math-and-meta

[opaque oak]

So the whole system stabilized within few train loads after I started with completely empty receiving side and matched production and consumption.

Only between the first and second trainloads did the gas run out. And part of that was that I didn't have the second train on the route from the start, like was needed.

[vapid gorge]

well there's only the base recipe and the 1 alt - and it's pretty darn good. The only reason to use the other really is if you're already making steel beams and can't be bothered changing it to pipes

[heavy gust]

oh there only was one alt for it?

nvm then :D

[vapid gorge]

.so is it actually stabalised?

[opaque oak]

Yes.
Trains are waiting at the sending side to get full load, receiving side never runs out of gas in the platforms themselves.
Receiving buffers stay full except while the unload flow stop happens.

The train cycle is stable.

I'll let it run some more.

[vapid gorge]

thats really cool 😄

[opaque oak]

As I said, based on old dev streams when the nitrogen was added, you need to invert the stuff because it wants to stay up.

Even this priorisation horror works. The single pipe has priority when nitrogen starts to run out.

Not 100% like with liquids, but close enough.

[snow dove]

gases work like liquids minus the headlift iirc

[opaque oak]

Prefer upper path to lower ones, but not 100% like liquids prefer lower to upper.

But if someone wants to check the sending side math:
I have a nitrogen node that can produce 780/min @ 100%, with single 60/min normal output not used.
Only using 120/min 6x pure output.
I have clocked the pressuriser to 225% so it says 1755/min.
So the one normal should then be 60*2.25=135/min
And so the used capacity should be 1755-135 = 1620/min.
And I have 9 nitrogen packagers @ 75%, one is 60/min normally and 45/min now, for total of 9*45=405/min packaged output (4m^3 nitrogen per single packaged one).
180/min input per each @ 75%, so 180*9 = 1620/min input to the packagers.
So the input of gas to the system should be 1620 and output to sink is 1620/min nitrogen gas, as 405/min packaged nitrogen.

And because people think the nitrogen gas is just fluid without headlift requirements, the buffers get built wrong at stations, and they hog the gas, because they are higher in standard setup. When they need to be lower with nitrogen.

To depriorize them.

Leading to those sending station buffers hovering around 140/400 when the station fill is in progress.

With them as the lowest point in the system.

So it would be better to say that Nitrogen behaves as inverted fluid without headlift.

[deft lichen]

Interesting discovery

[opaque oak]

IIRC it was said in the dev streams when nitrogen came.

[frosty owl]

Ohhhh, THAT'S why I keep not understanding why people say buffers are bad for gases: I always have my stations (and buffers jn general) lower than the factory

[fierce ruin]

can anyone describe to me the most efficiant way to do a coal and hydro powered power grid

[opaque oak]

Fluid vs. Gas incoming station buffer, you can understand why the fluid version doesn't work that good with the Nitrogen.

And the buffer gets too much gas and doesn't give it out properly.

[opaque oak]

Coal generator per upper stub, water extractor per lower stub, 120/min coal input with 15 per coal generator. The standard 8:3 ratio of generators to water extractors.
The pipes two pipes between the extractor manifold and the coal generator manifold can be any lenght and need pumps if the coal generators are higher than the extractors.

Just space everything out to fit your build but keep the relative connections. Exact outputs on each junction can change.

And the inverted nature of nitrogen gas was easy to see from the pipes when they filled up.
Upper segments always filled up first with lower segments only having small amounts of stray gas until the upper ones were completely full.

[deft lichen]

imho this kind of pipeline splitting/balancing is wholly unnecessary (it works, it's just unneeded)

if each extractor has its own pipe, you can connect 1 pipe to any single junction and it works

G G G G G G G G
+-+-+-+-+-+-+-+
W W W

[fierce cypress]

3 lines like above, or 2 lines merged in either way

[deft lichen]

any form of merging 3->2 works to reduce the amount of pipes on which pipeline pumps have to be placed 👆

[opaque oak]

Sending side balancing of 3 pipes of 540 to 4 platforms isn't working right, center two are getting more and the outer platform using packagers will run out of gas soon for moments...
Until the sending side middle platforms get full and the transport balances again.
But that is just because of that weird 3 > 4 split.

Currently the two sending center platforms have 2000 more gas than the outside platforms.

So put more oomph into the extractor and will get the whole system full and then downclock to usage again.
Because that would be the state I would run it once the system is actually used.

[fierce cypress]

@toxic solar example for a phase 2 sushi splitter with everything at 60/min input (will probably be less, but just as an example)

and you merge the 2x 5/min and 1/min line and i think that would work

you might need to do some sushi merging though - again, id ask ven

[toxic solar]

Since my trucks are delivering everything one part type at a time, it's going to have only smart plates in the system... then automated wiring... then (blank), it'll still jam

[fierce cypress]

you could also probably do larger factors of the ratio 5:5:1, for example 10/mins and 2/min

[fierce cypress]

then have a smart splitter storage system into buffers before the split system

but yea trucks might make it screw up too

with a non-constant flow rate

you could maybe use a flow-rate limiter to help?

but by that point it might just be overengineering

[tropic hawk]

Yes.

[tropic hawk]

Mind explaining?

[primal flicker]

I've built a very large coal plant. I've built smaller ones. The bigger it is, the more chance of a water feed problem. So multiple small, separate modules seems more efficient than one large facility, from my own experience. Comparing 16, 48, and 128 generator setups.

But generally, overclock the miner on a pure node, DON'T overclock the water extractors (unless there's not enough room for enough of them otherwise) and maintain a double-fed 3-to-8 ratio for each unit.

[rustic patio]

im still working on my oil plant, at first my plan was to turn 100% of the fuel into energy and the polymer just directly into rubber/plastic but after seeing the recycled alts i think ive changed my mind

what ratios for fuel->power and fuel->plastic/rubber would u recommend?

and how often should i duplicate it?

[wind spade]

enough power to last you for some time and as much plastic and rubber as you need for whatever you have planned

[opaque oak]

The gas rail transport is still going strong.
But I have to tweak the train settings because the receiving side stations had developed big imbalance in their contents over the night.
Would have started causing problems soon.

But that is from that stupid 3->4 on loading side and then then some shared usage from all the platforms on the receiving side that doesn't balance because Gas.

So switched from having to do full unload for all train cars to just unload and wait. Will see if this is better for the idiotic usage model I have going.

Or if I have to change how the startup nuclear cell gets its nitrogen. Current plan is just to leech from all 8 incoming nitrogen platforms. And then 3 blenders per platform as normal users.

But have to see if I can handle that leeching caused imbalance.

[last saddle]

do yall plan for leftover intermediate products when making factories or do you just use overflow for that
i've been realizing that planning to have some intermediate products exclusively produced after the tier where they are primary is futile because you'll already have a great supply of it by the time you've set up the next product

[vapid gorge]

why are you doing 3-4 instead of just 1 car for each pipe?

[vapid gorge]

if I'm not doing permanent factories then I just have a spliter set up to send some of every part to containers for use, it eventually all gets filled up

[opaque oak]

Because I'm stupid and want to see if I can make it work?
And the main problem is that combined leech use at the moment, not the 3 -> 4 on loading side.

[vapid gorge]

xD

[opaque oak]

I will see if I need to go to just 3 -> 3 in the loading side.

1-1 with platforms on both sides and 4 blenders per platform on the receiving side, but that leech usage would still cause problems with uneven unloading.

I might have to go to that and use the fourth platform for that currently leeching use from the extra normal node.

So I do have backup plans, but I first want to try the stupid thing.

But by using the extra normal on one nitrogen field and the extra pure on the other, I could feed the startup cell with the fourth platforms instead of trying to leech from all 8 platforms and causing imbalances.

Just would need to tune the transport chain to work so the trains stay at intervals, by having input and output balanced and sending side being the bottleneck.

But I can run this experiment for today while doing other stuff I have been neglegting and see what happens.

[last saddle]

the amounts i want to produce keep making a ~1.5 iron plate : 4 iron rod constructor ratio
i want to make my setup as modular as possible (as i move onto 960 and 1920 iron/min) but i don't know how to line up my constructors if this is usually the case

[vapid gorge]

Either leave floor space to expand the length of hte manifold or you could just build to acomadate mk5 belts and mk3 miners now and just upgrade the belts as you go

[last saddle]

my problem is that i would be completely fine if the ratio was even
i could place n plate constructors followed by 4n rod constructors, 4 times (1 for each node)
it would be nice to have an exclusive manifold for plates, but that would involve having 5 rows instead

with the fifth not being even with the rest

i think i'm gonna be lazy and solve it with underclocking

[opaque oak]

New setup seems stable so far, trains leaving with 350-400m^3 of gas still in tanks.
But we will see what the day brings.

[frosty owl]

For crying out loud

[opaque oak]

Remove the https://?

[frosty owl]

Given how often people ask for similar things, I think it might be worth pinning this tool that I can't link proopelry due to it not being whitelisted 👇 satisfactory.featheredtoast DOT com/rate-limiter
(proper link can be grabbed in the post below)

[opaque oak]

https://old.reddit.com/r/SatisfactoryGame/comments/zb5vys/rate_limiter_layout_calculator/

You can link that, and it contains the link to the actual place.

[opaque oak]

Still works, but need to change the system to the single to single.
Because the two trains would start to bunch up once there is other transit on the rails.
Currently only those two nitrogen train, bauxite train and crude oil train.

Will cause cold start to take more time, because the startup part cannot use all unused nitrogen in the stations but has to wait for deliveries to restart.
But will be stable.

So basically only because the train timetable settings don't allow suitable timing certainty with two trains to keep them not converging over time without "full unload" setting.

With the changed system, the Full Load setting will force the interval to keep constant, based on the production side.

So, because Nitrogen Gas cannot be balanced 100%, I need to switch to more sane system 😁

[vast jungle]

I also have a Nitrogen train, but its a funny one... front car is packaged Nitrogen... back car is for the empty containers going back ^^

[vapid gorge]

why funny? probably the most commonly used way to move gas

[vast jungle]

I always stumble about how many empty bottles I have to put in when I add another stop to the nitrogen train...

[vapid gorge]

oooh multi stop trains. I avoid that.

[vast jungle]

the alternative is multiple trains... which is not better

either way, I need to put in enough bottles so that even if the receiver freight stations are full with full bottles, there are still some left at the nitrogen source to fill new ones

[wind spade]

why not?

[vast jungle]

it just adds more trains, but delivers the same functionality... and I need even more bottles because now multiple full trains are driving around

and it blocks the source-trainstation more... because "more trains"

I am using the Nitrogen-Train like a Manifold... single injection, multiple outputs

[vapid gorge]

separate stations A - B , C - D ect ect

[vast jungle]

which would mean a lot more train stations... they are already large enough in my opinion 😉

[wind spade]

doesn't need more stations

more trains can go to single station

[vast jungle]

Cobalt suggested just point-2-point trains...

[wind spade]

just each train only has two stations in list

[vapid gorge]

Could also do that I suppose

I avoid the whole thing by centralising things and never needing to have more than 1x A - B situation

[wind spade]

things like feeder trains are great imo

[vast jungle]

I have central sources for some mass production items... e.g. plastic/rubber...

so I don't have to setup an oil factory everywhere I need some plastic or rubber

[vapid gorge]

yeah logistics become a pain that way

[vast jungle]

its also just not feasible to produce everything "on site"... there are not enough places where you can get all resources.

At the moment I am building my second Space Elevator Phase 4 factory... and it will import plastic, rubber (and quartz?) by train...

[vast jungle]

@wind spade you only transport raw materials via train?

[wind spade]

same can be for final products 🤷‍♂️

tho those are simpler, just factory (-> train) -> storage

[vast jungle]

I mean you don't transport some processed stuff (e.g. plastic) from one factory to another one? You would transport Oil instead?

[vapid gorge]

besides the uranium and 1 node of sulfure world max uranium rod made with only moving things 400m

[wind spade]

centralising production of intermediate resource sounds weird to me. if I need more, I need to track down the whole chain of factories and upgrade all of them, instead of just building new factory from raw resources, isolated from everything

[vast jungle]

I started doing it especially for rubber and plastic... you often have factories that need "a little bit" of it, so it doesn't make sense to create a dedicated production for them.

My "standard" plastic/rubber factory produces 540/min... because it allows for an easy design and some clean ratios. But then of course I have to distribute the produced plastic to multiple smaller consumers

[wind spade]

for plastic/rubber, I'd probably consider some sort of centralised production, but even there I'd probably prefer to have physically separated setups for each factory

but for most other things I don't really like centralisation

[vast jungle]

for me its the stuff I like to produce in certain amounts because the ratios feel sane... plastic/rubber... or Aluminium Sheets/Casings... most of the rest is produced (mostly) on site.

I think for Quartz I even ship the raw quartz, because it didn't felt useful to make Quartz Crystals or Silica on site ^^

just added my second plastic factory, so I am now at 1080 plastic/min and 540 rubber/min

[oak coral]

i want to run 32 coal generators.

What's the most effecient way to have the water spread over that

[fierce cypress]

4 sets of 3:8 systems

[vast jungle]

just consider them as 4 groups of 8 generators...

no special stuff necessary... coal can be a single manifold... or balanced across 4 manifolds...
Water... just 3:8 or 1:2 (with underclocking)

[oak coral]

i now use 1 extractor for 2 generators. But with the 32 generators i need 1440m3 water. Which means that i need 12 extractors.

[fierce cypress]

3 extrators for 8 gens

[oak coral]

what do you mean with 3:8 systems

[snow dove]

3 extractors for 8 gens

[vast jungle]

3 extractors at 100% give enough water for 8 generators at 100%

or (makes piping easier) 1 extractor (at 75%) for 2 generators (at 100%)

both works

[snow dove]

doesn’t really make piping easier

[oak coral]

but the flowrate of the pipe is 300. and i need 45 per generator. 3 extractors give 120

so i'll have 360 flowrate

[snow dove]

two pipes

[vast jungle]

or a ring-pipe

[oak coral]

can i still attach 1 pipe to them?

[vast jungle]

yes, but it will not work

[opaque oak]

Something like this.

[vast jungle]

thats one example of a ring-pipe system

[oak coral]

i have to make that every 8 generators.

[vast jungle]

yeah

[snow dove]

how so? it’s just dual injection into a manifold

[fierce cypress]

@oak coral ^

[opaque oak]

Because the extractor side is also manifolded and the mainline goes full ring.

Not always needed but helps in some situations.

[vast jungle]

water pipes work differently than belts...

you can visualize the solution from Baldur as a flattened ring, with both the extractors and generators attached to the ring

where you attach them doesn't matter

[oak coral]

What is better for future expansion?

That pipe line or what you typed

[snow dove]

well aware of the former. never heard it called a ring pipe

[vast jungle]

if you need a new "ring" for each block of 8, you don't need to expand the existing water pipes... you just build new ones

[fierce cypress]

you shouldn't expand - they're modular systems, build new ones

but the one i typed is simpler and works fine

[oak coral]

right

[opaque oak]

Yes, the one I put is for longer distances between the generators and extractors.

Better to keep them close and then you don't need the ring type.

[oak coral]

When the pipes are horizontal i don't have to use pumps right? no matter how long they are

[vast jungle]

the easiest way I have to visualize the solution is to imagine a circle/ring... and then attach 8 generators and 3 extractors where I want... I can also deform the ring any way I want as long as I don't "rip" it...

[snow dove]

headlift is only relating to verticality

[vast jungle]

normally the extractors are on one side and the generators are on the other one... which makes it easy to add 2 pumps

[fierce cypress]

the best setup is this imo:

G   G   G   G   G   G   G   G
+---+-+-+---+-+-+---+-+-+---+
      E       E       E```

[oak coral]

conclusion: I don't need pumps when the pipeline is horizontal.?

[snow dove]

yes

[fierce cypress]

pumps are only needed when going up - so yes

[oak coral]

Cool thanks

[snow dove]

not really a “best way”, they all require similar time and effort, and produce the same result

[fierce cypress]

but this one reinforces healthy fluid management habits and is simple to build and hard to mess up

[wind spade]

  G  G  G  G
E-+--+--+--+
E-+
E-+--+--+--+
  G  G  G  G

[oak coral]

What comes after coal power? Oil ?

[fierce cypress]

roughly this:

Biomass > Coal > Fuel > Turbofuel (optional) > Geothermal (optional) > Nuclear > Plutonium (optional)

[oak coral]

Gotcha thanks

[snow dove]

anything past biomass is optional

[oak coral]

ofcourse but it's more efficient to proceed to coal fuel etc if i'm right.

[vast jungle]

Geothemal is an odd one... you get it very late, its costly to setup, but it doesn't require input resources (except the place to build it)... and it doesn't give you constant power 😉

[fierce cypress]

technically its all optional - but i just did the mainstream/recommended progression

[snow dove]

it gives you constant power, just not consistent power

[vast jungle]

"biofuel only" is a challenge only for people know like the pain 😉

[fierce cypress]

handcrafting only

[vast jungle]

not possible unfortunately 😉

[fierce cypress]

screw the spelevator

handcraft everything

[vast jungle]

in a "handcraft only" game you never get steel...

[fierce cypress]

ok fine, how about this HUB burners only

[vast jungle]

hmm... MIGHT be possible... but you need a lot of underclocking to get products with refineries, blenders or manufacturers involved 😉

[opaque oak]

Everything but phase 4 would be possible.

Or can you underclock particle accelerators enough?

[vast jungle]

hmm... particle accelerator set to 1%? not sure

could be possible

not sure anyone would have enough time to finish Phase 4

[opaque oak]

Yeah, nuclear pasta only has 1500MW max consumption, so the 40MW can handle that with underclocks.

[fierce cypress]

yea with 1% isnt it only like 3.4MW

[vast jungle]

still, not sure a computer will run long enough to complete Phase 4 this way

[opaque oak]

The article even has numbers, but those might be old.
With 10% clock being still doable, for 20min per pasta, and little under 14 days for 1000.

So lower clock and more time now.

[vast jungle]

interesting optimizer question... what is the minimal number of items you need to complete the game...

[fierce cypress]

on a DS it could be possible?

[fierce cypress]

im guessing 2 based on no attempt at working it out at all

theoretically you could probably get through to phase 3 in 0 hours - time only needed to physically build

but im assuming that isn't counted in this

[oak coral]

Is there a limit of how far i can go vertical with pipes? or is it just a matter of placing more pumps on the lines then?

[fierce cypress]

no limit - if you reach the current headlift height, place another pump to increase it

[oak coral]

gotcha

[deft lichen]

have fun reaching the skybox 😛

[oak coral]

hehehe

[fierce cypress]

well technically there is a limit you can't go past the build limit (1997m iirc)

but otherwise do what you want

[deft lichen]

yep that is the skybox

it goes ~200 m below as well

[fierce cypress]

yup

with -18m being sea level

[deft lichen]

it'd take 109 mk1 pumps to go from the void to the skybox

[fierce cypress]

making the ultimate water tower

[cinder viper]

anyone have any good ways to split quarts into silica and crystal lines? The numbers are a bit odd at 37.2 and 22.5 respectively

[mystic moon]

Just overflow it

[cinder silo]

My method relies on overflow within the same facility.

[cinder viper]

ahh.

[cinder silo]

Have the crystal and silica constructors in the same line and have it as one production line.

[cinder viper]

smart, thats why I asked this chat full of far more INTELLIGENT humans, ty

[tropic hawk]

manifold

[cinder viper]

how many constructors of each do you guys reccomend for just starting out with quartz and that shit

like 2-3?

[cinder silo]

If you're going to do oscillators in the future you will need more.

[wind spade]

depends on what you need and how much of that 🤷‍♂️

[cinder viper]

i gotcha itll just be temporarily then

[cinder silo]

@cinder viper This is how I set up my quartz, the whole thing has a mixed output that a smart splitter sorts later.

[cinder viper]

yeah this is how im doing mine

[cinder silo]

Allow yourself room for expansion and you're good to go.

[tropic hawk]

one node, figure out how the splits match up then match them

[mellow sphinx]

Which one?

[tropic hawk]

use cases for each incoming

[mystic moon]

My overall pick would be 2 though

[wind spade]

depends which one is most useful to you and your plans/factories

there's no "good" or "bad" recipe, just a recipe you don't or do like

[tropic hawk]

1: Useful in early to Midgame when you need a lot of concrete at a high rate.
2: Solid Steel Ingot doubles your steel output iirc.
3: Fused wire is useful late game when you need a ton of wire and have some spare caterium

[mystic moon]

Greeny, does you calc take into account the full cost for partial buildings in it's materials tab?

[tropic hawk]

it does

[wind spade]

it assumes you underclock last machine and calculates accordingly

[mellow sphinx]

yeah i've chosen solid steel. Seem'd the most beneficial

[mystic moon]

Alright, that's what I was wondering

[cinder silo]

Solid steel is fantastic, the cost is needing to smelt the iron in to ingots before feeding the foundries.

[mystic moon]

Pure iron + solid steel is phenomenal

[tropic hawk]

it typically is

[wind spade]

but obviously if you find an issue, let me know, it's possible it's wrong (noone complained yet lol)

[mystic moon]

Lol alright, I think I just grabbed one stack to few or something

[balmy spade]

Why is this shit flactuating? I calculated the coal and water supply perfectly. I even overproduce a bit

Its 25 coal generators

[mystic moon]

Did you let the system fill?

[balmy spade]

I have the system running for about 35 hours now

they are all running on 100% as far as I can tell

[mystic moon]

What is running out? Coal or water?

[balmy spade]

I kept expanding it until I stopped at 25. So I cant really tell if it counts as letting it fill up

[wind spade]

also classic

• don't use buffers
• don't use valves
• feed from above or level
• loop the pipe
• don't exceed pipe limits
• place pumps only if you need headlift
• check numbers

[balmy spade]

nothing. And thats my question

[mystic moon]

Then they shouldn't shut off lol

[wind spade]

no gen is running out of anything?

[balmy spade]

yes

[mystic moon]

Watch the lights and when they turn yellow open the ui

[wind spade]

have you checked them all?

[balmy spade]

I've been standing here for 20 minutes

[mystic moon]

The water buffer should be completely full and it should have some coal in it

[balmy spade]

checking every single light

[wind spade]

are there biomass gens by chance?

in your network

[balmy spade]

nah. I got rid of all of them

[wind spade]

no other gens?

[balmy spade]

nope

I just unlocked the new generators like 2 hours ago

[wind spade]

what does your setup look like?

[mystic moon]

Can I see a screenshot of you plant?

[balmy spade]

the fuel ones

one sec

[wind spade]

how many extractors per 5 gens?

[mystic moon]

Oh boy, you're piping that water a long way

[balmy spade]

5 generators per water and 600 coal per minute

[wind spade]

extractors are just behind that setup

did you overclock anything?

[balmy spade]

we dont talk about my piping...

[mystic moon]

Oh yeah I see them lol
I didn't remember there was that little puddle there

[balmy spade]

nope

[mystic moon]

That looks like 10m to me

IIRC that vent is at 12

[mellow sphinx]

Im proud of my piping 😄

16 Gens atm - 1 Coal Node

[wind spade]

one extractor can feed 2.666 gens only. If you have 5 extractors, it's 13.3333 gens max

[balmy spade]

oh ye the water extractors are oc

[wind spade]

how much?

[balmy spade]

to 225

so they can feed exactly 5 I think. I dont remember anymore what I've calculated lol. But it should all be 100% eff and its still falctuating

[wind spade]

that should be exactly as you need. Does water reach the top pipe?

[balmy spade]

yes

as I said, the coal gens are working 100% efficiency

OHHHHHHHHHHHHHH

dammit

[wind spade]

they aren't

[balmy spade]

nono they are, BUT

one sec. I'm fucking stupid. Idk if thats the issue, but lemme send a ss

[mellow sphinx]

My pipes are Vertical 😂

many pumps needed

[balmy spade]

I forgot that I had these guys down there. They are all overclocked to 250% and the water supply should be working 100%. But I think it's flactuating, because I didnt have mk4 belts to get the correct coal supply and so I wanted to future "proof" until I got mk 4 belts, which I have now...

so I just found the mistake by myself I think

[mystic moon]

"no other gens on the network"

[wind spade]

xD

[balmy spade]

I forgot lmfao

Because I cut them from the network for most of the time and just recently as emergency supply I put them back in

well thanks for trying to help anyway

[shadow prairie]

Official Satisfactory Wiki

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

[thorny heron]

whoops wrong channel

[rustic patio]

i love oil

[bronze silo]

any ideas why it's not passing after the 2nd pump? I put it in the snap point

[cedar mica]

The snap point dont work 100%. You some times has to be a little below it.

[bronze silo]

k

[vapid gorge]

Are you using a daisy chain mod?

[fierce cypress]

it does appear so

[rare path]

Question, using Satisfactorytools.com - I'm using U6 (exp), how safe is it to use? what kind of bugs/inaccuracies would I need to worry about

[fierce cypress]

its pretty safe - i use beta most of the time which is notoriously buggy but its still fine 99% of the time

[rare path]

thanks!

[wind spade]

only things missing are:

• u7 recipes (pretty much none)
• overclocking changes (don't trust power numbers blindly but they won't be as off as they could)

[fierce cypress]

@north cloak

• using SFT

[north cloak]

OOOOOOO you can do alt recipes?

[fierce cypress]

yup

[wind spade]

||leaked beta pic|| just kidding

[north cloak]

since when?

[wind spade]

since... all the time lol

[fierce cypress]

i assumed as long as i didn't link it, it was fine

[north cloak]

Well ..... I suck

[wind spade]

nah, it's fine to show beta stuff, just "linking" should go through "join discord and read stuff before using"

it's not a fight club or anything xD

[north cloak]

ok looking at any other alts I can throw in here

[wind spade]

just fyi the tool will pick what it thinks is "most resource efficient" based on dark magic math, but you can disable recipes you don't like to "force" the "less efficient" recipes

[north cloak]

ohh ok cool

[fierce cypress]

its pretty easy to use once you get the hang of it

although i am running out of space

[wind spade]

something something rookie numbers

[fierce cypress]

those are only the ones on this computer - ive got plenty of tabs on two other ones as well

[wind spade]

I just make a new tab every time someone asks here for some math stuff or sends me a "bug report" url to check... I've cleared all the tabs like 3 times already when they get to like 5-6 lines 😄

[fierce cypress]

same

[vapid gorge]

just save the links to a file xD

[fierce cypress]

why be organised when you can just open a new tab

[wind spade]

I'm hoping I'll be able to solve this issue in new tools. Have a few ideas

[vapid gorge]

because your browser might lose track of everything youv'e saved and you'll cry

[wind spade]

that one will also get solved 😄

[fierce cypress]

99% of them i'll never open again

i use it to help someone here or to quickly crunch some numbers and then ditch it

only the few i use for my factories are frequented

[north cloak]

Here is my plan

[fierce cypress]

cool

ditched the stitched

[north cloak]

oh wait forgot to add in copper

[fierce cypress]

thats weird - why is the copper splitting?

why doesn't it pure ingot it all

[north cloak]

Probably to conserve power

[fierce cypress]

it shouldn't optimise for power

[north cloak]

ooo I never have really explored this tool

[fierce cypress]

are you meaning to do steeled and bolted frame?

[north cloak]

yes

[fierce cypress]

why so?

[north cloak]

Well I have both alts already

[fierce cypress]

does it give you more?

[north cloak]

why not?

This is what the system is saying to do

And yes if I turn off bolted frames it gives more

[fierce cypress]

yea, thats more efficient

[north cloak]

hold on....

can you share that

[wind spade]

I'm gonna guess maximise + ore limit 🤔

[north cloak]

ooooo you have more coal that I can supply right now

[fierce cypress]

limit coal forces bolted yea

[north cloak]

exactly

I have a ton of work ahead of me... but I have a feeling I will be very happy that I did this

[fierce cypress]

https://www.satisfactorytools.com/production?share=FyZ3oCNGvGPJAro3Rwk8

[north cloak]

thanks

[wind spade]

I can tell when you were surprised it can do alt recipes

[fierce cypress]

its only the best sf calc

[north cloak]

Well this is my new favorite thing ever

[wind spade]

[keen shard]

Can someone remind me whats its called, the thing where a belt cant really transport 780 items a min when it comes to big long builds?

i forgot its name

[wind spade]

that thing was fixed

well, not 100% fixed, but the belt can now carry like 99.99% capacity, so it's not an issue anymore for most people

[fierce cypress]

[wind spade]

(it was called b2b bug)

[cinder silo]

The 780 thing was very well patched, still imperfect but 99% of the time it will actually send 780 now, loads of testing was done for that.

[fierce cypress]

^ belt to belt throughput loss (b2b)

[keen shard]

is the wikipage for the problem still up?
I remember I actually cancled a giant build I wanted to because this was an issue

[fierce cypress]

if there was one you should be able to find it - even if its archived or no longer maintained

[wind spade]

I don't know if there ever was a page on it though

[fierce cypress]

i don't remember there being one

[cinder silo]

I had to seriously stress the game out using my oversized save and crash a huge train to get errors out of all but the longest b2b test tracks at the arena.

Once my fps crashed down to just 4 I managed the occasional error even on the welded belt test 🤣

[vapid gorge]

eh, there was always ways to get around it

[fierce cypress]

still was inconvenient at best

[keen shard]

non satisfactory ways

[wind spade]

so basically unless you really really really really care about 100% and you're not happy with 99.99% (pretty much no visible difference), then you can assume the bug was fixed completely

[keen shard]

it was posts like this

[vapid gorge]

like what? You could weld belts or link to merger/splitters, or both

[cinder silo]

A whole load of tests happened in here #1006573529183027401

[cinder silo]

I was able to cause thrown objects on welded belts per one of my test tracks, the really long range stuff I put a splitter on a mine and sent two belts to avoid the bug at all.

[keen shard]

Oh you buggers have been busy love you

[cinder silo]

Ever so slightly 🙂

[vapid gorge]

pre fix I welded this belt - great throughput. All one piece https://media.discordapp.net/attachments/558721941410807812/1052384920191897640/unknown_62.png

[cinder silo]

Test repeat destroy, it was the only real way to get data.

[keen shard]

Yeah great if it worked for you, but all it takes is one 'accident' and the whole line is gone

I play multiplayer

[cinder silo]

The 400 wagon train crash really made for edge cases of thrown items and belt errors 😄

[vapid gorge]

I mean if you play MP you're taking your life into your own hands there

[fierce cypress]

is that curve done with a mod?

[cinder silo]

I did that with a 10km welded belt (it was buggy anyway but still)

[vapid gorge]

that was was just the wan method

[fierce cypress]

that works with belts?

[vapid gorge]

or did you mean the belt curve?

[fierce cypress]

yea the belt

[vapid gorge]

oooohhh no it was MANY belt segments

that I welded together

[fierce cypress]

assumed so

looks great though

[vapid gorge]

thanks xD

[keen shard]

did you do it in a blueprint tho?

[fierce cypress]

pre-BP

[vapid gorge]

this was way before bps.

[cinder silo]

All of cobalts stuff is pre bp

[vapid gorge]

except for now that I'm getting back into it 😄

[fierce cypress]

and a BP for that would be pretty useless too, since most curves will be different

[keen shard]

Ah, see im not that paitient

[vapid gorge]

Yeah even if I didn't vaguely dislike bps I can't imagine how I'd use them in curves to my liking.

Almost every curve thing I do is bespoke for the situation

[cinder silo]

My old turbo fuel facility would have tested your patience, that was a 1.2km by 1.4km place built before zoop was a thing, let alone blueprints.

Took me a while to demolish it and move to a new location when the spire coast changes were announced too.

[vapid gorge]

I used area actions for the floors on this? Not quite the same. Was good when I discovered better/cleaner ways to do curves and I could delete all the floors and use AA to put up the improved versions https://media.discordapp.net/attachments/553550313533997057/1056024567614804018/unknown_-_2022-10-16T144310.523.png?width=1241&height=905

But that's a building that will never be built in my world again

[keen shard]

I doubt it, i build a rail+3 oil pipes pipeline from the oil islands to the spire coast

[vapid gorge]

yeah running pipes under existing rail is a fav thing of mine too

[keen shard]

how the hell? a mod?

[fierce cypress]

which part?

[keen shard]

idk it just looks so alien to me

[fierce cypress]

i think some mods were used to avoid rebuilding it all the time but afaik you can make that in vanilla

[keen shard]

I cant wait to start again, I wanna see how the blueprint designer is gonna help.

[cinder silo]

Done the long pipelines myself

I moved the spire oil , all of it, to the dunes via tunnels once they were gathered.

[keen shard]

smashing. I only got this to defend myself

[vapid gorge]

So the only part of that that needs a mod is the triangle above the hyper tube. Those are shrunk

[cinder silo]

Ahh cannon launchers, I use contained cannonways, less effort and proper fast.

I tend to build on a large scale.

[vapid gorge]

:shock:

[keen shard]

when you not flying over the world. Sad

not a soaring eagle, but a rat in a tunnel

[cinder silo]

I'm fine with regular supersonic travel and no effort getting from a to b.

[keen shard]

I could never, I love flying too much

[cinder silo]

Horizontal failure before my cannon testing yard was completed, flying kinda happened.

[keen shard]

*moment of realization *

[cinder silo]

Yeah got smashed through two walls and a land bridge about a quarter mile away.

There is an issue with supersonic travel though, the terrain likes to not load 🤣

[keen shard]

I noticed this problem. dont step on it you will flow through it 😂😂😂

[cinder silo]

I have platforms at each end to reduce that sort of death.

[keen shard]

smart man

[cinder silo]

I tend to build large structures around even the simple things like a stair way leading under the map.

[river fjord]

just got an aluminum setup with 240 ingots/minute (assuming everything goes well ofc), what's a good ratio of alclad plating to casing for my first aluminum setup? ill be siphoning off at least 40/min for a preliminary radio control unit factory

[cinder silo]

I'd say almost all casings unless you have classic battery plans, the few plates are for your mk5 belts.

[river fjord]

hmm, 2:1 or 3:1 ingots going to producing casings to sheets?

[cinder silo]

Even with classic battery I have a 2-1 casing to plate ratio on mine.

[river fjord]

id like a fair bit of plates because i always make belt highways instead of trains

[cinder silo]

Fair enough, you don't need a huge plate output though, just have enough storage to allow it to refill when spamming belts 🙂

There are a lot of the recipes that take casings, only a couple for plates, ensure you have the capacity to build belts for days and maybe classic batteries, everything else is casings.

I apologise for my typing, I've been on the vodka.

[river fjord]

alright, thanks

[oblique notch]

pretty sure all of that can be done in vanilla yet. Cobalt may have used Micro Manage to get things into place easier without having to do all the beam tricks, but ... yeah im pretty sure thats all buildable with out it.

[vapid gorge]

I did have to shrink some inverted ramps under the hyper tube. Other then that just vanilla

[oblique notch]

hard to tell that from the perspective

[vapid gorge]

I tried not making it obvious, but if you know your vanilla obj you can tell you can't get a point like that in such a small space

[oblique notch]

mmm yeah i can see now the width is too small for that.

[fierce ruin]

has anyone ever calculated the theoretical maximum amount of resources you can get from a node /minute? Beyond having a tier 3 miner with 3 shards, what would the rate be for, say a normal iron node, if all of the space not occupied by the miner was being used by portable miners, and all 127 players were also mining the node by hand?
Obviously a setup like this would be completely impractical under 99% of situations, but i think it would be hilarious to see an absolute wizard at the game doing something like this in order to eek out a tiny bit more production

[opaque oak]

Limit would be based on how fast the player(s) are in deploying/emptying the portable miners.
No time for hand mining.

[fierce ruin]

i know, but lets assume that the players are bots and are able to mine and do inventory management at the same time

[vapid gorge]

If you can clip the portable miners and have infinite people hand moving stuff? Near infinite I guess

[cinder viper]

If I got 240 copper comin through some miners, which ratios should I use for cable, wire, & sheets?

[opaque oak]

As always, depends on your specific needs and changes over time.

[cinder viper]

im just gonna wing it then

[opaque oak]

Build production for what you need now, and leave part unused and space for more production.

[vast jungle]

just build what you need for other factories plus a bit space for storage

[prisma kraken]

flap, just build as much copper wire as you can

the package 3 delivery needs 1.4 million wire assuming the base recipe for auto wiring

no matter your playstyle and what you enjoy about the game, you'll never go wrong with overbuilding wire production

package 4,, oops

its late here, lol

[cinder viper]

yeah just checked all the recipies I have right now

wire seems like the way 2 go!

[vast jungle]

or just build the necessary constant input factory for the automatic wire factory with it... no need to store anything before you build the factory

[cinder viper]

true but then you'd just need to source copper if your main smelting area isnt near what you wanna build

idk if anything im saying makes sense lol

[prisma kraken]

differnt play styles shrug

[vast jungle]

yes

but in the end there are LOTS of resources on the map... so finding a new copper source is not that difficult

especially with SCIM or he new U7 radar towers

[prisma kraken]

it becomes important in pasta making in the p4 delivery

but that's a different factory than intro copper

[vast jungle]

yes, Pasta is fun... my "3 pasta/min" factory (the last one I finished) had 37 Foundries for Copper Alloy ^^

that was LOTS of copper

[prisma kraken]

yeah, that's a build in itself, lol

[vast jungle]

it was... but the strange thing... it completed the 1000 Pasta long before I could finish the second factory for P4

[prisma kraken]

haha

you finish the package delivery long before you complete your design 😉

[vast jungle]

I am working on the Assembly Director Systems at the moment... "only" need to finish the Supercomputer part (now that I have extended my Plastic production to 1080/min)

[prisma kraken]

how do you get to 1080?

[vast jungle]

I have a standard design for rubber and plastic factories... 180 oil input, 540 output... so when I ran out of plastic production, I just added a second plastic factory

[prisma kraken]

like 900 is the number i find easy

[vast jungle]

my design works as this
6 HOR production refineries (or 8 at 75% for a nicer design)
8 D(P)F loops... (or 4 blenders at 120%)
3 Refineries to produce 60 rubber from the polymer resin
3 Refineries to turn the 60 rubber into 60 rubber and 60 plastic
and 2x7 Refineries as 3-staged doublers with recycling... turning 30 into 60 into 120 into 240.
Total output is 2x240 + 60 = 540

[prisma kraken]

300 oil turns into 900 rubber or plastic with the recycle recipes

[vast jungle]

you can easily build this design with MK1 pipes and MK3 belts... and there are no loops (and no unnecessary merging) in the recycling part

[opaque oak]

That is the often shown default, but the ratios are horrible for that, or the 600 crude to 1200 one.

[vast jungle]

the 180 input works well because the output of the Polymer-2-Rubber conversion is a multple of 60...

and I really tried hard not to merge anything (liquid or solid) unnecessarily to keep everything easy to debug and maintain

I don't even need to manifold any solid, so it doesn't need time to balance out

[prisma kraken]

i do the 900 build late game because rubber is something i really need for hmf's and nuclear's demands for concrete

[frosty owl]

A rare appreciation of Flexible Frames I see

[prisma kraken]

i don't find i need a lot of plastic

but yeah, rubber recipes, i find its always great

[vast jungle]

the standard supercomputer recipe consumes LOTS of plastic... and so do circuit boards

(unless you want to hurl in quartz)

[prisma kraken]

yeah, you can produce them sans plastic using quartz instead

i use the overbuild of rubber early with coke for circuit boards

set that up with an ad-hoc assembler as you get to the hmf problem

that circuit board recipe alt isn't very efficient

but it uses up the rubber early

[vast jungle]

yes...
but in my currently playthrough I jumped directly from 2x2 refinery rubber/plastic production (with coke sinking) to 2 of the 540/min. factories mentioned above...
so I never had much coke as a waste product

[prisma kraken]

i like using the steel and aluminum recipe alts

so like my initial petro thing turns HOR into coke, and i turn it later in the game into sorta a gasmask/weapons thing

and fuels steel for HMF's

[vast jungle]

I think I scrapped my first oil setup the moment the larger factories came up... and never produced any coke after that.
Thats the fun of Satisfactory... small design choices early on can form entire (and valid!) production chains 🙂

[prisma kraken]

i've made diffent choices in the early game

it really does sorta change your late game designs

i see how you got to 1080 as 540x2

it wasn't obvious to me, thanks for explaining that

i was assuming you overclocked your oil wells

[vast jungle]

some of them, not all

all three factories are in the "new part" of the northern coast... lots of oil there

and I only use 3x180=540 of it at the moment

[prisma kraken]

i've never considered the math of keeping them at 100%

[vast jungle]

I can overclock them when I need more and run out of unused oil wells ^^

still, I hope 1080 plastic/min will be enough until all P4 factories are finished

[prisma kraken]

yeah, i haven't don the plastic calculations if you use that for circuit boards & supercomputers

[vast jungle]

I think my Assemble Director System factory needs 580/min plastic

(something like this)

[prisma kraken]

i've never done the math on that

i think that chain requires tons of screws too

[vast jungle]

no screws involved I think

[prisma kraken]

i think the default computer recipe uses them

the number 52 sticks in my head for that

[vast jungle]

yes... but I use Caterium Computers in this factory

I even have (I think part of Turbomotor production) a factory with Crystal Computers 😉

[prisma kraken]

i still need to do turbomotors and supercomputers

i have all the autowiring pre-made, but need to build out nuclear to do supercomputers

[frosty owl]

TFW "tons of screw" just means "some steel beams"

[marble topaz]

Do you have a picture of your 540/min factory? Id be interested to see how you are doing it.

[vast jungle]

will take a few minutes, I am on a different computer right now...

the original was 8 Refineries for the HOR, and directly behind them 8 Refineries for DPF and a packager/unpackager on top of each other... the rest of the factory has only Refineries rotated 90° be the two...

left side the three Polymer-to-Rubber one, right side three Refineries to turn 60 rubber into 60 rubber and 60 plastic... the rest is just (left and right) 7 refineries for the 30 to 60 to 120 to 240 recycling conversion... (1+2+4 refineries = 7)

fits quite nicely when you get the trick of stacking packager and unpackager on top of each other... each trio (pack/DPF/unpack)has its own belt loop, so no merging/splitting involved

[marble topaz]

Ahhh, I built mine in a very straight line using blenders.

[vast jungle]

in my last factory I replaced the 8 DPF loops with 4 blenders (but kept everything else the same)

let me see if I can get a screenshot...

[vast jungle]

here

[marble topaz]

Yours is much more well thought out compared to mine haha

[vast jungle]

its the forth iteration of the same concept I think

designed to have no manifolds, nearly no mixture of liquids and no loops beyond the DPF one

this is a side-view of the DPF-loop

[marble topaz]

That looks amazing from that viewpoint. Impressive to not have to loop anything as well.

[primal flicker]

Bottom feed empty canisters back to the refinery right?

[vast jungle]

Yes, the packager at the bottom gets the empty containers from the left and the full water containers are then delivered below the refinery to the refinery

[left fractal]

Hi guys i need help

[vast jungle]

Unfortunately the dpf loop is to large for a blueprint

[left fractal]

I have a factory that make 1560 empty canister per minute

I split it in 2 belt
780 each belt

If i want to transfer them with drone how much per min can drone transfer

Or how much per trip

[fierce cypress]

it depends on how far the trip is

[left fractal]

And should i build 2 drone port or 1 is enough

[fierce cypress]

you will probably need 2 i think

you can check the throughput of a double drone one port and double it if needed

[left fractal]

Like without test can't be happening?

Oh also if i want to transfer them do i need one extra drone port to feed them battery?

[vast jungle]

You might even need two done ports on each side

[left fractal]

Like 2 to transfer?

780 per min each dorne port ?

[vast jungle]

Maybe even more ports and less per belt

You have to test, drones are not that great for mass transit i think

[fierce cypress]

they're good for low throughput - they struggle with high throughput

[left fractal]

Oh i should mention that

3.9 refinery for aluminum

And i need to package all

[fierce cypress]

package alumina solution?

[left fractal]

240 per min
240 * 3.9 if im right

It's 936 per min

[left fractal]

Yeah

[fierce cypress]

you have some interesting logistical choices

[left fractal]

How

Im lost in all that math lol

[fierce cypress]

most people never package alumina solution

its all at the same location

many people do sloppy alumina > default scrap because that has a 1:1 solution ratio

[left fractal]

936 / 120
It's 7.8 packager

[fierce cypress]

instead of transporting it

[left fractal]

120 empty canister need per min for one
So it's
7.8 * 120

[left fractal]

Um idk what 1:1 ratio mean

[fierce cypress]

you just did 936/120 = 7.8 7.9*120 = 936 you went in a circle

[left fractal]

Yeah told you im dead from math i need help 🥲😂

[fierce cypress]

sloppy alumina outputs 240 solution/min and default scrap consumes 240 solution/min

[left fractal]

Oh i just need to pack it and store some and use some like

I got battery factory made 60 per min

[fierce cypress]

why are you storing alumina solution?

[left fractal]

And it's already handle like 6 port

[left fractal]

I store everything 😐

[fierce cypress]

i see? not the smartest idea imo

[left fractal]

I know lol

I made mistake hand craft qll6

All till get nueclear now im starting to make factories

Now tell me what to do 🥲

[fierce cypress]

thats your descision not mine

and its like 2am and i should be asleep so probably not the best time to be planning a factory

[left fractal]

😂

Okay okay sorry

[fierce cypress]

not your fault i only have myself to blame for this one

[left fractal]

Why

[fierce cypress]

well you didn't make me stay up to a ridiculous hour did you

[left fractal]

I didn't

But your brain did

Oh wait what if i use train to transfer the empty canisters

[fierce cypress]

yup

[left fractal]

How to a loop trains

[vast jungle]

Why do you need a constant input of new containers anyways?

Exactly... Ein the containers in a loop... As short (the loop) as possible

[left fractal]

No i mean

How should the train face

I tried before it didn't work

Um can you join my word and help me?

[azure ingot]

quick question, does anyone know whether i can get some kind of file that has all the recipes written down in detail?
Such as:

recipe: xxx
machine: Manufacturer
input: {20x abc; 15x def; 5x ghi}
output 5 xxx
outputrate: 20

I don't want to use 10 hours of my life writing it all down manually...

[left fractal]

If found send it to me

Im still looking on such file

[azure ingot]

for how long?

[left fractal]

I even write like few of theem Excel

[left fractal]

1 month

[azure ingot]

...

[wind spade]

CommunityResources/Docs.json

[left fractal]

Huh 😐

[wind spade]

(or if you want slightly better formatted, I have one in my tools' code)

[left fractal]

😐can i have it

[wind spade]

warning, 36k lines (formatted for better readability)
https://github.com/greeny/SatisfactoryTools/blob/u6/data/data.json

[azure ingot]

you're a saviour

[wind spade]

does not contain ficsmas stuff

[left fractal]

Um

[azure ingot]

did you scrape it from the calculator page? or from the code itself?

[wind spade]

it's generated from the CommunityResources/Docs.json file

[azure ingot]

but where do you find that path?

[vast jungle]

The game comes with the list from css itself

[wind spade]

in Satisfactory install directory

[vast jungle]

In your game installation folder

[left fractal]

Should i download all .json file?

[wind spade]

I just parse it to this file so that the tools I made can read from it (it's slightly easier to read from it than the Docs.json format)

[azure ingot]

thats practical, but a one-liner imo

[wind spade]

.json is intended to be mainly machine readable, so if you want something human-readable, then it's probably not for you

[left fractal]

Nah i will make it human readable for myself

[azure ingot]

jeaa... i'm planning to trim the list down a bit tho, don't need each and every info

[left fractal]

But i need to know which one to download

[wind spade]

(with some code/script knowledge you can easily make anything out of it)

the one that was linked

[azure ingot]

this one

[left fractal]

Okay thank you

[wind spade]

https://github.com/greeny/SatisfactoryTools/blob/u6/data/data.json

(click the "raw" button and then do rightclick -> save as)

[azure ingot]

@wind spade is the file only formatted or did you add to it as well?

[wind spade]

there's a few extra things like slug fields (used for the tools) and a few others

but it's mostly just data from the game

[left fractal]

How many empty canister can drone transfer per trip?

[vast jungle]

you know... stack-size(empty canister) * slots-available(drone) 😉

[left fractal]

Is it 100 or 200 for canister?

[vast jungle]

https://satisfactory.fandom.com/wiki/Category:Item_stack_size

[left fractal]

Can you tell

[azure ingot]

100

[wind spade]

it's better that he tells you where you can find the info, because then you wouldn't have to ask again next time

teach a man to fish or something

[azure ingot]

yea that's kinda true... btw i've been looking through your tool for a bit (only the github & reddit till now) and it seems amazingly detailed. That's some great work, how long did it take?

[left fractal]

Thanks

[left fractal]

Yeah but the like i bary can get the message cus im downloading alot of file same time

[azure ingot]

is it in 8k?

[wind spade]

well, hard to estimate.

The tool started when satisfactory released (March 19, 2019) as a simple project for me and my friend to calculate stuff (originally excel, then https://satisfactory.greeny.dev/ - now outdated). Then I shared it here and got some feedback and a lot of feature requests and have been improving it since then. Around Update3, I rewrote the whole code into what is now satisfactory tools. Currently I'm working on another rewrite that would allow me to add tons of new features like user profiles, mod support, etc.

Obviously I didn't work on it those almost 4 years all the time, but it's definitely a few thousand hours sank into making that tool.

[azure ingot]

damn that's a lot of work... i can only wish you good luck with the rewrite, since that is far outside of my capabilities rn

you spent more time writing that tool than some spend on playing the game

[wind spade]

well, definitely more time that I spent ingame

I have 50 hours on Satisfactory 🤔

[azure ingot]

...what?

you're not missing a 0 are you?

[wind spade]

(at least half of that spent just gathering data)

no, not really

I last played in May 2019

[vast jungle]

wow

[azure ingot]

but if you don't even play, then why do you make the program?

[left fractal]

Oh wait 1 more questions
If i have 780 per min empty canister
How should i know how many stack can fill in drone

[wind spade]

(not counting ~30 minutes of testing if the game can run on linux through Proton in 2021)

[left fractal]

No

[wind spade]

I do want to play, but I like making the tool as well, people have been giving me feature requests and supporting me over the last few years, so I do what I enjoy. I want to do a playthrough of Satisfactory at some point, but right now I don't want to take time from the rewriting (which already is super slow). I'll hopefully play when I "finish" the rewrite

[left fractal]

I can help i think

I have knowledge in python

[wind spade]

the website is in javascript/typescript/angular, with backend api in PHP

[azure ingot]

hmm... did you learn coding along the way or were you already programmer before, you said you rewrote it a couple of times but was it because your knowledge increased or because the vision changed?

[left fractal]

Oh all worste then other 😐🥲

[wind spade]

I'm a website developer for last 12 years 🤷‍♂️

[azure ingot]

ohhh ok, so it's not a fully local client

[wind spade]

no, calculation is currently done on server

also, sharing functionality

[azure ingot]

wouldn't it be cheaper to offload the calculations to the client?

[wind spade]

in the future there will also be user profiles and stuff, so API will need more features

at the time when I did it, I tested several javascript libraries for linear solvers and none of them were fast enough

(also it lags the browser on weaker pcs)

[azure ingot]

ah since it's in a browser

that surely makes a difference

i was speaking more along the lines of an applet

[wind spade]

currently I'm using native C based solver that is running on the server (php just calls it and parses the result)

[left fractal]

Can i dm you?

[azure ingot]

C, the ultimate foe, i lost 40 hours on a single assignment

[azure ingot]

what for?

[left fractal]

Will tell there

[wind spade]

first rewrite happened because the first version wasn't supposed to be a long-lasting thing, I just wrote a quick'n'dirty website for calculating production, didn't expect it to get big (so e.g. data was hand-written, tons of stuff were quickly stitched together, etc.)

second rewrite that is happening now is because I didn't really plan forward and there are situations when two different versions exists at the same time (U5 and U6 for example) because of experimental branch, and switching between versions is not ideal, as well as migrating to higher version. So for now the tool has only a single datasource and it can't do multiple datasources, so I have to run two different versions of the tool, each one with one datasource, which isn't ideal. The new planned version will have support for any number of datasources, which will also allow me to support mods and other things like ficsmas.

[wind spade]

typescript and php are great, not sure what do you mean 🤷‍♂️

[azure ingot]

but now i'm wondering, do you actually need to worry about memory management on the C-side?

[wind spade]

the c library isn't made by me, I took existing one

[azure ingot]

oh that's a way to do it ig

but then that takes out the only thing that i understand (somewhat) about this

[left fractal]

i was trying to make a data base with php before for my phone app but i wasnt good at it and i couldn't make it

[oak coral]

How does that work actually?

[wind spade]

what do you mean by this?

[azure ingot]

i only know how to do a bit of c

[wind spade]

  G  G  G  G
E-+--+--+--+
E-+
E-+--+--+--+
  G  G  G  G

[oak coral]

8 generators require 360 water

but mk1 pipes only deliver 300

[vapid gorge]

look at the diagram above

[wind spade]

see the schema above

[wind spade]

I know a bit of c as well. The problem here was to do the linear solving calculations reliably 😄 the algorithms are pretty wild

[oak coral]

i used this

[wind spade]

that's basically the same, with the additional connection at the ends

[vapid gorge]

yeah as long as you are delivering water from two points and you don't gove over the 300 limit at any point

[wind spade]

  G  G  G  G
E-+--+--+--+
E-+        |
E-+--+--+--+
  G  G  G  G

[azure ingot]

actually what kind of calculations need an advanced algorithm?

[oak coral]

but how can 300 water fur fill the need of 360 water?

[vapid gorge]

it doesn't, 3 water extractors does 360

[wind spade]

the tool uses this https://en.wikipedia.org/wiki/Linear_programming

[oak coral]

the pipe line only flows 300

[opaque oak]

There is capacity for 600 fluid in that system, with two pipes.

[azure ingot]

and what calculations need those?

[vapid gorge]

but if you have 2 pipes feeding thats effectively 600 throughput

[wind spade]

pipes have a max flow rate of 300, but you have essentially two pipes

[opaque oak]

Without the extra end connection, each leg only handles 4 coal generators, and 1.5 extractors.

And the extra end is just needed in some specific cases if you have long pipes between the extractors and generators.

[oak coral]

ah so the pipe's 300 max isn't the same as a conveyorbelt with 60 max

[opaque oak]

Both are just for the segment.
Not the system

[wind spade]

well, basically anything more complex. You have things like loops, self-feeding, etc. Those can't be easily solved by just starting from the product and doing the classic tree shaking

also the tool optimises for lowest resource usage with planned updates for power/space optimisation. Since players can select multiple recipes, the tool needs to know which recipe to use

[vapid gorge]

it is.

[wind spade]

it is the same

[vapid gorge]

if 1 pipe can deliver 300 water
how much can 2 pipes deliver?

[oak coral]

600 but not over the same pipe right?

is it not the same as connecting 3 mk 1 belts in a merger and having that go to a mk1 output belt?

[opaque oak]

Where is that system moving more than 180m^3/min over single pipe?

[azure ingot]

ah so you use that for when a bauxite production feeds itself with water, that makes sense

[wind spade]

when changed to conveyors, you're moving 120/min using two conveyors

40 ->-M-->-- 60
      |
40 ->-S
      |
40 ->-M-->-- 60

[wind spade]

well I use it for any calculation

[azure ingot]

oh ok xD

[wind spade]

since the tool always finds the "most resource efficient" way to do stuff

e.g. if you select "iron ingot" and "pure iron ingot" recipes, the tool needs to know which one it should use (or if it needs to use both)

[vapid gorge]

this is 2 pipes delivering https://media.discordapp.net/attachments/558721941410807812/1066020707399643176/Screenshot20230115-10391200000.png?width=1609&height=905

[azure ingot]

so it's not just the player chooses recipes but the player chooses which recipes are allowed and the tool finds out which way is best

[opaque oak]

And the Greenys is just the same with the top middle pipe missing.

[azure ingot]

that's... pretty advanced ngl

[oak coral]

is there no maximum amount a pipe can deliver?

[wind spade]

yeah. That way you don't get things like byproducts not being used

[vapid gorge]

for the love of god.

yes. 300 for mk1 and 600 mk2

that's why you have multiple pipes

[vast jungle]

and 600/min in a MK2 pipe requires special care

so better plan with a bit below it

[azure ingot]

now that's pretty ingenious

[oak coral]

i get it now.

[opaque oak]

Left and right ones do 180/min, for combined 360/min.
So no segment goes over 300.

[wind spade]

basically byproducts are the main reason of doing that. Because if each recipe only produced one thing, you could calculate it normally. But if a recipe makes something extra, you can't just go from product to raw ingredients, because you may need to make different decisions because of the byproducts (e.g. not using the most efficient recipe for the final product, because you will have some byproduct that can be useful in different recipe)

[vapid gorge]

so in the picture you shared you have 2x mk1 pipes. Easily can handle 360 water.

In G G G G
E-+--+--+--+
E-+ |
E-+--+--+--+
G G G G

there is also 2 pipes doing the load

[wind spade]

you can't just let user pick recipe for each item, because of recipes that have multiple produccts

[oak coral]

I understand it now : )

[opaque oak]

So you can do calculations like this to check that no single belt or pipe segment has to go over their limit.

So even when the total is again 360, no single pipe does more than 90 in that one.

[azure ingot]

i understand the concept if it now, but the exact application is still out of reach for me

[wind spade]

you can check out this thread https://discord.com/channels/370472939054956546/1048004440914927628 if you want to try to understand how it works 🙂

[azure ingot]

yeah that part i pretty much got, i was more talking about the 'slack variables' wikipedia was talking about. And as far as i understand, you simply created a giant matrix with all recipes and if you want a certain amount of womething, you change the value next to it from 0 to the desired amount right?

care to share what the dimension of the matrix is?

[wind spade]

indeed

dimension is [number of items] * [number of recipes]

so roughly 150x150

(obviously depends on what the user selects, because unselected recipes don't go there and that may eliminate some items from the list)

[azure ingot]

but does it need to have all recipes in one row? i thought it'd be smaller since not all recipes have the same io

so not all appear in the same row

[wind spade]

well it has 0 in most cells

[azure ingot]

that makes sense

[wind spade]

but idk these details that much, since that's what's managed by the C library

[azure ingot]

something like elevator parts would only have 2 cells or so after all

[wind spade]

well, it has "power", "input 1", "input 2", "output 1", "machine", so at least 5

in the thread I showed simplified version, but in reality it's even a bit more complex

[azure ingot]

so it takes power into account as well, but i'd guess that has a pretty low weight right? or did you calculate it based on nuclear fuel rod materials to power rate?

since power is, in the end, also a part that needs to be produced

[wind spade]

well currently it doesn't weight power at all, it's mostly for future when power optimisation will be a thing

[azure ingot]

but i'd say that you can simply calculate the most eficient raw mats to MW ratio and add in in front, should work just fine right?

[wind spade]

well yeah, but that would assume the player actually uses that kind of power with those recipes

which is not always the case

[azure ingot]

but otherwise the player would necessarily have to input their power production line

[wind spade]

the optimisation would be either resources or power, not both

[azure ingot]

maybe you can just make a maximum power draw option

such as, optimize as much as possible with the given power

[wind spade]

yeah, that and much more is planned

[oblique notch]

Greeny does your site already have a prioritization for which recipe to use when multiple are selected or does it always calculate for something?

[wind spade]

It always had that, picks combination that leads to minimal resource usage

[snow dove]

i believe it takes the recipe that’s most efficient

[oblique notch]

thats what i thought, i just wasnt sure if you had shortcutted that with some recipe prioritization - ie you know recipe A is more resource effecient than B, so A is always prioritized - instead of actually doing the calculation each time

[wind spade]

Recipes aren't more efficient. It compares whole production lines essentially

Also depends on resource weights (those were supposed to be changeable)

And sometimes you can use multiple recipes for same item

[oblique notch]

what method did you get for picking the recipes so fast? I know the thing i was working was super brute forced and not very effective, but like mine can take 15-20 seconds and you have it optimized to no time at all.

[wind spade]

Linear optimisation

Or rather linear programming

Again, thread for more in depth look of how it works https://discord.com/channels/370472939054956546/1048004440914927628

[oblique notch]

oh very cool. those are the things I wish I dealt with more, but devops doesn't always care about optimizaton like that :p

[wind spade]

It's basically optimisation of set of inequalities based on target function

[oblique notch]

i really want to go back to school and get my CS bachelors and masters... some day

[wind spade]

I had to learn about this myself xD

Well with a little help of a friend

[primal flicker]

Would it be possible to add a m^2 optimization?

[wind spade]

wdym by that?

[primal flicker]

I've been shuffling alts around to get production from the smallest footprint. The resource and energy efficient alts generally take up more floor space.

[wind spade]

problem is you don't know how much space logistics will take

[primal flicker]

So steel screws, for example, are a godsend for me.

True, but logistics is typically stackable. And clip-able.

[wind spade]

machine m2/m3 is definitly possible

[primal flicker]

That's what I meant, yes.

I went through the wiki to make a list on my spreadsheet, but I'm not sure it's 100% accurate. I also picked a few alts that were almost the best for m^2 but are significantly better in terms of low building count.

[wind spade]

the problem with this approach is that you're looking at each item separately, instead of looking at the whole production line

[tropic hawk]

I think the best metric you need for m² is m²/items/min

[oblique notch]

It's also completely subjective by how much your producing. If you want smallest scale you can just have 1 machine of every item... sure it's slow as hell but it technically works.

@primal flicker did you take into account using alts that remove steps? Like going straight from ore to steel instead of ore to iron to steel? Or removing screws entitely, effectively removing all of those machines?

And what about factoring in recipes thst are not that far off a 1:1 machine ratio. With a bit of creative over/under clocking you can achieve a 1:1 ratio for part a to output b... meaning you might be saving a ton of space even if another recipe takes less machines for part b but many more than the difference for part a

[wind spade]

well I assume the optimisation is "lowest space per item/min produced"

[primal flicker]

I chose to use coke steel for the items/minute/m^2 it makes, and steel alts for iron parts to completely remove iron ingots. Similar with copper alloy ingots, fused wire and quickwire. Copper rotors, quickwire stators, rigour motors.

[tropic hawk]

Do you want maximize production per meter or minimize space per item?

[fierce ruin]

you can only have 127 players on a game

[tropic hawk]

It wouldn't be a continuous rate, there would be pauses for the miners to get emptied, and that would be a pain to do...

[fierce ruin]

that's abstracted away

[keen shard]

anyone know a place where I can look up and read on the object limit in satisfactory

[cinder silo]

2,162,688

[oblique hollow]

other that that there is no info as we know nothing else

theres a bit of vague info i know but literally nothing thats substantial or helpful

just a bit of trivia

[deft lichen]

https://satisfactory.fandom.com/wiki/Tutorial:Unreal_Engine_Entity_Limit

Snutt added some information to that page once, so we can assume it is accurate from CSS's pov

[keen shard]

Thanks lads, you are stars on my heaven.

[cinder silo]

The object limit stopped me for a few days until I found the way past it.

2162688 seems such an arbitrary number though doesn't it.

[keen shard]

thought it was a binary numbers thing, something something 64s idk im so bad a math

[last saddle]

should i

• focus on the most advanced products having pretty numbers to optimize intermediate production, but risk the modularity of my factory
• focus on the primary and intermediate products having pretty numbers so i can easily plan and expand my factory, but only know my capability of producing advanced products after everything else is established

[oblique notch]

Whichever.

Or neither. I build chains for a single product - a product I plan on storing for use with the build gun- and rarely ship from one factory to another.

[last saddle]

i just grab lower grade products from my factory's overflow
by the time i prepare for the most advanced products in a tier i'll have a ton of the previous tier's materials, so i don't worry too much about exclusively producing them

[oblique notch]

I don't build for massive numbers either though. I am usually producing of any given"Final" item 60/min or less. Tbf this is because I concentrate more on building cool structures and using the machines to add movement and purpose to otherwise empty buildings.

So I produce like 1500 concrete a minute... and like 30 copper sheets lol.

[last saddle]

fair
i like to optimize to a limited extent
i barely had more than 30/min rotors and reinforced iron plates last time i hit tier 5
and it was definitely enough to pass objectives at a desirable rate

[cinder silo]

Apparently five of four hundred train cars defy gravity.

[rustic patio]

are there any calculators for comparing production chain weighted point efficiency?

[vapid gorge]

you could get the same effect by selecting the final product you want sunk being the points you want?

[oblique notch]

not really, becaues the point values are generally just about the sum of the point values of the parts used to make it.

[rustic patio]

yea i missed the part where they always used the best sub recipes to weight the alt recipes

i thought they used the default recipe to weight higher level item alt recipes

[fierce ruin]

which one is meta

[rustic patio]

https://www.satisfactorytools.com/production?share=eVE2kyHkbpQVJvzstjTC

is this a good idea?

the numbers are kinda funky, but i plan to just export all excess resources to other places where they are needed

[vapid gorge]

whatever you think might be useful for your next build?

there's no 'meta' recipes

[fierce ruin]

dont need any of them

[rustic patio]

i dont like steel rotor cuz u cant make rotors and stators with the same items with that

[fierce ruin]

kinda got f'ed on those

[rustic patio]

turning something rare into something less rare is also not desirable, so id take 2

[vapid gorge]

whatever you think you'll need next then?

[fierce ruin]

are they needed after stage 2 of space elevator?

[rustic patio]

yes, the items from lower space elevator stages are needed for the higher ones

if you use the default smart wiring recipe you will need 1.4 million wires to make all the space elevator parts

[fierce ruin]

thats ez

[rustic patio]

its the highest number in the game

[fierce ruin]

i make 60 per min on one constructer

im taking 3

[rustic patio]

id use the alt that makes it so u can make them both from the same materials

[rustic patio]

are the numbers too weird?

[vapid gorge]

Weird?

I find it odd that you have the rare case where Heavy Flexible Frame recipe would be good and not using it?

[rustic patio]

well, oil is quite rare so i dont want to waste any rubber

and i dont like screws

[vapid gorge]

probably do sitched plage and wet concrete then?

rubber concrete is great output per limestone but wet is only somewhat less so and water is near free

sure there are times you might want rubber though - I don't know your exact situation

[cinder silo]

Due to a massive overflow of rubbers I went with adhered plate and still too much rubber.

[vapid gorge]

sure but that's not a super common problem xD

[cinder silo]

I went nuts processing all of the spire coast and dealt with the consequences later, all the plastic is used but at least 1000 rubber is going to sink.

[rustic patio]

how much oil did u use from the SC?

[cinder silo]

All of it, well except the two impures at the far west.

[rustic patio]

so uhh

wait, the ones near the desert too?

[cinder silo]

This is the place that processes it all.

[rustic patio]

not sure if that counts as spire coast

[cinder silo]

Yeah I grabbed those as well.

[rustic patio]

soo 3150, nice!

so you're good with pipes?

[cinder silo]

Due to a mistake in initial intake, I used default rubber/plastic in the 120 induction refineries, 80 more refineries recycle using fuel made from the HOR for final output, the leftover 1200 fuel goes next door to a 15Gw power station annex.

[rustic patio]

nvm i think i found the mistake

if that was it then ill catapult myself into space

[cinder silo]

Cannon experiment! , for science.

[rustic patio]

time to test it

[vapid gorge]

A big recycled plastic/rubber plant is a test of someone's pipe skills thats for sure

[cinder silo]

Mine took some major troubleshooting and identifying two broken pipes.

[vapid gorge]

yeah I did a 3000 oil recyc plant. That was a time xD

Spent too much time trying to make a vertical pipe manifold to work

[rustic patio]

the processing part is all done and works great, my problem rn is feeding the oil into the refineries

[vapid gorge]

are you shipping it by fluid cars or something? That should be the easiest

[rustic patio]

well, i built all of the refineries and blenders that turn oil into fuel first, then i hooked them up to plastic/power one after another

[cinder silo]

Vertical, yeah I had issues there. here is my fuel & HOR nightmare during construction, this pipework didn't fly.

[rustic patio]

as i got to utilizng 80% of them the oil wasnt distributed evenly anymore

im currently trying to figure out why

[rustic patio]

pipes

[vapid gorge]

ahh ok so you're not sure if the whole thing works yet.

[rustic patio]

yep

well, every 5% of production buildngs are theiyr own section

[vapid gorge]

I mean there's the basics

• Did you flood each step of fluid before activating the next?
  -Looped?

no weird merges/splits that aren't needed

[rustic patio]

so i 100% know that if i get enough from the first refineries everything will work out

[vapid gorge]

Do some overhead shots and maybe we can troubleshoot

[rustic patio]

im currently trying to see if the current version works, i moved from bottom feeding to top feeding

[vapid gorge]

ah yeah bottom feeding is possible but you need to take extra precautions

[rustic patio]

my aesthetics are ruined 😭

[vapid gorge]

for example

[rustic patio]

/hj

[vapid gorge]

not polished yet ofc

still needs a bunch of aesthetic work but it works at 100%

[cinder silo]

Problem with mine was all the HOR was connected, and gravity is a thing so the top two floors kept starving the blenders, and that made the rest fold up quite efficiently.

[vapid gorge]

yeah it might be possible to make a smooth functioning vertical pipe manifold but ... I'm not going to throw away hours on it xD

[cinder silo]

I blew about 8 days before isolating each floor, it was way too problematic.

[rustic patio]

all green

not

[cinder silo]

D'oh!, what went wrong 😦

[rustic patio]

ah, nvm. i turned on all the generators and some machines were still idle cuz they were still full with fuel. all green so far

[cinder silo]

Cool 🙂

Check in on it from time to time, just in case though, I found a huge flaw in my turbo fuel facility after more than a month of operation caused by intake issues at the oil>Hor end.

[rustic patio]

id never do turbofuel... even with normal fuel i havqe so much that i cant place enough fuel gens lol

[rustic patio]

how did you do it?

lik, what layout did u use

[cinder silo]

There is always room, I built 888 fuel generators as shown here https://www.youtube.com/watch?v=_rLV_YcVt8s

[rustic patio]

i know, it would cost me my sanity tho 😭

i have 320 rn

i plan to reduce it to 0 once i get nuclear, 100% rubber and oil

[cinder silo]

Tbh once the nuclear is up, keep the existing fuel either running or in mothballs, it only takes a recipe change to knock out your reactors.

I'm not saying I don't trust coffee stain or anything, oh wait I am 🤣 , I tend to hope for the best but plan for the worst.

[rustic patio]

id leave them standing anyway

if nuclear gets bricked i could just reconnect the pipes from rubber/oil to power

every section takes/makes exactly 345 so it fits together neatly

[cinder silo]

Leave them running, and have them also charging a fat array of power stores, it never hurts to have a backup in play even if you have the power disconnected from the grid with one switch that has huge billboards screaming out "this is backup power ~ don't fuck it"

[rustic patio]

ill charge a battery, yes

but i wont leave them running, i need rubber!!

[cinder silo]

Imho tap a different group of oil.

[rustic patio]

but i want to tap all oil and turn it all into rubber! i cant go to space for more oil. well maybe in a future update, but then that would need to be converted to rubber and plastic too!

ill just make a giant power storage, if nuclear dies ill use that to restart my oil and then i can use that to restart my nuclear

but since i sitll dont have nuclear, no need to worry

[cinder silo]

You must have a thing about rubber 😛 , tbh I have far too much spare but I can drone that to another site to recyc to plastic.

[rustic patio]

well, ill be producing most of the rubber via recycling

[cinder silo]

If you want huge production though, you want residual > recycled going, that'll produce more than dozens of players could use.

[rustic patio]

thats what im doing

heavy oil residue to diluted fuel to recycling

ill turn 3450 oil into 9200 rubber/plastic

[cinder silo]

Easily 2-3x what I produce with my default mistake going in to recycled.

I can fix the error, problem is it will be a huge pain.

[rustic patio]

how do you recycle?

[vapid gorge]

for the vertical manifold? I tried a dozen different ways. Couldn't get it to work

[rustic patio]

wdym vertical manifold?

[vapid gorge]

having htge pipe go up and down while branching off and feeding machines along the way. Just not a good idea

[rustic patio]

sounds complicated

[rustic patio]

i mean the recycling, what layout did u use?

[vapid gorge]

oh fairly straight forward layout I guess?

[rustic patio]

wtf is that 🫢

[vapid gorge]

that was one of the vertical manifold attempts xD

the 2nd picture is the working version

[rustic patio]

how does it work?