#math-and-meta

1 messages ยท Page 535 of 1

fierce ruin
#

overcloaking a generator to 250% consume the input a bit more than twice as fast.

split tulip
#

overclocking coal gens is pointless, overclocking coal gens mean you have to overclock the coal node and water extractors (in most cases) which will just use about the same % of power you would use on them if it wasnt overclocked. Overclocking would only be viable if you have more coal nodes than you need, but remember pipes can only hold so much water

bleak coral
fierce ruin
#

commie

magic shadow
vital quest
bronze silo
#

as ore to iron ingots is 1:1 is it ever a good idea to not smelt near the mining site?

fierce ruin
#

probably not

bleak coral
#

can't think of one, only if you're not smelting but using the pure instead

bronze silo
#

pure?

#

oh, nodes?

fierce ruin
#

pure recipes

bleak coral
#

no pure iron ingots

bronze silo
#

hmmm, just unlocked a HDD

#

I now have the option of alloys ๐Ÿ˜›

bleak coral
#

copper alloy can be nice, I don't see the point of iron alloy though

#

never have I thought "I have too much copper but not enough iron"

bronze silo
#

so mixing copper with iron is good?

fierce ruin
#

mixing copper and iron to get copper is good*

bleak coral
#

technically pure copper is better, but the alloy can be nice if you don't want to bother with the space & power of refineries and/or getting water

fierce ruin
#

there a super infographic on the wiki about it

bronze silo
#

I have noticed that with the iron alloy that it produces MUCH faster ๐Ÿ™‚

fierce ruin
#

but why?

bronze silo
#

wait, thats a different recipe

fierce ruin
#

you can just build moreโ„ข๏ธ

bronze silo
#

R Iron Plates

fierce ruin
#

RIP

bronze silo
#

yeah ๐Ÿ˜„

fickle scaffold
#

My liquid manifold is not looped, though it seems like it's working. Does not looped manifold somehow affected in the long term? Or should I not worry if it works

#

It's just hard to fit in my setup

frosty owl
#

It only risks failing when you get too close to max flow (over 300 or 600/min)
If you stay clear of the max there shouldn't be bugs

sand garnet
#

feeding from both sides of the manifold apparantly should prevent issues

proven prawn
#

I have no idea, I must have did it by mistake and didnt realize it๐Ÿ˜…

fickle scaffold
#

Gonna keep this in mind though

frosty owl
#

Is there any sort of analisys of loading time for a save without items/machines not running VS with items/machines running?

deep root
#

Are you thinking about turning power off before you exit???

fierce ruin
#

if you run on fluids, you'd have to close off the pipes too

glacial hemlock
#

Coated Iron Plate may be useful, after all.

bleak coral
#

Wait coated iron plate? Not coated steel plate?

#

I don't see what coated iron is doing that steel coated isn't doing better

glacial hemlock
#

Yeah, coated steel plate is good too.

#

The coated recipes do have lower weighted resource consumption and also reduced energy cost. What's better, they even have lower building counts and building space per iron plate

wind spade
#

I've been advocating for these recipes since they were introduced and everybody was like "nah they suck" ๐Ÿ˜„

glacial hemlock
#

you knew it since the beginning? pog

wind spade
#

well... my tool did ๐Ÿ˜„

fierce ruin
#

๐Ÿ‘ ๐Ÿง 

bleak coral
#

I've been saying it for a while too, because I saw it on greeny's calc jacelul

wind spade
#

plastic is super cheap, 9 plastic is 3 oil

glacial hemlock
#

it all come down to the question: "is it worth to use oil / coal just to produce iron plate?"

bleak coral
#

Yeah that too, and I think most of the time the answer is no

#

Not so much resource efficiency, but is it worth it to set up all the oil stuff compared to just some more refineries/smelters

wind spade
#

that depends on how much iron you have lying around ๐Ÿ˜‰

glacial hemlock
#

it use only a little bit of oil, not that much. The problem is the coal.

bleak coral
#

I think for most players, it's fine to be inefficient with iron, I mean there's so much of it

glacial hemlock
#

true

bleak coral
#

Coal is the 3rd most common resource

glacial hemlock
#

true

wary tulip
#

Another consideration..... oil and coal are power resources.... and in high demand

loud heron
#

Once you go nuclear, fuck coal and oil

fierce ruin
bleak coral
#

Side note: love seeing your analysis on the steel ingots and getting justified in calling compacted steel ๐ŸŽต the wooorst ๐ŸŽต

deep root
loud heron
#

What's wrong with compacted steel

#

Just gotta smelt the iron to get more steel

fierce ruin
#

sulfur use

bleak coral
#

You're thinking of solid steel

loud heron
#

Ooohh

#

Yeah u rite, compacted steel sucks

bleak coral
#

Solid steel is awesome

fierce ruin
#

solid steel = a solid deal

glacial hemlock
#

Compacted steel ingot is the all-round loser. haha

loud heron
#

Solid steel = a solid steal

wary tulip
#

Yes demand is different for different resources.... Iโ€™m saying coal and oil also have more weighted importance because they offer power options as well. Iron does not, and is more prevalent.

loud heron
#

I'm taking your joke and making it better.

fierce ruin
glacial hemlock
deep root
#

Guys! You are both the same size! It's okay

glacial hemlock
#

After seeing this, you may not want to use coal power for long term.

bleak coral
#

Yeah coal is a stepping stone, and oil turns into power very efficiently, and once it's not enough you have nuclear

fierce ruin
#

mad stacks

deep root
#

I personally only use enough coal and sulfur to get to turbo fuel. Then my coal plants stay built until I need those resources and I take them down

wary tulip
#

Size isnโ€™t the point ๐Ÿ˜‰
Itโ€™s efficiency. And yeah, coal is just a transition. But oil is much more stable. But also in much more demand.

bleak coral
#

I still think compacted coal needs an energy buff, I think it's kinda lame that it's basically the same as just getting an equal amount of coal as you get sulfur

wary tulip
#

To use oil for a simple recipe like iron plate makes 0 sense to me.

glacial hemlock
#

the compacted coal itself also used too much sulfur, which is kinda sad

bleak coral
#

I don't know how to fix compacted steel though lol

deep root
#

Fix it by making it use coke instead

fierce ruin
#

decompact?

bleak coral
glacial hemlock
#

if the compacted coal is balanced, the compacted steel will have its own use. If it use iron ingot instead of ore, it could be better

bleak coral
#

The killer is HOR + diluted + recycled recipes, without those oil efficiency falls apart and it absolutely becomes the rare resource it's had the reputation for

glacial hemlock
#

true.

oblique hollow
#

map general analysis complete jace_smile

unknown.png
#

time to set my horrible "only fluid alts" plan into motion

#

granted this is grossly simplified

#

but this is where general densities seem to match up nicely

#

this is much more detailed

unknown.png
fierce ruin
oblique hollow
#

the northern forest is really uninteresting

#

theres like only a bit of iron and limestone there

fierce ruin
#

if it's a bit is it 1 or 0

oblique hollow
#

a superposition of 1 and 0

fierce ruin
#

Madman

oblique hollow
#

the band of aluminum is real nice tho

#

you really gravitate toward the map center

fierce ruin
#

stick man rave party at the red forrest celebrating alums

oblique hollow
#

Dance till you die from aluminum poisoning

fierce ruin
#

but is that enough quartz tho

oblique hollow
#

for what

fierce ruin
#

alum

#

unless you're gunning for pure

oblique hollow
#

granted i did make the connections a bit odd

#

the upper left corner really doesnt need it

#

buuuut i could use the limestone from there for cheap silica

fierce ruin
#

who doesn't use cheap silica?

oblique hollow
#

top right is definitely a crystal oscillator location

#

funny enough for as bad as grass fields are, they are reall nice for the copper + caterium combo

#

its the quickwire location

#

top right seems more like its a nice normal wire location

#

and with oil there, coated cables is gonna be easy ๐Ÿ”Œ

fierce ruin
#

why coat when you can code?

oblique hollow
#

encodes cables in quantum binary

glacial hemlock
#

your ingots factories are not very close to the water, so you will have to build a lot of pipes

oblique hollow
#

thats the goal here lol

#

also, the resource wells exist

fierce ruin
#

pain

oblique hollow
#

its gonna be a funny map that acts on two goals:

  1. Make a save with only using fluid based alts, if possible
  2. Create a savegame that acts as a training map / challenge to solve some pipeline problems, teaching the player
fierce ruin
#

but does it have the 3:8 setup

oblique hollow
#

yes

fierce ruin
#

so we can copy paste the ss all over

oblique hollow
#

if only we had signs ingame

fierce ruin
#

crtlwin shift s

oblique hollow
#

i will try to include every possible "easy setup" for pipes

fierce ruin
#

I'm piped

oblique hollow
#

its gonna be a major project so ofc it will take a while

#

buuut the things the players will have to do will be rather minor, in fact

glacial hemlock
#

and of course, not FPS friendly

oblique hollow
sand garnet
#

oil islands not in the general oil area ?

oblique hollow
#

the west coast?

#

or what do you mean

fierce ruin
oblique hollow
#

absorbed by alu

unknown.png
#

again, the "general analysis" SS with the boxes is grossly oversimplified

#

the line graphs below are a more detailed version

#

but also not 100% accurate

#

as there are 0 interconnections between all this (drones, trains, etc)

#

i will definitely be using tool assistance here

fierce ruin
#

owo a TAS

glacial hemlock
#

how are you going to plan the train route? will it be network type or 1-1 type?

#

oh, you mentioned above, all trains will be individual

#

that's interesting

oblique hollow
#

while resource management will probably be a web or something

#

the idea is to build it all but to not include pipes in a majority of this

#

so nothing will work if loaded

#

then you just follow a small guide that tells you "go to checkpoint x and solve the issue there"

glacial hemlock
#

i see

oblique hollow
#

i find that solving given problems is a great way to teach some things

#

if i had signs i could include all the info on site

#

buuut for now i will just include a ReadMe text file that guides you through and gives you tips on how to solve things

#

i can definitely see that oil might be a bottleneck tho lol

#

since i use all the rubber / plastic alts xd

bleak coral
#

I suppose you don't want to make it dependent on mods. Cause there's a really good sign mod.

oblique hollow
#

id rather not, no

#

i want people to be able to play this without any possible issues

#

and mod requirement when you perhaps never even had the SML? no thanks

fierce ruin
#

you could mark it up in post po

oblique hollow
#

a seperate savefile with sign mod.... maybe

bleak coral
#

Seems like a lot of effort to do both a read me and signs, doesn't seem like a good use of time

fierce ruin
#

sign mod may not get a good angle for overview shots tho

oblique hollow
#

i want this map to be a practical learning experience compared to the Plumbing Manual that i also plan

bleak coral
#

Challenge mode: fuel generator tower isn't working, and there's like 10 things wrong with the setup

fierce ruin
#

don't build the left side then it'll be all right...

deep root
oblique hollow
fierce ruin
oblique hollow
deep root
#

Make a hundred of them!!!

oblique hollow
#

current game progression in fluids is how im gonna structure the savefile too

unknown.png
#

im having doubts on intruducing players to pipeline logic tho

#

i think that would destroy them

#

Challenge Nr.420: Build a Pipeline Arithmetic Logic Unit

fierce ruin
oblique hollow
#

oof

worthy copper
#

Among the math lessons satisfactory can teach:
Systems of equations, via plastic/rubber recycling chains

fierce ruin
#

linear solving, multi calc...

worthy copper
#

i dont think there's anything thats nonlinear in SF (aside overclocking power) but there's a lot of good linear math

#

and at the top end some sick linear optimization

fierce ruin
#

the volume of an inscribed rectangular prism in the range of the hover pack

loud heron
#

wouldnt it be easier to do a spherical check

#

I suppose a rectangular prism is more consistent for straight flying

fierce ruin
#

oh, Count Calculus was the one to solve that

wind spade
#

or just solving the math about least amount of waste to radiate whole map ๐Ÿคท

loud heron
#

Wouldnt you just need to add a plutonium waste every like cubic 2 meters

wind spade
#

was doing that math for U3, as it was like a day after U4 came out and I didn't have the values for U4 stuff

loud heron
#

I love programming

#

I hate mathematics

fierce ruin
#

ironic

loud heron
#

ikr

oblique hollow
#

.... Has anyone ever suggested new alternative recipes on the QA site?

fierce ruin
#

sounds like an alternative recipe for disaster

#

I'm sure there has

versed violet
#

Troubleshooting nuke (again).
So my nuke plant drops out every couple hours. I suspect water problems?

  1. Is it possible to run nuclear power at 250% clock? Wiki says it will default to 200% and some decimals, so it will run out of water (mk2) pipe when running max clock?
  2. I have clocked down my plant to 249,96, which should be below 200% of water usage, yet still got nuke to drop out after couple hours of play.
    My second nuke runs at 100% and never has the problem. Should I downclock even more?
    Previously the second nuke would have the water buffer stuck at 197,6 whenever I went to check the problem, but now it stays properly full (400) ater the first downclock.
wind spade
#

nuclear plants scale differently from other gens

#

250% shouldโ„ข๏ธ be possible

#

(producing 200% power)

versed violet
#

Wiki says it will be 2.00000009951 at 250%, but it eats 300m2 water at 100%, so will run out of water from mk2 pipe at some point I guess?

wind spade
#

I think game imperfection would ruin it faster than those decimals ๐Ÿ˜„

bleak coral
#

btw greeny, have you seen how kwjcool is measuring resources weights on the wiki? he's compared how much 1 unit is worth of the total of that specific resource, rather than the total of a specific resource to the total of all resources. Do you think that would change how weighted calculations are done compared to how you do it?

wind spade
#

hm

#

I wonder how much different it is

#

since I think it's pretty much the same

#

unless I understand it wrong

keen flame
#

well it would be the difference between iron being 1/~70k vs iron being 1/~150k (or whatever the total is)

#

one of my next goals I think is going to actually be working on a more in depth resource weighting system

wind spade
#

idk how that would change weights, can try

keen flame
#

I want to make it comprehensive and use all of the relevant factors - quantity, demand, how many recipes there are, even distance from spawn (since that affects bauxite and uranium heavily)

wind spade
#

but I don't think any such system can be objective enough

keen flame
#

I want to also convert power into a resource opportunity cost. May try and recruit @iron prairie for that too since they've done a ton of work on that

granite jasper
#

There will never be one right answer/formula. The more a user can customize it to their needs the better.

keen flame
#

hot take: I may experiment with Uranium being weighted as 0, since it's only used for power, so there's no opportunity cost associated with it

#

(I think that messes with calculations for point sinking though)

wind spade
#

yeah, but mine is at least objective enough, "complexity" is hard to quantify

keen flame
#

I think the way to do it is going to be to run an analysis on, well, every possible recipe combination and their raw resource inputs

#

there's a few recipes that are worse in every way (looking at you, compacted steel) that might be skippable

wind spade
#

I don't think there's any "worse in every way" recipe tbh

dry wave
#

Anybody have thoughts on this HMF setup?

HeavyModularFrame.png
cedar mica
#

Its a good chunk of oil, for a product that dont use it normally

iron prairie
#

More HMFs than are really needed IMO, and I'd put in steel coated plates if you have it?

dry wave
#

400 rubber is a lot of oil?

wind spade
#

also replace steel beams with steel rods

iron prairie
#

400 rubber is not a lot of oil.

wind spade
#

400 rubber is 133 oil

cedar mica
#

Shows 5000 rubber, but didnt see the arrow saying it drains 400

wind spade
#

yeah, that's input value, need to fix it ๐Ÿ˜„

dry wave
#

I'm going for 100-ish rockets.

iron prairie
#

I ran the numbers at some point, and the coal "saved" by the rods->screws path constitutes less energy than is saved by steel screws, by maybe a factor of 2.

cedar mica
#

Still, I would not be spending rubber on something made from just Iron, the most abundant resource. Maybe moving it a few steps up so the Heavy Modular Frame used it, it would be a different matter

wind spade
bleak coral
# wind spade since I think it's pretty much the same

From what I understand you use the inverse of the percent of total resources one resource reprensents. As in (total of all resource throuhput) / (total oil throughput) = the amount of points 1 of the resource is worth Is that right?

kwjcool for the recipe analysis is using ( (resource throughput used in recipe) / (total throughput of that resource on the map) )* 10,000 to assign it a number of points to represent how much that resource is "worth" compared to the total of that resource you can get. Scaled up for a more general analysis I'd assume you'd do (1 / total resource throughput) * some normalization number to get the amount 1 of a resource is worth and use that as the base numbers to compare instead.

iron prairie
#

Adhered iron plate isn't as effective as steel coated plate, but unless you're running short on oil, I'd say it's worth it for the excellent savings in iron.

cedar mica
wind spade
bleak coral
#

yeah that's what my question was, so far it seems to be the same but he's not done yet

wind spade
#

e.g. in my case iron has weight of 1, while copper has weight of 2.438 (so 2.438 times heavier)
in kwj's case, iron is 1/70380 and copper is 1/28860 (if I understand it correctly), which again equals to 2.438 times heavier

#

I think the weights relative to each other are the same

#

and the concrete numbers are never important, it's only the ratio between them ๐Ÿ™‚

dry wave
#

More like this?

HeavyModularFrame2.png
bleak coral
wind spade
#

the exact numbers differ, but as I mentioned, you never care if iron has a weight of 1 or 1000, you only care if iron is twice as heavy as copper or vice versa

dry wave
#

Did my update incorporate all of your suggestions?

wind spade
#

looks good, tho as people said, maybe the screws can be rolled back to the steel screws ๐Ÿคท I just looked at it from resource efficiency, but if you use coal for power, it can indeed hurt more than help

bleak coral
#

also it's a lot more buildings, steel screw is crazy fast

dry wave
#

I have turbo fuel for now just to bootstrap it all

#

I'm gonna sink the Plutonium rods because I want zero waste. So I need to keep it under like 200GW

#

My main considerations are: 1) Getting that 96 Rockets, 2) Reducing complexity, 3) Coming in under 200GW.

wind spade
#

the third one can be easily avoided by actually using plutonium power ๐Ÿ˜›

dry wave
#

NO WASTE! ๐Ÿ™‚

wind spade
#

waste was not a big deal in U3 and plutonium waste is even lesser deal ๐Ÿ˜›

dry wave
#

True. I should just abandon that dream. I just never wanted to have to revisit waste storage. In U3, I made a GIANT waste storage area under the map which of course I never used because I started over for U4.

#

I'm going for clean, lazy builds. Long ass lines of machines. Fugly.

bleak coral
#

you probably won't have to, plutinium waste is really slow so it's not hard to build enough ISCs to not worry about it

the only times I think anyone should sink plutonium are:

  1. you have an irrational hatred of waste and can't stand it
  2. you don't actually need the power yet, so you're sinking it until you need the power
  3. for some reason you want to leave the game on 24/7 (this will be more of an issue with dedicated servers once they exist)
dry wave
#

I'm definitely 1.

#

Though, I like the option to just use those if I need the power.

#

so maybe I'm a 2

frosty owl
dry wave
#

Anyway, I'm coming in way under max coal and oil usage, so I'm just gonna go ahead and splurge whenever using those reduces raw machine count because I'm lazy

wind spade
#

that's actually a good point. You can prepare it for full power and just sink all the rods that you don't need yet

wind spade
#

and yeah, the ability to change weights is planned

bleak coral
#

somethings broke on mobile where if you have a newline it can't autocomplete

#

so stuff like pings and emojis

wind spade
#

yeah mobile discord is crap

bleak coral
#

it's so black though ๐Ÿ‘€

#

I want that black on desktop.....

keen flame
frosty owl
frosty owl
wind spade
#

planned optimisation targets:

  • weighted resources
  • power
  • space? (if I can get my hands onto a good size data)

planned additions to production tool (additions to list of stuff you can produce):

  • power
  • sink points
#

all of those should be available when I finally finish the v2 of the API + tools

bleak coral
bleak coral
wind spade
# bleak coral it's so black though ๐Ÿ‘€

e.g. you want to put #math-and-meta in your message, you start typing #math, a list appears, where you can click on the channel name for it to be autocompleted, you click on the item and it indeed autocompletes to #math-and-meta, then you continue typing e.g. alternate and the autocompletion dissapears and you're left with #mathalternate

wind spade
bleak coral
#

oh that's on the wiki, it probably just needs to be confirmed to be up to date (most of it is, someone usually catches it when models change)

wind spade
#

the thing is, I can get the data manually at any time. I want automated way of getting the data

bleak coral
#

oh that's different

wind spade
#

there were some attempts by one guy and those attempts already proven that a few data on wiki were incorrect

#

though he's not responding to messages anymore ๐Ÿ˜ฆ

#

which is a shame, as he was building a tool to export those for me

bleak coral
#

oh that's unfortunate

wind spade
# bleak coral oh that's different

the reason being new updates fucking my datasource (what to put on new buildings? I'd have to wait until a reliable source gives out the measurements, which would delay the update)

bleak coral
#

I enjoy measuring stuff, so I could check the accuracy down to the meter (past that is guess work), but yeah I have no idea how you'd automate that

#

or even setup an export from the wiki

#

maybe it's something we could bother CSS to add to a public-readable file in the game?

#

not sure if that's a good Q&A post, or better for the community managers since it's a community/tool thing rather than a game thing

wind spade
#

I actually want to poke a CSS dev with some changes to Docs.json, so I may add that to list of "no pressure but if you could add this that would be awesome sorry for pings don't ban me plz kthxbye"

#

(I was talking to him before, reporting issues in Docs.json and giving him suggestions, so it's not a problem. I just wanted to wait a bit after U4 so that I'm not pinging while they have tons of work)

frosty owl
# wind spade planned optimisation targets: - weighted resources - power - space? (if I can ge...

Dunno how in line with your idea of the tool this is but...
What about making the nodes on the graph more interactable so that one can change more in-depth the plan?
Eg: plan for making rotors (standard recipe)
There is one node for all the rods production, but I want to split it so I can differentiate the rods for screws and the ones for rotors
Then, I change the clock on one node to get an even number of machines in that (or make it 1:1 with the next node, whatever) and the change is reflected in the power calculations (very useful for those who OC whole sets of machines to match numbers)
||I'm using "OC" for both Overclocking and Underclocking||

wind spade
#

improving the visualisation is one of the big plans for the futureโ„ข๏ธ as well. However, I first need to do tons of research to find a decent library that would support all of my crazy ideas (as the current one, while being pretty good, is not the best really and brings some issues)

frosty owl
#

Btw, an option to change clocks for all or each kind of machine in the "power" tab of the tool to see how that affects power would be pretty cool too
Sorry if I'm throwing ideas your way, I'm just hoping they might prove useful :sweat_smile:

#

So, no visualization touched this way

wind spade
#

overclocking is to be honest the feature that I'm most scared of

frosty owl
#

How so?

wind spade
#

it's on my todolist for a long time, but it brings so many issues with it and even though I've thought about tons of ideas, I'm still not sure if I'll ever manage to find at least a decent way to implement it

dry wave
#

The main thing I want is to add one Factory Tab as an input node of another Factory Tab.

frosty owl
#

You can kinda do it by adding an input manually...

dry wave
#

And the resource usage of Tab1 would be deducted from input resources of Tab2

bleak coral
#

btw is the merging of the U3/U4 sites waiting on the V2 to happen?

frosty owl
#

Oh my, text wall incoming? hehe

wind spade
#

basically, you have two different ways you can implement OC:

  • you define what do you overclock before you get the final result
  • you define what do you overclock after you get the final result

for the first one, it's not as bad. Only thing that's really limiting is that the user doesn't know how the final setup would look like, so all those people that want to overclock to a nice number or have a limited amount of shards and want to put them only to a few buildings would have pretty hard time setting this up. Not even sure how would the overclock settings look like

for the second one, it's way worse. For user, it's great, you can do pretty much anything and the tool will recalculate... oh. not really. Recalculation means you'll get a new result... without the changes you made to the previous result. In the current state, you could just copy the changes over after recalculation... but that brings many edge cases like people changing the input values and how you then copy all the changes over?
Also, the soon-to-be-implemented power functionality would break this. Imagine a request "I want to produce max X items, but I'm limited by Y power", so the tool gives out final numbers and then you switch OC on some buildings, making the whole result invalid (you either have leftover power, or not enough power).

It's way more complicated than this, but those are general issues with OC being implemented. And yeah, I have no problem discussing it further if you want ๐Ÿ˜‰

wind spade
loud heron
#

The overclocking thing isnt really necessary imo

#

You can just think in your head how many shard youll need to overclock a machine to make up for the amount of machines needed

#

if the tool says 2.35 constructor, just overclock one to 135%.

The input stays the same, the output stays the same

#

Only difference is power, and power is something you should be making in excess if you are a minmaxer

wind spade
#

well the tool is not just for minmaxers ๐Ÿ˜‰

#

I want to make it to widest possible range of users

#

which unfortunately means that I have to make a lot of compromises

frosty owl
# wind spade basically, you have two different ways you can implement OC: - you define what d...

So, to see if I got it (second OC implementation):
Tool makes the plan on input data (A). User changes the clock node X
Tool recalculated but without considering how node X got changed...?
It makes little sense to me atm, sorry ๐Ÿ˜…
Like, is there no way for the tool to:
Make a plan
User modifies a node, presses enter
Tool checks the node, adds the data and recalculate plans on that.
New graph is made (have to re-order nodes, but that's not a big deal for the user)

wind spade
frosty owl
#

Oofff

wind spade
#

the nodes are part of a result, but not part of the request

loud heron
#

It seems more like a foundational design flaw to make it easy on your development

#

Unless youre open to reworking how the entire logic of processing a factory line works

#

So that overclock logic is easier to implement

#

But then it turns into the SCIM tool

wind spade
#

example of a request:

  • make 100 HMF
  • use these recipes: [list of recipes]
  • use these weights: [weights] and limits: [limits]
  • also I have [input]

example of response:

  • you need:
    [list of:
  • recipe used
  • number of machines making that recipe
    ]
bleak coral
#

I do think there's something to be said for OC being a low priority. Once you have the production amount it's not hard to figure out what you can do with the number of machines, it's just some division. The thing that gets hurt in that is accuracy of power, though generally of course OC = more power and UC = less power. So power estimate is still a good baseline to work off of.

frosty owl
# wind spade the nodes are part of a result, but not part of the request

Some way to cheese it by adding constrains to the optimization?
Eg: recipe X must divide its input/output between A/min and B/min (or just force to display 1 node for each output on demand), resulting in 2 nodes (node-split)
Tbf, the clock thing could be purely VISUAL on the node and then plugged into the power calculation whenever one clicks on that tab

wind spade
loud heron
#

i got this many shards, how many buildings can I remove from the picture using them

wind spade
wind spade
frosty owl
loud heron
wind spade
#

because it's linear, it doesn't support exponential math ๐Ÿ˜‰

loud heron
#

So is that a limitation on the existing foundation

frosty owl
#

Just square it all~ ๐Ÿคฃ

wind spade
#

and I can't just list all possibilities, as you have 2500000 of them per building recipe

#

(and that's just assuming all buildings making that recipe have same OC)

loud heron
#

Either way its not gunna be simple to implement, so if its really just gunna hurt you more than it benefits us, theres no rush.

#

The way your system does factory planning is nice and efficient but has little room for modularity it seems.

#

since your result is fixed and generalized, rather than piece-by-piece

frosty owl
wind spade
loud heron
#

so is overclocking

#

afaik, theres no true way to properly overclock any building plot

frosty owl
#

All very fair points

wind spade
vital quest
#

is there a way to know how much water an overclocked coal generator is consuming?

frosty owl
#

I feel like I'm trying to overwork Greeny by pressing this so much xD

loud heron
#

You could overclock any set of machines in the node and they'd all be equally as efficient regardless.

#

so it sounds impossible to calculate on your tool, fundamentally wise.

wind spade
#

I never planned on the tool doing smart overclocking on behalf of the user

bleak coral
loud heron
frosty owl
wind spade
#

"support for overclocking" for my tool means "user can change clock speeds", not "tool will automatically overclock stuff based on random math"

bleak coral
wind spade
#

^ also that

loud heron
#

You could technically add an input for "overclocked machine by x"

bleak coral
#

so I don't think there is a way to do it objectively and is a good fit for people

wind spade
keen flame
loud heron
#

So how does one define "change clock speeds" cause I doubt itll be useful if every machine on a node is overclocked simultaneously

#

thats alot of slugs depending

wind spade
#

one option is to have it similar as in old tool

loud heron
#

Most people will probably want to know "what can I do with X many power shards"

wind spade
#

only with recipes instead of items

c9015dce-a77d-4236-aaab-ac2122bae1dd.png
loud heron
#

when it comes to factory design

loud heron
wind spade
#

beginners shouldn't overclock in the first place

frosty owl
loud heron
#

So who is this target demographic for overclocking features?
beginners dont overclock, and intermediate and veteran engineers can do the math in our head

wind spade
wind spade
wind spade
#

also, this topic is very closely related to another issue currently not solved by the tool - miners

frosty owl
loud heron
#

This just seems like basic addition if you know how much ore you need, and the types of miners, nodes etc

frosty owl
loud heron
#

oof

wind spade
loud heron
#

If anything, the feature I wanted for the tool is "I got X amount of ore, how many of a part can I make with this"

wind spade
loud heron
#

Would be nice to choose an option for "Fused frames - as many as possible"

#

Without having to play +1 -1 +10 -5

wind spade
#

there's literally a "maximise" setting

loud heron
#

I havent noticed.

#

Thanks

wind spade
#

and you limit your raw resources in Items, Input

loud heron
#

You have to type maximise?

wind spade
#

no, it's a select box

loud heron
#

Oh in the second column

#

was not obvious

wind spade
#

idk I think the arrow down that marks the thing as a select box was good enough ๐Ÿคท

frosty owl
# wind spade - different node purities - different node counts - overclocking miners (related...

My idea: have them in their own tab (left side of graph), the plan solved for max/min MK on max/min purity, then it's up to the user to change the amount and OC of miners, while keeping the amount of resources OR power used constant (again, for the whole power and OC ordeal)
Or just have them on a set list of OC level (belt limits maybe?) so the tool only has a limited number to choose from for the solve?

wind spade
#

tho currently maximisation is a bit broken if it you use input with it, sometimes it works, sometimes it doesn't. So I don't recommend using input+maximise. Maximise from raw resources works fine

#

issue comes with people that go like "I have 2 pure iron nodes and 1 impure"

loud heron
#

cause thats how I feel half the time

#

like an idiot when I figure these things out

wind spade
wind spade
frosty owl
loud heron
#

Even then your clients will find a way to submit support tickets

#

on things you didnt ever expect

wind spade
#

if I had a dedicated UI designer, I assume the tool would look better. But I'm a backend developer, almost never designed any UI and I just put together something that was "good enough" for me, so that I can work on the features

#

and tbh, there are very few people that haven't noticed the maximise option, so I don't think this is a big problem

loud heron
#

Thats fair

wind spade
#

more problematic is the maximise slider, but that's a story for another time anyway

frosty owl
#

I gotta bail the convo now, wish you all a good brainstorming ๐Ÿ‘‹
Thanks for keeping up with all us idiotic pioneers and still not give up on development, Greeny <3

wind spade
#

basically for OC/miners, I need answers to these questions:

  • how does user define what to OC and what to not OC?
  • how much does the tool solve regarding OC and how much is just visual?
  • how to define available nodes/purities?
  • how to behave if more/less nodes are needed than available?
  • how to solve stuff like "I have limited amount of shards"?

and there are no easy answers to these questions, especially if I want the tool to work for all most possible requests

#

so that's why for now OC and miners features are kinda put on hold

raven osprey
#

Do I want Compacted Coal or Solid Steel Ingots? I'm just about to set up my first steel production and have a choice between those two alternate recipes.

supple mural
#

i think solid steel is better

#

just cuz it doesnt need sulfur

deep root
#

Compacted is pretty much the worst recipe in the game

raven osprey
#

Oh, alright. I saw some places saying it was useful for Turbofuel or something but I'm nowhere near that

wind spade
deep root
#

OH!

deep root
#

Compacted COAL is excellent...Compacted STEEL is terrible

#

It's also good for coal generators

fierce ruin
raven osprey
#

Hm. Looks like I have a choice to make. Once I pick one of these, how hard will it be to get the other?

deep root
#

You can get them all, they aren't hard

frosty owl
bleak coral
#

Compacted coal is meh for cola coal generators

#

It's the same as just grabbing more coal basically

frosty owl
deep root
#

We all have opinions, I'm glad we can both share them and let others decide on their own if they want to use coal or compacted

bleak coral
#

There's like one place in grasslands where that's true, but I'd rather convince somone to just walk a bit lol

frosty owl
#

Huh, I just noticed I didn't even touch compacted coal last playthrough. Went blended TF directly from fuel

bleak coral
#

except you also used a rarer resource instead of just 300 coal

frosty owl
deep root
#

Meaning it's the same, and both our opinions are valid...........

bleak coral
#

no it's worse because you added complexity and used a rarer resource and got nothing from it

#

spend more, get the same

deep root
#

What do you use sulfur for in T4/5?

wind spade
frosty owl
#

Turbofuel?

deep root
#

And what do you use it for in T3?

frosty owl
deep root
#

Meant to ask T3/4 what do you use sulfur for

frosty owl
#

Just... For everything xD

bleak coral
#

if you spend the sulfur on compacted for coal generators, that's now tied up until you can replace its power, which you can't use that sulfur for

frosty owl
deep root
#

Because you are starved for sulfur in T4/5???

bleak coral
#

I mean it's not like a terrible thing to do, it's just a useless thing to do

#

play however you want, but you absolutely don't benefit from doing it

deep root
#

Which brings me right back around to my very first statement, I'm glad we can all share our OPINIONS on the worth of compacted coal

frosty owl
#

Yeah, I see Soranis' point
Let's say you overshoot your armory plans and max out 2 sulfur nodes on compacted coal, just 'cause you could...
Why would you NOT burn all that stuff?
By the time you get to turbofuel, you got a whole step taken care of ^^

deep root
#

Added benefit

frosty owl
#

Tbf, having to MAKE compacted coal was the reason I pushed with fuel up 'till the blenders xD

deep root
#

I skip turbo in 5/6 and go straight for blend in 7 now...

#

Compacted coal lets me build compact (haha pun) setups of coal gens and have enough power to do so

bleak coral
#

what? it's more space not less for the same amount of power

#

cause you need assemblers

deep root
#

The coal gens not the other space

bleak coral
#

it's the same amount of generators for the same amount of power

wind spade
bleak coral
#

you don't get bonus power, they just burn slower so they can support more gens

#

you can yell "it's my opinion" till you're blue in the face, but you keep saying shit that's just wrong

deep root
#

Sure

bleak coral
#

having an opinion doesn't make you immune to criticism

deep root
#

LOL

#

Yet it means mine is worth less than yours

#

Cool story

wind spade
#

relax guys

old tartan
#

Let's do a little thought experiment ok. Say and I dont know the numbers but that compacted coal takes less manufacturers to power coal generators. Ok no duh. Now you realize it is more efficient and producing more power then normal coal. Now you realize that when you get lucky you can turn those coal generators into fuel generators and it will be very easy to convert the compacted coal to super fuel so what Soransis is saying isn't and opinion it is actually if you do the math Sornsis is factually correct

wind spade
#

or you'll go to hell where there will be quickwire screws

old tartan
#

Is vencam writing a thesis?

frosty owl
# wind spade yeah but that still doesn't answer like half of the questions ๐Ÿ˜›

Toggles can be a way for the user to answer those question himself, can't they?
Issues with miners are
-"How does user define what to OC and what not": choice between "OC lowest purity" and "OC highest purity" (former is more miners and less power, latter less miners more power)
-How much does the tool solve for OC: IMO the visual part discussed earlier would be plenty enough as adding it to the planner doesn't seem worth the hassle. Having it for miners already seems a stretch
-"How to define aviable nodes/purity?" Select aviable nodes from a small "SCIM-like" map interface or just "select all pure/normal/inpure or let user input the number of each (within limit)
-" How to behave if more/less nodes are aviable": if less - > error, prompt to insert correct data; else - > Just like with unused byproducts from production, have a node notify of the excess output
-"How to solve stuff like limited number of shards": I don't see much value in this option, so dunno :man_shrugging:
These are my 2 cents, sorry for spreading them out over such a long text
:sweat_smile:

wind spade
#

How does user define what to OC and what not
was a question regarding OC more than miners

old tartan
#

JESUS

wind spade
#

Just like with unused byproducts from production, have a node notify of the excess output
which nodes will be used and which not? ๐Ÿ™‚

frosty owl
#

The ones requiring lowest power I'd say, unless there's a toggle like "don't bother about power usage". In that case, I think it'd be fair to assume one doesn't care about what to OC and would prefer the minimum number of miners

frosty owl
old tartan
#

I can see it being a huge headache for non spaghettis

frosty owl
#

What would?

old tartan
#

The additional everything

#

My brain is very small and so ne and my friend go halfway

wind spade
old tartan
#

Holy shit HELLBACON

frosty owl
old tartan
#

Oh I see

#

Thought that was a random person

frosty owl
#

||No, just me doing some funny... I guess the picture kinda gives me away though :/||

old tartan
#

Yeah at least you did it on an idiot who didnt think of that

swift robin
#

so i think i discovered that when i delete nodes on a truck's path that changes it's movement speed when it's out of full render range. i was watching on my map and when the truck reached the segment where i deleted some nodes it went super speed on that section

lethal crypt
swift robin
#

teleporting to each node sounds plausible

#

something to exploit so my little tractor can saturate a mk.5 belt? OK ๐Ÿ™‚

sand garnet
#

Yes it does teleport

oblique hollow
#

@swift robin Peak cable output simon_smile

unknown.png
fierce ruin
#

I'm not even cableable of that!

oblique hollow
#

im gonna slap you, you UnCableable HelpDesk

wicked tinsel
#

you can add caterium to increase wire production ๐Ÿ‘

oblique hollow
#

for 1 caterium ore/min extra i actually get a lot more out of it

#

which is funny af

fierce ruin
#

interesting

swift robin
#

so spreading out resources real thin and mixing them all together is the way u get insane efficiency

fierce ruin
#

sushi belts are viable?

oblique hollow
#

2.88 and 3.59

#

not bad

swift robin
#

wow

oblique hollow
#

but i think diluted fuel is gonna destroy this

swift robin
#

i like that one a lot

fierce ruin
#

faht stacks

swift robin
#

less than 3 copper ore/mine for 100 cable is XD

fierce ruin
#

XD

#

of oil the things that could happen

swift robin
#

oil makes everything OP

oblique hollow
#

lmao long live diluted plus recycled

unknown.png
#

2.4 and 3 xddd

fierce ruin
#

OP = Oil Products rolljace

wicked tinsel
#

๐Ÿ‘

oblique hollow
#

10 oil, 3 caterium and 2.4 copper = 100 cable

swift robin
#

anyone who actually makes their cable like that? lol

oblique hollow
#

jup, makes sense

fierce ruin
#

It's nerf or nothing

oblique hollow
#

i will be simon_smile

swift robin
#

nicee

#

i guess we all learned something today

oblique hollow
#

in a nutshell Oil = OP

swift robin
#

so i guess i will actually split that caterium ingot belt and use it to make wire now

oblique hollow
#

recycled and diluted invalidates any recipe using raw HOR

wicked tinsel
#

yep

oblique hollow
#

theres 2 recipes

swift robin
#

yeah

#

factory are become too complex now

#

brain overload

fierce ruin
#

greeny releases 2.0 messing this all up

oblique hollow
#

i recently made a factory that produces fused quickwire and fused wire

#

is real nice

fierce ruin
oblique hollow
#

its where screenshots like this came from

Screenshot20210429-21395900000.png
#

yello'n'orange

high wave
#

nuclear power makes me cry

unknown.png
#

also ik a lot of the numbers there are wrong thats the problem

#

or maybe they are right im tired

toxic hamlet
#

I understand this is a channel for satisfactory math, but i'm getting desperate

unknown.png
fierce ruin
#

stackflow hehe

oblique hollow
toxic hamlet
#

yes

#

i don't know if it's possible to solve i'm dying

fierce ruin
#

best guess is that it's imaginary ||(but it's been sooo long)||

wind spade
oblique hollow
toxic hamlet
#

epic, thanks

wicked tinsel
#

๐Ÿ‘

unknown.png
oblique hollow
#

log5(-125) = ln(-125) / ln(5)

#

Though realistically the solution should be - 3

#

Oop

#

I mean.... - 5^3

wicked tinsel
#

but that would need a base3 log

oblique hollow
#

Ah well

obsidian oak
#

Main condition that number under log have to be more than 0

oblique hollow
#

Just do some funny substitution and equivalent exchange

obsidian oak
#

Lets wait for the moment when input/output rate will be numbers with log lol

oblique hollow
#

This is Log BASE 5

#

And 5 ^ 3 = 125

wicked tinsel
#

yeah

oblique hollow
#

If only we could magically move the -1

wicked tinsel
#

which is why you get 3 as real part

#

log isnt defined for negative numbers so it wont work

#

you need to switch to complex logarithm

oblique hollow
#

If it was the log of base negative 5 then it would work

wicked tinsel
#

yeah, since (-5) ^ 3 = -125

#

but you cant create a real exponent that turns positive number to negative one

#

well, the root question is, where this comes from and whenever there are no typos in it

#

if its calculus 2~3, then probably polar coordinates should be used and complex result provided by wolfram is the right answer

#

if its basic math then there is likely a typo somewhere

oblique hollow
#

Huh..... Why tho

IMG_20210506_165237.jpg
wicked tinsel
#

well, this just flips the log around y axis

#

now negative numbers are positive so its defined for negative x

#

aghr, doesnt copy right

#

this is the solution for whole real axis

unknown.png
#

ie. you would split it like you did into ln(-125) and then apply this

oblique hollow
#

Oh wait.... I seee

#
  • x, but x is negative itself
#

So - 1 * (-something) = +something

wicked tinsel
#

yep

oblique hollow
#

Sneaky log hackery

wicked tinsel
#

nah, you just flipped the function on y axis ๐Ÿ˜„

oblique hollow
#

Its still weird if you are like a math beginner "duh, why does everyone complain that there is no log for negative numbers. Look, i did it!"

wicked tinsel
#

yeah

#

well, its easiest explained like i wrote before

oblique hollow
#

I dont wanna deal with complex numbers outside of work though

wicked tinsel
#

you cant make a real number that will turn positive number into negative using exponentation

#

its just not possible => no negative log

oblique hollow
#

I know i know, rotation around the imaginary plane by multiplying with i, yada yada

wicked tinsel
#

not sure if Frigid is convinced ๐Ÿ˜„

#

im 90% sure its just typo tho

oblique hollow
#

Tranfer function and frequency responses.....
Shudders in Laplace

wicked tinsel
#

i was asked if im interested in going for phd, but "meh, more calculus" :x

#

and in last ~11 years of working i had used it once and it was a problem that wasnt covered :<

#

and wolfram had sufficient answer in few seconds

oblique hollow
#

Heres what i deal with. Replace s with i * w (omega)

IMG_20210506_170333.jpg
#

Thats one of the more tame things

#

Its just a PID controller

lusty oxide
#

moves his poles to the left side of the plane

abstract thorn
#

Anybody know the numbers for 100% efficiency for saturated mk4 belts of iron ore into pure iron ingot?

#

how can I get it into ideal outputs

wind spade
abstract thorn
#

That number I can come to

#

I mean balancing it onto belts

wind spade
#

don't balance ๐Ÿ˜„

abstract thorn
wind spade
#

balancing is useless

#

based on where the ingots go, make as many refineries output to one belt as the target needs

manic oak
abstract thorn
small light
#

what is the most optimal of efficient speed to run a fule gen

#

fuel gen

wind spade
#

there's no difference

#

run it at 100% so you don't get weird decimal numbers

#

you don't get any efficiency boost from overclocking or underclocking

small light
#

i have 10800 fuel/min do i 250% or do i lower it to a normal number that has no decimal's

#

becouse i dont want to place 900generators in my world

abstract copper
loud heron
#

No way to burn 10800 fuel/min without some giant fuel gen plant.

glacial hemlock
loud heron
#

Building nukes would probably be more your pace lol

#

You only need like 50 of em

small light
#

540gen= 137160 MW of power should be enough

loud heron
#

Then go make the 540 generators.

glacial hemlock
#

that's a bit too much. You are over expanded considered you need that much power yet not unlocked nuclear

#

seems like you have to really build that much generators

loud heron
#

They really should just make Fuel Gen MK2

#

Or just...nerf oil recipes altogether.

#

Being able to 4-6x your fuel output with a lil water and some blending is really OP, but OP in a bad way.

bleak coral
#

I think they should just buff fuel gens overall, 75MW to 150MW seems a paltry jump compared to 150MW to 2500MW

#

fuel/turbo fuel is capable of easily doing 10s of GWs, and I think that's perfectly fine

#

the generators just should match that capability

#

right now roughly you go from thousands of MWs (coal) to tens of thousands of MW (fuel/turbofuel) to hundreds of thousands of MW (nuclear), and I think that's a nice progression with our power needs

#

fuel and nuclear can do more, but that's roughly the curve

wicked tinsel
#

the root problem is that fuel generator has fixed power

#

it should scale with input to begin with

#

also someone with saw should take care of that pointless stack =.=

#

one of my wishes for this game is to be able to saw off stacks from buildins and replace them with pipes

#

and some dedicated collector stack building to get rid of exhaust gas

loud heron
#

Just give us yuge fuel gens

#

Or eliminate turbo fuel as a recipe I dunno.

#

This game has a really screwed scaling when it comes to I/O numbers

#

Too high, and they start breaking their own game (conveyer mk6)
Too low, and we're forced to create an artificial venezuela

#

I think we could use a squash update.

wicked tinsel
#

a lot of stuff was designed to look cool on its own but in result looks bad in large quantities

#

probably art and gameplay direction deficiency

keen flame
#

I fully expect we'll see an update way down the line that adds tons of aesthetic customization and improvement

#

They're just focusing on gameplay and feature development first

#

Gotta remember we're still in early access xD

loud heron
#

We're not talking directly bout aesthetic

#

We're talking about the unnecessary large quantity of machines.

#

I feel like they really should target a "maximum" machine count for balancing their factory line diagrams.

#

It still kinda feels like they've been pushing more for content and less for actual polish before adding the content.

#

Which I guess is a design decision they made early on, just feels a lil exasperating at times

keen flame
#

Yeah....polish is the last 10% of a games development

#

That's true of any game

loud heron
#

Adding the finishing touches at the very end is fine

#

But if it's going to be in early access the least they could do is do a bit more iterative feedbacking

keen flame
#

Polish is also more than just finishing touches. There's a saying among devs that the last 10% of the project is 90% of the work

keen flame
loud heron
#

And that's great, but fuel gens have been kinda sitting here for a bit.

#

Unless they have future plans for em

keen flame
#

They're a small team with limited resources :P

loud heron
#

I don't remember if anyone asked during their stream QA about the fuel line

keen flame
#

I'm pretty sure they've mentioned so, yes

#

It wasn't in a QA stream though

frosty owl
#

@solar pond Would you mind sharing some details behind the balancing and beltwork in your screen? ^^

solar pond
frosty owl
#

That'd be awesome! ๐Ÿ˜

#

I never tried having a mixed belt without sink and relying only on balancing yet, but I'm really curios as to how well that works ^^

bleak coral
#

is the idea if there's always room in the buffer it won't get clogged up? it might work if you could turn everything on at the same time

frosty owl
#

Balancing offers quite a good amount of buffering usually :P

#

Too bad you noticed, I wanted to just at you at the end of a spicy sushi convo with screens (do check out Providence's screen in #screenshots though!)

solar pond
#

its a big image if you open the link

fierce ruin
#

tight

solar pond
#

lets me keep the inputs much more compact since i dont need to do multi level splitters

frosty owl
#

But I'm wondering... If you were to balance the assemblers' inputs... Wouldn't this setup work even without smart splitters along the assemblers? superexcited

sturdy loom
#

Food

solar pond
#

it can clog without smarts

#

in theory it shouldnt but it does anyway

#

each of the 2 mixed outputs has 75 cat and 375 copper so 10 fused quickwire assemblers should consume it all exactly

#

ofc the splitter and merger set up is completely different for every product

frosty owl
solar pond
#

another downside is that even though its not consuming 780 copper there needs to be 780 copper coming in or weird shit can happen, so if you want to skim off the 30 extra copper to use somewhere else you have to take it from the excess output and not the input

#

ofc theres a hundred different ways to balance belts to exact amounts

bleak coral
#

so are the machines sending down 180? or is the 180 after the loopback?

#

but this is really clever, nice

solar pond
#

theres actually 270 caterium on the caterium in put belt but the first splitter only has output for 180 so it strangles itself

bleak coral
#

theory: with the loop back you should be able to get away with as little as 150, cause that's what's actually getting taken out

frosty owl
#

I'm totally not gonna steal take inspiration from this, if it's OK with you Providence :)

bleak coral
#

it would be the same mechanism as prime splitters

solar pond
#

yeah if you make the first care splitter a smart splitter and have it force onto the 120 belt first and overflow onto the 60 belt you can have 150 as the input

#

you also wouldnt need the 30 30 splitter in that case

frosty owl
#

@bleak coral You remember who was it that had issues and disbelief towards sushi belts?
I wanted to at him/her, but can't recall who it was... (I think we all had a convo about it one time)

bleak coral
#

I don't think you even need the smart splitter, just leave it as is and after the first loop back it would work

solar pond
#

hmmm yeah actgually

bleak coral
#

cause that's the exact principle under which prime splitters work

solar pond
#

it would split 60 to 90 and then the 60 would loop back in to the 150

#

30 of the 60 would

#

and it would cycle itself down

bleak coral
#

exactly

solar pond
#

however if i had 150 going in none of this would be relevant

#

because i would simply split it into 2 75s

bleak coral
#

true, though I figured part of it was to clear out an backup that can happen randomly

solar pond
#

the point of this is to take a belt that has a random amount on and force it to become 2 75s

bleak coral
#

cause of low FPS or setup or bugs or whatever

frosty owl
#

Btw, it'd be interesting to see how long this setup lasts
I wonder wether the belt issues could affect it on the long run

solar pond
#

it will last forever

bleak coral
#

I'm on the train now of giving us exact splits, cause imagine this kind of cool stuff but for any numbers thinking_helmet

solar pond
#

just put a grinder on the end of the mixed belt if you dont have faith ;p

#

and that will fix any hiccups

bleak coral
#

that's what I do for my aluminum setup to guarantee no water backups, I don't trust the game so I grind any errant copper/aluminum ingots

solar pond
#

but ive been watching it for an hour and nothing has got through into the grinder

frosty owl
# solar pond it will last forever

I'm referring to how later in the game (when FPS start dropping) weird empty zones can form on belts after smart splitting. I'm pretty sure that would F up the balancing :/

bleak coral
#

unfortunately I work in messy decimal land, so stuff like this is out of my reach for now, best sushi belts I can do are those with sinks on the end

frosty owl
# solar pond and that will fix any hiccups

For testing sake: would you mind having a storage at the end of the line and connect only the overflow from that to the sink?
So you could easily check if any overflow actually happened by checking the container ^^

solar pond
#

i mean i can just fire up twitch stream and show you it directly if you want lol

frosty owl
#

It's not the kind of issue that shows on such short notice, unfortunately
It's the kind of thing you check just whenever you have the chance to pass by it

#

Let it endure load/unload of map and savefile and, most importantly, frame drops in late game (or whenever your PC starts suffering due to base size)

solar pond
#

1800 quickwire a minute feeding directly into 90 AI limiters a minute

rare surge
#

Are you just sinking them after storage is full?

rare surge
#

@high wave

2100 uranium/min

84 manufacturer's producing

1680 EUC/minute

84 manufactuers producing

50.4 UFR/minute

NC Reactors producing

504 Uranium Waste/minute

9.1636363636... particle accelerators producing

274.90909... Plutonium pellets/minute

27.490909090 particle accelerators producing 

137.454545...  plutonium fuel rods/minute

4800 / 137.45454545 = ~35minutes
#

My maths, may expand info if you want it ๐Ÿ™‚

high wave
#

interesting

#

where does NF uranium come in?

rare surge
#

1 moment

high wave
#

this is my math

unknown.png
rare surge
#

Wait, I found an issue in my calculation. sorry about that

#

I'm using the base blender recipe for NF uranium, and used 15 UWaste a minute instead of 37.5

high wave
#

the rate of uranium to U waste is 4:3, meaning that the fertile uranium is more efficient than the base recipe

rare surge
#

Ah, nice.

high wave
#

and the optimal rates are 1200 uranium to UFRs, 900 to fertile NF uranium

solar pond
rare surge
#

I assumed going pure waste route was more efficient as it didn't require a secondary input of uranium ore

high wave
#

yea i thought that first before i ran all of the numbers through spreadsheets

rare surge
#

I really should sit down and finish my recipe optimizer. but with all the other tools out there I don't really have the motivation

bleak coral
#

spending that ore instead on uranium fuel rods, making more waste, and using instant/plut fuel unit yields more power

sturdy loom
#

lets game it out has a good idea for storing uranium waste, store it all on mk1 belts to radiate your factory

high wave
#

my spreadsheet says that using all of the ore on uranium rods makes a lot less power

bleak coral
#

what alts are you using? that's based on all uranium efficient alts for making uranium fuel rods

high wave
#

yea im using all the uranium efficient ones

#

actually maybe im not idek anymore

#

i just found a completely different error so

bleak coral
#

here let me get some examples, given I haven't tested this on plutonium setups not using all the alts

#

just all alts vs all alts minus fertile

wind spade
#

๐Ÿ‘€

high wave
#

yea now after fixing 2 slight mistakes you are right

fierce ruin
#

how does a line with a positive slope decrease?

#

as x->infinity ofc

wind spade
#

what do you mean by that?

fierce ruin
high wave
#

is using instant plutonium cells better for power?

bleak coral
#

I believe so, doesn't hurt to double check the math though

#

also I go away to make two examples, and come back to see they're not needed ๐Ÿ˜›

high wave
#

i changed something and now the spreadsheet says its going to make me 3 tw so idk anymore

#

and yes ik that is not possible

bleak coral
#

yeah something definitely went funky there lol

high wave
#

i dont think 1260 plutonium fuel rods per min is possible

#

just a hunch

#

oh i fixed it

#

idk how but i did it

bleak coral
#

now that I've got these examples put together, think I might do an analysis on the rest of the resource usage

#

cause I suspect fertile might actually save a bunch on other resources, which would be neat if you're trying to save stuff for points or whatever and not just make as much power as possible

high wave
#

nah i dont need other resources

bleak coral
#

mostly my own curiosity too, plus it's good to know if fertile is the way if you want to max points

high wave
#

i think ive decided on what i want

unknown.png
#

im using mostly fertile NF to conserve nitrogen

frosty owl
#

BBBRRRRUUUUUUUUUUHHH
I finished a nuclear factory (2100 uranium) a while ago. Very nice project, heavily decorated, load balanced, all the jazz...
Only now I remembered the best recipe for uranium cells is not made in blenders...... :man_facepalming: :tired_jace:

high wave
#

rip

high wave
frosty owl
#

I se up the WHOLE factory based on that...
And it runs at 100% too...

high wave
#

ooh that really sucks

frosty owl
#

Well, I got some nice screens out of it...
But damn......

#

Feel so dumb

high wave
#

ive been working on my spreadsheet all day and i still dont know if its optimal

frosty owl
#

If it says fertile uranium gives more power, you can be sure you got skmething wrong hehe

high wave
#

yea but ive decided to use mostly fertile anyways because nitrogen

#

and really 110gw isnt a lot compared to 1.08 tw

#

so its not a huge loss for the extra resources saved

#

fertile takes half the nitric acid and that stuff is expensive

frosty owl
#

Why you wanna save so much on nitrogen? ^^

high wave
#

late game aluminum products

frosty owl
#

Even without using fertile, you still have enough leftover to make over 50 fused frames/min

high wave
#

idk, its also for other things like cooling systems and turbo motors

bleak coral
#

oh good timing I just finished the comparison chart, did it at 1100 cause that's nice for the uranium ore split for fertile

#

so in the end, it's more bauxite and waste while reducing everything else, but also less power

#

and it's only a 21% reduction in the nitrogen, not half thinking_helmet

#

ohp I formatted that last line wrong lol

high wave
#

21% is half if you are as bad at math as i am

sturdy loom
#

50% is half

#

Lol

high wave
#

wow mr. math genius coming in here with his big math facts

sturdy loom
#

Maf :)

bleak coral
#

fixed:

Screenshot_111.png
#

turns out less power is bad, and should be red simon_smile

sturdy loom
#

I plan on getting to nuclear with only bio burners

#

With refined power mod tho so I dont have to use hand fed ones

bleak coral
#

oh well that's just cheating lol

#

do biocoal like a champ!

high wave
#

do it vanilla and use liquid biofuel in fuel gens

sturdy loom
#

It's a mod not a save editor

high wave
#

or biocoal yea

bleak coral
#

liquid biofuel = extra sadistic mode

high wave
sturdy loom
#

Shut up I'm not hand feeding that many bio burners

bleak coral
#

I mean that's basically the challenge

high wave
#

the real impossible challenge is se4 on only biofuel burners

sturdy loom
#

I'll need a army of people to just keep it running

#

and I'll need a LOT of batteries

bleak coral
#

biomass slaves friends I call to you in my time of need!

#

power storage doesn't work with biomass burners in vanilla

sturdy loom
#

Damn

#

Wait a minute

#

I'm in vanilla and I charged one with a bio burner just to test it

bleak coral
#

it effectively wouldn't do anything, cause it'd just change it from energy stored in biomass to energy stored in power storages

#

unless you sat there and hand fed it for an hour or something

sturdy loom
#

Ok? Energy is energy

#

Remember I'd have a team to hand feed them

bleak coral
#

yeah that's my point, it's like moving water from one bucket into another, it's still the same amount of water

#

ah right of course the "friends" ๐Ÿ˜›

high wave
#

it would make a difference if they werent using mods

sturdy loom
#

You mean random people I find (slaves)

high wave
#

because there can only be so much stored energy in biomass in the grid

#

but there can be as many batteries as you can build

#

so they are better for storing power than biomass

sturdy loom
#

Well if my power shuts down the batteries can keep the factory running while I find the one without fuel

fierce ruin
#

how much power and how many rods do i need to make for 600 uranium am minute?

solar pond
#

time to REALLY test my sushi belt balancing

Screenshot20210507-17453700000.png
wind spade
#

oh noes

solar pond
#

taking in inderterminate amounts of rubber, ai limiters and crystals and forcibly strangling it down to 6 belts that each contain exactly 150 crystals 15 ai limiters and 105 rubber (per minute)

#

each of the 6 out put belts should have exactly that amount of each per minute

#

in ~t h e o r y~

vivid abyss
#

lol and here I am barely getting 60 rubber per min

wind spade
#

that's like 20 oil per minute lol

viscid shadow
#

What is the solution for automating water overflow?

daring crow
#

@viscid shadow I use coal plants and sulfuric acid factory to use my excess water from Alum factory runoff..

viscid shadow
#

Too bad there isn't a way to just dump water

tardy moth
#

there is always the possibility of packaging it and sinking it

#

we just dont want to do that

viscid shadow
#

well thats a waste of plastic too isnt it

tardy moth
#

ye, exactly

viscid shadow
#

you can use steel to make packages though. So you could use the water on iron to make packaging to sink it

tardy moth
#

i just make exactly the right amount of water i need to do what i have to do

viscid shadow
#

the problem is my factory is backing up so the water stalls

tardy moth
#

fluid buffers can help, as you can flush them

#

with the whole network

viscid shadow
#

I'm going to play with valves until I make it work

#

there has to be a way to keep it from backing up

tardy moth
#

everytime I make excess water, I feed It back into the system where I'm making less water than I need, because the excess water becomes a part of the whole

#

takes like a minute longer to start but that doesnt matter

solar pond
#

well i mean if you need 1000 water in and your machines output 500 water

#

just put 500 in and use the output 500 to make up the difference

tardy moth
#

exactly

bleak coral
#

you might need to flush some pipes to give some room for the recycled water to come in though, and also if the system ever stops it'll flood and sieze up until you fix it

tardy moth
#

that's why I have some buffers to flush if I have to

bleak coral
#

I like to keep the recycling and fresh water separate, but setting it up requires some seed water in the recycling refinery

#

the only advantage to doing it this way though is that it allows the system to stop

#

which I don't do, so ๐Ÿคทโ€โ™‚๏ธ

tardy moth
#

๐Ÿคทโ€โ™‚๏ธ

normal owl
#

is it possible to balance 6 belts into 9 belts?

#

I'm really having a hard time trying to divide up my screws into 9 780 belts

#

from my 6 lines of constructors

#

I have 6 lines of constructors making 1250 screws per min and want to balance them into 9 belts of 780 with some extra left over

regal kayak
#

if you are making 1250 per line, then you already have to split the output lines to get everything out, so it cant be 6 to 9 split

normal owl
#

well the way I did the math was 60 constructors making 7500 screws which is a little over 9 full belts

#

or should I just limit each constructor line to only 780 and just add more constructors to handle the full belt of iron ingots?

regal kayak
#

well there's no point in having more than 780 production per "line" (assuming by line you mean single conveyor belt) so yes

normal owl
#

yea cause I'm trying to get my single 780 belt of iron ingots into screws with the alternate recipe

#

without it backing up

regal kayak
#

screws always compress worse (ie. take more space in belt) than their ingredients, so if you are using the cast screw recipe, then you need to split your iron ingot input line evenly into four belts, since 1 ingot => 4 screws

wind spade
#

Ideally you want to not use screws (except for rotors)

regal kayak
#

or rather just manifold and have 4 output screw belts

wind spade
#

If you still want to use screws, then make them where they are needed, don't make a central screw factory

normal owl
#

so 4 195 line of iron ingots into 4 sets of constructors with on the 4th being underclocked?

regal kayak
#

basically yes

normal owl
#

ok then I went way overboard with 60 constructors

wind spade
#

Why do you need so many screws anyway?

normal owl
#

cause I have a full belt of iron not doing anything and the only thing I really could use it for is screws cause there's nothing else for me to make

river night
#

so now you have several belts full of screws and your problem got worse? ๐Ÿ˜„

normal owl
#

just gonna send these screws to add to my reinforced plate and modular frame factory

#

cause I'm low on screws there

wind spade
#

Just don't use screws, get alts and screwscrews

normal owl
#

I know but I'm trying to make everything and normally I would avoid them but hey, I like a mind numbing migraine inducing challenge lol

rare surge
#

Just a sanity check:

Crude Oil -> Refinery(Heavy Oil Residue) -> Refinery(Residual Fuel)
Is more efficient inputwise than
Crude Oil -> Refinery(Fuel)

And the numbers are Crude->Fuel(4:3) vs Crude->HOR->Fuel(9:8)

#

(yes, I know there's diluted packed fuel; I'm not there yet)

signal nimbus
#

Using the common multiple of 36 for the input crude oil:

36:27 versus 36:32

#

It's about an 18.5% increase in total production.

full forum
#

Hey guys, i am starting my first large factory, in which i want to have multiple part production. I was wondering what method to use for the component production.

#

I am thinking of creating a lot of smaller parts and then divert them where i need them, or also i could just create separate production lines using up the exact resources needed.

any thought about it?

cold snow
#

so if you for example make a multi floor factory you have one floor that produces all the rotors for the whole factory and just split the outputs for all the other floors in the factory? sounds fine, just plan ahead how you will split it, as you otherwise might need to rely on belts filling up completly

full forum
#

yea, something like that, will keep thinking about how to plan it, thanks

uncut pine
#

hello! good idea for alt recipes for making aluminium?

glacial hemlock
#

based on the number, the Residual fuel recipe is slightly more efficient in oil consumption, but consumes a bit more power, and takes up much more space.

frosty owl
#

Just thinking about not having to use stitched plates feels like praisethesun

frosty owl
# full forum I am thinking of creating a lot of smaller parts and then divert them where i ne...

I like to divide my buildings into arrays of machines, trying to tie together as possible to have them feed one another 1:1 or 1:2 ^^
Eg: Floor 1 takes in iron ore, with arrays of smelter-constructors-assemblers to make reinforced Iron plates, plus one area
Floor 2 takes in RIPs and steel beams to make bolted frames with arrays of constructor(screws)-assemblers

I think this way the inside of buildings become much more interesting then rows of the same machine over and over again ^^

frosty owl
glacial hemlock
frosty owl
#

And the design is scalable

glacial hemlock
#

Direct feeding (or direct insertion in factorio) is my fav

glacial hemlock
#

@quiet oar pipes and belts both have throughput issue when approaching 100% usage. So it is better not to fully utilize them. That means the actual flow rate of pipe is lower than 600

quiet oar
#

ou, thats good to know, i had problem with this problem already few times and i was thinking that there must be some problem with it... curently i manage to run it somehow so everything looks good even at 100% usage but thx i will keep my eye on it ๐Ÿ™‚

frosty owl
# quiet oar ou, thats good to know, i had problem with this problem already few times and i ...

I'm not saying kwj is wrong, but the "max throughput" issues shows only when you try to split from a full belt/pipe
Eg: a full buffer can feed 600 water/min into a generator. 2 max OC extractors can too (just 1 junction with 2 sides providing the same amount)... More than that I can't say, I haven't looked much unto how merging pipes run into issues, but I think that the more junctions, the higher the chance of issues
Eg2: trying to feed a junctions manifold with a 600/min pipe will generate flow losses

quiet oar
#

in my aluminium production i have some crazy piping where i on one side producing 600m3/min water and feeding 5x rafinery for solution that go to rafiery for scrap production that create another 300m3/min of water that go back to solution production... that create a lot of connections and splitting of pipes ๐Ÿ˜„ it was nightmare to make it run but now it working... i probably underclock a bit something i dont remember now.
But for a while i had this issue on nuclear reactor and it is litterally one pipe to one reactor. But i underclock reactor and take it back to 250%, then it start working properly...
I dont know, its weird to work with piping, its better to make connection as easy as possible.

frosty owl
#

The last issue you mentioned (stopping and restarting reactor) sounds like you just needed to let the pipe fill up ^^(which is generally a good practice ๐Ÿ‘)

quiet oar
#

funny is reactor was full of water when first fuel rod come in

#

but i dont watch it whole time, i just saw it almost empty, so dont know what really happen

frosty owl
#

#Pipes-be-weird

high wave
#

i made a setup for using all of the oil on the map just for plastic and rubber

unknown.png
frosty owl
#

Must've made a mistake somewhere, max plastic should equal max rubber, @high wave

oblique hollow
#

unless you use a funny resin setup

#

and split it 50 50

#

instead of only turning the resin into rubber

frosty owl
#

There's only one way to max rubber or plastic out after all ^^

oblique hollow
#

bless recycled

high wave
#

resin is turned to rubber

#

and then some fuel is diverted to turn that to plastic

#

and the rest goes to an equal loop @frosty owl

frosty owl
#

Check your numbers with greeny's tool ;)

oblique hollow
#

ratios seem to be skewered.

#

its 50% more plastic being produced from rubber

#

so thats why its not nearly 50 / 50 here

#

17550 is what greeny outputs for rubber

#

and 17550 * 2 = 35100 (the plastic here)

frosty owl
#

That's not a 50/50 plan, that's a MAX plastic OR MAX rubber plan

oblique hollow
#

well.... if you recycle more rubber to plastic, its still max

frosty owl
#

Plastic is fine, but rubber should be the same

oblique hollow
#

just.... shifted

high wave
#

rubber doesnt have to be the same

frosty owl
#

If you max out, yes

oblique hollow
#

its max, not 50/50

uncut pine
#

so, with 800 turbofuel/m i can make 60 turbofuel generators to work? ๐Ÿ˜ฎ

oblique hollow
#

so they just recycled more rubber to plastic

oblique hollow
frosty owl
uncut pine
#

177? D:

oblique hollow
#

177.77777

#

800 / 4.5

uncut pine
#

177 at 100% or?

oblique hollow
#

yes

#

you assumed gens consume 12 turbofuel per minute

#

thats incorrect

uncut pine
#

its usefull to overclock this generators?

oblique hollow
#

...nope, its inaccurate

uncut pine
#

okay thank you!

oblique hollow
#

very hard to work with

frosty owl
#

Eh, you save space, but need to find the exact number to get the "true" 200% OC (and need 3 shards to do so)

uncut pine
#

oki^^

#

thanks for your help with that ๐Ÿ˜‹

frosty owl
#

Always welcome~

glacial hemlock
#

@high wave haha residual plastic go brrrr

high wave
#

i found my mistake

unknown.png
#

wait maybe i didnt

#

actually no i did find it

#

@frosty owl

#

first screenshot had wrong building counts

unknown.png
glacial hemlock
#

It is good to separate plastic and rubber. Unless you have plans that one won't cause the other to back up

viscid shadow
#

using the natural effect that water cant go up without a pump, valves that force one direction, and a system that needs and produces water, how would you design a system of pipes to allow constant input of water while removing output during a back up

#

The only thing I can think of is a valve at the output to prevent back up, and extra water storage feed high above the input

high wave
bronze silo
#

hey guys, I am using a pure oil node ... then 8 refineries (4 producing 10 oil residue each / 4 20 oil residue) with totals 120 per minute. After that I have 3 refineries requiring 40 per minute each for coke, yet my fluid buffers are maxing out?

viscid shadow
#

your math is wrong then, recheck everything