#math-and-meta

1 messages · Page 518 of 1

vast jungle
#

so 75% more power?

torpid robin
#

for a total of 689062

#

thats a dam good number

#

with only 295 waste?

#

the waste is the part i confuse myself on

bleak coral
#

no alts for plutonium makes 31.5 plutonium rods from 3150

torpid robin
#

ok. wtf is my calc saying hold up

#

shit. it had one alt clicked 🤬

#

yea but ok.

#

so thats even better

frosty owl
#

I'm on mobile, so not the best thing to run the calculator 😅

torpid robin
#

so now its only 236 bauxite and 1890 gas

bleak coral
torpid robin
#

thats amazing. thats a huge savings and not many loss of rods

frosty owl
#

With NO ALTS?

torpid robin
#

yea

#

like thats amzing

frosty owl
#

Ye, it's pretty good
Tbf, the alts do cut down a lot both on machine count and number of steps

torpid robin
#

so now you get 196,875mw from plutnonium rods

#

at the cost. of well. absolutely stuff all

frosty owl
#

Can you compare with the instant recipe, pressure cube recipe, then both?
I'll just do it myself later otherwise :P

topaz hedge
#

hiya ven :p

#

while we talking about nuclear.. here's my little setup for uranium fuel

unknown.png
torpid robin
#

pressure cube

topaz hedge
#

the pressure cubes are super expensive and takes a pretty big chunk out of what I could use to make thermal rockets tho x.x

bleak coral
#

I'm still hoping they give another efficiency alt for uranium power, though honestly I'm not sure we actually need it

torpid robin
#

1k bauxite
3220 gas for teh instant

topaz hedge
#

I doupt it. I kinda like uranium how it is. we have the one alt to make fuel rods

torpid robin
#

you get 46 rods with the instant

bleak coral
#

what do you do with the people who have the alt and spent a hard drive on it?

torpid robin
#

so if you want a boost. probs the instant would be better than the cube one

topaz hedge
#

if you want/need max power build a plutonium setup. I think that's what they had in mind

torpid robin
#

you pay the same amount of gas adn 1500 less bauxite to get about the same

topaz hedge
#

I dunno. maybe another plutonium alt that makes a cleaner ratio?

torpid robin
bleak coral
#

default is a really clean ratio, 1:1 uranium rods:plut rods, 100:1 waste:rods, and 100:20 u.waste:p.waste

vast jungle
#

whats the problem about "having the Alt" ?

bleak coral
#

those are some clean numbers

topaz hedge
#

that's the default.

#

using all the alts it's kinda werid? I couldn't make it work in the calculator

bleak coral
vast jungle
#

not every alt is worth the effect...

torpid robin
#

they have more hdds than alts

#

and they added more hdds anyway

topaz hedge
#

Yeah, but it's not fair to be stuck with a dead recipie.

bleak coral
#

barely, it's like 2/3 more than the recipes if you missed the christmas thing

vast jungle
#

I would guess they will patch out anything that doesn't work anymore later

#

that might be why they give us more HDs than necessary... because you might have wasted a few HDs because of patches

bleak coral
torpid robin
#

ok so from what i can tell. its fertile uranium that does the damage

bleak coral
#

I'm sure they'll come up with a solution, just hope it doesn't get overlooked

torpid robin
#

makes you use more uranium .

#

more gas

#

so id say full nuclear. then use the instant alt. and no others

vast jungle
#

so what are the comparisons for the three cases now?

  • nuclear with waste to dump
  • nuclear with waste sinked
  • nuclear with plutonium power
#

I would assume that if you want to sink nuclear waste, you will go for the "most resource efficient way" but if you go for "full power", you might want to go for "max output"

#

if I understood you right "max nuclear power" is "Non-fissile Uranium (Default)" + "Instant Plutonium Cell (Alt)" + "Plutonium Fuel Rod (Default)"

torpid robin
#

max nuclear->plutonium isnt possible.

#

you run out of resources

#

so sure go waste to dump. but with it now producing 100pm thats rough. if you go max nuclear at 31.5 rods. thats gonna cause some issues

vast jungle
#

so "maximum electric power" is nuclear without plutonium?

torpid robin
#

so best bet is turning it to plutonium

torpid robin
vast jungle
#

(okay, have to be a bit more clear)

#

"max nuclear" is a bad term

torpid robin
#

cause id advise turning it to plut rods

torpid robin
#

so lets say max nuclear(usage of all uranium on the map using only nuclear processing)

vast jungle
#

options:
1.) original "max nuclear electric power" without plutonium
2.) "max nuclear electric power" while turning waste into plutonium and sink it
3.) "max nuclear electric power"... which means running some powerplants with plutonium

torpid robin
#

then id say you have max plutonium power( not feasible due to map resource limitations)

#

i mean 1 when i say max nuclear

vast jungle
#

wouldn't you get more electric power by running the same plutonium setup and using the rest of the uranium for normal nuclear power (and store its waste) ?

frosty owl
torpid robin
#

it depends on what resoruces you try to save

#

the plut(gonna spell like this cause im over writing plutonium) processing uses alot of other resources that are needed in other production lines

#

one that uses the nitro gas uses uranium. but also alot of gas.

vast jungle
#

so we got even more options depending on the situation... that makes life difficult but the game more interesting I think

torpid robin
#

you remove that and it allows you to turn all uranium into uranium fuel rods.

torpid robin
#

mind you i havnt comapred power usage or machine count in any oif this.

#

its based solely on resource usage

frosty owl
#

I think the options can be summed up as:

  1. Have nuclear, use plut alt recipes to deal with the waste however you prefer: use many resources for few machines and many rods (instant plut+pressure cube); use less resources for fewer rods but more machines (standard recipes); get a TON of rods with fertile uranium alt
  2. Have max uranium rods and expend a lot of resources to get a good number of plut rods (alt recipes)
  3. Have maximum uranium and expand as few resources as possible to make the minimum number of rods (no alt recipe)
torpid robin
#

i think 3 is best

#

you could even use the instant recipe for a boost. and not much resource usage

frosty owl
#

Using the rich uranium recipe with case 2 and 3 is madness IMO, since if you need max nuclear, you'll probably need the materials used in that alt for other things in the first place

torpid robin
#

thats dead right

#

i think either way. if you go nuclear. you are stupid to not do the plut processing

frosty owl
torpid robin
#

yea i wouldnt disgree with that

vast jungle
#

pressure-cube might be a nice way to keep the setup compact when you don't want to max out anything

torpid robin
#

came with the same amount of plut rods

#

but higher gas and bauxite count

frosty owl
torpid robin
#

hence why i felt instant was better

frosty owl
#

It's a lot of machines to deal with all that waste after all

torpid robin
#

like. thats alot

#

compared to like 300?

frosty owl
vast jungle
#

you don't need that many machines to deal with uranium waste (per nuclear powerplant)

frosty owl
#

302.4 minutes to be exact

torpid robin
#

yea i know. i just feel you are also missing out on a decent amount of extra power

#

fo no real cost apart from time and machines

vast jungle
#

or did you revise the numbers 😉

torpid robin
#

depends what alts you use

#

3 options IMO

frosty owl
torpid robin
#
  1. no alts
  2. instant alt
    3.cube alt
#

actually 4. cause you could use both

#

2 and 3 create the same amount of extra power

vast jungle
frosty owl
#

"Many" is relative. I'm just saying it takes more machines then the uranium part
Eg: if you have 10 machines to make uranium rods, you'll need 20+ machines to deal with the waste

topaz hedge
#

if you want max plutonium rods per waste it's quite a few

vast jungle
frosty owl
#

Ye, I'm referring to that too, in this case

#

We could just check on the calculator... But again, I'm in mobile, don't wanna 😅 😆

topaz hedge
#

then it's not too bad

vast jungle
#

so for "least resources and power" the default recipes offer a cheap way to get rid of the waste... I think its 200 waste for 1 particle accelerator... so the power requirements for this one don't matter that much

torpid robin
#

so assuiming you have made all uranium into uranium rods which is 31.5 which makes 393,750

for the plut processing

no alts at 31.5 rods makes 196,875 mw

instant alt / cube alt(pretty close to the same) makes 291 687 mw with 46.67 rods

both cube alt and instant alt will make 70 rods with 437,500mw

topaz hedge
#

This doesn't include any raw resource input, silica, al, nitrogen, copper, steel, etc

unknown.png
vast jungle
#

makes sense that "no alts" is the best for "cheap waste recycling" with these numbers

topaz hedge
#

That's for 9 a min

torpid robin
#

adding fertile just isnt worth it imo as it costs uranium and makes the numbers stupid and impossible anway

topaz hedge
#

900 waste in, 9 rods out, with no alts

torpid robin
frosty owl
torpid robin
#

each time you add an alt. you add bauxite and nitro gas

#

il get the machine and power count for those options

vast jungle
#

so "just store the uranium waste" is a really stupid strategy now... you don't gain much of anything with it and you deal with an increasing problem

topaz hedge
#

to be fair, you could think of it like a backup too. if some how you do use all your uranium power.. you can burn plutonium for a little more, even using the cheapest way

torpid robin
#

no alts

unknown.png
vast jungle
#

names for the three options:

  • clean nuclear
  • efficient nuclear
  • max power nuclear?
    😉
torpid robin
#

this is from max uranium fuel rods fyi 31.5 rods 3150 waste

vast jungle
torpid robin
#

instant alt

unknown.png
frosty owl
torpid robin
#

pressure cube alt

unknown.png
#

both alts

unknown.png
frosty owl
vast jungle
#

14466 MW for 3150 waste... means ~ 23MW per nuclear powerplant for waste removal

#

thats less than 1%?

#

is this WITH the particle accelerator?

#

no its without the accelerator...

torpid robin
#

yea it may be without

#

just work on an avg of 1000mw

#

cause thast what it is suppose to be

frosty owl
torpid robin
#

its not overly expoensive in power usage

vast jungle
#

so its an additional 12.5 MW per nuclear powerplant

torpid robin
#

i was meaning for the accelerator

vast jungle
#

so for the "clean nuclear" option you loose ~ 36 MW per nuclear powerplant...
or ~ 1.5% of power

#

machine count per Nuclear Powerplant: 1/20 Blender (Non-fissile Uranium), 1/40 accelerator (Plutonium Pellets) and 1/20 Assembler (Encased Plutonium Cells)

#

so setup groups of 20 or 40 nuclear powerplants up for nice waste-removal numbers 😉

torpid robin
#

im not gonna be able to fit it all in the swamp 😱

topaz hedge
#

Nope.

vast jungle
#

just stack two layers of powerplant on top of each other 😉

topaz hedge
#

Nope lol

#

I went with 45 nuke plants. So my numbers will be a little wonky. I'm used to it by now.

vast jungle
#

(for the waste removal)

topaz hedge
#

OC is going to happen regardless haha

sullen cloud
#

Remaining question is if plutonium waste will someday be required for another item or task

vast jungle
#

so it might make sense to build a nuclear setup (with waste removal) somewhere AND have space to replace it with a "plutonium afterburner" in the future

topaz hedge
#

Doubt it

torpid robin
#

kinda doubt it tbh

vast jungle
#

hmm
how fast does radioactivity goes away if you stop feeding the machines and the powerplant?

torpid robin
#

i think best way to go about it. is use all uranium rods into plut rods. with 590,625mw thast a decent amount

vast jungle
#

@topaz hedge thats tiny... 🙂

torpid robin
#

add my 133000 mw of turbofuel and im set

vast jungle
torpid robin
swift robin
#

with tier 10 i hope they add in something that requires massive amounts of energy to use

topaz hedge
#

it's a small setup, only uses one uranium node (: although that's 600 uranium ore.

vast jungle
swift robin
#

no lol

torpid robin
#

yea 100 of them and il run outta power

swift robin
#

i do wonder what kind of recipes the accelerator might get tho

torpid robin
#

i think my pc will also run outta power

swift robin
#

maybe we get to make mini black holes or something

vast jungle
#

the nuclear pasta might be the start of building a ship with a warp/FTL drive

torpid robin
#

they have super position oscillators and quantum computers in the game

#

i asuume things to do with them

sullen cloud
# topaz hedge Doubt it

yeah, it’s just a guess. But it would also be quite unlogical to produce an item with an dead end just for some surplus of power

#

that would only attract max orientated players

topaz hedge
#

Well, they already had that with nuclear waste, and people either hated it and didn't touch it, or they loved it.

#

now they've added an option b to nuclear power.

sullen cloud
#

Yeah, but I am talking of the final meta, not a semi-final of the game

topaz hedge
#

I still highly doubt it

sullen cloud
#

i have like 800 hours playtime of uranium waste and it was not an issue at all storing it at a distant place

vast jungle
#

nuclear waste storage is easy to do even for really long games... so its only a matter of taste

sullen cloud
#

I am just speculating but the reward of extra power just for all the effort with an dead end item seems not rewarding enough for this tech tree

vast jungle
sullen cloud
#

Yeah, but power in general is no big issue, even for huge builds

torpid robin
#

its a bit closer now

#

both vencum and i have checked it over for a big factory. thats what kicked this discussion off. and you can get pretty darn close to using all the power you have for a very end game factory

#

and thats without them having a t10

sullen cloud
#

I produce at least one machine at 100% for every item in the game and the constant power usage is like 25GW pre u4. And I guess that 35-40 will be enough including all U4 items

torpid robin
#

yea but in the way of things thats pretty small

#

for the people who can spend 2.5 -3k hoursa on a save and use most resources. you need that power

sullen cloud
#

Your end game factory is maximizing for certain items and/or inputs. I am considering a build as described above

#

which I still consider an above average huge build

torpid robin
#

if that is the size you go. you can easily get away with out using any nuclear at all

sullen cloud
#

yeah, absolutely

vast jungle
#

which is nice... because its another valid option

#

"produce everything" doesn't contradict "no nuclear"

torpid robin
#

nuclear is a choice at the end of the day

#

like hell i could make another 2k turbo fuel and that would be waaaaaay more than enogun for 90% of the people who play

vast jungle
#

not sure if "coal only" is a valid strategy for the endgame 😉

topaz hedge
#

if you have one machine producing at 100% of every item in the game, I'm sure you can pull it off lol

torpid robin
#

you could.

#

itl be tough

#

you may need to make some steel with coke

topaz hedge
#

use half the coal in the world for power, that's roughly 75,000

torpid robin
#

or or or. use it all and get 140gw 😂

#

oh the pain

topaz hedge
#

I'm afraid I don't like building coal gens that much lol

torpid robin
#

most i built in a save was 14000

#

that was enough for me

#

so this time i only built 128 😂

vast jungle
#

14000 coal gens?

torpid robin
#

whoops

#

14000mw worth

#

188

#

stuff that

vast jungle
#

ahh... okay

#

thats sounds unreasonable but still "okayish" 😉

torpid robin
#

if you think thats unreasonable. come help me lay all my ore refineries

frosty owl
vast jungle
frosty owl
#

Btw, with how clock is now, can we overclock power gens to get an effective 200% usage exactly?

wind spade
torpid robin
#

greeny the man himself comes to the rescue

wind spade
#

the closest we can get is either 246.2289% to get 200.00001% or 246.2288% to get 199.999948%

torpid robin
#

😂

#

its soooo close

frosty owl
#

Goddammit disappointed_snutt
Thanks, Green

wind spade
#

the OC number to get exactly 200% seems to have infinite decimals

#

246.22888266898325689987861383354862197522755222016355886741310764 ...

#

exact number is 200*2^(3/10)

frosty owl
#

That's enough decimals to warrant a "if I had one penny per decimal... " comment hehe

wind spade
#

also what I am doing with my life

frosty owl
#

Counting pennies? ^^

frosty owl
sullen cloud
torpid robin
#

well consdiering i have 3600k worth of rubber made. and 3600plastic half made. i think i know 😂

frosty owl
#

Same deal. 1 machine per space elevator part is a pretty basic setup, imo

torpid robin
#

never wanna see another refinery again

frosty owl
#

Wanna work on my pure caterium for me? ;)

torpid robin
#

oh hay! now lets go do ore refinery wooooo!

#

please shoot me

torpid robin
sullen cloud
#

And it is 100% for each and every item in the game, not only some of them

wind spade
frosty owl
#

Thank you very much magic math man <3

wind spade
#

100% is a convenient number tbh 🙂

frosty owl
#

I just wanna cut down as much as possible on the total number of machines in my saves 😅

frosty owl
torpid robin
#

im doign my own. why would i wanna do yours lol

#

are you using pure copper?

torpid robin
#

waaaaay bigger. costs higher tier parts.

#

alot of power. id be happy

frosty owl
wind spade
#

a lot of power = more machines to make power = more fuel needed to make power = less resources available to produce stuff = big sad

frosty owl
torpid robin
torpid robin
wind spade
torpid robin
frosty owl
#

Btw, all "copper ingots" setups in picture are on 2 floors xD

wind spade
#

grammar nazi inc

frosty owl
wind spade
#

we need :grammarhammer:

#

you swapped the iys and also added one extra, that's like failing as much as possible xD

frosty owl
torpid robin
#

teeny tiny 😉

frosty owl
#

How much machines you think that is...? XD

#

||Totally not trying to trigger greeny ☝️ ||

torpid robin
#

oof. lets say 1k

frosty owl
#

Oh, that's a good estimate!
Average clock is about 200% though :P

torpid robin
#

so many slugs

#

dam i needa get a doggo farm

vast jungle
#

Vencams motto is "OC everything!"

torpid robin
#

not completely a bad one

unkempt acorn
#

finally completed my nuclear fuel rod set up. should i encase the place in a building thinking_helmet

unknown.png
torpid robin
#

even now with the copy paste

wind spade
torpid robin
unkempt acorn
#

Hmmmm

wind spade
torpid robin
#

whoops

vast jungle
unkempt acorn
#

low key flirting i see?

wind spade
torpid robin
#

i have mine opn wee platforms with the extractors for each one dfirectly below

vast jungle
unkempt acorn
#

you kow what... i will build the reactors ontop...

torpid robin
#

i mean pumping water is gay

vast jungle
#

hmm... when I finally build my nuclear setup I should build a TOWER of nuclear powerplants 😉

torpid robin
#

but its gotta look pretteh!

unkempt acorn
#

actually nvm... i need 20 reactors...

frosty owl
hybrid dune
#

1+1=

torpid robin
wind spade
#

another person that thinks they're funny with 1+1 but instead it's just spamming

frosty owl
#

Tbf, I think the idea before was valid: OC everything to 200%, save on half the machines and still have enough power to finish all ores (roughly).
But now.... hehe hehe hehe hehe

hybrid dune
unkempt acorn
vast jungle
unkempt acorn
#

Why thank you

vast jungle
#

luckily Chrome allows me top stop GIFs from moving

keen oyster
vast jungle
unkempt acorn
#

finally! nearly took me 3 hours, finished my factory for nuclear rods

frosty owl
green mango
#

ok so i have a problem... 1/2, 1/3 1/4 etc up to 1/10th splitting of from a conveyor belt isnt a problem.... however i kinda got stuck atm since i need 1/11th?? anyone got a good setup for that?

wind spade
#

manifold

forest minnow
#

just use a manifold at that point mate

wind spade
green mango
#

i need 30 to one machine and 3 to the other tho

forest minnow
#

still applies

wind spade
#

still, make one splitter to two belts

green mango
#

would remerging the first 10 achieve that on a manifold?

wind spade
#

it'll balance itself

# 
--S--> 3
  |
  V
  30
#

if you provide at least 33 items/min to it, it'll work

green mango
#

ahh yes facepalm if i fill the belts it won't use more than 3 on one machine thus rest go to the other!!

wind spade
#

since the 3/min machine will get more items than it needs, filling it's input and the belt before it, when it's full up to the splitter, all the extra will overflow to the other machine

#

yeah, it's the same technique as in manifold

forest minnow
#

that's the beauty of a manifold

green mango
#

lol feel kinda stoopid now 😛

wind spade
#

just make it overflow 🙂

green mango
#

too many belts and splitters going everywhere i guess... was so totally focused on needing to set up a complete system for it

#

tnx

frosty owl
green mango
#

Yeah I'll fill both from start

wooden pond
#

You don't

#

why did you delete your msg?

pearl gazelle
#

because I hadn't finished typing, hit enter by accident

wooden pond
#

then just edit it...

pearl gazelle
#

Howdy math folks! Got two questions for the channel!

First - I know in the past if your manifold goes deep enough (50+ buildings in the line or so?) you start losing some efficiency... I know Greeny had a calculator for that in the day, wondering if that was still a thing.

wooden pond
#

Either way no form of splitter/merger setup has inefficiencies

#

It's impossible for it to be less efficient

pearl gazelle
#

Second Question - does anyone know if anyone's done the math on sink points for getting the most points out of any particular raw resource?

#

Great to know, glad they fixed that

wooden pond
#

because it's impossible to BE an issue

#

i think your getting confused with the time it takes to build up resources

pearl gazelle
#

could have sworn there was an issue with splitter and merger code that could cause an issue from that way back when

frosty owl
abstract thorn
#

can someone help me balance some pipeline stuff for aluminum?

versed violet
#

Generic advice: put the refineries turning solution into scrap slightly higher than the previous ones and put valve on the water output from them before merging to inputs

abstract thorn
#

So more specifically, my water input is 900. I have 300 coming from scrap output, and 600 coming from water extractors. Should I balance the pipes somehow, or should I underclock?

versed violet
#

underclock so input = output. the elevation and valve are to ensure the output gets emptied first. Otherwise, the extractors will overfill and whole thing deadlocks.

#

or, you can feed one of the stage 1 refineries with all the water coming from stage 2.

#

No connection = no deadlocks

austere notch
#

Anyone know if a splitter calculator exists?

abstract thorn
#

I underclocked all of my refineries to take 150 water/min

austere notch
#

not so much for the few levels deep stuff - dealing with a few fractions is fine... but more so for things like determining flow if you feed part of split output back to the input and how that impacts the other outputs down the line

oblique hollow
austere notch
oblique hollow
unkempt acorn
#

how many power storage's are considered to many? thinking_helmet

oblique hollow
unkempt acorn
#

well im just about to build my nuclear factorys x 20. but currently i have 12900 MWH saved

oblique hollow
#

show a screenshot of your power graph

unkempt acorn
oblique hollow
green mango
#

how long time do you need for corrective measures if power crashes? id say that is the aim for how large your backup needs to be.

unkempt acorn
unkempt acorn
#

some day in the near future, ill cover the entire grass area with power storage's

green mango
#

lol easy then 😛 keep on adding till your FPS drops 😛

deep root
#

Then box it up and keep going!

unkempt acorn
green mango
#

hmm thats a FEW 🙂

austere notch
# oblique hollow ?

Yah basically - was thinking a little more complex because with your example I believe your output either equals what it would have been originally or it approaches it as a limit

#

versus

unknown.png
torn imp
#

Guys, how do you choose the best place for your bottom floor?

#

on a large factory with multiple production chains?

#

from the most bottom spot on the map or a relatively higher spot and transport resources up?

bleak coral
#

high enough so that any terrain it might bump into won't bother you

#

or higher for aesthetic reasons

austere notch
bleak coral
#

oh right I guess water is a consideration too, I don't think a few pumps are much of a bother though, and terrain doesn't always lend itself to building on the same level as your water source

#

so depends™️

torn imp
#

The terrain is not flat and the space near the water source is lower than most of the mines in that aria, mainly it will be my first base

#

but the terrain is uneven...

#

i could go with the water level or a little higher, and start building everything on the next floor

#

since it is the first factory on this map i guess water will not be necessary until much later, where i can handle everything with outposts.

#

right?

bleak coral
#

yeah, don't need water right away

kind spindle
#

you need water for coal power relatively early though

torn imp
#

all my coal is 800 KM away

bleak coral
#

that doesn't need to, and probably shouldn't, be the same factory as your first factory

torn imp
#

i will need water close to those nodes, the base will use the water on late recipes (max) the rest will be produced near larger water sources. The one near can fit 6 pumps max.

#

Thank you for your feedback! Good gaming to you all!

#

I almost forgot... should i focus on a N-S grid or should i build the tiles based on the terrain? mainly i will use sky lanes to travel.

bleak coral
#

Mostly just an aesthetic choice. I prefer molding around the terrain cause it gives me building challenges and makes stuff look less floaty and blocky.

But, if you want it so everything is working off the same foundation grid (like where all the foundations are turned the same direction), it's better to orient to the cardinal directions as best you can. I've heard there's less deviation over long distances that way.

wind spade
torn imp
#

ticks on the map, sorry...

green mango
#

sooo... smart plating, versatile frames and auto wire... are they used for anything else than unlocking tiers?

#

VC, but close enough 😛

iron prairie
#

Three things:

  1. Sink points. They're relatively complex items, so they're relatively nice for farming sink points.
  2. Smart plating is used for modular engines, required for the third Elevator tier.
  3. Auto wires are used for adaptive control units, required for the third Elevator tier (and is also one of the best items for sinking)
green mango
#

the sink it is then got plenty of them in stock now so overflow can be sinked untill i get the last tier going

iron prairie
#

Huh. Apparently, with U3, even with max turbomotors, you can still get 586 ACUs/min, which is roughly half-ish of the sink points of max turbomotors. It's a nice way to "clean up" on other resources after exhausting bauxite.

tepid gull
#

whats the best way to balance normal and inpure node to output 780

austere notch
tepid gull
#

yeah, mk3

austere notch
#

Overclocking not an issue?

tepid gull
#

no

bleak coral
#

probably 200% on the normal and 100% on the impure, the power efficiency is better on the higher purity nodes so you want to overclock that one

austere notch
bleak coral
#

sorry wait, 780, 250% on the normal and 150% on the impure

austere notch
#

yah mine doesn't actually give you percentages and I didn't calculate for efficiency

bleak coral
#

I haven't run the numbers exactly, but since they both use the same amount of power you get more stuff per MW outta higher purity nodes

austere notch
#

but it would translate to 200% impure and 225% on the normal

#

Yah true...

bleak coral
#

so you want to overclock the normal one first

austere notch
#

makes sense

austere notch
bleak coral
#

there might be a more precise balance, that's just my first gut feeling

#

cause the overclock isn't linear in power usage

austere notch
#

Nah I think you're right - you get more items/MW on the normal so you have to max it out first and then there's only overclocking the impure once that's done so...

#

no other option really

bleak coral
#

but mathematical proof is always good

austere notch
manic oak
#

You can also do this by hand. Solve for y in terms of x, sub out, then differentiate with respect to x to find the value of x that leads to an extrema.

austere notch
#

@bleak coral @manic oak - you have any idea how to express this mathematically?

unknown.png
#

It's basically just the splitter problem I posted earlier with non-relevant numbers put in

fierce ruin
#

Maybe a tree or a matrix

versed violet
#

Isnt this simply a set of equations?

austere notch
#

well - because it feeds back into itself I was thinking maybe a recurrence relation?

fierce ruin
#

you can just go the sheets and find all the combos of 2 and 3 then find their product

#

feeding back limits throughput, might be a infinite series thing

#

also I believe you're referring to prime splitters

bleak coral
austere notch
#

How would we determine the final output of the single belt on the last splitter once the system settled out?

#

ok looking

#

oh wow - that's cool

fierce ruin
#

sum(x = y percent of n) then set n=x and repeat

#

so yes, recursive infinite series I think

austere notch
#

@bleak coral, @fierce ruin - that's exactly what I was looking for, thank you!

#

that prime #s wiki

bleak coral
#

the major drawback is not all the prime-split configurations can do full belts, it lowers the max throughput

austere notch
#

now if I really wanted, I need to review the math behind it so I can do random stuff

bleak coral
#

if I was doing prime splits regularly that'd be math I'd be most interested in: how to figure out the max throughput of a loop-back splitter

versed violet
#

Wouldn't it be simply 1 - r/n, where n represents the total outputs, and r is the number of return belts?
Or wait... me needs to think.
⏳
Let x be input and Y the input belt capacity.
For an array of o outputs and r loopback belts, the total throughput is given as
X = Y * ( 1 - r / (o + r))
Ex, for 780 belt and 11 outputs, one loopback belt

X = 780 * ( 1 - 1/12)
X = 780 * 11/12
X = 715 42

bleak coral
#

no I think that's right, cause it needs room for that much to be put on the belt

fierce ruin
#

⏳ thinking

#

I probably janked it up

versed violet
#

If I were a teacher, I would give such questions at final exam, and watch the students squirm

manic oak
#

@austere notch You can write the input as a sequence in time, assuming each time step is how much time it takes for the full set of items to travel through it, and thus you obtain the amount traveling through as a function of time. The recursion relation for the input would be:
a_(n+1) = a_0 + (2/9)a_n
Where a_0 is the amount flowing through the input initially.

versed violet
#

We were only given easy ones at university, like "What is the full name of the course you are writing exam for"

manic oak
#

The amount flowing through the question mark belts is (1/9)a_n

#

This is where the fun begins

austere notch
versed violet
#

There is 2 loopback belts, so it would be 2/9 ?

austere notch
manic oak
#

^

austere notch
#

He was probably referring to the 3rd feed that flows out from the middle splitter

manic oak
#

Things are about to get a little tricky, let's see if this works:
a_n+1 - (2/9)a_n = C (eliminating the a_0 name to make it look nice)

austere notch
versed violet
#

I prefer to think of it in term of limit, where Y goes to 780 as t goes to infinity, cause we are not concerned about the intermediate steps, are we?

austere notch
bleak coral
#

same, generally I discard things like warmup times because they're usually trivial

manic oak
#

let's split up that (2/9)a_n as a_n - 7/9 a_n, giving us:
a_n + 1 - a_n = C - 7/9a_n
And now we can use calculus:
dx/dt = C - (7/9)x
=> x(t) = De^{C-(7/9)t} = x_0e^{-7t/9}

versed violet
austere notch
manic oak
#

This doesn't seem right, as this asymptotes to zero, which I don't think can happen here

austere notch
#

But I was wondering if it would lead to calculus...

manic oak
#

Everything can lead to calculus

#

Calculus is just a continuous limit of a finite system

wicked tinsel
#

it was just geometric series wanst it 🤔

fierce ruin
#

discord notation uck

versed violet
manic oak
#

Hold on, I'm checking my work and I need pizza

#

Ah, I see what I did wrong

versed violet
#

The equation for volume of Pizza, with radius of z and height of a is
Pi * z *z * a

manic oak
#

tau*dx/dt = x_0 - (7/9)x

austere notch
manic oak
#

Redefine variables:
y = x_0 - (7/9)x => dy/dt = -(7/9)dx/dt
=> -tau*(9/7)dy/dt = y
=> y = Ce^{-tau(9/7)t}
=> x_0 - (7/9)x = Ce^{-tau(9/7)t}
=> x = (-9/7)Ce^{-tau(9/7)t} + (9/7)x_0
=> x_0 = (-9/7)C + (9/7)x_0
=> C = -(2/9)x_0
=> x = x_0[1 + 2/9e^{-tau(9/7)t}]

austere notch
#

Not sure if I put that into the calculator correctly...

versed violet
#

x_0 should be belt capacity here?

manic oak
#

x is the throughput of the first belt

#

note that this assumes unlimited belt capacity

#

You can actually use this to calculate how long it will take for the belt to reach 780 items/min

torpid robin
#

I feel you guys complicate belt throughout so much lol

austere notch
#

So can we use the to calculate the output of any iterative feedback system I'm thinking?

river quarry
#

this is so complicated i came here to have fun and do conveyor noodles what is this

manic oak
#

Yes, it depends on the split, though

#

But if you keep feeding a belt back into itself, it will eventually reach the max throughput

austere notch
#

Just calculate the first iteration of output and phrase it in terms of the fractional amount getting fed back into the system? (sorry I know my math language is attrocious)

manic oak
#

This all stems from the fact that you can approximate f'(t) as (1/tau)*(f(t+tau) - f(tau))

austere notch
#

I don't know what tau is ><;

manic oak
#

some small time step

#

like 0.001 seconds

versed violet
manic oak
#

in the limit of finite time steps, the continuous model breaks down and one should refer back to the recursion relation

#

Actually it might still work

versed violet
#

Do you have the calculated value for the given 780 belt and 7 output + 2 feedback system?

fierce ruin
#

at this point I don't even know I'm rusty

#

odd way to express a derivative, never used tau as delta h

austere notch
#

It even gives us the equation to use in this case

manic oak
#

It would be in the limit of continuous time

#

😄

versed violet
austere notch
manic oak
#

I'm not sure what you mean by 89, but the point of the graph is that eventually the belt hits 780 and growth will stop. The only question is how long that will take, and is controlled by how much is fed back into the system.

austere notch
#

In this case, we knew that the amount in system increases by 2/9 of previous iteration's input value with each iteration

#

That's why x(n+1)=x(n)+2/9(n)

versed violet
#

[pulls out his hair]
When belt max is 780 and we feed back 2/9, how much can the system put through?

manic oak
#

780

#

the 2/9 will essentially be trapped

austere notch
#

where x_0 = 780/9 ---- which is why you see 86.6666667 in the first field for my input

manic oak
#

x_0 is the initial belt throughput. If it's 780 then the system is frozen

versed violet
# manic oak 780

Thats not possible. The belt before first merger cannot take 780 input AND recycle 2/9th of the input

fierce ruin
#

wrong math again oop

manic oak
#

Exactly. If it attempts to recycle 2/9th of the input, it will still be stuck at 780

#

The amount grows by 2/9 of itself at each step, but since there's a finite throughput of belt the system will eventually reach 780 and be stuck

#

At that point the belt reaches 780 throughput and 1/3 of that is sent out through the middle line

austere notch
#

I think it's pretty cool it solved it

MSP760623fgg16ba7gc617i000057h96271f89779g4.png
manic oak
#

So actually this system is brilliant. You can force a build up to 780/min on the first line, splitting into thirds and one third can be sent to another assembly line

#

Of course all this depends on how much the side lanes are consuming. If they're consuming 1/3 of 780, then we reach that equilibrium

versed violet
#

Its not a prime splitter if you sent it away instead of looping

austere notch
fierce ruin
#

9 isn't prime

versed violet
#

7 is

fierce ruin
#

ah

austere notch
#

Right - the question originally wasn't about prime splitting arrays, but when they linked me that wiki, it gave me the iterative notation I was missing and put me down the right path finally

fierce ruin
versed violet
#

Well, the original drawing shown a 9 split with 2 outputs fed back, so a prime 7 splitter array.
My educated guess is that you can feed 607 units (using 780 belts) through the splitter, due to loopbacks.
Wouls love to see the 'exact' calculus answer you came with and how much it differs.

fierce ruin
#

I've past my initial thought, so I can't explain it anymore

versed violet
#

Replying to own question of "can you fit 65 smelters on 8x13 foundations" ->
Yes.
But its cramped.
And assumes no walls (use fences or overlap mergers)
And assumes non standard floor height of 2,5 wall minimum.
And assumes you can pick input belts location.

Still damn proud.
Just hoping it actually works ™️

65_is_fiveold_unlucky.png
oblique hollow
#

i could try and build a test setup

#

to measure the output

#

this reminds me A LOT of this here

1553749700.png
#

and i actually managed to solve this one

#

1/6th is feedback,
so the outputs are only 5ths.
blah blah, weird math:
Output Ratio = (1/6) * (6/5)

fierce ruin
#

for a 1:7 ?

oblique hollow
#

so 0.2

oblique hollow
fierce ruin
#

he was talking about a 1:9 with 2 feeding back making it a 1:7

oblique hollow
#

for 1 to 7, it SHOULD be:
(1/8) * (8/7)

#

however, thats for a 8 way splitter with 1 feedback

#

not the 9 with 2 feedback

fierce ruin
#

I got twice that

oblique hollow
#

That should be.....
(2/9) * (9/7)

fierce ruin
#

matches for a 780

#

can't prove it myself tho

oblique hollow
#

so.... i did it?

fierce ruin
#

IDK countcalculus was try some calc stuff

versed violet
#

222 goes through with 780 belt?

fierce ruin
#

assuming priority merge

oblique hollow
#

it is feedback after all
and also, the merger priority is important here

#

since the 2 sides are are less than 33.333% of merger output

fierce ruin
#

.2857 comes out

oblique hollow
#

yep

austere notch
#

This is what I ended up with after I went through @fierce ruin 's code --- not certain why your didn't iterate properly; I wracked my brain on yours some but coding/debugging isn't my forte so I started fresh: https://onlinegdb.com/H1vphygHd

{
    public static void main(String[] args) {
            double input = 780;
            double feedback = 0;
            double insystem = input + feedback;
            double output = insystem / 9;
            
        for(int i = 0; i<100; i++){
            System.out.println("Iteration " + i + ":    " + output);
            feedback = 2 * output / 9;
            insystem = insystem + feedback;
            output = insystem / 9;
            if (output > 780){
                output = 780;
            }
        }
    }
}```

Iteration 0: 86.66666666666667
Iteration 1: 88.80658436213992
Iteration 2: 90.99933953157546
Iteration 3: 93.24623680396004
Iteration 4: 95.54861302134178
Iteration 5: 97.90783803421442
...

GDB online Debugger

Online GDB is online ide with compiler and debugger for C/C++. Code, Compiler, Run, Debug Share code nippets.

oblique hollow
#

which is 1/3.5

fierce ruin
#

gross why comma

oblique hollow
#

because differen laguages use different things for decimals

fierce ruin
oblique hollow
fierce ruin
#

I took me awhile to get it on desmos so that code was very improv

austere notch
oblique hollow
#

final value?

austere notch
fierce ruin
#

780

austere notch
oblique hollow
#

thats nonsense

fierce ruin
#

it's a converging geometric sequence

oblique hollow
#

according to my numbers, limit should be
780 * 0.2857 = 222.857

fierce ruin
#

so it goes down that's all I know

manic oak
#

Why would that be nonsense? You're continuously putting more onto the belt at each time step, which means you put even more on at each time step

oblique hollow
manic oak
#

If you're taking 2/9th of the belt and putting it back on the belt, there's more on the belt

austere notch
versed violet
oblique hollow
#

because I did long ago

austere notch
#

Mk5s?

oblique hollow
#

yea sure

#

any mk will do

#

theres also this funny thing here

unknown.png
austere notch
fierce ruin
#

156 by the McGalleon Method of a 780 belt

oblique hollow
austere notch
# 
Iteration 1:    133.61111111111111
Iteration 2:    137.32253086419752
Iteration 3:    141.13704561042522
Iteration 4:    145.0575190996037
Iteration 5:    149.08689463014824
Iteration 6:    153.22819725876346
Iteration 7:    157.48453607150688
Iteration 8:    161.85910651793765
Iteration 9:    166.35519281010258
Iteration 10:    170.97617038816097
Iteration 11:    175.72550845449874
Iteration 12:    180.6067725782348
Iteration 13:    185.62362737207468
Iteration 14:    190.7798392435212
Iteration 15:    196.0792792225079
Iteration 16:    201.52592586757757
Iteration 17:    207.12386825278804
Iteration 18:    212.8773090375877
Iteration 19:    218.79056762196515
Iteration 20:    224.86808338924197
...```

Hit's 780 by iteration 66
oblique hollow
#

yeah, code is unusable then

#

thats not correct

fierce ruin
#

he changed my code to work very differently

oblique hollow
#

the thing you just calculated was a 1:5 splitter

#

i tested it myself

#

and output is 1/5 of input

austere notch
#

Post your math?

fierce ruin
#

my code was pretty much a iterative sequence

oblique hollow
#

(1/6)*(6/5)

#

thats it

fierce ruin
austere notch
#

Oh I calculated the output for a single belt out - not 5...

#

that might make a bit of difference w/ your diagram lol

oblique hollow
austere notch
#

my code was solving for each iteration of output in a single output which was correct for what I was working on

oblique hollow
fierce ruin
#

wouldn't that cap at the 33.33%

#

since it doesn't converge

oblique hollow
#

since its only one input (2 in total)

#

not 3

fierce ruin
#

thought you said mergers cap at 33.333 from the side

oblique hollow
#

if you use 3 belts total

#

the cap changes depending on how many belts you connect

fierce ruin
#

you got me

#

so any more results?

versed violet
#

YES

manic oak
#

I'm dead in an ARAM, so I'll weight in. Let's consider 600 + (2/9)*600 +...

#

Hold that thought

austere notch
# 
Iteration 1:    487.5
Iteration 2:    609.375
Iteration 3:    761.71875
Iteration 4:    780.0```
manic oak
#

Okay, they surrendered. 600 + (2/9)*600 + (2/9)*(600 + (2/9)*600) + ... is our sequence

oblique hollow
#

or any other mk

fierce ruin
#

mk5 is buggy

austere notch
# oblique hollow try with mk 1

it will always cap at the max speed of the belt - splitter overflow sends it through the belt of least resistance which is the last output in this case

manic oak
#

soo 600*(1 + 2/9 + 2/9 + (2/9)^2 + 2/9 + (2/9)^2 + (2/9)^3 + ...)

#

Assuming 600 was our starting point

#

After 9 steps you would have doubled your input, at 18 steps you've tripled, etc.

#

sorry, actually faster than that

#

at 9/2 steps you double, at 18/2 steps you triple

#

It's not a geometric series

austere notch
#

Thanks @manic oak - I've spent too much time today mathing the game instead of playing the game... so now I'm going to checkout for a while. You all have a pleasant evening.

fierce ruin
#

closure not found

#

could've just limited the the return belt to a mk1

oblique hollow
#

im booting up my game and running that old test again

fierce ruin
#

limit the potential bottleneck

#

but why use prime splitters uck

oblique hollow
#

for fun jace_smile

versed violet
#

Result insocming:
test setup - 2 boxes, one with concrete and rotors, second one with assorted plates.
Top belt is vanilla mk5, bottom contraption is the prime 7 splitter.
Both inputs are gated by leaves.smart splitters. when leaves are removed, both outputs start running concurently.
results are accrued at the end by observing container. The concrete box has rotors as 10th slot, to mark when to finish counting.
Final results as attached - mk5 belt (780min) pulled 905 items, while prime 7 splitter managed 708 plates.
After scaling to belt capacity of 780, the final result is ~610.
Which is in line with predicted 607 I mentioned earlier.
I rest my pencil now.

unknown.png
#

Tl;Dr;
Prime 7 splitter allows throughput of 606 units from 780 belt.
It appears the final throughput is given as Belt_capacity * Number_of_outputs / Total_number_of_lines_split.

#

Eg. 780 * 7 / 9 = 606

oblique hollow
#

who would have thought

fierce ruin
#

curb your prime plitters

versed violet
#

But...
Its...
simple?

oblique hollow
#

would you look at that, 50% output

unknown.png
manic oak
#

Ah, there's the confusion

#

We were solving the amount flowing through the initial input belt, not the output belt

fierce ruin
#

how do you have such good lighting

oblique hollow
#

its daytime

#

look, same for mk 1 input

unknown.png
fierce ruin
versed violet
oblique hollow
#

my equation stands.
but Tomtores's final conclusion too

#

Total output = input * (Number of outputting belts / Number of split belts)

#

@austere notch you must have messed something up

fierce ruin
#

I'm out of shape

oblique hollow
#

Hehe, next time, try doing math with pipe junctions and valves

#

that was fun back then....

versed violet
#

let me guess, 150 at both because fluids don't behave like fluids?

oblique hollow
#

thats what i thought too, before testing and doing the math

austere notch
#

Sorry - that's not accurate

#

It's an iterative sequence

oblique hollow
austere notch
#

as indicated on the tutorial page for prime splitter arrays

#

it's just not a prime splitter

#

No - you act like I didn't post the code w/ relevant variables. If you think there's something wrong with it then you find it. All you seem to do in here is just discount whatever anyone says because it doesn't match your math. Maybe your math is wrong. Go ahead an post a mathetmatical proof for your math instead of just telling other people their wrong.

wind spade
#

what's the problem here?

fierce ruin
#

prime splitter

wind spade
#

what are you trying to calculate?

fierce ruin
#

IDK anymore

oblique hollow
wind spade
#

hm, so you're trying to calculate max throughput of a loopback balancer?

austere notch
austere notch
oblique hollow
#

Testing this stuff ingame beats doing math for 5 hours any day

austere notch
oblique hollow
#

but your argument just doesnt hold against the test data. its not 780 output

austere notch
#

Here's the code again if you wanted to check it:

{
    public static void main(String[] args) {
            double input = 780;
            double feedback = 0;
            double insystem = input + feedback;
            double output = insystem / 2;
            
        for(int i = 0; i<100; i++){
            System.out.println("Iteration " + i + ":    " + output);
            feedback = 2 * output / 9;
            insystem = insystem + feedback;
            output = insystem / 9;
            if (output > 780){
                output = 780;
            }
        }
    }
}```
manic oak
#

you are again forgetting that we were testing total belt input, not final belt output

oblique hollow
#

i could send a video of the 1:5

#

a short one

wind spade
#

This seems like another case where simple equation just won't do the trick and simulation would be needed

#

What's total belt input?

manic oak
#

the input at the beginning of the system

wind spade
#

And you want to know how much slower the main belt would go after it settles?

austere notch
#

I think a picture might work better?

#

think I still have it

fierce ruin
#

why loop back the middle one that causes entanglement

wind spade
#

well the obvious solution would be to merge after first split 🤔

austere notch
#

initially I was trying to determine what the rate of output for the middle splitter's outbound belt just as an exercise

pulsar edge
#

So there’s really no way to harness the full power of a pure node, is there

austere notch
pulsar edge
#

I find that interesting

versed violet
#

Not with current belts

wind spade
#

yeah I think it's either some limit function or no function at all and would need to be simulated

pulsar edge
#

Wonder if they’ll just deal with that with a new conveyer type or if they’ll find some kind of other way

#

It wouldn’t be to far fetched for full capacity to have a cost of some sort

oblique hollow
fierce ruin
#

belt might not be practical given lag but an attachable splitter maybe

oblique hollow
#

Mk 4.
Input backs up

pulsar edge
austere notch
versed violet
#

Tootal throughput for that prime 5 splitter should be 5/6 of belt used. Can you measure that?

austere notch
#

It's backing up at the same time as each piece from the feedback hits the merger.

#

So the system is still being fed at the max rate.

#

you want the total output rate?

#

i don't have code for that

manic oak
#

You don't need code for it, it's 1/9 of the input

fierce ruin
#

we're measuring two different things here

wind spade
#

if you're talking about the belt after merger, then yes, it's always max rate. If you want to calculate speed for the input belt, it's less

austere notch
oblique hollow
manic oak
#

Ah yes, Greeny is correct, technically the belt I'm talking about is just after the merger

austere notch
#

So what are we talking about then - the limit on the input? the total output? the amount in system? -- now I'm just confused.

#

I originally wanted the math to be able to measure each final output belt according to however many belts I fed back to the beginning of the system.

fierce ruin
#

That solution was posted awhile ago

oblique hollow
#

theres a limit to all this i forgot.....
it only works with full belts......
feeding a mk 1 into a mk 5 prime splitter will just yield 60

#

since these work with backing up

versed violet
#

I would be surprised if it yielded more

#

we are limited by the first belt after loopback merger capacity

oblique hollow
#

you cant turn 60/min into more than 60/min.
If thats what you meant

oblique hollow
#

aaaaagh, damn it, thats right

#

you need to mix belt mks

#

to actually get prime output

#

no wait....

#

agh, forget ti

iron prairie
#

Hm? I've seen prime number splitters which let you do arbitrary splits without mixing belts or choking output.

oblique hollow
#

yeah, im done with that topic again.
I don't ever consider needing them or using them either.....

iron prairie
#

You don't have to pay attention anymore?

Like, for a 1:5 splitter, a simpler version is just "1:6, and loop one output back to merge with the input", with a full-capacity version being "split the input 1:2, split that sixth output 1:2 and merge, then do the 1:3".

oblique hollow
iron prairie
#

That's the simple version, yes. A more complicated solution deletes that first merger, then inserts mergers just after the first splitter. Then, that sixth output line gets split 1:2 and merged in with those mergers.

So:

      ----S---
      |  /   |
      M-S-O  |
     /   \O  |
In->S        |
     \   /O  |
      M-S-O  |
      \  \O  |
       \-----/```
#

If you merge before the first split, you can't run max-speed belts. If you split, then merge, you can.

vernal patio
#

gentlemen, fully clocked nuclear reactors consume x2 rods for x2 power, but do they produce x2 or x2.5 waste?

oblique hollow
#

Its all 1 to 1

vernal patio
#

so no downside to clocking at all?

#

provided slugs are available, of course.

#

(and it's honestly hard to imagine all hundreds of them them running out)

oblique hollow
vernal patio
#

ah, i see. thanks.

short swallow
#

this is not really satisfactory related at all

unknown.png
#

but i need some help

wind spade
short swallow
#

okie

tiny sentinel
#

anyone got some good designs for elevated train tracks that incorporate conveyor buses and vehicle roads?

#

or is there a subreddit for sharing of designs specifically?

fierce ruin
#

Can anyone figure out the dimensions of the max volume rectangular prism that is inscribed in the ellipsoid defined by (x^2/32^2)+(y^2/32 ^2)+ (z^2/46 ^2)=1

#

That’s the ellipsoid formed by the range of the hover pack

#

(Pure curiosity)

green mango
#

7.... !?! 😂 😝 😂

fierce ruin
#

My best guess is a block 26.128m x 26.128m x 26.558m assuming x=y in the best case

frosty owl
#

Bruh... Imagine reading the whole chat, only to find out no new dark magic was developed jace_happy

vast jungle
#

what do you mean with "new" ?

frosty owl
#

Something like "ah, we can do this even in THAT weird magical way"

vast jungle
#

this "merge then split" vs. "split then merge" is not really that new... the "belt limit" problem of the back-merging splitter has been well known here

tiny sentinel
#

any suggestions on feeding 80 coal gens with 4 belts of 300 coal? I'm thinking I just put the belts in a stack of 4 along the 10 rows of 8 gens I have. the bottom belt would split off to go down each row of 8. then I'd use conveyor lifts to merge down one of the other 3 belts on rows 3, 6, and 9 to replenish the bottom belt. Thoughts?

calm flax
#

use compacted coal and feed it 2 belts

#

you will need 2 sulphure nodes but one would be freed again late game when you get mk3 miners and you also end up being able to free all but one of the coal nodes doing it that way

tiny sentinel
#

or I just do an overflow where at the end of each row I have the excess coal flow back to the main coal lines.

#

well I just started over fresh and have no power to speak of at the moment. I used the map editor to mass delete everything from U3 and keep all my materials.

#

and I already have to figure out how I'm getting the 4 coal nodes near the lake to my gens

calm flax
#

if you want to run them on coal just injected manifold or a manifold fed for the middle out by 4 belts (2 rows)

#

if your playing U4 (you will be in 2 weeks) you won't have any coal remaining if you do it right

tiny sentinel
#

I'm on experimental right now and yes I know that. I've done the math already. just trying to figure out what the optimal belt layout would be.

calm flax
#

optimal in what way? In terms ot build costs?

unkempt acorn
#

wish i checked to see if i could over clock nuclear plants before building my set up >.>

frosty owl
#

Well... You can

#

... Wait, why would you build a SET of something before checking out that something? 😂

unkempt acorn
#

it never clicked in my head that you could over clock them 😆

#

so i spent a couple hours yesterday trying to gather resources for 20 nuclear plants

wind spade
#

well it's not like you want to overclock them anyway

#

it only saves space and we have tons of that

iron prairie
#

Automation is your friend for accumulating resources. Even with just single manufacturers, it takes less than a couple hours to accumulate the HMFs/supercomputers for 20 nuclear plants.

unkempt acorn
#

my nuclear fuel rod back was nearly backed up... so i had to try and build a bin from afar

unknown.png
unkempt acorn
#

would it be worth putting pumps on this set up? or is that just a waste?

unknown.png
wind spade
#

as long as the height difference isn't too big, you don't need pumps

sinful helm
#

If the sight glass is full, you are good to go 🔵

torn quarry
#

im not sure you need so many pumps, but i could be wrong

copper imp
#

is crystal osc. altarnate worth?

verbal hemlock
#

so like.. the Coal gets fed in a single line via Splitters.. has Industrial Storages in between, that would also be connected on another Industrial Storage. This way, there won't be much bottlenecks below, and the Coal would also be redistributed by the other Industrial Storage.... then at the end of it all are Industrial Storage again that would loop back all the Coal to the front of the loop.
EDIT: Oh my.. its 4hrs ago. mb

unknown.png
tiny sentinel
verbal hemlock
tiny sentinel
#

oh neat, I'll take a look!

verbal hemlock
#

here's a diagram.. dunno if that makes sense.. the straight lines are conveyors connecting the splitters.

unknown.png
#

i made use of this also on the production line. it makes it simplier imo... and kinda removes the clogging to an extent with is the common issue with manifolds(just learned its the term used here).. with this, I can cycle all items back to mergers to their supposed Storage spots after they run their course to other production line that they might be needed... The storage ones are also connected and will provide for whatever needs it.. this way, the only time the factory stops producing things is when all storage for the most advanced items are full... right now, there's few since I'm still at t3-4. New to the game 🙂

unkempt acorn
#

How do peeps equally distribute 1 input line of nuclear rods into 20 nuclear plants? should i use splitters... or just get a direct line from the manufacturer?

manic oak
#

@fierce ruin V = x*y*z, f(x,y,z) - C = 0. You can use Lagrange multipliers to solve this optimization problem with constraints.

verbal hemlock
#

When I'm bad at math but I'm playing Satisfactory... it's like I'm torturing myself.... in a good, masochistic way.

wind spade
wind spade
verbal hemlock
wind spade
pseudo tangle
verbal hemlock
#

just leave it for a few minutes, they'll "balance" anyways. hahahaha

pseudo tangle
#

yes, you can "push" this a little bit, so it doenst takes hours, denpending on your items and size

tiny sentinel
pseudo tangle
#

powerbuildings are stil best, ebacsue they can even prefill when not active

verbal hemlock
#

i have tiny pp.... tiny production points.

pseudo tangle
#

that should be posbile for production too

unkempt acorn
verbal hemlock
#

@tiny sentinel hop on and chat a bit on twitch, when you do, i'll show you what I'm using.. no mic though. so I'll reply via chat 😛 I just stream so I get "free" storage and just download them if needed before they get deleted.

wind spade
frosty owl
#

Manifolding fuel rods at 4/min? evildoggo

#

That manifold would take 3h and 45 min to fill up with the generators offline hehe

frosty owl
manic oak
#

@fierce ruin The solution is:
x^2 = y^2 = a^2/3 = 341.33, z^2 = b^2/3 = 705.33
=> x = y = 18.5, z = 26.6

frosty owl
#

As long as the generators aren't far away, ofc hehe

unkempt acorn
wind spade
#

which can be solved by pre-filling them or turning them on after they all are filled

unkempt acorn
unkempt acorn
wind spade
#

and most of the machines will work in first 10 minutes anyway

unkempt acorn
vast jungle
#

first time that I drive around in a 4-wagon build train... time measure how many Fuel Generators fit onto the roof

unkempt acorn
#

another question tho... train loading/docking... can a single platform unload and load in one stop?

vast jungle
#

no

#

one freight platform can do loading OR unloading... but you can switch the mode for different platforms attached to the same station

#

load wagon 1/2, unload 3/4

unkempt acorn
#

damn 😦 thanks

#

just means ill have to build 4 stations in total for this nuclear stuff

jade minnow
vast jungle
unkempt acorn
vast jungle
unkempt acorn
#

ahhhh

vast jungle
#

50+ Fuel generators per floor... sounds like a nice number

#

I need a total of 166... so 55-56 per floor would be perfect

keen totem
#

looks awsome tho

wind spade
#

manifolds look better 🤷‍♂️

bleak coral
#

Best way to keep things moving is to get smart splitters and end lines in overflows to sink after storage.

verbal hemlock
#

@bleak coral i'm not even at that part regarding the Smart Splitters :3

bleak coral
#

smart splitters are in the caterium tree of the MAM, you can get them at T2

verbal hemlock
#

it's working right now and for sure things will change anyways when I get to that point. kinda new to the game still, got the game just this 25th so still exploring.
EDIT: yeah, I have that, haven't researched them yet and haven't built too so I'm still oblivious of those.

bleak coral
#

sure play around with it, and if it keeps working good, but I just wanted to tell you about possible pitfalls cause how they work isn't exactly intuitive

manic oak
#

@fierce ruin Ignore my previous messages, the method is correct, but x, y, and z can’t represent both the dimensions of the prism and the coordinates of the ellipse

verbal hemlock
#

so far, from my experience, the issue I get with the simple manifolds is until you oversaturate the input, the ends usually gets starved. and by placing the storage in between a few Splitters, with the storage also connected by storage that feeds it raw too, it helps on the starvation.

#

even then, if you oversaturate but the belt speeds are not that quick yet (i really wanna next conveyor), they still get stuck to an extent.

wind spade
verbal hemlock
#

this must be the one you're talking about, I'll play with it after I unlock it.. after I automate everything I currently have unlocked then go for T5-6.

unknown.png
verbal hemlock
wind spade
#

oversaturated input to manifolds doesn't break it

#

undersaturated doesn't either, but makes a few machines starve

verbal hemlock
#

right.

#

kinda lost it there a bit when I'm trying to explain, sorry 🙂

bleak coral
#

if machines are starving you're not feeding them enough or the manifold isn't done warming up, it doesn't need anything extra to function at 100% efficiency besides time

#

by not feeding them enough that could also mean you're not underclocking machines that need to be underclock in order to run at 100% efficiency

verbal hemlock
#

I don't underclock. Yes.

#

I just let the game run it's course. 🙂

#

i do overclock my Generators.

bleak coral
#

I hope you're looking at the right number when you do that, the overclock UI is wrong on generators

vast jungle
#

I wonder, is it worth to build a fuel-reserve for restarting for U4?

#

fuel-reserves are more problematic because you don't have an easy "overflow" from turbofuel generation

#

so you have to build less generators..

verbal hemlock
#

@bleak coral I placed my 16 Coal Generators at 195%, supported by 10 Water Extractors at 100%. it's break-even on Water while 2 MK2 Pure Coals are more than enough to supply the Coal.

vast jungle
unkempt acorn
#

what are peeps thoughts on this? sub station where i can control the nuclear plants, lights and pumps

unknown.png
wooden ruin
#

for reinforced iron plates i need a input of iron plates of 30 per min but my factory makes 40 per min and I cant over clock because then my screw input would get messed up
I tried to find a way to fix this but It didn't work

deep root
#

20 from one machine, 10 from the other

wooden ruin
#

If i did that the iron plates would get too many ingots

#

im trying to waste nothing

deep root
#

Why is that bad?

#

Oh I see... Under clock to get the numbers you need

#

Technically speaking nothing is ever wasted, just backed up

wooden ruin
#

If I had smart splitters I could do this so it wouldn't be wasted

#

but I dont

bleak coral
#

you don't need smart splitters for that

wooden ruin
#

its just annoying I cant overclock or underclock because I already have the perfect amount of screws per minute

#

the over flow splitters so I can have the plates not wasted if the assembler gets max plates

bleak coral
#

resources are infinite, there's no such thing as wasted materials

#

it sounds like you're making too many ingots, make less ingots and plates

#

and mine less ore

#

which can all be fixed by underclocking/deleting machines

wooden ruin
#

This is my first time trying load balancing so Im trying to get max efficency for everything

deep root
#

I mean... We have both given you the answer to make your "perfect" amount. It is up to you to use the information. The power is yours!

wooden ruin
#

I think I thought of a way maybe but thx for the help

bleak coral
#

this advice is for either manifolding or load balancing, load balancing does not mean running everything at 100% clockspeed

wooden ruin
#

I know

#

My screw input is perfect thats just why I needed a different solution for the plates then changing the clock speed

deep root
#

You do you

unkempt acorn
#

so a single nuclear plant produces 100 barrels of waste every 300 seconds?
20 barrels a minute?

bleak coral
#

on experimental, yes

vast jungle
#

Wasn't it 5 waste power minute?

bleak coral
#

on U3 yes, they increased the amount of waste per minute

unkempt acorn
#

yay... so i produce 400 barrels a minute >.>

deep root
#

Holy crap

unkempt acorn
#

this is going to be a pain in the back side lol

deep root
#

Turn it all into plutonium rods and sink it

unkempt acorn
#

i can atm 😆

bleak coral
#

yeah large scale uranium fuel rods is very unappealing without plutonium

#

or I guess any scale, cause 4 plants isn't that much lol

unkempt acorn
#

cant*

tall stratus
#

If I did my math correct, blended turbo fuel yeilds 20% more turbo fuel than turbo heavy fuel if both uses diluted fuel and are made directly from HOR. Can someone confirm?

bleak coral
#

turbo heavy fuel doesn't use diluted, do you mean regular turbo fuel?

tall stratus
#

Oh, yeah. My bad

bleak coral
#

I can't remember the exact numbers, but for the same amount of turbofuel you use more oil and less sulfur for turbo blended fuel vs turbo fuel

#

so if you mean 20% more turbo fuel from the same amount of oil, that's definitely wrong

tall stratus
#

Oh, okay. May have miscalculated then

bleak coral
tall stratus
#

Yeah I see. Idk how i calculated

#

May have calculated heavy turbo fuel vs blended

#

So regular turbo fuel is preferred as long as I have enough coal and sulfur at my disposal

bleak coral
#

yeah it depends which one you're limited on, though sulfur is rarer than oil so ultimately less sulfur is probably good

#

I think greeny's calc prefers blended because it reduces sulfur and eliminates coal entirely, but eliminating the coal isn't usually a huge bonus cause it's the 3rd most abundant resource

tall stratus
#

Yeah, well for simplicity's sake I'm gonna make a bit of regular I think. I have 1800 oil to work with and all power I don't need is gonna reroute the fuel to plastic and rubber.

manic oak
#

Someone should do the math on how much power it takes to produce one normal turbo fuel and one turbo blend fuel. The wiki says there's a difference of 4 Mj/item, but I'm not sure what those calculations use.

bleak coral
#

wiki numbers are just per step, not the whole chain

#

cause there's no standard chain to base mj/item on

manic oak
#

Yeah, I imagine that it costs more energy for the standard turbo fuel since you need more refineries than blenders to process, and you also need more water extractors, but I haven't done the math.

bleak coral
#

the power numbers are in the links above

#

they're average & max, so everything at 100% clockspeed, so it could be tuned down more but it gives a rough idea

#

and blended definitely looks like it does better energy-wise

manic oak
#

Yeah, and the reduced sulfur usage as an added bonus, plus a majority of your oil being left as heavy oil means a simpler production line

bleak coral
#

the power usage is even worse if you do diluted packaged fuel instead of blender diluted fuel too

manic oak
#

Yeah, but no one's gonna do that once they have blenders. I did do the math on that one and you need like two refineries and a packager instead of one blender, and then you need 5/3 more of those setups since each blender can process more heavy oil at a time than a refinery

#

Actually it's one refinery and two packagers or something like that

unkempt acorn
#

right... sooooo...
48 slots x 500 stack size = 24000 waste per industrial storage
I produce 400 a minute... so each container should last me 1 hour irl...
80 storages = 80 hours....

bleak coral
#

each one is worth 1 & 2/3 sets of 1 refinery and 2 packagers

unkempt acorn
topaz hedge
#

you not recycling to plutonium fuel?

bleak coral
#

I think it's a retrofit on an existing system

pine rapids
#

Hey! For performance... trains == bad?

bleak coral
#

not universally, no

topaz hedge
#

everything = bad

#

Don't move past biomass lol.

pine rapids
#

🤣🤣🤣

bleak coral
#

belts vs trains performance wise is all wild-ass speculation, and also depends on your system and how you set things up

#

the most concrete performance thing we know is spreading out your production is better

topaz hedge
#

^

smoky nexus
#

The more efficient your system, the easier it is on your system as well.

bleak coral
#

so the more stuff you have close to together, the worse it is

unkempt acorn
topaz hedge
#

don't do what kibitz did and build a megafactory in dunes.. last I heard he was starting a new save

bleak coral
#

he is, it was too much to do to fix his save

unkempt acorn
pine rapids
#

I both love and hate trains... they’re so so easy to scale but then train stations are kind of the worst, and turning around is annoying

unkempt acorn
#

this is my nuclear train set up... thought id sacrifice some extra effort for the cool aesthetic

unknown.png
pine rapids
#

Eventually I’m thinking I’ll make mostly self- contained factories for each advanced item, then use drones to make my production follow me around

topaz hedge
#

@unkempt acorn Man, plutonium processing makes for an interesting setup. here's mine XD

unknown.png
#

poor kibz. he had quite the impessive world.

#

since I don't need the power I'm not turning my reactors on until the processor plant is completed, and then I'll have to find a place to.. store them until the plutonium reactors are built.

#

I think one stack of plutonium rods = 5000 waste if you're using the 100:1 setup

unkempt acorn
#

ffs. i forgot i was wearing a hazmat suit instead of a jetpack and i jumped of the edge of my platform >.>

tiny sentinel
#

Am I crazy or should I just skip setting up turbo fuel until I have the blender recipe and T7 access? I could set it up now but I don’t want to have to redo everything again when I get the blender variant

unkempt acorn
#

dat view tho

unknown.png
deep root
#

We got computers and heavy frames automated, then went straight to automating space elevator parts

manic oak
#

@tiny sentinel You could build a temporary turbo fuel facility that you can easily tear down later, generate like 5000MW to get yourself to T7/T8 (which is what I did), and have enough power to start up your new Turbo Fuel power plant

tiny sentinel
manic oak
#

Then you probably don't need to

unkempt acorn
#

264 hours worth of nuclear waste...

unknown.png
#

huh? has the last few hours of chat been deleted?

magic shadow
#

no?

#

i can see everything

fierce ruin
manic oak
#

My previous answer is almost right, but I'd have to change my work a bit

#

The volume is 8*x*y*z, and then you can apply Lagrange Multipliers under the Ellipsoid constraint you gave

fierce ruin
#

Hmm ok

#

I gotta learn some multi calc

manic oak
#

I can voice chat if you'd like, it's not complicated

fierce ruin
#

It's fine thank you

manic oak
#

Actually I just solved it again. The coordinate of a given corner is my previous answer (x = y = 18.5, z = 26.6). The length, width, and height are double that: L = W = 37, Height = 53.2

fierce ruin
#

thanks

short swallow
#

i have a question

#

would you guys consider 1000 megawatts alot (1 gigawatt)?

bleak coral
#

not really, but it's all relative

fierce ruin
#

late game no

bleak coral
#

that's like first coal plant setup level, should be ok for T3-4

#

probably want more if you build bigger and/or once you start doing oil

short swallow
#

hehe boi

honest lotus
#

if i have 16 coal generators, how much coal per second and water per second do i need?
i have access to 2 pure coal nodes, 11 power shards and a giant lake of water
i'm using mainly Mk 2 conveyors (i can use Mk3 but not many) and Mk1 pipes