math-and-meta

[frosty owl]

Sounds slightly familiar 😆

[sterile kayak]

That is smart! I can just turn off One of the factories that has a huge stockpile already! Thank you!

[wind spade]

Or overflow to sink

[sterile kayak]

yeah but then the factory will keep running, and i dont want that, because it takes more power that way

[wind spade]

Then get more power 😛

[manic steeple]

Huh, I was just looking at the pinned message for guides and realized that it was posted exactly one year ago.

Neat coincidence

[wind spade]

it was reposted a few times

[frosty owl]

Unused power is the only way to effectively waste resources in the game ^^

[sterile kayak]

True, but I’m already way over my max production

My coal power plant can go into meltdown any moment now

[zealous tide]

Delete awesome tickets

[frosty owl]

Or delete anything, true that :P

[thorn bane]

sinking anything thats not efficienct for points is a waste imo

[frosty owl]

You still get something out of it, even a fraction of a ticket
If you burn fuel and don't use/store the power generates, that's more of a waste imo

[thorn bane]

more sure
but its still a waste in my opinion

[prime night]

hey i need to underclock something based on it's input, not it's output. i think i need to find the ratio of input to output

[river night]

the ratio is right in the window, and you can enter a formula in the clock speed too, so it shouldnt be too hard

[wind spade]

Nah, just do [desired input] / [normal input] * 100 and you have clock speed

[prime night]

ok ty

[robust forge]

would it be a smart idea for me to setup a spreadsheet for power useage of certain machines and parts required?

[ornate shoal]

what would you do with that information?

[robust forge]

just as a useful help thing

[ornate shoal]

hmmm in my experience power consumption is either irrelevant or ther isn't enough

i mean that i never build power to specifically satisfy some consumption number

[robust forge]

I mean just jn general other than power having jt for what parts are needed for crafting and what I need for machines themselves

[ornate shoal]

yeah that could be more useful perhaps

i just use wiki.

[brazen oxide]

Satisfactorytools (located in pins) includes power usages.

Oh wait just realized the recipes are incorrect. From Update 3 I assume?

[wind spade]

which recipes?

[brazen oxide]

Alumina solution and aluminum ingot, from the ones I checked.

Makes me think there are more?

[wind spade]

are you using U3 or U4 version of the tool?

[brazen oxide]

Gotcha, didn't see that dropdown

Had a mini heart attack that the past 24 hours of gameplay were for naught

[oak atlas]

my sadly unused U3 concept when power was variable.

my oil plant was meant to have a purge stream that sink extra oil with recycled plastic/rubber to maximize oil production efficiency.

[thorn bane]

thats exactly what i do early on when i dont have fuel generators yet

[oak atlas]

I was building it one weekend, then U4 hit the next weekend.
ended up scrapping the plan

[odd forge]

anyone in here who wanna take a look at a sheet i did for a plastic/rubber/power project, im getting some crazy numbers on ammount of fuel generators/power

imma just throw it up here, hopefully i made a mistake or i might have to do something else. 700+ fuel generators seem a bit overkill 😛 https://docs.google.com/spreadsheets/d/1H41n59GF-vKkG3qhaNK-CnbzdNWFZCPagVbtiDrEwH8/edit?usp=sharing

[river night]

fully optimized oil processing factories are very efficient. 300 oil can make you 22GW of power alone

[frosty owl]

I had such a system in U3, quite fun to run
I took away plasti/rubber and packaged fuel all made from the excess fuel, it was very fun being able to guess how much spare power I had by how much extra plastic was coming from the recycled factory
(More spare power>More overflow fuel>More rubber/plastic)

[oak atlas]

Jesus Christ your fuel has more power than my nuclear setup

[river night]

it does use 778 fuel generators though, so thats massive =p

[odd forge]

im not sure i actually wanna build this thing though, it will take ages, it will probably break my potato of a computer and getting that ammount of coal and sulfur to one spot is gonna be a pain

[river night]

probably

if you want to build fuel power plants using that recipe, i would generally recommend making one that uses 300 oil for 22GW

if you need even more power - build more then one, far away 🙂

[ornate shoal]

nuclear would definetly be easier for this kind of power

[odd forge]

might just go with alternate turbo heavy fuel, will probably make it simpler and smaller, imma do some maths

[ornate shoal]

even turbo blend fuel would make it simpler

the biggest nightmare i see here is collecting all the sulphur and coal

you basically need a map wide railway network for sulphur. or drones. need about 5 nodes of it

[odd forge]

my only problem with that recipe is having to balance fuel/heavy oil res

skipping diluted fuel and going straight to heavy turbo fuel alternate seems a bit more doable tho

[ornate shoal]

it divided quite cleanly

[odd forge]

ofc most of the sulfur nodes are on the opposite side of the map...

[thorn bane]

dont use that turbofuel recipe
it uses waaayyyy to much sulfur
like that planner you linked uses 50% of the maps sulfur while using 10% of the oil

[ornate shoal]

yeah. i think turbo blend fuel would take about 3 sulphur nodes for that kind of power and no coal but tons of more oil

[odd forge]

alright, going back to plan a, going oil>normal fuel recipe>normal turbo fuel recipe

[thorn bane]

which isnt really a problem
youre more capped trying to find sulfur than oil
also oil is mostly clustered together in packs of 4 or more nodes

[ornate shoal]

need more than 4 nodes for 100gw probably

4500 oil for 100GW of power with blended recipe

no more like 2500 actually

[odd forge]

alright, i got the turbo fuel blend balanced and it ends up with 200 generators at 30gw, seem much more comfortable and ill only need 450 sulfur

[thorn bane]

i normally recommend between 50 and 250 generator so ye thats more reasonable that 700 xD

[odd forge]

okay so i did a miscalculation on the coke, apparently this one is 400 generators...

[ornate shoal]

from 1800 oil?

[thorn bane]

with new trains and the option to set the delay so you can time trains
you can always get 780 throughput using an ISC and multiple trains per station right?
no matter the round trip time?

you just need to figure out the wait time
probably roundtrip/trains on that track

[ornate shoal]

but wouldn't the trains eventually still possibly run into each others tail and overload the station

[thorn bane]

no because it would wait untill the trains is halfway around the roundtrip (if you use 2 as example)

[ornate shoal]

how would it know where the other train is

[thorn bane]

it doesnt it just waits long enough until the other train is far enough away

[ornate shoal]

but if your goal is max throughput then waiting trains are inefficient. you might as well get along with one train that doesn't need to wait at all

it could just wait until the car is unloaded

there seems to be some other preset conditions available as well since it's a dropdown list

[thorn bane]

hm ye waiting doesnt make sense i guess

[muted crypt]

tl;dr

tag me in coach

[frosty owl]

Finding the best balance between resource efficiency and machine count

[strange urchin]

if i am trying to make rotors, and i need 5 screws for every rod, BUT rods are made at 15pm and screws are made at 40pm and i have 8 constructors making rods.. how many of said constructors would have to be connected to screw constructors to maximize output? (no overclocking rn)

[wind spade]

you can underclock screw constructors to only take 5/min, which can help you

but I'd generally just suggest to ditch screws altogether

[strange urchin]

wym ditch screws

[wind spade]

use alternate recipes that don't require screws

[strange urchin]

im not that far yet

[wind spade]

you can get alt recipes at any time

[odd forge]

unless this is your first playthrough and you enjoy the exploring, mam harddrive research you can just load your save into satisfactory calculator and learn all alternatives

so after doing several different calculations on different setups for the past couple of hours, i end up going back to square one (which i build over 50% of already) which i actually like more than any of the alternatives. #justsatisfactory things i guess

[muted crypt]

I find that I'll never want to skip the exploration aspect of getting all alts, because it feels unfair to do so

the world was designed to be explored - if it was meant to be ignored, it'd be more like factorio's map, with no emphasis on exploration

[thorn bane]

ive done it once ive seen the map
no point in doing it 7 times for all playthroughs

[wind spade]

what's your goal for plastic / rubber / turbofuel?

[odd forge]

plastic and rubber goes to turbomotor factory with like 10-20 excess of each for storage/sink and the fuel for power untill (hopefuly) im ready to go nuclear

[odd forge]

i totally agree, but after getting all harddrives possible two times, im done with that tbh 😄

[supple belfry]

The map exploration and hard drive hunting stuff is also…not very fun, mechanically? The radar systems are pretty useless, many pods require materials that require you to backtrack for…it’s just a bolted-on mechanic that needlessly lengthens out the game.

[wind spade]

nobody forces you to hunt for drives tho

[zealous tide]

I almost completely disagree haha

[swift robin]

i like the hard drive hunts because they are often in cool locations that show off the majesty of the masterfully crafted map

[versed violet]

Opinion needed.
My factory requires additional 3034 Aluminion Ingots / min for current manufacturing needs. The initial plan was to use the 4 nodes in middle of red forest (1 pure, 2 norm, 1 impure). The other pure node is used in current factory and excluded form calculations.
Currently debating 2 recipe chains:

Option 1:
SloppyAlumina + Electrode + Pure ingot, as it looks the most energy efficient and the least hassle in building count. But the 4 nodes will not be enough, and I would have to ship additional bauxite from pure nodes on left to compensate (and since I'll be building a transport anyway, I'd actually drop the impure node from center and grab remaining nodes on left instead for power efficiency).

Option 2:
The alternative is to use SloppyAlumina + Electrode + Base Alu ingot recipe, and ship a truckload of silica to pad the production. I would have to tap 1 unused quartz node + limestone pures in middle and ship them 100m vertically up onto a cliff, plus tap the limestone node in top right + probably grab some extra quartz on the bottom right as well, then tile a large area with 150 assemblers for making cheap silica.
When padding with silica, the initial 4 nodes in red forest are enough, but it looks like a lot of hassle to make all that silica and ship the quartz/limestone/silica around.

I'm leaning toward option 1 - while I may need more aluminion in future, there are still unused bauxite nodes on the map, and less buildings = less lag.
Any thoughts or tips?

[frosty owl]

I don't see any reasons to go for 2 other than being worried of lacking aluminium rather then silica/quartz

[versed violet]

I have a lot of unused quartz nodes.

[drifting plover]

red forest is spooky

but after youve cleaned the cats out you are fine

[versed violet]

There is also uranium droppings all over the place, which makes me iffy about going to three nodes on left.

[drifting plover]

yeah i was going for the uranium node in red forest and then i got jumped by like 5 elite adult cats and there was constant radiation

not a fun place

[frosty owl]

Maxing alu production with silica still leaves enough quartz for nuclear, plutonium and some more, so it's no lt unfeasible...
At the same time that means cutting down on the total oscillators one can make, so... It's all about what you wanna make more: Rockets or Assembly systems? ^^

[versed violet]

Go to waterfal. Still some spiders, but nice sulfur node just next to it, and all other useful materals on the swamp. The red forest doesn't have any copper or limestone at all

[frosty owl]

Well, the latter from a coupon perspective

[versed violet]

My coupon plan is actually turbo pressure motors.

[which require aluminium for the radio units, aluminium for the fused frames, and aluminium to bottle the damn nitrogen]

[frosty owl]

Then max that allu!

You're in for some work

Ngl, I also have a preference for motors/rockets over ADSs and similars xD

[versed violet]

I think I'll just grab all bauxite nodes in red forest and the high-up area and use the option 1. If I need more and run out of bauxite, I can always train-in some silica and replace tower of smelters with tower of foundries, yes?

[frosty owl]

If your design accomodates for it, it shouldn't be too hard

[frosty pawn]

i noticed something interesting today. the pure iron ingots recipe has 35 ore input, 65 output. this means from a single miner the max number of refineries required is 780/35=22.285714... If i build 22 refineries, the amount of ore required is 22*35=770. the total output from this is 22*65=1430.
Now for the interesting part: the full throughput of 3 mk4 belts is 1440 which is only 10 more than 1430 and I also have 10 extra ore available that isn't going to the refineries, a single smelter can do this with the default iron ingot recipe, taking overflow ore from the refinery manifold.
with 3 mk4 belts, i can feed 3 floors in a building that has 12 solid steel ingot foundries per floor, each floor would output 720 steel ingots, a single mk5 belt each. that one smelter fixes all the numbers 😄
1 pure iron node, 22 refineries, 1 smelter, 36 foundries, 3 freight platforms in the blue crater 😄

[frosty owl]

Yeah, I noticed that when making a Flexible Modular Frame setup with a pure iron node. I ended up with 2 wings of 11 refineries with a lonely small smelter between them, like a dwarf between giants ahahah

[frosty pawn]

yeah my build is too compact and i noticed that if i put a 90 degree lift on the smelter output, i cant connect a belt to the top of the lift because it encroaches on the smelter's clearance 😠
need soft clearance asap 😫

[stark bronze]

Is there a calculator capable of figuring out how many automatable items per minute can you squeeze from all nodes if they are all the same amount

[blissful copper]

where is this tool to plan buildings out?

[fierce ruin]

Check the pins

That's SCIM.

[blissful copper]

thank you, i always forget pins exist

[fierce ruin]

Greeny's tool is better for production calculations, but SCIM is very valuable for the map function

[rotund sand]

d

[zealous tide]

There's a tool someone made to layout factories, I can invite you to the disc if you're interested. Can't link it here

[frosty owl]

SaLT should be whitelisted by now, if you're mentioning that

[zealous tide]

Yeah I tried to link but it poofed

[bleak coral]

the whitelist attempts don't always seem to go through

[deft lichen]

You can ping mods about it

They don't bite

[versed violet]

"Hey guys, I've learned how to reload rifle veeery fast!"
#screenshots message

[deft lichen]

wait until you find out that each jetpack stores the fuel individually

[versed violet]

jetpacks can't be easily hotswapped midflight tho

[deft lichen]

true, but you can potentially fill your inventory with them and overcome any vertical difference

impracticality at its peak

[brazen oxide]

Honestly... might have a use for two rifles at once.

[versed violet]

<- going into red forest for couple nights. I feel justified.

[empty glade]

I have to assume you're using it for casings. Alclad casings also gives you more casings with less aluminum ingots.

[versed violet]

2k goes into Heat Fused Frame factory actually

[empty glade]

Mod?

[versed violet]

Nope, alt recipe for Fused Frame.

[ornate shoal]

how many cubes in the end?

[versed violet]

42, the answer to the universe and everything

I only need to solve the problem with aluminion delivery and build a heavy frame factory

[ornate shoal]

nice i plan for 9 with also that alt. takes more aluminum than expected

[bleak coral]

so the solution to slow reloads is just moar daka rifle?

[versed violet]

If it takes too long to replace a magazine, replace whole rifle ¯_(ツ)_/¯

[ornate shoal]

fortnite double pump intensifies

[versed violet]

After having a very big spider charging at me, I really appreciate the psychological comfort of knowing I have another clip of bullets ready

[tidal briar]

ok so

i want to split 6 constructors into 2 belts of rods, one with 36 rods/min and another one with 50 rods/min

what should i clock each one at?

https://u4.satisfactorytools.com/production?share=qgNHtdRS5JZErwhbCCEa for reference

[drifting plover]

for most times that i have to seperate it i just manifold it

[tidal briar]

whats that

[drifting plover]

basically having 1 belt line that connects to all the machines with a splitter for each machine

[bleak coral]

you just do (desired output / default output at 100%) * 100 to get the clock speed for that individual machine

[tidal briar]

yea i guess i can do that, i just feel like it would be more messy tho

well

i guess i can just split it into 2

[drifting plover]

this is kinda what i mean sorry that it sucks i have to use ms paint

[tidal briar]

and just use overflow to fix it

[tidal briar]

ik what u mean i have that for my smelter setup

[drifting plover]

that is what i mean

[tidal briar]

yea ig overflow does kinda fix the problem of the rates so i could do that

[tidal briar]

i dont know what my desired output for 1 individual machine though

if i did i would just put it in

[drifting plover]

cuz basically the rods will fill the first assembler, then go to the next, and then go to the next ones

so the first one will fill up super quick, and after like 2-3 minutes then the entire thing should be filled with the last one being slightly starved which will take another 5ish mins to be fully stable

[tidal briar]

yea ig ill do that

[drifting plover]

and what i mean by slightly starved is it will make about 3/4 cycles

[tidal briar]

it just feels more messy to connect everything to 1 path

[drifting plover]

imo it makes it cleaner

cuz otherwise you are gonna have belts curving everywhere

it makes it super compact

[tidal briar]

well i would have each constructors path going directly to the next machine is what i mean

no need for mergers or splitters

also i realized that i was wrong on the first part i need 3 lines of each which means exactly enough for each conveyor

[drifting plover]

well yes but then if one requires 3.2 constructors to be joined and another one needs 2.8 constructors of that material then it will suck

i use manifolding in pretty much every conveyor line

and i dont run into problems with them being starved

and it usually conserves a lot of space

[tidal briar]

ok so

[drifting plover]

smart mod is also really nice for building them

[tidal briar]

i have smart mod

[drifting plover]

shift scroll forward with a merger or splitter

[tidal briar]

i would actually need 2 manifold lines on the same place since im using assembler

[drifting plover]

makes it super easy

yeah it does that

it makes 2 manifolds stacked on top of each other when you go near an assembler

[bleak coral]

@livid forge btw I did the math, and you fell about ~1.132KM

max accelleration is 40m/s, acceleration is 11.85m/s^2, so about ~3.38s of acceleration for about ~67m of travel ( 1/2 * (11.85m/s^2) * (~3.38s)^2 ) plus about ~26.62s of 40m/s travel for ~1,065m

so you definitely got more than halfway up there, but had a ways to go

World size and gravity numbers on the wiki here btw: https://satisfactory.fandom.com/wiki/World#World_size

[livid forge]

thats far... dude i dare you to build a functional factory that goes to limit

like at least one production unit per floor

im using 44 belts per floor with just this tiny thing, so that'd be alot of belts 😄

[bleak coral]

yeah I can't imagine you could use the whole vertical space, it's crazy high, and like you showed stuff starts looking weird way before you get anywhere close to the limit

[bleak coral]

especially belt lifts, since those only have a max height of 51m 😳

[livid forge]

oh id not even be worried about belt limit

production on every floor man 😛 you will never get close to capping the lift lenght

[bleak coral]

true

except you'll probably have to have things skip floors to get where they need to

[livid forge]

why?

[bleak coral]

like once you've saturated one belt, then need another

or the next floor doesn't need the material from the previous floor (cause production chains are webs rather than straight shots)

[livid forge]

oh right ye you do need a couple megalifts, do the math, iron plates for instance.

30 items / min input

each constructor takes 12m in height

an mk5 ofc is 780 items / min

how many mk5 lines do you need?

say 1 constructor per floor

[bleak coral]

it's also like, say you're making reinforced iron plates, you need to take the ingots and make both of the different materials (plates + w/e), and then take both of those to the reinforced plate levels. You're gonna have some ingots skipping levels cause they're for one material and not the other, and the one material will need to go past where the other material is being made.

Just as a simple example of not ever belt goes to every floor

[bleak coral]

And also this would only take ~312m to saturate a mk5, if one constructor per floor

[livid forge]

how about this, you make a vertical 1x1 factory with smelters.

Then on top of that, you make two 1x1 factories next to each other, each producing plates and rods

then on top of that, you make an 1x1 assembler factory to make reinforced plates.

Now, you start all over, next to all of this, a similar factory to make whatever else you need with reinforced plates to get to the next stage

so you literally end up with a tree looking thing with tons of slim towers going up to where ever 😄

[bleak coral]

that sounds kinda neat lol

[fierce ruin]

2+2 = 4

[woeful onyx]

Does anyone have a 4:5 balancer design?

[magic shadow]

!wikisearch balancers

[shadow prairie]

https://satisfactory.fandom.com/wiki/Special:Search?query=balancers

[woeful onyx]

I already looked through there and th sources and there isn't a 4:5 design

[magic shadow]

there is tho
merge 4 -> 1 ... 1 -> 1/5 + 4/5 (split into 4x 1/5)
^^^^^^^^^ bit from the wiki

little messy but it works

[woeful onyx]

oh ok

Thank you

[zealous tide]

finally found the perfect HMF factory, super simple, really low building count, pretty efficient with resources

[prime night]

IBeams use 24 steel beams, and steel beams produce at 15 per minute.

I just need to google "how to find the common denominator" or something like that right?

[wind spade]

no, you can just underclock to match your ratios as you wish

[lapis compass]

was ther ea website that offered build guides in blueprint looking like images?

[wind spade]

maybe don't ask in multiple channels 🙂

[lapis compass]

figured the other channel was too busy and not likely to get a response

[wind spade]

no channels are busy at this time, nobody is talking in any of them, most people are asleep

[lapis compass]

QandA/help was chatty when i asked

[wind spade]

two people talking is hardly busy

[zealous tide]

5 for IBeams fed by 8 machines

[fierce ruin]

hi m8 - have a look at= oldshavingfoam he has heaps of blue prints available

[lapis compass]

what do you mean?

[fierce ruin]

check this = https://www.reddit.com/user/oldshavingfoam

[prime night]

hey i'm interested in learning the maths and terminology behind that though. son is playing with me and since he's into the game it's a great time to put maths into real value here.

what's it called that I'm looking for?

[wind spade]

probably least common multiple, e.g. https://www.calculatorsoup.com/calculators/math/lcm.php

however as I said, you don't really need that. The way most people calculate things is by starting at the end (e.g. "I want to produce 10 reinforced plates/min"), and calculating stuff back to raw ores. You then get results like "I need 4.82 constructors", which means you build 5 of them and underclock the last one to 82%.

You can also use some of the online tools (check pins in this channel) that would calculate stuff for you.

[zealous tide]

What greeny said, or another way with ratios, 15(made)/24(needed) =5/8 (5 machines fed by 8)

[prime night]

it's funny, i finally figured out how to get the answer seconds before you posted.

although i'm not as good at math and started with 24 divided by 15, got 1.6 as an answer. spent time thinking about how to use that information knowing that somehow it was info i needed, and eventually figured out that was saying i had some sorts of ratio of 1.6 to 1. later i figured out that i need to convert that decimal to a fraction and got 16 over 10. then simplified it to 8 over 5.

your way, makes mine look stupid complicated lol.

[prime night]

these factory games is the only time in life i had a use for knowing more than just the simple maths.

[dusky quail]

ah.... vertical manifolds.... first project i tried to design years ago

typically i do 120/240 throughput per floor cuz building up gets tiring

then depending on the in-out ratios i do two-sided in, two-sided out, or one-sided setups

so like screws typically have 2 full belts coming out of my manifold

[flat plaza]

does anyone got good designs for a 3:4 even splitter?

[livid forge]

i built it primarily for astetics 😛 i wanted three tall buildings in the distance

it was a pain though.. god it took alot of time

60/min throughput per floor

actually not even.. 2 constructors per floor

4 constructors on the wider building

[dusky quail]

yea 60 throughput is crazy, i did it cuz i wanted something scalable from the beginning without having to rebuild

that's like 1-2 buildings per floor lol

but yeah it's a nice tower aesthetic/style that i would say isn't that popular

[livid forge]

for a full 780 throughput i typically try to keep it to as few floors as possible, and rather stack different factories on top of each other, taking up more space overall

such as this brick, not very pleasing, but i plan on putting something on top to break up the shape. This building produces all of my mid tier items, from reinforced plates, to AI limiters

[dusky quail]

yeah most peeps prefer all-in-ones like that

i do local production then move stuff via trucks/trains

arguably harder but it's much more bite-sized

[livid forge]

trucks are banned on my world

[dusky quail]

well i'm happy truck stations are getting a buff

double the throughput per station is solid

[frosty owl]

There's quite a few at...

!wikisearch load balancing

[shadow prairie]

https://satisfactory.fandom.com/wiki/Special:Search?query=load balancing

[frosty owl]

!wikisearch balancer

[shadow prairie]

https://satisfactory.fandom.com/wiki/Special:Search?query=balancer

[frosty owl]

@flat plaza the second one 😆

[flat plaza]

ty!

[steady ivy]

Not a math q, more of a meta q, but I'm just getting my first oil set up, and it's ~2km from the base. I've been quite tempted by the tractors, is it worth building a simple tractor line until trains, or should I bite the bullet and make all the belts?

oh I guess this should be in #old-questions-and-help , my bad

[oblique hollow]

eh, here is ok for that too

its kinda meta

general meta is: refine your oil where you extract it. if you want, you can truck the solids back to your base

[steady ivy]

ok

[thorn bane]

ok pipeline it to your main base
nobody can stop you

[steady ivy]

that's what I was thinking, last time I brought up trucks I was advised to just wait for trains

[oblique hollow]

if you are willing to deal with trucks, go for it.

[steady ivy]

but getting trains and everything online without some kind of larger p2p transit seems painful, especially now that my initial and nearby iron is getting pretty stretched

Are trucks that bad?

[oblique hollow]

trains are kinda expensive especially if you first set up oil

and trucks can be a bit iffy to manage

[thorn bane]

its more that they will be alot better in 2 weeks xD

[steady ivy]

ah gotcha

eh, I'll give em a try, see what happens

[oblique hollow]

a tractor should suffice for now

[steady ivy]

it's just one point to point thing, I have no intention of larger distributed logistics until I get trains

[oblique hollow]

cheaty trick: once you recorded the path, remove some of the path nodes on straight segments. that makes it faster

at least for now. in 2 weeks thats gonna be fixed

[steady ivy]

ah thanks 🙂 It's from the beginner starter area to some oil 2k away (I guess the closest node?) so it's not super straight

[oblique hollow]

oh theres many beginner areas

grass hills, rocky desert, dune desert.....

[steady ivy]

ah whoops. The grass hills, the beginner area for beginners 🙂

I think. Kinda under the cliffs of the red bamboo death forest

[oblique hollow]

i guess your compass pointed you somewhere northwest?

[steady ivy]

yup

[oblique hollow]

thats the golden coast

small islands, 4 oil nodes

[steady ivy]

yeah

[oblique hollow]

neat spot to set up a Oil Refinement Plant

[steady ivy]

I'm getting set up on one and making sure I have feedback of the petro byproducts into fuel to boost my grid

[zealous tide]

Best looking* spot in the game

[steady ivy]

once that's straight, we'll see about the others and getting rubber and byproduct refinement straight

[oblique hollow]

heavy oil is really easy to deal with. unless you already unlocked fuel gens, turn it into coke

you can then burn the coke in coal generators

and since theres water aaaaaaaallll around you, that shouldnt be too hard to supply

[steady ivy]

yeah that's the current plan

uh

I'm not sure a stack of coal is enough fuel to get the tractor from station to station 🥲

does the truck get better fuel efficiency? If not I might just make a tube and carry back stacks manually until trains

[frosty owl]

Dunno, but you can check out the numbers for fuel consumption and average speed on the wiki

Something one will have much less need to do after U5

[fierce ruin]

What’s the burn per minute and how far is it?

[steady ivy]

it's a tractor, and going about 2km

good news is, I'm about to get trucks

maybe I should just wait until trucks

[frosty owl]

!wikisearch trucks

[shadow prairie]

https://satisfactory.fandom.com/wiki/Special:Search?query=trucks

[frosty owl]

Damn "s"

[steady ivy]

looks like 4s of burn time, and twice the max speed, so that would be better once it gets up to speed

trucks it is

[frosty owl]

They also drive better overall, but do require much more space to manouver

[steady ivy]

who needs space when you can have unsupported sky highways to ignore any non straight terrain?

[frosty owl]

As long as you give yourself enough space it's all good 😆

[steady ivy]

oh yeah, other important question I guess. Do trucks and tractors drive the same speed as your programmed route, or do they go as fast as possible (I assume thery might slow on turns etc) through a set of points?

[deft lichen]

afaik as fast as possible, only stopping where you stopped (a pause path node appears there)

[empty glade]

If you can wait to use coal for power until fuel power, you can make packaged fuel at your refinery, and ship back only resin for processing into plastic and rubber at your base. An alternative may be to use liquid biofuel in your tractors; I don't know if that lasts longer. Also note that you can use carapaces for more biofuel.

[faint ember]

Is there anyway to create an overflow for solid goods? Like if I want a constant consumption of rubber but my factory consumption is variable?

Any method I can think of involves a splitter mechanic, which would just tap off the main line.

[frosty owl]

Couldn't a smart splitter do the job? Specifically, by sending one of the outputs to the factory and leaving another output free for overflow

[faint ember]

Ah, I thought those were only output item filters.

[frosty owl]

They have more power than that

[dark summit]

18.2

@fierce ruin pinged so you can see it if you still want it

[livid cipher]

need help with that .1%

[blissful sky]

ya playing on capped 60 fps?

[livid cipher]

me

[blissful sky]

try unlocking to 120 in settings

[livid cipher]

i got it on Unlimited @blissful sky

need help with that overclock

tho

[bleak coral]

Do you mean the efficiency being 99%? That's probably just the bug. It can just get stuck sometimes, especially if it was on standby for a while since it's an average over time.

Reloading the save or flipping the machine off and on with the switch will get it to restart the measurement and fix it usually

[livid cipher]

I need the intake to be 50 per minute exactly

[frosty pawn]

default input is 30/min and you want it to be 50/min. output is 20/min. under "target production rate" click the orange numbers and delete them, type 20*(50/30)

when you press enter it will work it out

[wind spade]

you can't get it exactly

you'll just get very close

[tough stratus]

If I wanna split a single input 4 ways on a 2:2:2:1 ratio, do I just do a normal 4 way split (split in half then split each of those in half), but then add an additional splitter to the end that gets less with a belt merging back with the initial input before the first split?

[zealous tide]

Try 166.667, might work idk

[wind spade]

either that or just do a manifold

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

[tough stratus]

Trying to avoid manifold here since it's radioactive and I'm trying to avoid stacks building up

[wind spade]

radiation of a few items isn't too bad anyway, unless you're building it in the middle of your base

[muted crypt]

hey gamers, here to present a not-so-daily tidbit of random information you didn't ask for: these two wrecks are effectively the exact same, with the same opening requirements and the same loot on the ground

(This is the dunes desert cliffs area, in the northwest of the biome)

bonus tidbit: the looted wreck near the bottom is the exact same as the other two as well, save for missing one ground collectable of ten AI Limiters

This has been your "why did you investigate this?" moment with maroon

[trail sail]

In the per minute type 50 and as long as you have enough power shards it will go to 50 per minute

You have to do that with the furnaces

It might not work idk but it is just .1% it shouldn’t hurt anything

[muted crypt]

update: I found another wreck that has everything that these other three do with the exception of the 10 AI Limiters again lol

[deft lichen]

all of the U4 crash sites are obviously rushed, they all look really similar and have no enemy encounters or other hazards near them

same goes for new resource nodes as well as all resource well

[bleak coral]

Wdym about the new nodes and resource wells? I haven't looked too closely at them. Unless you mean how the water wells all kinda suck.

[deft lichen]

resource nodes and drop pods often have mobs guarding them

but none of the resource wells do

the north coast oil nodes have alpha spitters and crab hatchers all over, yet the oil resource wells... nothing

[bleak coral]

Ah I see, guess they'll add them later, a good number of them are in zones that are set to be reworked (that area between swamp and blue crater)

[ornate shoal]

i thought only desert and northern forest were to be reworked

[oblique hollow]

yes for now

but later on spire coast gets completely reworked

(or so it seems)

[deft lichen]

note that while northern forest looks completely different, it's two orders of "completely different" less than the spire coast rework

[ornate shoal]

wtf have i missed some devstream

[deft lichen]

not a devstream, a video

[oblique hollow]

check their YT channel

[deft lichen]

https://www.youtube.com/watch?v=MxvGbCCc6DI

[ornate shoal]

i watched that video, it was just a map with desert and nothern forest highlighted

[deft lichen]

you didn't watch it whole then

they showed 3 maps in total, one for U5, one for U5+until 1.0, one for after-U5-until-1.0

[oblique hollow]

this is the first map

and this the second

third map just changes the desert and forest color to green

indicating "very little future changes"

[versed violet]

Oh my, they plan to change landscape in red jungle?
Guess my aluminium factory location has been decided. [red forest with nothing in red jungle.]

[ornate shoal]

i think swamp is perfect now

i have most my factory in swamp and titan forest but guess i'll start a new save anyway

[bleak coral]

Swamp has that water with a weird texture, and there's a decent amount of foliage that's floating or in the terrain. Definitely not finished.

[frosty owl]

I'm trying to fit the silica assemblers necessary to smelt alu ingots right above the alu foundries and oh boy is the design getting tight

3 rows of 4 assemblers, about 3 foundations per row for 10 total aviable.
I think I can still fit all the splitters on ground level though

[hidden scarab]

O_O das all?

[frosty owl]

That's to cover a single 750/min silica belt to process 900/min scraps in a footprint of the same width as the refineries and the same lenght as the 4x2 foundries placed in front of the refineries (in the same width)

That's 3 scraps refineries

[hidden scarab]

Build bigger D: jk compact design i c ;o

[ornate shoal]

if yo wanted to be truly space efficient, you would drop silica

[frosty owl]

I considered it, but I do need that alu :/

It's also becoming quite the fun challenge in and of itself xD

[ornate shoal]

i'm sure it's challenge. all my alu setups are coke sloppy and pure ingots

[faint ember]

The most efficient aluminum recipes are soooo space demanding.

Is the blender/sulfur recipe worth it?

[hidden scarab]

I dislike the pure aluminum ingot recipe but people really love it, it seems

[frosty owl]

Eh.... If you really got the resources to spare for it, it surely makes the plumbing simpler

[hidden scarab]

Instant scrap? I love it

[ornate shoal]

takes 60 scrap/min so loadbalanced manifolds

[lapis compass]

that is exactly what i was asking about thank you!

[bleak coral]

The big problem with instant scrap is since it's one of the 100% recycling loops it's particularly susceptible to the load bug

Though I guess you can delay it a lot with an industrial buffer

[ornate shoal]

recycling loops, my favorite things in this game

[versed violet]

My personal supply wagon of the building train (for storing 'I ran out of X' mats). Anything you would add/replace here? Concrete/plates/rods are in other wagons.

[frosty owl]

VIP the water, ez
VIP the acid too for extra OCD

[bleak coral]

Instant loop doesn't need outside water, though yeah you can add extractor too I guess

oh wait I forgot it takes water too

I'm thinking of the acid part

theoretically, you can not need outside water for the acid, just use the recycled, that's what I was thinking of

[frosty owl]

If you VIP either the water or acid you at least have a system that won't fail ever, regardless of the fluid bug

[bleak coral]

yeah that's probably best for now, but makes me sad 😦
the closed loop part with acid is what makes instant neat for me

[frosty owl]

Yeah, but luckily a single VIP as fix isn't too huge a thing

If one even cares

[ornate shoal]

you still need to overproduce water and acid to compensate for fluid buffer bug

[oblique hollow]

its not the fluid buffer thats at fault

its all machine connectors

[ornate shoal]

yeah, that's what i meant

[versed violet]

Assuming the slowest train can go on a track as 60 km/h, does that mean it can do a 1km in map distance per minute?
Attempting to establish how long this route will take - 5 stops total, one at each bauxite node and one at unloading station.

[wind spade]

I'd bring 2-4 to one station to save train trip time

[versed violet]

Does that also mean single cargo platform, or I will still need a platform per miner?

[faint ember]

#screenshots message

So, every time the train unloads, there's a huge delay where the conveyors do not load off the train station

Meaning you can't actually do a full 780/min

This is working for now, with two stations receiving/supplying the resources, both dumping into a single storage crate

So there's a constant 780/min output

[versed violet]

I always put ISC with two belts connected as a buffer, and assume 1 platform = 1 mk5 belt + a bit.

[faint ember]

Fair enough.

Two platforms is prolly overkill.

[versed violet]

I assume it is not physically possible to transport 3x780+600 Bauxite with two wagons?

[bleak coral]

sure, just as long as you don't try to do 2x780 in one wagon

just gotta split it up so it's <1560 in any single wagon

also the train is gonna need to have a pretty short round trip time

It'd have to be less than 2 minutes and ~10.6 seconds round trip to keep up that throughput with just two wagons......

[versed violet]

My math says if I can stretch the train roundtrip to 7 minutes, the final stations would manage to unload cargo in time, but...
as you say, the stations will fill before train returns again, so nope, doesn't seem possible.

Actually wondering, if I can have the train return fast enough if I use 3 platforms

[bleak coral]

I mean you could just use the formula on the wiki, that's all I did

[oak atlas]

So it turns out my clustering example can ignore Z-coordinate if I build train station high up.

here's one way to have a train station setup, which should reduce run time of my world harvesting train (WIP).
So two regional train collect from local cluster (black cross) and unload at regional cluster (pink cross), then the world harvesting train just went to 1 place to load everything to my megafactory.

It do also means that everything is built in the sky though, even my local cluster factory to process ore into ingots are built at least 1-2 foundation above ground.

I can see that this will be a problem later when it comes to parts requiring water.

[frosty pawn]

"when it comes to parts requiring water" cant you just load up a train with water? 😄

[frosty owl]

Fluid trains, everyone's love
I'm curious to see if U5 will help with that too

[frosty owl]

Turns out the total will be 7200 bauxite

[hidden scarab]

Das not bad

[frosty owl]

I might "silica" half of it

[hidden scarab]

Atleast u haven’t used all of it

[frosty owl]

Enough to play with however I want xD

[hidden scarab]

Yeap and bauxite has a lot of uses for not using a lot of it

[frosty owl]

What di you mean?

[hidden scarab]

Bauxite can make a lot of stuff with little of it

[frosty owl]

Oh yeah, no kidding. I'll probably be exporting most of the ingots (or sheets/casings)

[hidden scarab]

Yeah, I need to finish my main pc so I can get back on that grind

[sharp drum]

I'm going to start getting into trains soon and was wondering what ratios people use for trains? 2 heads, 4 wagons?

[thorn bane]

1:4

[gloomy palm]

can a walkway be considered the same as a footbridge? like a pedestrian overpass...
is it usual to find players using walkways flush with the ground?

[frosty owl]

Can't the secondary one handle SF??

[frosty owl]

I'd say "Yes, but no" but U5 will probably change my opinion

[hidden scarab]

It can but I’m focused on building my main pc and didn’t bother send the save over to the 2nd pc

[onyx plinth]

I always slide-jump into the Hyper Tube Entrace
What other ways (aside from the Hyper Tube Cannon of course) are there to increase speed on the Hyper Tubes?

[thorn bane]

multiple entrances
mk5 belt into it

[onyx plinth]

multiple entrances would be the cannon, but a high speed belt is not a bad idea, I like

[thorn bane]

you dont have to make a cannon you can just use multiple entrances into a long tube so you travel faster

*with cannon i mean shoot you in the air

https://www.reddit.com/r/SatisfactoryGame/comments/jo1n7u/from_one_side_of_the_map_to_the_other_in_eight/
something like this

[onyx plinth]

gotcha

that is fast

[frosty owl]

You can still slide-jump on the belt too

I think that's about as fast as 3~5 entrances

[frosty owl]

In the end, 4 rows it is. Compact, but I still want it to be walkable (belts won't go higher then ground level)

@thorn bane Can you imagine the beltwork? 😏

(Cheap Silica)

[thorn bane]

imagine not having manifolds until you cant even see that far

[frosty owl]

Oh, did I mention I still plan to load balance them and not have each assembler connected to lifts? :hehe:

[thorn bane]

oh god

[frosty owl]

@thorn bane @hidden scarab @bleak coral (Lund as I think you like this sort of weird things/designs xD)
This is the belting, 600 Limestone + 360 Raw Quartz in, 750+Overflow Silica out. There's even an overflow connection for the sushi, connecting with the silica output to be filtered out later (this took some clever alignment)
Inputs come in the "uderfloor" (inside the 4m foundations' floor) from the left, opposite of output

How it looks while at work. About 30s before 100% efficiency

[thorn bane]

damn i really gotta do more sushi factories but i end up building a bus all the time xD

[bleak coral]

Oh the mixed belts are driving me wild. It's a visual, perfectionist nightmare

shiny shiny shiny, dull dull, small small, medium!

[frosty owl]

The difference mks on the belts are a must of the sushi though, not a personal preference

[bleak coral]

Oh yeah I figured that, but the visual result is certainly...... unique

how does that limestone overflow work? Is that a programmable splitter? Wouldn't an overflow + limestone filter still send down quartz if it couldn't go anywhere else?

[frosty owl]

They are all programmable splitters set to (limestone+quartz) on 2 sides and Overflow on the other.
Honestly, it wouldn't really prevent any jams, but it would help clearing up the belt in case a jam did happen (this normally happens during testing, the point of the system is to run perfectly as long as you don't F up the inputs)

[bleak coral]

so more of an emergency valve than anything else, like hook up a temporary sink and let it run till you just see silica

[frosty owl]

Kinda, yes. Since the silica production would stop if a jam happened anyway, the output belts would actually be free for overflow whenever a jam happens. Then I can cut the power and solve the jam (clear machines' inventories), without having to manually clear the belts

[upbeat tide]

Its better IMO to keep your inputs and outputs seperated and not use programmables to mix n match belts if im understanding

[frosty owl]

I couldn't find a way to have the items separated without having the belts in more than 1 level (stack them a over the floor)

The underfloor is just 4m high, so can't accommodate for 2 belts balancing

Well, not without mixing some stuff, that is

[upbeat tide]

Yea thats why I avoid those situations 🙂

Like this is my silica production center

[thorn bane]

oh manifolds for days
i love it

[frosty owl]

I'd rather not have to wait over 10 minutes (or even 1, really) to see if a simple silica setup reaches max efficiency or not 😅
That makes manifold out of the question, unless I want to also add buffers to help prefill, increasing the footprint by a lot

For context, that silica setup has the same footprint as the foundries needing the silica (floor below)

[upbeat tide]

Issue is cheap silica is a massive setup no matter how you do it. Manifolds work best in this game befause of input/output limits. Anyway this example is its own standalone factory structure apart from what it feeds.

Balancing in general using programmables is almost always going to end in a jam eventually.

The only time I do it is with fluids like oil setups or aluminum, the like. Stuff that has byproducts and simply can never stop or deal with jams.

[frosty owl]

If balancing jams, it just means somebody didn't plan carefully enough :P
BTW, "manifold work best" as in...? I can agree it can make building simpler, but pure belts manifolds are arguibly as good as mixed ones when it comes to number of belts (thus FPS, quite important late game or in big builds)

[bleak coral]

unmixed manifolds are the fool-proof solution, but you are no fool

[frosty owl]

I'm a fool squared

[upbeat tide]

Big builds is all I do. I just have found too many situations where mixing a single belt leads to some error or another.

Like thos is my latest, not even fully ready to online yet.

Non-Fissle uranium

[bleak coral]

yum yum spicy candy bar

[frosty owl]

I'm not trying to be offensive 😅 , but I'm lead to believe that those issues came from a misunderstanding of the smart/programmable splitters use or a small mistake somewhere (the stereotypical MK1 belt choking throughput or similar), rather than the fault of mixing belts itself (which can be perfectly safe, even taking game's bugs and imprecisions into account)

[upbeat tide]

Maybe. Not saying its impossible, cant be done, etc.

[frosty owl]

"Almost always going to fail" sounded kinda definitive xD
On the other hand, I'm not saying mixing is easy and should be approached nilly-willy, but it's no pioneer-eating monster either~

[thorn bane]

well both achieve the same result
manifolds are way easier

[frosty owl]

Load-balancing is faster at getting efficient though, reee

[frosty pawn]

this is really cool but i will literally never do it. i can't stand the idea that i would need to ever worry that i got some numbers wrong, plus mixed belts are yucky to me 😛

[hidden scarab]

OC'ing assemblers ;o

[frosty pawn]

in a lot of cases i do load balancing (unbalancing?) mixed with manifolds, like when i have 12 machines it's very easy to group them into threes and then manifold the 4 groups

here is one such mess

needed 14 machines, didnt want to fill the building with splitters and mergers

[sand epoch]

You see that as a mess? o0

[frosty owl]

You can get numbers wrong on normal belts too 😆
What is that you find yucky about mixed belts though?

[frosty owl]

What would you OC? ^^

[hidden scarab]

mines o3o

[hidden scarab]

u cheating in OC shards?

[frosty owl]

... No

[hidden scarab]

_> woooooooooooooooooooooooow

[fierce ruin]

me and my mate want to start experimenting with sushi belts, any tips or recomendations?

[dull bolt]

Expect pain and suffering.

[fierce ruin]

well that was a given

[frosty owl]

Be sure to know what you're doing or always have a connection to the sink ^^
A central storage is great to have more freedom with how you merge stuff (since it's gonna take care of the sorting anyway)
Try not to keep too many "loose ends". Example of a "loose end": you mix items in a heavy Modular frames setup on multiple belts. All the items are consumed and you output a belt with HMF + overflow, making this a "loose end" until you properly dispose of it (reach next processing step or storage/sink)

[fierce ruin]

Ohhh

[frosty owl]

Thats what I could think of the top of my head

[fierce ruin]

thanks dude!

[frosty owl]

Most welcome~
Feel free to at for any doubts, there can be some trickeries to smart/programmable splitting or sushi feeding (if one gets into the latter, extra special care is needed)

[frosty pawn]

i like to know roughly how healthy my production line is by quickly glancing at a belt and seeing if it's continuously moving items. for some reason i find it unsettling to rely on splitters to do what we know they do but is not written in any official documentation.

[frosty owl]

I can't agree more about the moving items, it's great rather than having to check machines individually
Only having to just check the power draw can rival that for me

[frosty owl]

Btw, what kind of documentation are you referring to? Wiki or CS stuff...?

[frosty pawn]

something from CS to state exactly what a splitter does. we know that when an item is received at the input it outputs on the next available belt connection, in order, one at a time like 1,2,3,1,2,3,1,2,3... and if any belt cannot take an item it does the next one, so when belts are different speeds, that's how they get more items on the faster belts. also there is a cache of 8 inside the splitter. i dont know where this information came from, so i don't want to take it as gospel and spend hours building a factory based on it :S

[fierce ruin]

not sure because I'm a little bit of a newbie by couldn't you just treat it as a perfect split between belts? or is that not how it works?

[frosty owl]

I trust this sort of stuff only after I made sure it works the way I think it does through testing~
Ie: before saying that splitting sushi could be done reliably I had such a system run for tens of hours in different FPS scenarios (and by different people)
After all, I too dont want to build a factory inky to see it crash and burn, which is part of why I don't mind making parts of it particularly easy to tell if something isn't working exactly as expected as a... Proof of sorts 😆

[frosty pawn]

if i think about how to write the code i would have an array of outputs numbered 1, 2, 3. output belt ticks as fast as the belt speed (mk1 belt would tick at 1min/60, mk5 at 1/780) every tick the belt would tell the splitter "i'm ready" and the splitter would see if the number is in the array; if not, add to the end of the array. every time the input belt ticks "i'm ready" because an item has reached the end of the belt, the splitter would check the array and if it's empty, check the cache, if it's less than 8, take the item from the input belt and add it to the cache. if the cache is full, stop the input belt until the next time an output belt is ready. then it would look at the array and if it's not empty, move the first item in the cache to the belt that is represented by the first number in the array and remove that number from the array.

[frosty owl]

Afaik, you can do that as long as none of the outputs back up. Else other rules come into play~

[frosty pawn]

it's never actually mathematically a perfect split because one item going in doesnt make 1/3 of an item going out of each output

[fierce ruin]

"other rules" FEAR, WORRY EVEN

[frosty pawn]

i probably missed some stuff because it's all text, but i could probably write it in code pretty easily

[frosty owl]

It is a perfect split between certain parameters, though, don't you think?
Eg: if the output isn't backing up, there is never uncertainty on who will receive the next item

[frosty pawn]

while that is true, i dont want to be concerned with what order the items go to which belts, i just want my 30 iron ore to make 20 iron plates 😛

[frosty owl]

Mostly behaviors you can control via smart splitters luckily :P
Hopefully there'll be a comprehensive guide about such behaviors soon™️...

Until then, I'll gladly help if any question arises~

[frosty owl]

Well, I concerned myself with such an order, and found that not even saving, loading or passing the save to others could break such order (or if it did, it was to a truly insignificant degree ||like tens of milions of minutes before any error could pile up enough||)
So I trust sushi to a... Good degree by now, I'd say

[frosty pawn]

so from your experience it seems the splitters' state, including which output is next, is saved in the game save file

[frosty owl]

It's just one more tool in the arsenal, and can be a very powerful/convenient one

[frosty owl]

Even a "dumb split" made with a normal splitter to filer a limestone:copper belt survived loading the save (despite it being prone to failure)

In a similar scenario, a smart splitter avoids the uncertainty completely (avoidable anyway by using higher tier mk belts for the outputs) by filtering the items directly

[frosty pawn]

also, on the topic of save file, i think the way the game is saved is like when a save is triggered the game collects all the information needed to be saved and then dumps it in a file... personally i would have the save data updating a save object continuously whenever something changes. any time we need any data during gameplay we fetch it from the save object rather than from the belts and buildings etc, so when we need to save, we pause the ghame and duplicate the save object in memory because that should be really quick, then we can unpause it earlier than otherwise and asynchronously save the duplicated object to file while the original is back to being used in gameplay

[frosty owl]

Sounds RAM-unfriendly and #off-topic-tech, ngl

[frosty pawn]

yeah but it's not textures and sounds, just a big ol' blob of numbers lol

[frosty owl]

I'm unsure how big a ~5MB save gets when in RAM, but I doubt that's an insignificant amount of overhead

[frosty pawn]

that's how databases work because its fast AF

i assume the save file on disk is probably bigger than the raw data in memory because all the data needs to be decorated in a way that the file can be validated on load

[frosty owl]

I assumed the save got compressed, thus taking more space in RAM...
But whichever the case, I can only attest for the results of my experience. Anything else is just me guessing

[frosty pawn]

me too. i'm sure asynchronous (or at least faster) saving is on the roadmap somewhere

[hidden scarab]

download more ram

[wind spade]

Reading a big save file in realtime to load stuff is weird, not to mention that in sf everything is loaded

[frosty pawn]

the game never needs to read the file in realtime, but reading all the things from memory in a centralised location rather than decentralised locations would mean not having to fetch and reorganise all the data when it's time to save

[wind spade]

when does the game need to read something from a save file other than when loading a game?

[frosty pawn]

it doesnt

[wind spade]

so what's the point of what you're suggesting?

[frosty pawn]

in unreal there is a save object. its purpose is to contain all the data that needs to be written to a file and read from a file. generally when a developer uses this object they write a function that collects all the bits of data from all the relevant objects in the game and organises them, and then writes itself to a file.
what i'm suggesting is that all the data from all the objects is stored in the save object already, so that step can be skipped.

[wind spade]

not really

(not to mention that the step you want to skip doesn't take too much time anyway, longest is IO)

[frosty pawn]

if you duplicate the object and do the IO stuff on a separate thread, that long time thing can happen in parallel to the game running

[wind spade]

you can do that by forking the process that runs the game without having data in one object

[frosty pawn]

ok, sure, so dont store everything in the save object. but i still want asynchronous saving 😦

[wind spade]

I don't think the benefit it gives is enough to outweight the massive amount of time needed to make it work

at least now when the game is in EA

[frosty pawn]

there is already a massive amount of time going into dedicated servers and i think part of it is rearchitecting the save function so the server doesnt hold up the clients too much or freeze up the entire server

just speculating

[empty glade]

So they should be able to do a snapshot save, eliminating the pause when saving?

[mighty bolt]

i cannot wait for dedicated servers

its the main thing keeping me and my friends from playing

i have a nice ryzen server with extra space left over.. it better have linux support 😄

[deft lichen]

Multiplayer needs to be fixed up in general so the experience is actually bearable

[drifting plover]

Yeah multiplayer has some super annoying bugs like not being able to pick up leaves that respawn when you rejoin

So theres just bushes or trees in ur factory

[oak atlas]

https://tenor.com/view/modern-problems-solutions-dave-chapelle-gif-13450651

Don't have friends = no multiplayer problems

[empty glade]

It's only for certain players though; the host can't see them because they're not actually there, just showing up for the guests (either only that guest or all guests I don't know).

[wicked tinsel]

trees regrowing for clients is like smallest of pile of multiplayer bugs

hopefully at least some will get addressed in u5

[empty glade]

How about autosave desynch while falling causing a guest to fall through the map?

[wicked tinsel]

well, shit happens 😄

autosave also disconnects you if you are inside train

moving toward largeish factories cause massive lag spike that may send you into void if you are in vehicle

everything rubberbands af

because there are no priorities on entity streaming, if you are running inside a hypertube its possible it will fail to load when approaching large factory

[empty glade]

I hope they fixed a lot of multiplayer bugs in U5

[versed violet]

Layout advice needed.
I need to build 4 of the bauxite refining modules like one on picture. Second one will be on the platfrom on left, but debating where to put another 2.
On top of existing ones? I dislike clipping, so it would have to go quite far up (refineries are tall) + need to find some place for supports.
Or should I just build 2 more platforms to the left and deforest some more of the forest?
Any obvious advantages or disadvantages?

[frosty pawn]

yes. maybe not completely, but at least mostly.

[sweet flower]

can a mk 1 belt handle 60 parts per minute

[limpid lance]

Is there any use for drones? I'm running everything on trains, and i don't see anything from long-range that I don't already have hooked up with my train (it's close enough.) So is there even a use for me to use them? They seem cool and I wanna try it out, but at the same time they seem pointless.

[fierce cypress]

yes

thats its max throughput

[limpid lance]

Also, what about geothermal generators? It's hard to find on how much they cost (the building itself, not the research). I'm saving up supercomputers for the research however, because 600mw from a single pure-quality water geyser node seems really good.

[fierce cypress]

https://satisfactory.fandom.com/wiki/Geothermal_Generator

[limpid lance]

Oh, i must've missed this part. Thanks.

[frosty owl]

What if...
You make the second module with some distance from the first (say 3 free foundations)
Then make the third and fourth module on the second floor, having the middle free space used to place supports for the next floor and logistic between floors (belts, hypertubes, pipes, ladders...)

[versed violet]

That was the idea, but too late, I've already cut a piece of forest to accomodate modules 3 & 4 🪓 . There seems to be plenty of space without stacking tho, and farther left I go, the shorter my bauxite delivery lines will be 😄

[limpid lance]

Does anyone know the amount of water you need if you feed water from the alumina/aluminium process back into the part that needs water? nothing is overclocked or underclocked

i mean like, the amount of extra water from an external pump

[frosty owl]

That's (water consumed)-(byproduct water produced)

[limpid lance]

yeah, ofc, im just wondering if anyone else has figured it out yet because im too lazy to calculate lol

[frosty owl]

Could it help if I told you you can open up a calculator in-game by pressing N? ^^

Just chuck in the numbers you see when you interact with the machines and set the extractor/s to that

[stark bronze]

How do they manage byproduct merging without valves
Was priority junction necessary back then?

[frosty owl]

If you have the numbers right and connect everything properly the system can clog only if the refineries stop while the extractors keep going (eg: full with alu scraps, thus also full with solution)

So while not impossible to run, it's pretty delicate and relies on continuous input-output

[thorn bane]

well you just go miner into refinery and have an overflow sink
but the thing is the system can still break

[stark bronze]

A valve seem to solve all the issue but you reminded me i need to go underclock water

[sand epoch]

I just feed the water byproduct into wet concrete and sink ..o0

[pastel gulch]

Then you feed too much and AAAAAAh it all broke.

It seems like anything fluid-related works as long as it's not...big.

Like mk1 everything. Scaled to the lowest denomiator. (impure node, lowest mk., lowest mk. pipe, whatever will fit on those as long as there's not too many)

(wrong channel, sorry)

[high wagon]

Just a philosophical sorta question... Is there any reason to use the refinery recipe for iron ingots over just... making more smelters? I wanted to have one main iron building that would service the rest of the world, but i feel like going full on into massive refinery builds for just iron maybe overdoing it?

Asking in here because i figure the meta heads would know if it's worth it or not.

[wind spade]

yes, running out of iron 🙂

power is nothing in lategame with nuclear builds, so it's a nice boost of iron for free (just add water)

[high wagon]

I'm a little before that point, but i just wanted to get the iron build out of the way so i could move on. Blegh. Back to laying down refineries then.

[wind spade]

you can always just do smelters

it's basically a decision between "build smelters and carry iron from far away" and "build refineries but use less iron so probably just the nearby one"

[high wagon]

But then i'll have to fix it later and i'm to lazy to tear down something that works. x3

[wind spade]

you can keep the factory there, don't have to rebuild it

[sterile kayak]

right, im going to start a turbofuel adventure soon, and i am doing some research. there seems to be 2 ways to make it, directly from HOR, or refine the HOR into fuel, and then make turbofuel out of it. wich one is better? i want like 70/100 turbofuel gens i guess, but im debating wich option to choose.

[ornate shoal]

the second way is a bit better

but this is if you have a diluted fuel recipe

[frosty owl]

That's because the bigger the scale, the easier it is for you to make a mistake in building the system :P

[wicked tinsel]

one of major problems with pipe systems is that you cant see if some part hadnt connected properly

and as you increase amount of pipes, such bad spots probability increases too :<

its a problem with rails too i guess

there should be some prop rendered at end of terminated pipe/rail segment to make it visible something went wrong

[molten ice]

check this
#old-questions-and-help message

[frosty owl]

It's the same as belts: when stuff doesn't go through, you must've F up the connection

I always go through big pipes system to check for those "fake" connections after having some fluid in the pipe. It's a must 🤷‍♂️

[wicked tinsel]

yeah, but with belts, at least you see where its broken

with pipes, its pretty damn painful

[frosty owl]

Same with pipes, there's a fluid indicator

[wicked tinsel]

you need to inspect each pipe segment manually

and then there can be a height issue too instead

[frosty owl]

You don't need to interact with them, just check if the flow indicator is full or not. If not, pipe is not connected

[frosty owl]

That's completely on you :P

Funfact: out of ~110 refineries' inputs for my latest factory, I had to fix ~10 borked input pipes

[jaunty smelt]

and one able to help in the best way to make the most plutonium out of nuclear waste

[oblique hollow]

instant cells i think

[jaunty smelt]

ok thanks ill look into that

[frosty owl]

If you're willing to use up some uranium too, the first step could be fertile uranium
Else, it goes: Non-fissile Uranium -> Instant Cells -> Plutonium Fuel Unit (the one with the pressure conversion cubes)

[bleak coral]

fertile uranium makes more plutonium, but if you're using both of the uranium alts you get less power out of it

[jaunty smelt]

ok thank you all

[steady ivy]

kinda a meta question, what's the train-car ration people are running?

factorio I usually did standardized 1-2 or 1-3, but there's no inclines and it's a different game

[stray willow]

Four cars per locomotive is almost always safe to go with

Five or six if your network is relatively flat, but personally I say do four so you don’t have to worry about it

[steady ivy]

ok cool

other question is about logistics structure as an overview. Should I look at more point to point production, or should I do distribution hubs. I've got a picture

values are per minute, oops

[stray willow]

I think the game lends itself better to point-to-point logistics, but both methods can be made very efficient if you know what you’re doing

[steady ivy]

I see

[stray willow]

Do what makes sense to you

[steady ivy]

I think in the beginning it will be the left, but might shift to the right later on, I think it depends on how centralized I decide to go for first round ore refining

shipping screws for example doesn't seem worth, so those will be on site, but centralized smelting (especially with alts) might be worth it down the line, and definitely worth for coal/steel

[stray willow]

Centralized smelting can make life easier, but I prefer to smelt on-site and ship ingots around, especially with steel. There are a handful of locations on the map where coal and iron exist in close proximity that are my go-to for steel

[steady ivy]

that's what I've got now, and I have to admit that shipping ingots around seems really appealing

at least to cut down on the stuff I have to build per factory

[stray willow]

It’s also worth mentioning that resources that don’t smelt 1:1 (e.g. caterium, steel) have better throughput per train car/truck/belt in ingot form since there are fewer items to move

[frosty owl]

I think it's far more convenient to use the left one for low volume productions (usually low-tier ones), and try using the right one only for the final steps of processing, the lowest volume ones (usually manifacturers and such)

[sour wyvern]

at that point you might as well make plates/pipes on site too.

[stray willow]

I often do, if that's all I need the steel for

[sour wyvern]

that's what I do - an entire coal/iron setup for plates, for example

[stray willow]

My spaghetti logistics skills often don't allow for that level of pre-planning though

[sour wyvern]

coke steel is better though

coal has more uses

[stray willow]

depends on what's nearby, I don't often train in coke from some far-off oil plant if there's coal a stone's throw away. I don't build to the scale that the map supply of coal is a problem

[sour wyvern]

see I do blocks, so I'll do a massive block of X - say wire, or cable, or quickwire

then route between them

[stray willow]

hm, that makes sense

[sour wyvern]

#screenshots message like that

[stray willow]

my struggle with doing it like that is managing how much of what I'm producing versus using

How do you keep track?

[sour wyvern]

I generally work it out when I want to do something - and I try to have extra. I don't keep track of it all as such, I do a flyover now and then to check status

[stray willow]

fair enough, ill give it a try sometime

[bleak coral]

like what? the opposite is why I usually say solid steel eeks out a little over coke steel: coal has little use outside of making steel

coke steel is fine too, there's plenty of oil, just no reason to save coal

[sour wyvern]

I needed coal for.... something else.

[bleak coral]

well that's ominous

[sour wyvern]

Aluminium.

[bleak coral]

oh, well electrode is better than the regular recipe, and instant is bottlenecked by sulfur hard, and they're all bottlenecked by bauxite anyway

[sour wyvern]

yeah - I mean, as per all of these, it was really that I had coke available, and more coal was a significant complication.

so I made it work

plus I haven't found solid steel yet. Need to do some more exploration

[bleak coral]

yeah that makes sense, like I said coke steel is just fine, but as a rule of thumb there's not really a reason to save coal for... well... anything

none of its uses really strain it

[sour wyvern]

My issue is I had one coal node for two bauxite nodes

because all the others in the area were already used for steel production

[bleak coral]

you don't gotta justify why you did what you did, was just disputing the coke steel is better cause you need to save coal part, which isn't generally true

[thorn bane]

electrode scrap uses oil to save on bauxite
coke steel uses oil to save on iron
but you have alot of iron and only limited amounts of bauxite so most people use oil for bauxite and coal for steel

[bleak coral]

I mean honestly there's enough coal and oil to use them for both, you're bottlenecked by the bauxite in the end anyway

[frosty owl]

You can always chuck down a few beacons to mark your factory's input and/or output after you finish it or whenever you rework it

[versed violet]

[whispers] X3-Signs mod [/whispers]

[frosty owl]

The future is still a week from now

[crude coyote]

https://tenor.com/view/hack-the-planet-gif-15624885

[gloomy palm]

Did they say if there will be signs in U5?

Signs are cool 🙂

[river night]

they did not

[gloomy palm]

🙂 oki

maybe in U6

[crude coyote]

yesterday's stream suggested it might be coming (edit: no express confirmation though, but I felt it hinted strongly that it was confirmed and they did not expressly deny it like they do other requested features - ALSO it gels well with the QOL focus of Update 5)

[gloomy palm]

Oooo

[dark summit]

we had drones in update 4, we get signs in update 5 as the end reveal

[empty glade]

They did say there are many things in U5 they will not cover before the launch.

[gloomy palm]

Hmmm

[sour wyvern]

I am excited for update 5. to celebrate I bought myself a treat - a satisfactory pioneering statue

oh god. after 362 hours of game time

[sour wyvern]

coke doesn't save on iron, it saves on coal

[versed violet]

Is there a way to 'pick' full belts from manifold if total items exceed the max belt throughput? I've got 8 reineries outputting 300 scrap each, and would like to shove it into full mk5 belts by picking items inline. Doable or not?
Currently I have to manually put splitters/mergers to achieve mk5 belts.

[wicked tinsel]

no, you need priority merger for that and its not a thing

you probably can split more aggressively and merge smaller belts together

at least thats what i did

someone ping Ben and request priority merger

https://questions.satisfactorygame.com/post/5e5dae05c672159e0cdf8f10

[versed violet]

There is not enough space to add a second line for mergers I'm afraid.
Wish I could just smart split it in-line, but That would require the belt feeding splitter to exceed mk5 capacity 😕

Transport question:
1 pure bauxite node - train it? belt it? Trucks don't seem plausible.

[bleak coral]

Wait till next week for viable trucks + zooping a highway?

but also I've totally run belts that far before

[wicked tinsel]

there are like 4 baux nodes in that region

connect them together and train to your destination

its like 1 pure, 1 normal and 2 impure or something like this there

around the circled node

[brittle terrace]

is pure ingot recipies worth it? it would be 52 refinerys for 3.3k iron per min or 60 smelters for 1800 iron per min

[versed violet]

Power wise, just build more miners and use smelters. Iron nodes are plenty.
You are likely going to hit another iron node when building refineries to process that iron.

[versed violet]

Only need 1 pure for even match at my main factory.

[wicked tinsel]

well, you can always expand

possibly later

one issue with that pure node over there is that its on top of spire iirc

while normal nodes are like 70m under it

on the other hand, it should match height of your X

[versed violet]

6 000 baxite should be enough for now 🤣

[wicked tinsel]

so train should be fine if you dont mind it floating somewhere

[versed violet]

There is a very nice rock 'walkway' all the way up to node

[wicked tinsel]

not wide enough for trucks or train line tho

[versed violet]

trucks, without foundations, nah. But train only needs supports now and then.

But youre right about 'future use', I'll just build a train in the unlikely event I want another 1k bauxite from the area.

[the player quietly realizes he tapped over half nodes on the map by now]

[brittle terrace]

im not worried about power

[versed violet]

It also uses more space and requires laying pipes, which is more time consuming than just splitters. Not to mention you need to supply water.
Personally, I'd just slap a row of splitters + mergers and be done with it - with mods you just need to scroll up to get whole row connected at once vs having to lay pipes for refineries.

[frosty owl]

I went for belts to get stuff out of that area. Can make for some nice wavy spaghetts around the bamboos

[pastel gulch]

Isn't that what nobelisks are for? 🤔

[vapid gorge]

On that location I recall there being several nodes. I combined them then trained them in

[vapid gorge]

Similar to the shot in #screenshots

[empty glade]

What do you mean?

[lilac skiff]

sorry posted in wrong channel

[frosty owl]

@thorn bane Small challenge for you, if interested ^^
Given rubber, copper ore, caterium ore, copper sheets and quartz crystals, would you be able to set up a non-overflowed sushi oscillators setup with your merging method?
I tried once in my early days by feeding somewhat "exact" amounts but failed, haven't tried again on it since then ^^

[wind spade]

@bleak coral moving here since #satisfactory can't do image uploads.

For "rare" items you just won't add overflow to sink because you don't want to sink them anyway

top smart splitters are set to overflow to sink, bottom ones are just for filtering

which as I understood doesn't bring the issue, as it's related just to overflow

[bleak coral]

oh yeah sure you can work around it, but that's like a lot of extra space

[wind spade]

that's how I would do my storage even without the bug... not too much extra space, the top part is already built anyway, I just add the sorting belt (bottom half)

I'm not?

I'm using the "any undefined" option in the bottom splitters

at the end of the bottom belt it's just a container of unsorted items

[bleak coral]

I just feel no need to have an auto-sorter, I've got equipment and limited material boxes, and then I've got main storage which is just constantly being filled. Feel no need to complicate it more than that or greatly inflate the number of belts I've got

[wind spade]

I can also process e.g. slugs to power shards before putting them to storage

[wind spade]

yeah that's fair, the discussion shifted to "is it possible to evade rare items getting stuck in smart splitters in sorting builds" anyway 😉

[frosty owl]

Careful with bottlenecks though~
Eg: if the main belt is mk5 but the shards output is MK1, dumping 10 shards in the system could clog it temporarily due to the MK1 making the items on the mk5 lag behind (shards queued for the output)

[bleak coral]

fair enough, I'm gonna go test if any undefined doesn't produce the issue now to double check that this is safe

[wind spade]

is there ever a reason to use anything other than mk5 if you have mk5s tho?

[frosty owl]

(It is, but testing doesn't hurt). Only overflow makes items pile up in splitters without before any belt backs up

[frosty owl]

Being able to see the items going through
If your input is below ~10/min, with mk5 you might think the belt is empty until you actually go close to inspect
Also other misc preferences

[wind spade]

yeah fair, so no real reason but for some visuals it may be chosen

[frosty owl]

There are some technical things too, but those only involve careful mixing 😆

[wind spade]

anyway the design I just showed is my planned storage design. I feel like it's super versatile, you have overflow on all items and you can process them in any way you like. I'm not forcing it on anyone, I'm just suggesting that it may be a solution if you need one 😉

[wind spade]

so nothing any normal person does anyway 🤷‍♂️ 😛

I would just attribute that to generic "decimals being decimals" tbh

all the planning for perfect ratios and production is nice, but deep in your mind you know you won't ever reach that anyway bcs computers are dumb

[bleak coral]

floating point imprecision haunts everyone though, just usually you don't notice

@wind spade like vencam said, I'm not seeing the same behavior of stuck items unless I'm using overflow

idk, it's definiley just items that have to go through the overflow exit, when I was testing it if nothing had to go through there, nothing got stuck

or put another way, other exits can clear the internal buffer fine, but an overflow exit can't fully clear it

I was also only seeing 3 items, but I've definitely seen up to 6 before

so another fix could be this

Cause any items you care about getting stuck won't fill up storage

You could also flip any undefined and overflow, and I'd merge the specific item line into the storage line, however you're doing that, just leave this for auto-sorting of inventory

oh wait this is exactly greeny's thing in a different shape....

[frosty owl]

If you want to test that theory, you could try letting my nuclear save run and see if it ever jams

[wind spade]

something working for limited amount of time on a limited selection of computers doesn't prove it works everywhere and anytime 🤷‍♂️

[frosty owl]

I wonder...
That argument was once "It probably just works if FPS don't drop", then "Just because it works on your PC"... There might be a pattern going 😆

Though the mistery won't be solved easily if not enough people actually try

[wind spade]

my argument is that given enough time and specific conditions, the setup will break up

which has nothing to do with whether or not will it break on your pc in your lifetime 🤷‍♂️

[floral ice]

What is the any undefined line used for?

[deft lichen]

overflow should already catch everything else than the specific item

unless it's to prevent the 3-item buffer?

[bleak coral]

It's for that

You don't need this in 99% of cases

Only if you care if something gets stuck in the buffer, which the only time I can of for that is an inventory dump auto-sorter

To be clear that won't prevent the 3-item buffer, it'll just cause only the sorted item to be in that buffer

[floral ice]

How does / spoiler work?

[bleak coral]

Wdym?

[floral ice]

You can type /spoiler and then it says the text afterwards is being shown as a spoiler

[deft lichen]

||like this||?

wrap the text as ||text|| to get ||text||

[floral ice]

Thank you

[faint ember]

||Y||||o||||u|| ||c||||a||||n|| ||s||||p||||o||||i||||l||||e||||r|| ||t||||a||||g|| ||e||||v||||e||||r||||y||||t||||h||||i||||n||||g||.

Aww I fucked it up.

[frosty herald]

Not sure this is the place to ask but redirect me if its not. So i got a question, is there a way to make an even 5 way splitter?

[empty glade]

Split 6 ways and merge one back in?

[drifting plover]

Honestly i dont like splitting items unless its the base material that needs to get places

Most of the time i just manifold

[shadow prairie]

Official Satisfactory Wiki

This is the totally awesome community driven wiki for Satisfactory!! Come on in and check it out!
<3 @drifting plover

[drifting plover]

!manifold

[frosty herald]

What is that

[drifting plover]

Idk the command

[bleak coral]

!wikisearch manifold

[shadow prairie]

https://satisfactory.fandom.com/wiki/Special:Search?query=manifold

[drifting plover]

That

[bleak coral]

it's a search function, that embed happens to be a good one

[drifting plover]

S is splitter, m is merger ofc and you can do it for more than 4

My entire factory is pretty much manifolds only for belts

[wind spade]

--S--S--S--S--S
  |  |  |  |  |
  X  X  X  X  X

[frosty herald]

That relies on overflow though right?

[drifting plover]

Yeah

[frosty herald]

damn

[bleak coral]

you don't need a smart splitter though, just need for stuff to fill up

[drifting plover]

If you produce 60/min and 3 use 20/min then they will all get the materials they need

[bleak coral]

otherwise yeah, you can just do splits and merge-back splits

[frosty herald]

let me try

[drifting plover]

I thought it wasnt gonna work at first either

[bleak coral]

keep in mind it's not gonna work right away, it has a warmup period

[wind spade]

yeah, but even normal splitters work and it's not that bad actually. it just needs some time to warm up (or you can fill machines first)

[drifting plover]

But it does cuz its only starved for a bit

[bleak coral]

it works cause splitters don't split evenly, they alternate between the exits and if an exit isn't available (no belt or belt is backed up) it skips that exit

[frosty herald]

makes sence

[bleak coral]

so once a machine is filled and belt to it backed up, it will only take as much as needed

[frosty herald]

alright

mine cant really use overflow since the resource is only coming through once in a while

[drifting plover]

Well that just makes the warmup time longer

Might want to fill it up by hand

[wind spade]

what's your usecase?

[drifting plover]

Just make sure you arent using more than making

[frosty herald]

I have build a fuel rod production that is barely enough to supply 5 power plants and now I need to supply them evenly

[wind spade]

ah, if it can supply them evenly, then you can still use manifold, it'll just take a long time for all of them to start working. However if you don't need all the power now, you can just manifold it and do something else while it's filling, it'll eventually work at 100%

[frosty herald]

Alright thank you

[bleak coral]

oh yeah that's actually a case I'd use a balancer, cause the ppm is so low

[frosty herald]

whats a balancer?

[wind spade]

a thing that makes sure all buildings get equal amounts of resources

[frosty herald]

so

[bleak coral]

and you can do prime balancers with merge backs: https://satisfactory.fandom.com/wiki/File:Balancer_odd.png

[frosty herald]

just what i need

[bleak coral]

dammit that ones wrong too

[frosty herald]

yeah

[bleak coral]

(embed is wrong)

click through

[frosty herald]

what?

oh

[wind spade]

well you don't really need it. It's your decision if you will build manifold or balancer, both will work (eventually), if you don't need all the power now, then it doesn't matter, otherwise if you really need all the power now, go for balancer

[frosty herald]

Alright yeah thank you

so

a five way splitter is possible

kind of a cool puzzle honestly

[tepid sail]

Technically you could manifold and put the belt for the last one first then wait like 30 seconds or something and move like that up to the one closest to the source

[frosty herald]

maybe I'm just dumb but i tried to draw one and really struggled

[tepid sail]

That would accelerate the warmup time

[frosty herald]

yeah that should work

[wind spade]

all balancers are possible

[frosty herald]

really?

[wind spade]

yeah

[frosty herald]

thats cool

am i dumb? Is the solution very easy?

[wind spade]

all M:N balancers are possible, because worst case you just split each belt to N belts and then merge M belts together to each output

but most of the time there's easier solution

and there's always the manifold way which is easiest (although technically not a balancer)

[frosty herald]

manifold seems janky

[bleak coral]

they're just super unwieldy, because prime balancers are possible so all 1:N balancers are possible

[wind spade]

manifolds are what majority of players use imo

[frosty herald]

I still dont understand how, but interesting

[bleak coral]

and M:N just work by splitting all M inputs N times and then merging one from each M input into N outputs

[frosty herald]

M is merger?

[wind spade]

maybe not for nuclear fuel rods, but for pretty much everything other, if you build a manifold, it will most likely fill by the time you build the next section

[bleak coral]

but think about how unwieldy that'd get for say a 5:6 balancer, you'd have 30 separate belts to merge back into 6

[wind spade]

no in this case it's M and N variables

[frosty herald]

so

M is just a variable

that isnt N?

[bleak coral]

yeah

[frosty herald]

okay

[frosty herald]

That would look cool though

especially in a power plant

[wind spade]

like 4:9 balancer -> M = 4, N = 9.
you split each of the 4 belts to 9, giving you 36 belts. Then you merge each 4 belts together and get 9 belts output, giving you 4:9 balancer.

[bleak coral]

we just don't have good tools for belt balancing beyond 1:N, cause honestly we don't need them because of over/under-clocking

[frosty herald]

Yeah thats true. The usecase is pretty small but i still think its an interesting math question

[empty glade]

M is number of inputs, N is number of balanced outputs

[quaint rampart]

This is exactly how I built the 8:5 god damned monstrosity in my smelter

[bleak coral]

But why did you build it lol

[quaint rampart]

Four ranks of 12 pure copper refineries plus four ranks of 8 copper alloy smelters. Needed to load it into trains reasonably. Five platforms at full load beat eight platforms loading kinda slow.

It’s a mid game setup and the alloy stage exists to soak up spare iron. Later I’ll tear it down and do pure only.

[frosty owl]

Any setup can be broken under "some specific conditions"

[fringe pawn]

Geyser powered uranium ore -> sink wouldn't break ever?

If I ever did a planetary build I would probably concoct some sort of multi level power grid. Geysers powering a small turbofuel setup, powering a big turbofuel setup, powering nuclear infrastructure, powering the rest of the planet.

[frosty pawn]

and 10000 power storage lol

[fringe pawn]

Well, power storage is probably mandatory because of the fluid loss bug, unfortunately.

But hopefully that will be gone in U5, and maybe we'll get more particle accelerator recipes that make people want to do max nuclear.

[frosty owl]

I'm slightly tempted to not deal with the last 300/min uranium and use them up with the fertile uranium recipe in plutonium processing instead

[rapid iris]

how do i calculate over clock stuff. Like for example if im going to make reinforced iron plates and i have it at 100% (30 iron plates pm and 60 screws pm) how would i tell what materials i need if i set it to 63%?

[bleak coral]

it's just linear, you'd need 63% of the rate

[quaint rampart]

0.63 * 30 iron plates = 18.9 iron plates hth

[rapid iris]

oh

ok thanks

[wind spade]

don't think so. Mixed belts without overflow sinking can break due to decimals or lag or something (however unlikely it may be). Non-mixed belts can't really break

[proven prawn]

I have had such concerns myself of such setups but despite my testing even under even worst cases I haven't been able to prove such setups can be broken, assuming perfectly ratio's of production vs consumption of course, so until I can definitely prove that it can break even with perfect consumption and production numbers for it all I'm not going to rule out such setups from being viable. Though it is possible under some x amount of time it could happen, but thats to much theory and nothing I can put into practice as far as real issues I can test for in-game.

[wind spade]

yeah, my point isn't "setup is broken, don't use it", it's rather "I don't like to use this setup because it can break"

[proven prawn]

yeah though for me its more important to prove something can happen before dismissing a setup simply because I have concerns about things like belt imprecision somehow causing something, though I have already had setups break in the past because of the very issue of belt imprecision, but such issues have to really be tested in the long term which is what i plan to do with such setups in the future, though yes if I can prove they will eventually break I will most likely move away from not using them, only one specific plant is going to really use that setup anyways aka rod production, so for me having to redo this setup later isn't a big deal and it accomplishes my interest to see if such setups can work in the long term without investing to much into a system.

[jovial crescent]

What's the most efficient way to make concrete?

[wind spade]

efficient how? power? resources? space?

[fierce cypress]

in terms of limestone or power?

[wind spade]

that's a valid point, but the simple fact that mixed belt setups can break and are usually harder to build, I don't see a reason to even try them. If they would reduce complexity by a lot or have some other benefit, I'd definitely consider them

[fierce cypress]

@jovial crescent if its in terms of limestone its something like this

[jovial crescent]

I meant limestone thanks

[fierce cypress]

idk why anyone would set that up for concrete production tho

[wind spade]

you can also do oilless

[fierce cypress]

the 3 concrete alts

theres also default and wet concrete

[proven prawn]

for sure they can break, especially if something happens to a production line and then the real issue is when your entire production chain issues such a setup it isn't easy to get it back up and running, after all you can't have machine being in overflow state, though it uses things like smart and programmable splitters to give it many aspects of a overflow system, it still isn't exactly the same, not to mention all of the problem of literally going through all of the machines again when such setups break because you have to reset it all, which is definitely far from a ideal experience

[sonic blade]

Sorry I’m newbie, can you explain please where can i made this scheme myself?

[fierce cypress]

https://u4.satisfactorytools.com/production
check the pinned message in this channel for more helpful sites

[sonic blade]

A lot of thanks

[frosty owl]

Of course a non-mixed belt setup can break. Just disconnect a belt somewhere or pause a machine, the system is likely to stop 🤷‍♂️
The difference between a pure belts system and a mixed system is just how many points of possible failure you're willing to deal with, it says nothing on wether one can control such possible failure points or not
Eg: the only way to brake my system is to manually F it up somewhere. That's the whole point. Just like you can't F up a miner-smelter combo unless you manually try to (like putting one on standby or giving a too low mk belt, whatever)

[wind spade]

I meant break without intervention obviously

[frosty owl]

Then it simply doesn't if you set it up right 🤷‍♂️

There are no "specific circumstances" one cannot control that can affect its failure

[wind spade]

mixed belts can break due to game not being perfectly accurate, non-mixed belt can't

[frosty owl]

I'm not sure what you mean by "can break"
Are you referring to the precision issues with full belts?

The precision issue is avoidable though and doesn't only affect mixed belts

[wind spade]

precision issues of decimal numbers, not just full belts. And any imperfection leads to just temporary slowdown of production in non-mixed belt setup, while in mixed belt setup it can lead to complete stop due to overflow of one material

[frosty owl]

So something one can avoid as long as not too many decimals are used

[wind spade]

considering the whole game works with decimals, it's impossible to avoid them 😛

[wicked tinsel]

not really tho

most of recipes use pretty natural ratios

something like 30:60 input is easy to sushi

[wind spade]

I'm talking about game engine

[frosty owl]

You mean any decimal would lead to one item being over/under produced between (eg) plates and screws for RIPs?

[wicked tinsel]

from engine pov it doesnt matter if there are decimals or not

as long as splitters arent catastrophically bugged, it will work fine

[wind spade]

I mean that game uses decimal variables to track some stuff and decimals in computers aren't 100% precise

[wicked tinsel]

those dont matter tho, as long as splitters work fine

even if there is some minimal gap, it wont break

[wind spade]

we're talking about a mixed belt situation, where small imperfections can lead to items being sent onto wrong belt, which can lead to belt being stuck because it has more items than should / different types of items than needed

[wicked tinsel]

again, this is not a problem as long as splitters work

and splitters themselves shouldnt be affected by decimal issues

[wind spade]

and again, I know it works for [insert player who tried it and it works for them], and it worked for [insert timeframe], but the possibility is there and that's why I don't want to deal with that setup personally

[wind spade]

not sure what do you mean by that but if a splitter sends more plates into a machine that needs screws, the plates can back up there and therefore screws won't get to it, permanently locking the machine and manual resolution is needed

[wicked tinsel]

that would mean that splitters dont work properly

[frosty owl]

You've been "cutting me" off for a while Amelek xD
(I'm joking, I just agree)

[wind spade]

not necesarily, that could also mean that due to decimal error, extra plates or not enough screws got produced or items are in different order than expected, or whatever

[frosty owl]

As long as they produce the correct amount of items each cycle, it's fine. Machines' internal buffers can accommodate for such minor fluctuations
And I don't think you can prove those pile up to a disturbing point either...

[wicked tinsel]

due to how splitters work they arent affected by any decimal errors - they just split items into lanes in sequence

it means that as long as average ratio is fine and there arent too long gaps of only single items, it cant break

[wind spade]

again, splitters are irrelevant if the error happens before them

[frosty owl]

I've tried to look for errors. But the only issue I found was with trying to use too many decimals (so clock rounding issue)

[wind spade]

and again, I know it probably won't happen to anyone, I'm just saying that it can happen and that's why I personally don't want to use that

"[X] works for me" isn't proof that X always works

but "X doesn't work for me" is usually a proof that X doesn't work everywhere/anytime

[frosty owl]

Sure...
But that's an empty argument... One could say that about anything: McGalleon' VIP, hypertube cannons, you name it

[wind spade]

how's that empty argument? if something doesn't work 100%, then it isn't 100% reliable 🤷‍♂️

[frosty owl]

There's still no "X doesn't work for me" case to validate this level of doubt at this point

[wind spade]

what are you trying to say even? I don't want to use a feature that is more work than the other way, has more breaking points and has a possibility to deadlock.

not sure why are you trying to persuade me otherwise

[frosty owl]

Sure, I get that, and never tried to push you to use it...
But that's not a good enough reason for me to accept your comments on how it could possibly break when the whole point of it is that one can make such a system, and it'll work reliably

[wind spade]

the same as some people aim for flat power graph or never underclock/overclock machines, I'm aiming to make a system that has least amount of breaking points 🤷‍♂️

[frosty owl]

Just like I don't argue about how manifold is slower than balancing or whatnot anytime you mention manifolding, I think it'd be worth discussing wether something works or not when there's actually something to be discussed about that point, else let us preface it by saying it's just our preference

[wind spade]

I said it several times that it's what I personally think and why I personally won't build them 🤷‍♂️

(and fyi I know manifolds take longer to boot and it's usually what I mention to the person)

[frosty owl]

Yeah, but I don't make a point of it anytime someone mentiones manifolds. That'd be quite rude and repetitive on the long run, wouldn't it?
The whole convo started when you said that 'just because it works for you it doesn't prove it works", when you can't (afaik) bring forth an example of it failing to make sense of it
While would I not argue against it? 🤷‍♂️

While discussing the feasibility of the system requires such talks and examples to be brought forth, just dropping ambiguous comments about wether it even is feasible or not only keeps people away from even considering the subject

[wind spade]

it started by me sending a message completely unrelated to any mixed belts or something, also with slight exaggeration:

all the planning for perfect ratios and production is nice, but deep in your mind you know you won't ever reach that anyway bcs computers are dumb
then you replied that I can test that theory by running your save, to which I replied that I don't think testing on single computer can verify that something is running (again wasn't meant to be only for mixed belts, but as a generic statement).

then it continued by you (and other people) asking me questions and talking about mixed belts. It wasn't me who brought mixed belts to this conversation, I was just replying

and I don't recall me ever getting into a conversation just to say that mixed belts are broken or something. I recall maybe one occasion, but there I also said that mixed belts just require a bit more thought for them to work correctly

I think you're missing my point in this conversation, it isn't "mixed belts are broken" or "mixed belts will break", it's "mixed belts introduce another breaking point and that I do not like"

for example, I wonder how well would some setups work in case of power outage in specific moments, it's possible that it may work, but I also think there's a possibility that it can break. I don't really want or need to talk about this and figure it out because I wouldn't use mixed belts anyway even if they were safe.

but I'm open to discussion if you (or anyone else) want to discuss stuff about mixed belts or anything related to them (or anything else for that matter).

[sand epoch]

well that was several walls of text.. o0

[upbeat tide]

Soo im deciding what to do with my waste water from non-fissle uranium. Im thinking coal gens. Maybe best?

Gonna be making about 1008 waste water

[bleak coral]

You don't want to recycle it in the nitrogen gas or sulfuric acid steps?

[upbeat tide]

Nope

Learned recycle dependence the hard way isnt soo fun

[quaint rampart]

"I ain't reading that, I'm happy for you or I'm sorry that happened" ? 🙂

[bleak coral]

Pure recipe to use later is probably just recycling dependence with extra steps

[quaint rampart]

Recycle works pretty well if you have a separate recycling stage rather than feeding it back into the primary production stage, ime

A bank of machines only fed from recycled water, with no other sources, separate from the bank of machines fed from water pumps