iron" equivalent, default costs less by that math

[wind spade]

if you want to force it, disable the default recipe

[copper seal]

Ye it works now thanks

had to disable the default recipe for it to work

[frosty owl]

They can at least split mixed belts evenly (with some limits though ), unlike all other splitters

[sleek jackal]

test

I have a question about Pipeline Mk2 . Lets say I have 1335 Fuel needed to be delivered to 14 blender. Since Mk2 is 600 liter max, do I need to do 2 pipeline or can I combine them all?

I means, 3 pipeline...

[brisk smelt]

any suggestions on how to make the main line have through priority

that part comes off a really nice downhill segment and i dont wanna hve the trains burn their speed slowing down for a red path signal

hmm delete some block signals downstream?

neverrr mind i figured it out

[spare jolt]

600 m3 is flow rate, not volume. if you connect them together, all blenders will still receive 600 m3 cumulative at most

[umbral current]

Hi fellow pioneers! I'm working on a web tool for production planning & logistics tracker, answering the good old questions: where are my somersloops? Do I have enought batteries? Am I producing enough caterium ingots?

I started with an Excel to track all my trains & drones, but lost hope after a while; so this time, since I wanted to finish this addictive game once, I made a tool since an Excel sheet is not something to be proud.

We reached v0.7.0 this weekend adding Somersloops & Overclocking support in the calculator, Transport Graph (to see where all your plastic is going) and Import existing plans from SatisfactoryTools. You can already find Power Generation, share games with friends, and many other things!

If you want to take a shoot or leave some feedback, I'd be more than happy!
You can find it here: https://satisfactory-logistics.xyz/

[sleek jackal]

thanks, so i'll connect them all to one pipeline and manifold

[spare jolt]

no, you will have to separate them. your blenders need 1335 m3 of fuel per minute, and the pipe can take 600 m3 of fluid per minute

so one pipe or even two pipes will not be enough

[brisk smelt]

seperate into 5x3 and pipe each seperately

think of it as electricity, the amperage of the pipe is 600

you cannot force more electrons ( fuel) through it

[amber umbra]

A tip, electricity is definitely not easier for most people to understand.

[bleak ivy]

does anyone know where the cave entrance for this uranium node is?

[sleek jackal]

got it. I'll separate

[brisk smelt]

waterfall

left side, 15 meters down from the river

[plush gulch]

Just think like them are belts. Not entirely, but it comes close.
Each m3 is 1 item. 1 pipe can transport 300/600 items

[warm bane]

I can't remember, when in a chain you need like, 7.5 assemblers
is it more efficient to just underclock the last one to 50%, or to "spread" the underclock over the 8 assembler you need ?

[plush gulch]

In theory to spread the underclock

I'm the end it doesn't matter that much, cuz unless you are building 1000 assemblers the difference is neglectible

[bleak ivy]

i cant find it, please can you mark it on the screenshot?

[spare jolt]

have explored this cave many times without even researching uranium

[worthy thistle]

Can anyone point me to some Fluid dynamics 101? New to the game, and someone pointed me here for a 'manual'?

[spare jolt]

https://satisfactory.wiki.gg/wiki/Pipelines

[brisk smelt]

oh im very blind i thought that was the swamp uranium node mb

[prisma kraken]

it also is linked in the channel pinned messages

[turbid patio]

the time has come, the very first factory (rotors + RIPs + frames) needs to be disassembled and turned into a motor factory

with 30 smelters instead of 9

[unborn dome]

I really wish train stations wouldn't block their inputs. Like just have a second internal buffer that everything gets transferred to when the train arrives, and block THAT.

[prisma kraken]

yeah, building them is kind of repetetive, i'll agree

[unborn dome]

Like I'd want to just see everything vanish from the platform inventory as soon as the dock sequence starts, and then incoming belts just start refilling it again without missing a beat.

[bleak ivy]

yeah truck stations do that so i dont get why trains arent like that

i usually just build that second buffer myself, with an industrial storage container with 2 belts to the station on each input/output station

[unborn dome]

Yeah that's what I'm currently doing, just seems really unnecessary and doesn't look nice

[brisk smelt]

is there a document for the math on slooped oil recycling loops

somewhere, perhaps

[unborn dome]

This is only going to be ~280 fuel/minute. Is this sufficient, rather than that big double-input-with-pump setup on the wiki?

[brisk smelt]

whyy are you transporitng fuel in the first place

[unborn dome]

(Or maybe with a pump under the frame?)

[brisk smelt]

but i recommend always packing fluids for travel

[unborn dome]

That fused frame alt needs 30/min, and tank cars are cool lol

[brisk smelt]

ah fair enough

that should be enough, that isnt even pushing mk2 capacity so you're good

[unborn dome]

I might toss a pump in under the frame just to be safe, but thanks!

Actually maybe I don't need a pump, this is flowing down from like 30m higher than the platform

[warm bane]

Anybody here knows what that figure means ?

[unborn dome]

It means the miner is only running 66% of the time

Because its output is not being consumed fast enough

[warm bane]

thx, figured it was smth like that, wanted to make sure

[brisk smelt]

"how to make plutonium fuel rod: turbofuel!"

[unborn dome]

Does anyone know if there's a mod that does this for train platforms?

[heavy gust]

do nuclear nobelisks clear more foliage than the normal ones, or is it just more damage to creatures?

[brisk smelt]

i think more foliage

theyre bigger in general

[fringe pawn]

Also more damage to creatures, bearing in mind the game's physics are kind of wonky about explosions and cover reducing damage

But you can totally kill 4 cliff hogs with one nuke if you can catch them spawning in a group.

[heavy gust]

i just want nukes to clear trees

[fringe pawn]

They're definitely better than cluster for that

[heavy gust]

and those pesky gas pillars

i hate having to spam like 30 nobelists at a time to somewhat clear an area

[amber umbra]

@unborn dome It’s unlikely there are mods to change train station docking behavior in that way. Changing it effectively just increases throughput which can more directly be implemented if modding.

[unborn dome]

Yeah true. I'm not really interested in a throughput increase here, I just don't want to have to manually buffer each station because the internal buffer is wonky.

[amber umbra]

Technically the belt(s) feeding the station can buffer if the throughput is low enough.

If you’re playing Satisfactory you need to be comfortable dealing with some jank, lack of QoL.

[humble jolt]

So I'm playing around with grid isolation/priority power switches, and something strikes me as not very user friendly

In the UI for the priority power switch, why do the graphs "marry" as they do? Wouldn't it be more useful to show the individual properties of the subgrid?

In this example, power grid A is the upstream main grid, and power grid B is this subgrid for manufacturing one item. WHen the switch is OFF I get relevant stats for the sub grid, but not when the switch is on.

Its nice to know Power grid B demands 264MW. When the switch is on, that information is lost

[unborn dome]

Yeah the hope is maybe they fix the train stations eventually, but it's been a thing basically their entire existence.

[unborn dome]

Yeah it'd be useful if they show what's on each side, but they don't 😦

[amber umbra]

Seems exceedingly unlikely train docking behavior is changed at this point.

I also think it’s an un-elegant design in its current form.

[unborn dome]

It seems like it'd be an easy fix. And it wouldn't break existing buffer setups, just make them redundant.

[sleek jackal]

if I produce 2070 rocket fuel per minute, and it is 4.16667 per minute in a fuel powered generator, that means i need to build....497 generator?? damn I didn't calculate that part of my plan. I finished everything exept the generator

[unborn dome]

Yep they're pretty efficient. You can OC the generators to reduce the number you'll need by 2.5 times at least.

If you have the power shards to do so.

[violet halo]

Drop it own to 200

Make it a nice even number.

[sleek jackal]

and all generator would need 3 shards?

[unborn dome]

Yep

[violet halo]

At 200 it would 198 x 3

[sleek jackal]

i only have 150 shard approximatly

[unborn dome]

Do you need this much power immediately? Or could you build a fraction of the generators now and then wait until later?

Or build all 200 generators and only shard a number of them?

[sleek jackal]

yes good idea, i'll do that

probably would need to use a blueprint for the generator

[unborn dome]

I haven't tried it, but I've heard you can BP with the shard in the generator already

[unborn ermine]

25/6 is the consumption fraction, makes it nice to calculate, aka no rounding

[unborn dome]

And it just adds shards to the cost of building the BP

[unborn ermine]

also yeah BP for generators is great, I would just make sure you want a specific clock for most if not all gens.

[torn shoal]

What's the best way to grind coupons?

I've bought everything else, now i want the achiement for getting the nut statue

[unborn dome]

https://satisfactory.wiki.gg/wiki/AWESOME_Sink#Points_generated_per_item

Make space elevator parts and sink them

[torn shoal]

word

[fringe pawn]

Silicon circuit boards, AI limiters, also uranium ore directly into the sink.

[halcyon yoke]

hey, this is more or less my first time posting in this discord and i'm doing it because i've run into the first thing in the game that i'm really scratching my head about.

how should I be thinking about miners and energy? once I unlocked the mk. 3 miners I started using them and upgrading my old mk. 2 because i assumed double the rate without needing to overclock would be a boon, but i didn't actually look at the energy consumption until today because i just assumed that since overclocking is exponential, that the mk. 3 would be more efficient.

I get that we can extract more for higher energy costs, but I'm surprised that unlocking a higher tech version isn't a strictly better situation.

[ember fractal]

Getting twice the ore is a better situation

Also higher level techs allow you to produce more power, so mk3 miner VS Mk2 using more power is not an issue

[fringe pawn]

At one point they changed the clockspeed power curve so that overclocking uses less power (becoming better), and underclocking saves less power (becoming worse). I'm not sure if the scenario you're describing was true back then.

[viral patio]

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

[wind spade]

[want] / [base] * 100

[fringe pawn]

As greeny says. Note that you can enter the formula into the clockspeed box and it'll calculate it there as well. So you can apply this same thing to all machines.

[viral patio]

thanks

[wind spade]

generally better to calculate yourself if you care about decimal precision

[lean fable]

Do machines such as Refineries have a Headlift?

[spare jolt]

the game literally tells you the headlift of the machine in its description

[lean fable]

Missed that one! Sorry

[wary storm]

Guys, is this overkill?

Apparently that's enough to power about 800 fuel power generators

[edgy leaf]

no.

its never overkill.

[brisk smelt]

3000 TF?

or 6000

[wary storm]

what's TF?

[brisk smelt]

isnt 3000/18.75 = 160 gens

[brisk smelt]

turbofuel

[wary storm]

rocket fuel

[brisk smelt]

ah ic

[wary storm]

3450 I think? /4.17

[brisk smelt]

is it not 10.42

ooh wait thats 250%

yeah so 350 gens i think

u gonna pipe allat manually?

sounds painful

[wary storm]

is there any other way to pipe?

[brisk smelt]

hmm, you can do it as 20x18, then build 20 gens, copypaste 17 more times via SCIM, thats a lot easier imo

[thorn trellis]

i think if anything i went overkill

[wary storm]

damn, that looks cool. I have no idea how to make things look pretty

[thorn trellis]

i mean urs functions rn mine isnt on

[wary storm]

Turns out 4 fit just fine on the blueprint designer, so I'll just spawn them in a row

[dusky dust]

Okay, take 2; been poking at this a bit more this afternoon and am still not really making progress.

I'd love to figure out how to arrive at how to compute the ratio of Recycled Plastic to Recycled Rubber refineries you need in the usual most-efficient petrochemical loop

I do know that so long as you've only got Rubber and/or Plastic targets, you can figure out the amount of fuel to spend on them with a couple of equations, but I don't yet understand how to figure out those equations for myself

Namely, with a target rubber production R and target plastic production P, you can get the total fuel spent on Recycled Plastic with: (P * 17/27) + (R * 8/27)

And likewise the total fuel spent on Recycled Rubber with: (P * 7/27) + (R * 16/27) (or just take the total amount of fuel (which is easy to get) and subtract the previous value, of course)

I don't yet really understand how to derive those fractions for myself, though

It's easy enough to compute the ratios iteratively by bouncing back and forth between Recycled Plastic + Recycled Rubber until you're sub-4-decimal-precise, but I'd love to know how to figure out what the actual formulae are

[fringe seal]

I have not delved into it
but I notice that it is asymmetrical
so it must be taking into account the residual rubber

[dusky dust]

Yeah, the "seed" rubber you get from Residual definitely has an effect on it

Example of the iterative generation you could do: https://gist.github.com/apocalyptech/ed6072f875d2ef8f4d067f752ede55ea

The formulae are definitely related to that 81-as-common-denominator thing; 27 is 81/3, of course. The reddit post where I ended up finding those formula is here, btw: https://www.reddit.com/r/SatisfactoryGame/comments/x94t7i/could_use_help_with_the_math_behind_the_oil_loop/ -- The OP basically goes into doing the iterative approach, which is also the only way I knew how to get to it. The comments contain plenty of useful info but they skip some steps to arrive at the fractions in those formula, and my brain's just not been able to make the leap with 'em.

[fringe seal]

got it

the four equations are:

P = 2Fp - Fr
R_bonus = C/3
Fr + Fp = 8C/3```
Fr & Fp = Fuel used in (Recycled Rubber / Plastic)
R_Bonus = Rubber obtained with Residual Rubber
R & P = Target Rubber / Plastic
C = Crude, which is not important

R_bonus resolves to (Fr + Fp)/8

so the set of equations are

P = 2Fp - Fr```

let's find out Fr

Fp = P/2 + Fr/2

R = 17Fr/8 - 7Fp/8
R = 17Fr/8 - 7P/16 - 7Fr/16
R = (34-7)Fr/16 - 7P/16
16R = 27Fr - 7P
Fr = 16R/27 + 7P/27```

[glossy wagon]

One constructor will work for copper powder right?

with the appropriate belt?

[fringe seal]

...for what

[fringe seal]

sure

[glossy wagon]

just basic level copper powder

[fringe seal]

@dusky dust you have sucessfully nerdsniped me

[fringe seal]

Wiki says it is 300/min input 50/min output
and particle accelerator @ 100% needs 100/min copper powder

[glossy wagon]

one constructor does 300/min

[wary storm]

any idea how the hell this is working without a single pump?

[glossy wagon]

I need 600/min for particle accelerator holy cow

[fringe seal]

turbofuel I assume

[wary storm]

rocket fuel

[fringe seal]

ah then it is a gas

[wary storm]

oh

gases don't need pumps then?

[fringe seal]

nope

[wary storm]

huh. well, good to know

[dusky dust]

Hah, apologies. :D

Just taking a look at your stuff now. :)

[unborn ermine]

They dont need a pump, but depending on the flow needed, you might want a loop if they dont feed correctly once full and you start it up.

[dusky dust]

Awesome, many thanks, btw! What I'd been missing in the end was representing P and R by those first two equations

(R = 2Fr - Fp + R_bonus and P = 2Fp - Fr)

Just hadn't had the proper things to sub in without those. Cheers, appreciate it!

[trim beacon]

I did some math to make heavy modular frames with recepies i have, can someone double check if my math is correct?

heavy modular frames(2):
-(10) modular frames
-(3.34)reinfored plate or(3.75)reinforec plate
-(11.25/12.6)plate or(20/22.5)plate
-(22.5/25.125)wire or(40/45)screws
-(17)pipe or(15)rods
-(40) steel pipe
-(60) steel ingot
-(10) encased industrial beams
-(60) pipe
-(50) concrete
-(240) screws
-(60) iron ingots

[unborn ermine]

this can also help with double checks
https://www.satisfactorytools.com/

[trim beacon]

Oh, good to know

[prisma kraken]

kind of hard to say if that looks right, how you've split stuff into either/or is tricky

[trim beacon]

I used / to make diffrence between 2 recepies of same item

Cause first number is for steel pipe frames and second is default recepie

[prisma kraken]

your math seems a little off, just tossed this comparison together w/o drilling into making rips

[trim beacon]

Good to know

What is that program? Cause IT doesnt look like one recomended above?

[prisma kraken]

sp.runesun.com

it just lets you lay out a factory plan in sort of a barebones way and does a lot of the same stuff you'd use a spreadsheet for

the roll-up view is great for recipe compares though 🙂

[trim beacon]

Thanks

[latent swallow]

does this issue with having 600 fluid in a pipe but not recieving 600 still exist?

[opal pivot]

why is this pipe blocking when its loaded from the top?

[prisma kraken]

i use it a lot more than sftools because it doesn't choose recipes for you - it assumes you know what you want to build and lets you pick

[opal pivot]

the water extractor is the pipe going in from the top...

[prisma kraken]

yeah, pushing pipes to 600 is still problematic, the pipes aren't broken, it just is that they get weird when at max speed (belts do the same)

[prisma kraken]

from the looks of it, you're disobeying pipe rule #2 : keep pipes simple

[trim beacon]

There are pipe rules?

[prisma kraken]

1. keep pipes full

2. keep pipes simple

[vapid gorge]

would need @latent swallow this is a little false:

the issues about pushing pipes at 600 exist at much lower throughputs as well. It's not a 600 problem

[unborn ermine]

Tools only does that if you let it though, and thats spamming everything

[vapid gorge]

unclear what exactly is going on, you look like you have a ton of merges and connections all coming from different elevations

[opal pivot]

they are simple?

[prisma kraken]

yeah, but you have to fight with it

[vapid gorge]

Have fluids go from point A to point B, no merges or splits

[unborn ermine]

"fighting" ok lets limit the water for pure ect

[vapid gorge]

thats simple

[opal pivot]

one extractor feeds multiple machines, it goes down into this pipe, then up into the machine

[latent swallow]

any way you suggest to remedy this problem?

[opal pivot]

That's impossible when it needs 2 inputs...

[prisma kraken]

if i asked you if liquid flows in your pipe network 'up or down', answering 'both' is the opposite of simple, sorry

[vapid gorge]

output manifolds count as Point A , input manifolds count as Point B

[vapid gorge]

keep pipes simple, loop the input manifolds, pre flood the system

[opal pivot]

Thats what I have

but I need more

so A goes into B and C joins into that half way through

[latent swallow]

by loop the input manifolds you mean connect the end of one to the end of another so it fills it from the back? outside of this i have kept it all very simple

[vapid gorge]

also you're gonna hve to get a much clearer shot of your pipes in the image , which may be impossible, or draw a diagram

you have elevation changes in your image and it's unclear at what part of the system its in. That can murder the flow

[vapid gorge]

this is probably the most stable loop you can do. If you're having a specific problem with a system pls do some over head shots of the entire layout

[opal pivot]

the pipe alone the ceiling is the water extractor, the one along the floor is the byproduct

[prisma kraken]

i can tell that it is an aluminum build, that you're bottom feeding your refineries and that you're playing with pipes at different several different elevations

and that you're using mk2 pipes

[vapid gorge]

ah you're doing waste water for bauxite? yeah don't direct feed.

[opal pivot]

what?

I saw people say that if you feed from above it won't block...

[prisma kraken]

all of the things i've stated are things i avoid when building aluminum

[vapid gorge]

a lot of people have lots of bad advice it's true

[opal pivot]

there's only 2 elevations

[vapid gorge]

any elevation change can be an issue

and the person you were talking to were probably talking about VIPs in a very poor way

cause a VIP junction has to be built in a very specific way

[opal pivot]

well 1 is just the input for the machine

[vapid gorge]

and bottom feeding is a danger zone

[latent swallow]

the problem is these two blenders at the end of the line where the 600 runs out (top pipe is fuel, lower is ~100 HOR, bottom is 600 HOR), the dodgy curved pipe is a because ofthe small spacing and is not doing anything jank, making turbo blend fuel

[vapid gorge]

just on it's own, let alone when doing more elevation changes and waste water

[opal pivot]

I did bottom feeding so I COULD feed from above :/

[latent swallow]

the manifold im using also has had time to fill properly, and nothing prior in the system is getting bottlenecked

[prisma kraken]

@vapid gorge think it might be good to create a curated gallery of known good aluminum builds?

[vapid gorge]

yeah it's so compact and things overlapping it's hard to tell, but 600 systems almost always need loops especially when it's not very clear and flat

[vapid gorge]

possibly yeah?

[vapid gorge]

in any case you're going to need to rethink your alum processing

[latent swallow]

appreciate the suggestion thank you

[vapid gorge]

fluids are one of those things where you'll probably need to sacrifice some compactness to be able to appropriately build the logistics :\

[prisma kraken]

idk, i feel like after a bunch of screwing around people figure it out, but there's a lot of trial and error and we all sort of converge on our own solutions

[vapid gorge]

some simple general rules to follow #math-and-meta message

[latent swallow]

its not too compact the problem was my insistence on using a vertical bus for my manifold, but thank you again

[prisma kraken]

yeah, rule 3 of pipes: all consumers on a pipe network should be on the same horizontal level

[opal pivot]

I mean, it looks like this from above, and took me all day 😅

maybe Ill test it with pumps xD

[vapid gorge]

well, you'll have some issues trying to do loops for all those manifolds that are stackd on top of one another.

it's very common for people to hit the pipe wall head first on their first big fuel station, imo it's reasonable to just accept your system running at 90% power output and just take it as a learning lesson rather than do huge amounts of rebuild, but it's up to you

[opal pivot]

I really wish we could dump water :/

[prisma kraken]

lever on each pipe to do that

[frosty owl]

We can sink it, at the measily cost of packages

[vapid gorge]

there's very simple ways to reuse waste water, you seem to have been told about 1 way in a stupid way from someone. Cause what you're doing is not technically impossible, but it has to be done in a specific way that the person didn't explain appropriately it seems

[latent swallow]

this isnt even a major problem im only making packaged turbofuel out of convenience for drones/time crystals, the simplicity of old fuel setups that i got used to have dumbed me down a bit

[vapid gorge]

for example, this is the most reliable way to do it

[opal pivot]

I could just bring Iron here and smelt it with this I guess...

[prisma kraken]

they're annoyingly inconvenient to make, lol

[vapid gorge]

in the image the blue water is fresh, the red water is waste, they don't mix.

[opal pivot]

oh god, thats a lot of piping...

[vapid gorge]

that's like 1/2 the piping you were doing

[vapid gorge]

it's 780 bauxite being processed in that image

[vapid gorge]

yeah my image is 10x simpler

[opal pivot]

I have 2700

[frosty owl]

Just turning some refineries around and having them side-by-side the length of most pipes would be halved

[opal pivot]

Math time I guess xD

[vapid gorge]

yeah but looking at it as a unit of processing? per your unit? much simpler

[opal pivot]

I even have my Sloppy Alumina refineries at 75% to make them all straight lines xD

so they're 105 out, but 150 in...

[vapid gorge]

the top is an example of sloppy solution and Electrode scrap. It's what's in the first image

[opal pivot]

I mean, Theyre all in straight lines and Ive spent hours on this so Id rather not change the machines xD

[frosty owl]

2700/150 is a nice split

[opal pivot]

Ill probably just use the water for Iron ngl

[vapid gorge]

that's still going to be 'changing all this' right?

because you need to redirect teh waste water and put in more piping for more fresh water

[opal pivot]

I have 18 refineries at 75% for Sloppy Alumina and 18 refineries at 100% for Electrode Aluminium Scrap

[opal pivot]

its only piping if I use the byproduct at least

no machine changes

[prisma kraken]

a different design that reliably works

[vapid gorge]

you could proabably do minimal to no machine changes if you split the fresh and waste too, just clocking changes

[opal pivot]

Oh thats basically what I have now

[vapid gorge]

in any case did you end up fixing the 20 trains at one station problem?

[opal pivot]

but I have 18 instead of 4

[vapid gorge]

that is absolutely not.
elevation matters

[opal pivot]

yeah, I needed more trains xD

[prisma kraken]

note that it uses mk1 pipes

[vapid gorge]

... did you not understand taht more trains made less throughput at the station?

[opal pivot]

Throughput went up 🙂

you can see the throughput at the station

[vapid gorge]

if you have 20 trains stopping at 1 station, throughput goes down. unless you have a mod

[opal pivot]

Nope, it went up

no mods

[vapid gorge]

then you don't have a single rail station

[opal pivot]

I do

[prisma kraken]

station reports of throughput are a running average calculation, it waxes and wanes as different trains come in with different amounts at different intervals

[opal pivot]

yup, thats why I have buffers on the station

[vapid gorge]

buffers don't stop 20 lock out times

[vapid gorge]

if in a 5 minutes space of time at a station you ahve 1 train stopping? that's 30 seconds it's not unloading
so you only have 4.5 minutes of throughput.

if you have 2 trains stopping in 5 min? that's 4 minutes of throughptu
6 trains stopping? 2 minutes of throughput

[prisma kraken]

if those output belts ever run dry, your alum processing will have problems

[opal pivot]

Yup, thats why I added trains

[opal pivot]

only if those 6 trains all stop at once

If one train is never waiting for the next one, then throughput is fine

Im using less resources per belt than the belt can carry

[prisma kraken]

i'd advocate for you to make some popcorn and watch the station and buffer fill levels for an hour or two

[opal pivot]

Im draining the buffers quicker than they are filling atm, so still need more trains

[prisma kraken]

i'm not being snarky, you'll learn some stuff in how it all works by watching

[opal pivot]

Well it was transferring 50 items per minute with one train, which is obviously not enough

[versed dawn]

Guys just a question, how do you normally approach, rods, iron plates and screws? do you produce them all in 1 factory? do you separate them? like 1 for plates, 1 rods and 1 screws?

[opal pivot]

so 2 trains is just under 100 because of potential wait time

but the journey is so long that they rarely hit that wait time

and if they do, the buffers support an input of 2400, and Im only taking out 900

[scarlet sky]

Has anyone noticed that lower pipes don't necessarily take priority when they're at max capacity? I seem to be observing this behaviour since 1.0...

It could be just that I've never tried some particular build configuration though

[prisma kraken]

early production of them is really just set up some smelters & constructors around an iron node to feed the phase 1 elevator stuff & fill some crates with building materials. later in the game, you need the parts at much higher volumes as components for other parts and at that point i start making the components as part of a factory for the higher level part. i follow that strategy through most of the midgame components and then start rounding up and centralizing stuff for endgame stuff since it tends to be much lower volume

my rod production for building materials is still the same stupid constructor i set up very early on in the game that's faithfully been dumping 15/min into a storage crate and overflowing to a sink, lol

i'm making tons of them elsewhere as a component for modframes feeding into 90 hmf's/min, but that's just stuff inside a factory that you could look at from the outside as a black box

as for screws, you really don't even need them for building material (only things that you can make with them are the awesome shop & some equipment), so really you don't even need more than a stack of them on hand in the game ever

[tidal shuttle]

That’s assuming an alt recipe for reinforced iron plates

[prisma kraken]

find me something that uses screws as building material. only thing is the awesome shop

i wasn't speaking of 'as component used to make other part'

[tidal shuttle]

||Modular engines|| and smartplates will either require alt recipes, or a source of screws in your supply chain. I think @prisma kraken is saying he incorporates that need into his larger factories later in the game. I often overbuild and stockpile instead of sinking which has usually been beneficial in a more scrappy less planned runthrough. It means all your time exploring is spent banking. Ultimately you will likely enjoy some planning and balancing. If part if your giant modular frames plan requires X screws, feel free to make a smaller factory and send them over. Or make a mega factory.

[prisma kraken]

one other thing i'll add is that there are some extremely high volume components like screws, wire, quickwire & concrete that it often is better logistically to make localally than deal with the logistics to move them

and of course, alt recipes often provide more efficient or easier to build means to construct factories

[tidal shuttle]

Not sure how far into the game and spoilers you are. How much youtube you’ve seen. Do whats fun for you. If you like exploring and being “scrappy” and less planned, you can learn a ton making spaghetti and trying to upgrade small builds. I vote for that first time around. If you wanna always plan how many of everything to make, that’s a whole other side of learning and challenges. Iron is a common resource so really its up to you. If trains stations or truck stops excite you, build smaller spots and connect them. If transportation logistics stress you out, just make ultra long conveyors like me 😬

[prisma kraken]

or both...

[spice radish]

If im not wrong a normal iron node with 250% mk.2 miner should be able to support 300 iron sheets a minute right?

[violet halo]

What's the conversation rate for ingots to sheets, 1:1?

If so, then yes.

[dusky dust]

FYI, I did go back and add in Fuel as a possible output on there, now that I actually understand how that works. Doesn't really change much since there's no actual new variable per se.

The only change initial-equation-wise is F + Fr + Fp = 8C/3 (ie: adding the fuel output target)

And then it just gets carried on down the line

[inner flame]

anyone know how to stop my reinfroced iron plates from backing up, and to send more of them to storage? ( it was literally working fine earlier, idk if me setting up coal power somehow affacted it)

[dusky dust]

In the end: Fr = 16R/27 + 7P/27 - 2F/27 and Fp = 17P/27 + 8R/27 - F/27

So that gets you any arbitrary split between the three outputs

I suppose it'd probably be worth putting all of this on the wiki somewhere

[brisk smelt]

anyone have a solution for stuffing 8 refineries and a constructor in a mk3

[brisk smelt]

overflow splitter?

[prisma kraken]

30 ingot->20 plate

[prisma kraken]

i just read over what you wrote, interesting to see it all in equation form. that 27 denominator is why i could never resolve the 9 out of the division, lol

how i've come to understand it is that recycled rubber or plastic convert fuel 1:1 into one or the other

that's more of the 30,000 foot view of the process though 🙂

[dusky dust]

Heh, yeah, it's a persistent bugger. Goes back to the 81-45 rule thing

[spare jolt]

i'm now not so sure which recipe I like more - Diluted P. Fuel because I have to deal with liquids less, or Diluted Fuel because I don't have to deal with packagers.

this is a mix of recycled plastic/rubber + turbofuel + diluted fuel

[unborn ermine]

Honestly, the blender version works better if it dosent have to be messed with (aka splitting 1-2 blenders)

I miraculously got my setup to work with a split, and im happy

[spare jolt]

no, it's just I was saying here "I lIkE dIlUtEd P. FuEl BeCaUsE lEsS lIqUiDs". now I'm harvesting fruits of my labor I guess 🚬

[prisma kraken]

which i've never understood as a precept

[prisma kraken]

i like the blender version when you're making fuel en masse and the packaged version for when you're making vehicle and jetpack fuels

building the loop for cannisters when you're using the fuel for recycled plastic/rubber or heat fused frames is just more ugg on top of a lot of ugg

[unborn ermine]

more unga on top of bunga

[spare jolt]

I see little other choice because I need at least some gigawatts if I want to proceed with decent-sized base

[pure crow]

So I'm thinking of using plutonium fuel rods for drones... how tall do I have to make a tower so they don't radiate anything at the base of the tower? o-o

[spare jolt]

i'm not going with 120+ hour build like I did last time, I learn my lessons, but simply 20 fuel-powered fuel generators won't cut it this time

[prisma kraken]

the recycled loops aren't too bad to build for me anymore, and i don't mind using dpf for early power (small scale)

50m?

[spare jolt]

(why did they even rename them to X-powered generators instead of just X generators?)

[dusky dust]

One final ping and I'll shut up about it, btw: https://gist.github.com/apocalyptech/cc99a7467e4a10f7543d41c4a924a729 Thanks again! Really appreciate your help. :)

[prisma kraken]

the 50m was for @pure crow

[pure crow]

50m? Huh, that doesn't seem too bad.

[prisma kraken]

they radiate for quite a distance, but it isn't that bad farther away

[unborn ermine]

Fuel is a hard word, the base fuel is "fuel".

so its not just "fuel" powered now

[spare jolt]

well it kinda makes sense for fuel generator, but say if you need to get rid of HOR and you don't have those, it's a bit unintuitive that you can also burn coke

[unborn ermine]

technically its coal coke, and compact coal for the coal generator

[dusky dust]

I believe it was basically just some manual work done to sort of find a lowest-common-denominator between all the various recipes, so that you'd be able to know what values you can clock stuff at to avoid repeating decimals

[spare jolt]

yea compact coal is obvious

[dusky dust]

Most stuff in the game works with 45, whereas oil is 81.

[spare jolt]

but petroleum coke - not quite

unless you get a "Eureka" from "... less efficient coal replacement"

[dusky dust]

And you don't have to use that whole number; the usual advice would be to divide by two a bunch to construct values based on those numbers

so like 81 -> 40.5 -> 20.25 -> 10.125 (beyond that point you start going beyond the game's rounding limit, so it's generally best to stop there

So like if you wanted to produce 50 plastic/min, you'd get repeating decimals and sort of ugliness

But 40.5 + 10.125 is 50.625 -- If you make 50.625 plastic instead, all the machines will have nice "clean" clock numbers and you wouldn't run into rounding problems

(Which is really only a problem if you care about total efficiency -- when you do run into rounding errors it's only gonna pop up like every few thousands of cycles, so generally most folks aren't gonna care. :)

Personally I don't tend to take it into consideration, but it's nice to know it's there

[prisma kraken]

Sev always used to argue with me about this when i said that anything in the game can be expressed as a rational number where the numerator and denominator are each a multiple of of the integers [0-13]

i'm stating that poorly because the exact formula for the numerator and denominator are 2^n x 3^m x 5^o ... x 13^z

that's why he always listed a few recipes as exceptions - the ones with 7, 11 or 13 as prime factors in the numbers 🙂

[glossy wagon]

does sloppy alumina solution always need to be every now aand then or is there a way to make it %100 efficient

[spare jolt]

It doesn't need to, it's just sloppy alumina is one of the ingredients for removing silica from the equation

But afaik sloppy + default ingot gives you more ingots, but you spend more silica

[violet halo]

Yeah, normal is 100% of bauxite, sloppy is 90%.

[glossy wagon]

does anyone have a picture of a %100 efficient sloppy alumina factory

[prisma kraken]

it is dependent on getting your pipes (and everything else) correctly built

[prisma kraken]

#math-and-meta message

i've found that for aluminum, when you are building it in modules that process 600 ore, it is better to use mk1 pipes in a loop

[glossy wagon]

You can't do it with 1 machine?

[prisma kraken]

wdym? you need at least one refinery to eat the alumina solution output

[amber umbra]

Idk what the original question meant. Sloppy alumina is just a recipe like any other.

[glossy wagon]

i'm just having trouble with the math. I have a refinery with 200 water/min and 200 baux/min coming in outputting 480 solution per minute.. I am not sure how to balance it how many machines i need with the first 1 doing %100

[spare jolt]

Apparently alumina is just a disguised dark matter residue

[prisma kraken]

yeah, unplayable, demand refund, lol

more annoying of the fluid issues is the lack of packaging for dissolved silica

[prisma kraken]

are you using the sloppy+default scrap combo?

[glossy wagon]

yes

[amber umbra]

This is my sloppy alumina, default scrap production line. This is the style where a dedicated machine consumes the recycled water.

[prisma kraken]

gimme a quick sec

[amber umbra]

The electrode scrap alt is a bit more bauxite efficient which is why it’s popular. But default isn’t too different.

[prisma kraken]

this is pretty much how you'd balance sloppy+default

it needs extra water as input, which is where the ugly happens

why it all is problematic is that the total water needed for the sloppy refineries is 600 - max for an mk2 pipe

[amber umbra]

Shouldn’t you just scale down the production line until it fits within the pipe, belt limits?

[unborn ermine]

Its what I ended up doing, 2940/min divided by 5 groups with 5 refineries each. (sloppy and electrode)

[prisma kraken]

i just build it in modules of 600 baux in and 1200 (or 1080 if doing default) scrap out

[unborn ermine]

it ended up being 1 pairs of the refineries were a 1-1 sloppy-electrode machine, meaning it really was 4x a group

[amber umbra]

Increments of 300 bauxite seem nice per above for sloppy, electrode. Then do 4 of them for a 1200.

Although I really haven’t seen what the meta style is.

[prisma kraken]

yeah, doing it 1:1 with sloppy at 75% and electro at 100% is my preference, but it works if you do a crossbar pipe connection 3 refineries to 4

@vapid gorge advocates clocking it all unevenly so that the wastewater supplies certain refineries & the fresh water supplies the rest

that's a more robust solution, but i don't feel that extra safety necessary

[amber umbra]

Yes

[prisma kraken]

regardless, a lot of problems with aluminum also come because the consumers down the line of the scrap and ingots are complex, and math errors there can cause production to stutter and eventually back up into the water loop

[unborn ermine]

Whats funny is I ended up underclocking a lot of my aluminum to just work better
4x 75% sloppy + one at 67.5%
4x 100% electrode + one at 90% (which paired with the 67.5% sloppy)

[prisma kraken]

best to throw an overflow sink somewhere on the scrap or ingot belts so that prod never stops

[unborn ermine]

Yep! Thats what I did

[amber umbra]

Afaik the Al scrap needs to never stop producing for those stops to prevent the water extractor refineries from jamming the pipes.

[prisma kraken]

yeah, if you're recycling the waste water, any pause in the scrap refineries causes the fresh water to fill too much of the pipe and that causes it to lock up

[unborn ermine]

Im still loving the fact that it recovers all the way to the miners when it backs up. aka powered off, given enough time.

Miners need unclogging, but its all smooth

[prisma kraken]

the extractor doesn't pause until AFTER it has locked up, by which point everything sort of is already dead

it really took me a lot of trial and error to figure out an aluminum build that didn't suffer from problems

[unborn ermine]

I think the only fix I need is a place to dump ore

[pure crow]

I just used the VIP, worked the first time, never went back to question it.

[prisma kraken]

like i probably spent 6 months chasing that 9x% stuff

what i realized is that around the time that i'm building aluminum, my fps starts degrading

and that some things like mk2 pipes don't work right anymore

it isn't just mk2 pipes; mk5 & 6 belts do the same 😦

[pure crow]

ya I heard mk6 belts don't always run at max speed? is that because of fps problems?

[unborn dome]

Ok I know people say not to use valves, but are they actually just broken? I put one on a pipe that I wanted to limit the flow of, and it just stopped flowing entirely, even with the valve facing in the right direction...

[prisma kraken]

sometimes they'll skip items - watch one running at full speed as your autosave triggers

[prisma kraken]

if you define 'sometimes work' as just broken, yeah, they've always been broken

nothing new 🙂

[unborn dome]

Not trying to use it to prevent backflow, just want to limit a pipe to 240/min...

Apparently they can't even do that

[prisma kraken]

never could, lol

[astral warren]

Yeah

[unborn dome]

Oh no, valves were added in U3 and they still don't work right??

[prisma kraken]

a valve's flow setting internally as x/256 - 255 unique values, so flow when you set it to a fraction is always an inexact approximation of the value entered

[loud trellis]

Well then.. this'll be a project and a half

[prisma kraken]

if you set it to something like 150 or 75, i suppose it might correctly meter out the flow, but that isn't very useful

i still think they ought to just remove them from the game entirely

(or make them a decorative piece with a on/off setting)

[unborn dome]

Yeah on/off would be nice even

[dusky dust]

I'd heard that was improved in 1.0, so you do at least get the exact value now. (Though I have still not tested that for myself)

[glossy wagon]

this is what i'll use.. thank you

[prisma kraken]

that's a question for the devs i've not watched the dev streams in a while

[glossy wagon]

It's ugly, BUT, I wont have to come empty the pipes anymore because it stopped producing aluminum scrap cause the water was clogged.. Thanks again @amber umbra

[fringe seal]

and the R_bonus equation too

[dusky dust]

Well, as-wrtitten, the R_bonus equation stays the same, since that's based on the Crude Oil input. (R_bonus = C/3)

[fringe seal]

ah yeah I forgot that it was expressed in terms of C lmao

[dusky dust]

Though if C were expanded, then yeah, it'd have to be tweaked too, to account for F. :D

[fringe seal]

I'll blame the fact that I just woke up

[unborn dome]

Anyone know if there's a way to nudge conveyor lifts height? I'm having a weird issue where the conveyor floor hole makes a huge gap in the lift, but makes the little click to say it's connected when I build it..

Because it's like 0.5m higher than the lift

Or like this lol

[astral warren]

Are those 1m floor holes?

[unborn dome]

Yes, but it's actually not exactly 0.5m height difference. I've tried a few alignments, including attaching the floor holes to those very short floor frame things.

[astral warren]

I think it’s a bug exclusive to those, you shouldn’t use them rn anyway because they’re currently deleting a portion of items that go through them

[unborn dome]

Do they? I've used them heavily in my HMF factory

[astral warren]

2m or 4m ones should align fine, otherwise you can just not connect the lifts to the hole and just phase them through if you want to keep the 1m foundations

[astral warren]

So I’ve been told

[unborn dome]

Cheesed it by just clipping the lift through the floor and then dropping a floor hole on top of it without it actually connecting

[prisma kraken]

that gap thing can be solved by building the lift in the opposite direction

[unborn dome]

If I start it at the hole, it won't snap at the right height at all

[prisma kraken]

that's one i don't have an answer for then 🙂

[unborn dome]

Haha np, I think it's probably a bug

Rendering bug, that is. I assume it's actually connecting, since it makes the little alignment click.

Just hiding the slight height difference under the floor where I can't see it...

[prisma kraken]

eeewwwww

[unborn dome]

Yeahhh

It's ever so slightly off. Apparently just enough to mess up the lift snapping.

[prisma kraken]

we all have those little things that we bury 😉

[unborn dome]

It's all because my pipe girder structure thing is shifted ever so slightly in the Z-axis so that pipes running in it don't collide with the top and bottom.

And so now belts are all messy trying to run them in the same girder

So I can do this

[prisma kraken]

i've got similar in a few places, best not to think about them

[spare jolt]

you guys think this piece of lake can fit 30-ish extractors?

if consdering as much water space as possible

[unborn dome]

It's hard to say without actually testing it

[astral warren]

30 more or 30 including the ones already down?

you might be able to get 30 more in but they definitely wouldn’t have room to be aligned in any sort of neat manner

[wraith timber]

does gas have the same issues as fluids at getting fully 600/m out of a pipe?

[spare jolt]

Excluding the ones built

[wraith timber]

IE: can I safely run 600 pipe at full with gas?

[rotund summit]

based on my personal save i had problems running 600 nitrogen gas in one pipe as everytime i load into my save it will drop below 600

[wraith timber]

gotcha! thank you

do you think that having a tank would solve this for you?

[eager solar]

Making the usual output loop should help you

[prisma kraken]

no, buffers and valves never solve flow issues, they just cause more oscillation in the pipe

[eager solar]

*input

[wraith timber]

what is the loop?

[eager solar]

Loop the manifold with a parallel pipe above it, feed from the top pipe

[prisma kraken]

nitrogen is a little funny wrt to pushing it to 600, usually you're making a 600 line from 2 well points and piping it into a very small number of blenders and it doesn't have headlift constraints

[eager solar]

Image in the pinned pipe manual

[wraith timber]

oh thank you!

[prisma kraken]

i think if everything on the pipe is the same clock rate, you'll be ok, but you might want to split the 600 into (or leave it as) 2x300 pipes

[wraith timber]

this is my first time working with a proper factory using gas

[wraith timber]

gotcha

[prisma kraken]

FYI, all nitrogen/nitric acid recipes take amounts that are easily consumed by small numbers of machines except for heat-fused frame

i've been doing some nitrogen work today in getting rocket fuel cooking and dealing with a lot of other nitric acid pain

[wraith timber]

gotcha, so just avoid using 600 pipes

a related but random question: should I use 600 pipes and only fill them to 300 or just use 300 pipes?

[prisma kraken]

my general feeling is that mk2 pipes should be used very sparingly

[wraith timber]

gotcha

[prisma kraken]

i'd use the mk1 pipes

[wraith timber]

what a bummer. Pipes always make me so sad

[prisma kraken]

lots of the stuff about them being broken is people's misunderstandings and bad designs, but you still should prefer using the mk1's when you can get away with it

mk5 & 6 belts suffer similar issues

[wraith timber]

gotcha. TBH I really think its at best a poorly implimented system but thats a discussion for a different channel and is very likely over talked about anyway

[eager solar]

Wouldn't it be better to use mk2 for 300 to avoid nearing the limit?

[prisma kraken]

i don't understand the game internals much at all

[wraith timber]

do they really? I never had issues with 780 belts and havent gotten to 1200 belts

[prisma kraken]

i do however have a few 1000 hrs of building in the game, and just know what works at this point

[wraith timber]

that is very fair

[prisma kraken]

every now and then you'll catch the belts when full have an empty slot

oft times it coincides with an autosave 🤔

[wraith timber]

ohhh gotcha

[eager solar]

Yeah I had that mess me up, you can also have a beat of error with mergers causing a slow back-up at max capacity

[prisma kraken]

there's something with how the simulation tick occurs that can cause belts to miss items. i assume similar happens with pipes w/o the visual to witness it

any time you push things to capacity though, strange things happen. machine buffers can't ever empty, etc

[wraith timber]

I love this game, its my favorite factory builder but there are def some odd choices/problems/bugs/etc and some that are quite frustrating for this type of game. but meh...still rather play this than factorio lol

[prisma kraken]

when 1.0 was announced, i switched over to play DSP for a while, figured i didn't want to sink anymore hrs into this when everything would break. that game has a lot of flaws as well, but well worth a playthrough

it does some things so well, and others just make you so mad, lol

[wraith timber]

I love DSP!

[prisma kraken]

pretty sure that's the case with any of these games that you spend tons of hrs in

[wraith timber]

I did the same thing. About 1000 hours of this pre-1.0 desided to take a break until 1.0 and played DSP during that time

[prisma kraken]

i kind of think one thing that SF is missing is the DSP style endgame

[astral warren]

I didn’t get all that far in this game pre-1.0, I think we stopped playing midway through setting up our first nuclear plants

[prisma kraken]

that's pretty far, lol

[astral warren]

Actually yeah I guess it might be considering we had one less phase back then

[latent anchor]

it does feel so much better now that nuclear doesn't create permanent indestructible waste

[prisma kraken]

it doesn't much change how you should approach nuclear though

[astral warren]

Honestly it hasn’t really changed anything for me considering how insanely expensive ficsonium is

[prisma kraken]

it isn't as expensive as it seems

[astral warren]

A vast majority of my plutonium rods are still getting sunk

[wraith timber]

yea for me I am going to be rocket fuel instead of nuclear cause I just dont wanna deal with all of that this save lol

[astral warren]

I made a line for it on ST to get an idea of how much id need to reprocess all my plutonium and it was just straight up impossible to make

[prisma kraken]

what is honestly the limiting factor to it all is the number of sloops you have

[wraith timber]

I would love a good "mega project" for this game. Beyond just...the space station. Like something that put back into the game (like the dyson spheres do in DSP)

Like if we finished the (first?) space station it would give a big bonus to some capacity of the game

[prisma kraken]

agreed, or some sort of point leaderboad, etc

[astral warren]

Yeah I’ve seen a lot of people asking for some sort of endgame milestone or repeatable objective to work towards

Maybe they’ll look into it for some update in the future

[wraith timber]

usually I start to run out of steam when i get close to finishing because it feels kinda like I have already finished? but not quite and the last bit is as long to do as the rest of the game lol

Blueprints are HELPING SO MUCH with that thoug h

[prisma kraken]

i'm hoping that there will be something like that or a challenge mode, difficulty setting etc

[astral warren]

Difficulty Tuner or whatever that mod is called will do for now ig

[astral warren]

I’m just taking my sweet time refactoring pretty much everything I’ve ever made now that we have MK6 belts

Saving all the project part stuff for last

[wraith timber]

for me these games are to chill to so personally "challenge" modes dont do much for me.

I like my DSP/factorio with infinite resources and minimal interaction with enemies

[prisma kraken]

something else i feel is in the same ballpark since we're talking about it... more achievements!

[wraith timber]

yea achievements 100% are good to have

[prisma kraken]

well, with DSP all resources eventually do end up being infinite

[wraith timber]

yea I just didnt wanna wait so I set them to infinite right out the gate lol

[prisma kraken]

i think that takes a little fun out of the game, but u do u 🙂

[wraith timber]

for sure

[fringe seal]

ah, ok. I should have mathed it out

for processing ALL bauxite nodes on map, using sloppy alumina and electrode scrap, 4920/min coke is enough

I am making roughly double that lol

also it requires 12300 water and 20500 silica, so that's a problem hm

I should give up making megafactories lol

[past reef]

but you can convert coal for quartz?

[pastel obsidian]

Sounds like you need a silica and quartz megafactory

While you are at it we can get through Petro coke, plastic and rubber online as well

[dark star]

You can ask satisfactory tools limits like this, cause it defauts to max resources the map can provide and automtically strives to use as little resources as possible through alts

[toxic hemlock]

Are there any weird things I need to know for bootstrapping a factory that circulates fluids

without input or output

I'm using instant scrap, which outputs as much water as it intakes sulfuric acid, so I'm planning to just route it around

[leaden cosmos]

fill the water loop until the sulfuric buffers are about 1/4 to half full and then stop filling, 🙂

[toxic hemlock]

That was my estimate

[astral warren]

I usually just fully fill everything, I think either way it should be impossible for it to fail if it’s running purely off byproduct

[frosty owl]

BTW @wind spade (my apologies if you've answered this hundreds of times already...) how's it going with including Somersloops on the planner?
Any plan for a "quick fix" like... "checkbox that doubles the output of one production nodes in the visualization graph" or something? 😅

[wind spade]

old tools won't get updates, new tools may or may not get them

[frosty owl]

Ok
I hope this is due to how you feel about it rather than a lack of support...

[wind spade]

old tools solver can't do sloops

because it doesn't understand it can't be split

[frosty owl]

Yeah, I was talking about the new one ofc :D

[wind spade]

unsure about new tools, purely because I'm not sure how to implement it properly yet

[frosty owl]

The functionality where you can add existing factories to a new plan seems like it could (maybe) be repurposed to easily manage slooped productions

[wind spade]

I have a few ideas but for now unsure

[frosty owl]

BTW, that was a great addition!
I can't believe I didn't see it coming 😆

[wind spade]

what addition?

[frosty owl]

It adds a huge amount of QOL for my purposes

[frosty owl]

The one where you can import existing factory plans into another one

Tbh, I haven't used it yet, I'm just assuming how it works based on UI 😅

[wind spade]

you mean the export/import?

[hard tendon]

I need the divide 175per/m 35x5

is there any good video for balanced divide for 5?

ory any explanirs here?

[tawdry cradle]

top result from a google search:
https://www.youtube.com/watch?v=IPE-BAlg0xA&themeRefresh=1

That looks really cursed though ICL

Ok so they basically make a 1 to 6 splitter, then feed one of the outputs back into the splitter

[obtuse leaf]

I honestly don't really do balancers 99% of the time, manifold is plenty for almost everything

[leaden cosmos]

1:6 feeding back 1 😉

[tawdry cradle]

yea

[hard tendon]

i didn't want to straight connection cause ofI don't want to wait for the machines to fill up.

[obtuse leaf]

Yeah - you can manually fill the machines but if having them all start simultaneously is important for you and don't want to fill them manually, then load balancer is a good choice

[wind spade]

you can also just do

--S--S--S--S--S
  |  |  |  |  |

[wind spade]

can prefill manually

[tawdry cradle]

That's a manifold though

not a balancer

[wind spade]

and after filling it will work the same as the 1:5 balancer

[tawdry cradle]

I mean, yeah

[warm bane]

Can you create a "throttle" by chaining splitters and sendingback the other lanes to the start ?
e.g. 60 -> 20 ->10
| |
(40) (10)
If I want to extract a small part of a belt

[wind spade]

not sure what do you mean by if you send things back to start you end up with more per minute on the belt so the numbers change

also, if you want to extract small part of a belt, there's many better solutions - manifolds, smart splitters or just having the small part be made separately

[warm bane]

I have a 480/min belt
I use a splitter who output to a mk4 and a mk1
So now I have a 420/min and a 60/min belt
I then split the 60 belt into 3 belts, who should get 20/min each.
I then merge 2 of those 20 back into the original belt
I should have one 460/min and one 20/min belt
Would that work ?

[wind spade]

theoretically yes. Practically.. who knows.

If you are splitting it to a machine that needs 20, the machine will naturally bottleneck the belt, so you don't need to do any of this

[patent blaze]

^this

thats how manifolds work, they fill up all the machines one by one until everything evens out and it just works

[warm bane]

I know, it was more to send to a drone port
The facility that would use those 20/min is nowhere close by, so I was hoping to divert a bit of the production to export it, without impacting the source too much

[hard tendon]

Thank you everyone

[wind spade]

it will work the same way. The facility will naturally bottleneck the belt

but if you don't want to do that, just split the production so that there's a separate machine (or machines) that make 20 in total, and ship that to the drone port

[warm bane]

Fair enough

[blissful epoch]

hmm, if you insist on 1 to 5, id go for split in 3, then each in two, merge the sixth back into the main

still, i havent used any balancers myself, only a sortof-half balancer where i balance my uranium fuelrods over 4 tracks, then have them do a little manifold for 10 reactors each

[ember fractal]

I've just used manifolds

If I have like 1 manufacturer's output into a splitter feeding 3 reactors, that would be nice balancing, without too much fuel rod build-up

Althought, that means I'll have my fuel rod production right near the reactors

[frosty owl]

It's funny how that works for all splits. No matter how complex, in the end they can all come down to "split to the closest multiple of 2 or 3 and merge back all extra" 😆

[analog meteor]

closest multiple of 2 or 3 is not correct

mathematically

[frosty owl]

Right, should be the closest BIGGER multiple (ofc, one needs to select a number of outputs equal or bigger than what needed, so that excess outputs can be looped back)

[analog meteor]

i think its more accurate to say 2A + 3B > number of outputs.
and keep A and B to a minimum. there may be multiple combinations of a and b but thats your decesion

i hope this formula is accurate. i used to be good at maths

[frosty owl]

What would A and B represent?

[analog meteor]

nothing. jsut two numbers

2 times any number plus 3 times another number should be more than the number of out puts but still be close to it

well maybe they do represent something

umm.
A represents how many times you split each output by 2 to. and B means you multiply by 3.

[blissful epoch]

dunno if thats it, as mixing 2 and 3 splits at the end would give different outputs

[analog meteor]

wait i think theres an issue.

ok mb im not that good at maths but i think im better than the average person

[frosty owl]

I don't see why you'd want to mention any other variable than the desired number of outputs
The number of outputs the balancer should have will just be a multiple of either 2 or 3... There's no other parameter needed to consider

[analog meteor]

a and b arent parameters.

[frosty owl]

Aren't they variables/parameters (I'm using these terms as synomims here) in an equation?

[analog meteor]

idk how to turn this into maths format. your trying to minimise a and b to make things simpler and it jsut need to be just barelyu more than the number of desired outputs

[blissful epoch]

thing is, for proper splits you cant mix 2 and 3 splits

[analog meteor]

wait does it multiply like 2a * 3b so its 6ab. that dosent make sense. i need help with this maths

[blissful epoch]

this has impacts on the equal outputs

[analog meteor]

yeah but you can layer 2 and 3 and each layer/step should be only 2 or only 3

[blissful epoch]

youd need to recursively work down the chain of 2 and 3s

[analog meteor]

is that really the only way to do it?
you saying divide the output number by 2 or 3 until you cant?

wait hold on is the formula 2^A*3^B ? i think this might be correct
edited it to be multiplication but still cant make sure this works

yeah bcs if the outputs is a power of 2 then you can just split by two and gewt it perfectly done

but i think the mixing of power of 2 and 3 dont work with simple maths stuff

ok i give up trying to make a simple defenition. easiest that i can think of is dividing the output number by 2 or 3 until you cant or smth.
ill leave this to someone else to try figure out a simple way of doing this

[frosty owl]

Eh, I think you're overcomplicating it... Explaining it through text alone can be simple enough ^^

Trying to write that in math terms make things a bit more complex to convey via text. Eg with N being the number of outputs you want, X the balancer outputs calculated before loopbacks: X = MIN( (3A>=N), (2B>=N) )
Cumbersome to read and write 😅

[blissful epoch]

yeah, youd want a min

[analog meteor]

i think powers of 2 and 3 is correct?

[frosty owl]

I don't think so

[analog meteor]

like change the 3A and 2B to 3^A etc..?

[frosty owl]

No, you want multiples 👍

[blissful epoch]

i dont think 2*16*3*25 would be relevant

[analog meteor]

3A where a is 2 means you split into 3 twice which means 9 outputs which isnt 3 times 2

[analog meteor]

no ignore thhe plus sign

[frosty owl]

A and B can be either 2 or 3, but can be multiplied by any number from 1 to infinity, not just by themselves

[analog meteor]

?

[frosty owl]

We agree that A and B are either 2 or 3, right?

[outer vale]

Assuming you're looking for even splits the whole way (a load balancer) then yes it's powers

[analog meteor]

no

[analog meteor]

ok

[frosty owl]

Dah, sorry, I got tangled up in terminology. Let me reread 😅

[outer vale]

total outputs = 3^a * 2^b, and that >= the number of splits you actually want

[analog meteor]

a and b are integers that are to be minimised somehow idk its so confusing

[analog meteor]

yeah thats the formula i thought would work but idk if it acutally works

[outer vale]

sure it does

not sure what you're actually trying to do with it mind

[frosty owl]

Yeah, lemme rephrase what I meant:
You want to multiply 2 and 3 by any number, not just by themselves. Eg: if N is 10, the closest multiple of 3 to that is 3x4, not 3^(something)
As long as we're only considering integers, ofc

[covert venture]

Asked ChatGPT to calculate a perfect zero loss Nuclear Power Plant. It failed. Satisfactory is smarter than AI. And all my calculations are wrong.

[outer vale]

nah ChatGPT is just bad

[analog meteor]

if A=2 and B=0 then that means we split into 3 twice which and dont split into 2 any time and the formula does give the right answer as 9 outputs

[outer vale]

parrot without understanding

[frosty owl]

It's not surprising that text generators are incapable of planning ^^

[analog meteor]

screw that i want to test my maths skills for the exam

[covert venture]

I currently have an excel spreadsheet to do it using alt recipes when applicable and available.. holy maths

[analog meteor]

ohj wait im rushing i havent read the whole msg mb im sorry. i thought you were telling us to use caht gpt. sorry

[spare jolt]

No, it's just you're asking a language model to serve you as a mathematical model 🗿

[analog meteor]

yeah if the trainign data stuffs dont give it the information on how satisfactory works then it cant answer questions about satisfactory.

[spare jolt]

I think it has some data about satisfactory, but it may not have a calculator integrated into it

[analog meteor]

yeah i think the formula 3^A * 2^B >= number of outputs. works. at least on a few examples i could solve in my head. and A and B stand for how mayny times you split into 2 or 3 in each "layer" of this balancer thing

[analog meteor]

jace made a video about chat gpt and how much it knows about sf. it was funny to see it say bs like "alien invasions"

[frosty owl]

I think you're focusing on splitting either in two or three rather than combining them freely.
I'm using A and B just to distinguish a multiple of 3 from a multiple of 2, not a balancer made only by splitting 3-ways from one made by splitting 2-ways. The objective being: between multiples of 3 (3xB) and multiples of 2 (2xB) bigger or euqalt to N, which one is the smaller number (closest to N, the one leaving the least loopbacks) ?

[analog meteor]

i think we have 2 diferent ideas on what the goal is

[frosty owl]

There's planning tools made by people too ^^

[covert venture]

To be investigated at a later time!

[analog meteor]

the goal was to make a system to split a belt into what ever number of outputs we want equally. and we can do that by making something split into a number of outputs equal to or more than the desired amount. and to make a splitter system split items equally, we takea belt and split it into 2 or 3 and split all the outputs 2 or all of them 3 times. and each time you split that is a layer of the splitter system and A and B is how many layers are splitting into 3 or splitting into 2 doubling or tripling the outputs.
idk if i did a good job writing that but this is what that formula does

[outer vale]

so what's the goal now that you have the formula

[analog meteor]

make sure it works ig? my goal was to test my own maths abilities for fun

[outer vale]

oh

then yes, ezpz

[analog meteor]

yiay im so proud. i need to flex this math to my friend bcs he thinks hes smarter. (hes not, trust me)

[outer vale]

the real challenge would be working out how to optimise your splitter array

with the first step being deciding what exactly you'd be optimising

[analog meteor]

i would think putting the 3 splits at first and then split into twos at the end

would reduce splitter or belt count or something.

[outer vale]

other way around for minimising splitter counts, want all the 2s earlier in the array

[analog meteor]

huh. oke

[outer vale]

simple case of one of each

• 2 first, goes into two 3s. Three total
• or 3 first, goes into three 2s. Four total

[heavy gust]

@past reef no its very much 15k

[past reef]

I'm calculating with 50.4 uranium rod into 12.6 plutonium rod into 63 ficsonium rod

forgot the trigon SAM so that's 6720+2520 raw SAM

[heavy gust]

12.6 uranium rod -> 16.8 plutonium rod is 168 waste a minute.
So 84 ficsonium rods

[past reef]

as you see the limiting factor would be the trigon, maybe reanimated SAM unless you setup a forward system that can supply 20 dark matter residue per fuel rod

[heavy gust]

im just not gonna do it unless it gets a rebalance patch>

[past reef]

which would limit down the plutonium rod, so I base the calculation on minimum plutonium rod (no alt), and that works until max unconverted uranium rod

[heavy gust]

and if it doesnt, there already is a mod that rebalances it to what it should be in the firstplace

[past reef]

understandable that you want to use the good alt to get better energy gain per rod though

[prisma kraken]

i've done a lot of the math on ficsonium

there's a couple of limiting factors

you're going to be stretched on SAM & aluminum whatever you choose to build for the aforementioned reasons discussed

addressing SAM first. the amount of sam on the map can be effectively doubled with 34 sloops

[heavy gust]

nuclear rebalance patch my beloved

[prisma kraken]

second... ficsonium really wants to be slooped as well

when you sloop either the particle colliders for the ikea lamps or the qencoders for the rods, the recipe chain ends up making as much dark matter as it consumes

because the trigons are pretty fricking pricey, it makes more sense to sloop the encoders

which also effectively doubles the number of rods you make

[frosty owl]

Yeah, you explained it well.
Sorry for late as I had a few things to take care of, but in the meanwhile I mulled over it some more and realized that you are correct in that we need to use exponents too. My initial analysis was too superficial ^^

[prisma kraken]

funny thing about doubling the power made from ficsonium rods... they now make as much power as the plut rods you burn

[past reef]

hm slooping the ficsonium rod machine doesn't sound bad

I can't do max matrix with all ficsonium anyways might as well use some to help this chain

[prisma kraken]

by my tally, if you sloop reanimated sam and sloop the fics rods, you end up needing 72+34 = 106 sloops

[frosty owl]

BTW, have you noticed how smaller the production plans for maximized coupons have gotten? 😆
We went from >40k machines to ~17k

[heavy gust]

i dont see the point of slooping the rods

the only reason for ficsonium to exist is to take care of the waste

[past reef]

not sure about slooping reanimated SAM, I wanna grab some matrix as well so probably not putting all into SAM

[prisma kraken]

you need to make dark matter from somewhere

in any event, where all the match sort has led me is that you can make a max of 225 fics rods/min, or 112.5 unslooped

[past reef]

or I can make an excess of 105 power shards and "sink" it with ionized fuel

wow choices

[prisma kraken]

that's enough to support a uranium 2500/min power build

[heavy gust]

im using 1155. Even if i sloop all sam thats 75%

[prisma kraken]

yeah, there's some sam leftover

[frosty owl]

Don't we have "just" 2100 Uranium/min?

[prisma kraken]

not anymore 🙂

[frosty owl]

So that's the max ficosnium possible limited by... Sloops?

[prisma kraken]

then aluminum, then sam

[frosty owl]

So 106 Sloops all for nuclear stuff, right?

[heavy gust]

it still is very dispoportionate

[prisma kraken]

looks like the numbers just magically fall right there

changing the math around a bit, if you hold the amount of uranium fixed at say 300/min

a 'small' build like that can yield 250gw

[frosty owl]

Just to make sure I understood your factory plan corrreclty...

[prisma kraken]

is it worth it to build ficsonium? probably not. Can it make some decent power if done properly, yes.

[heavy gust]

sadly its not

but i would like them to rebalance it so it actually is worth

[frosty owl]

@prisma kraken ?

[prisma kraken]

it ends up being somewhere btw 72 and 106 sloops

really that's depended on how much sam you need for other stuff

[frosty owl]

If it needs less than 106 Sloops, there should be room for some improvement

[prisma kraken]

yeah, the problem is that any scaling up from there really eats aluminum

[heavy gust]

they should just reduce the trion and singularity cell usage

fixes all the progblems

[prisma kraken]

i can't figure out how to increase the ficsonium yield past 225 up to the next quanta where the clock rates are all the machines at 250% for efficient slooping

in any event a 'max' build that processes 112.5/min plut waste into rods yields 1.875 tw

[heavy gust]

i have 168 waste

[prisma kraken]

use non-fissile uranium -> plut pellet

[heavy gust]

im pretty sure instant cell has higher yield

[prisma kraken]

why do you want to make more plut rods to burn? that isn't the goal

the goal is to maximize the ufr+pfr+ffr burning generator number

[heavy gust]

might be yours

[prisma kraken]

when you decrease the uranium->plut yield, you can fit in more uranium plants

[heavy gust]

i dont want to do anything that decreases yields

[prisma kraken]

so in essence your problem is the same as 'i want to make 6 million screws but the map doesn't have enough iron for me to do so, game broke plz fix'

[heavy gust]

not realy

i want wasteless nuclear without having to jump through ridiculus hoops

[prisma kraken]

dude, that's always been part of building large nuclear

[heavy gust]

yes and then 1.0 anouncement comes with "the solution to waste"
(does not actually solve it)

[prisma kraken]

sink the plut rods. that's wasteless

[heavy gust]

unfun solution

[prisma kraken]

better: built a u2100/min build. sink plut rods. take some rods you make from it sometime later and convert to ficsonium after you've built everything else

[heavy gust]

no point in trying to convince me.
have my plan and frustrations, i already posted on QA, so now we wait

[spare jolt]

let's hope the rumors are true and the loop will indeed make my fuel gens work 100% of the time

[heavy gust]

it did for me

[spare jolt]

though i still have no idea why and how this fucking works

[prisma kraken]

it doubles the pipe capacity

[spare jolt]

ah

i thought it just abuses the wacky flow direction or something

[oblique hollow]

In short: It adds a bypass for flow rates so you dont suffer any flow rate loss on the input

[spare jolt]

ah, i see

[oblique hollow]

As the extra pipe capacity can accomodate for "pressure spikes"

[sage void]

not sure if this is the right chat, but how do you calculate the items per minute for trucks and such?

[dusky dust]

I just set up the routes and then check to make sure it's providing the throughput I need once it's up. It's difficult to calculate the round-trip times ahead of time

With vehicles, if you need more throughput, you can always just add more vehicles to the same route

[sage void]

alr thanks

[dusky dust]

If there's a route that I am concerned about throughputwise, I'll generally time the automated route and then do a bit of napkin math based on the stack sizes to know how much material will accumulate inbetween pickups. (I generally follow new routes through one cycle anyway, to make sure there's nothing in the route that needs tweaking)

Though that's somewhat unnecessary since just watching it clear out the pickup spot once or twice gets you the same thing. :)

[static coyote]

Pre- 1.0 release the wiki had a useful weighted table for every recipe to give you an idea how much resources it used etc,is that still somewhere? I am really missing that shit

[vapid estuary]

TBH I'm glad it's gone because it was iffy the higher up the tree you went, it greatly depends on the lower tier recipes involved. There was abstract concepts of difficulty or space consumed. Personally, I need to be talked out of using the max-output recipe. I aim to make too much of everything, and piling up a couple containers worth while working on other things.

[static coyote]

It was good for a rough estimate for me personally

[amber jacinth]

It was removed due to recipe biases, but an alternative is bring looked into, AFAIK

[wind spade]

it was very misleading and subjective. You'd get better answers by asking random person on the street

was removed for those reasons and there are talks and discussions about how to implement something in a better way

[static coyote]

I see. thanks

[magic island]

basically, it would always assume you used the most WP-efficient precursor recipes, which would throw off the comparison if that's not what you're optimizing for, or if you've already picked different precursor recipes

for example, it would always assume you were using Pure Caterium, since that was the most WP-efficient Caterium Ingot recipe. But that recipe is also a massive power hog. So any alt recipe for a more advanced part that needed Caterium would be treated as having that high power cost baked in. The table wouldn't show you how much less power it would cost with default smelted Caterium

[wind spade]

or use iron wire everywhere, so you wouldn't see copper requirements in wire-using recipes

[eager solar]

regarding liquid loops for alu, am I correct to say that I should make it so the refinery receiving the water byproducts should be clocked a bit higher than needed just to make sure I'm using it all?

[wind spade]

not really. Easiest is to split fresh and recycled water

[eager solar]

I do

[magic island]

the main thing is just to make sure the byproduct water has somewhere to go that won't/can't get jammed up. over-clocking (over-overclocking?) isn't necessary

[eager solar]

I mean the one receiving recycled

since I have a lot of things connected with sometimes 'impossible' clocking like x.33333... I'm worried about a backflow over time

[wind spade]

clock it to .3334

[spare jolt]

I wouldn't do that if I were you. If the amount of water consumed is the same as produced - i don't see a reason to worry further

[eager solar]

I jsut had a problem with the water getting stuck after hours or normal operation, so if overclocking a bit too much can deal with that I'm just going to do that

the number themselves should be fine, but it might be a small rounding error accumulated over many machines since pretty much everything is interconnected in said factory

[magic island]

the most surefire solution for absolute safety is to send the byproduct water to a different process (such as a pure/wet recipe from a nearby ore, or to coal generators)

a loopback with separated fresh/recycled water should in principle be able to run indefinitely, but it's hard to guarantee, especially if there's a repeating decimal in the mix

[eager solar]

like, currently the clock speed of my recycled water ref is of 64.3435%, but with the rounds up the true number might actually be something like 64.343566666 or something, is it going to cause me any trouble if I jsut clock it a bit higher for safety sake?

[magic island]

From how you've described your setup, the deadlock scenario probably looks something like this:

• your alumina pipes get full, so your alumina refineries can't produce any more
• this means your alu refineries can't consume recycled water, so those pipes get full too and block the scrap refineries.
• end state: scrap refineries can't consume alumina because their water output is blocked, alumina refineries can't consume water because their alumina output is blocked

I've had loops like this that were able to run indefinitely, and functionally similar loops that would clog up after some hours, and I couldn't for the life of me tell you what the fatal difference was.

But I don't believe this is a situation that can be corrected by clocking refineries on either side to be a little extra greedy.

[eager solar]

my setup is actually more complex and involves sulfuric acid too

I think the situation can be solved as long as I make 100% sure that the recycled water is fully used up

[magic island]

oh, you're using instant scrap as well

the most success I've had with instant scrap is via chaining rather than looping back. using one blender's byproduct water to make the acid for the next blender, and so on

[eager solar]

and to that effect, and to 'compensate' for rounding drifts, I'm thinking of slightly ocin the recycled refinery

I'm not using instant scrap, but electro as well as 1 of normal recipe

this is quite a big factory, hence why I'm worried of rouding errors snowballing

[spare jolt]

Wait, then how is your aluminum setup takes sulf. acid?

Leached ingots?

[eager solar]

my factory involves acid

as in as a whole

[spare jolt]

Ah, the whole factory

Ok then

[eager solar]

and since, one thing leading to another everything that invovles water, alu, acid and oil are interconnected

it's an equilbrist play

for reference sake: https://www.satisfactorytools.com/1.0/production?share=X9L3PYDUYkR5lKbbpZcg

[quaint condor]

OK, riddle me this ... I have 42 Nuke generators, output for waste should be 420/min, have 5 blenders setup to handle the waste and have them set to do a total of 468.7 nuke waste/min, yet the nuke gens are still backing up with waste ..... Is it due to not load balancing the waste accurately or something?

[eager solar]

(does it share the right tab or all of them?)

[spare jolt]

Just the tab you copy

[magic island]

ah yeah, this would be a real pain to manage

nothing here consumes water in a rate that matches the byproduct rates, so you have to split out the clock speeds in ugly ways

[eager solar]

yeah, that's pretty much a megafactory in its own right, I want to learn how to do them right as a 'training' of sort before nuclear (and also because big is fun)

[amber jacinth]

Granted, that entire production chain has ugly numbers... have you considered making them nicer at all? Would make it slightly easier to manage.

[eager solar]

I will for future factories

but I wanted to challenge myself

And challenging it is

I'm quite a bit miffed tho, I troubleshoot the whole thing for hours and honestly tought I got it right for hours on end

[magic island]

in this sort of situation, I would not loop the byproduct water back into the alumina whatsoever (or any other interdependent chain tbh). the numbers aren't clean enough to feel comfy tuning that sort of setup

I would just send the water to some ore node (wet concrete or pure ingots or whatever), and clock those refineries a bit extra high so they're guaranteed to consume whatever rate of water they happen to get

[eager solar]

I don't actually have leftover nodes in a wide radius, everything has been used by that fact and another big one

guess I'm just clocking a bit higher just in case hten

like one decimal higher

actually, do you mind doing the math for the alu sol? I'dd like to double check

I think I need 3 normal ref at 100, 1 at 39.6566 and the recycled one at 60.3435

the total being 400%

[prisma kraken]

i'd highly suggest using classic battery instead of the default recipe

[eager solar]

as mentioned earlier, I wanted to challenge myself and tried to be really efficient hence why I use so many diff recipes

[prisma kraken]

well default battery uses more of everything that matters

[eager solar]

yes and no, I planned my prod assuming a mk4 belts of every mats I needed, this is hte result the planner sent me

[prisma kraken]

on top of that, i have to say that i've never figured out how to get a recycling loop to work properly with that recipe

[magic island]

I don't really know which math to do for you TBH

part of the issue is that you're using a mix of recipes for certain things, and mixed in odd ratios at that. which means none of the numbers can really start or finish clean

for example, you have two different Radio recipes, which means you need a weird number of casings, which cascades back to needing a weird number of ingots

and to produce that weird number of ingots, you're using both default and electrode scrap, in an also weird ratio

[prisma kraken]

to the point where I haven't tried in a year or two. when i've needed to use that recipe, i've just sank the water with coal gens 😄

[vapid estuary]

it would help if you share a link to your planner session

[eager solar]

I'm going to do that for future prods but I no longer have available resources beside oil and nitro in a wide radius

[vapid estuary]

there's a share button in SCIM

[eager solar]

it was linked above

[vapid estuary]

that loaded a cluster nobelisk factory for me

[eager solar]

it is, amongs other things

it's a decently big factory

it is meant to be a big and compelx factory to challenge me before going nuclear

so I went all in

[vapid estuary]

this is absurdly complex

my first nuclear plant used trucks to bring materials in which were made from first principles. it wasn't hard, just a tour of the swamp

so, effectively, stand-alone

maybe i don't understand the question, we were discussing batteries just now?

[eager solar]

originally: I discovered a very slow problem of water backing up, which lead me to ask if ocing the recycling refinery a bit higher than what the math returns me as a safety measure would be a rpoblem

and my slow as in everything went perfectly fine for hours on end

[vapid estuary]

am i right in understanding the 'recycling refinery' is not connected to any fresh water?

[magic island]

the way I would adjust this is

• turn off the "maximize" entries. now that you can see what you can get from maximization, you can replace them with cleaner values in the same ballpark. (side note: "maximize" aims for an equal output for all items set to be maximized, and I have no idea why you'd want to balance an equal number of supercomputers / nobelisks)
• untick recipes until you're only using 1 recipe for each item (there are exceptions, but letting the solver recommend two RCU recipes in wacky proportions will give you ugly results)

[eager solar]

it was, but I changed that after dicoverign the problem

I initially used a vip

[vapid estuary]

Bexy's description of 'maximize' is exactly right btw

[eager solar]

I'm going to do so for future factories

but right now, I want to see if I can fix this one

and obviously, I'm not going to rebuild it from scratch

[vapid estuary]

how far into construction of this monster are you?