#math-and-meta
1 messages ยท Page 444 of 1
i need 37,5 iron for a constructor, but can only go with 37,4 or 37,6. Whats better? a bit more or less?
ok, Im just afraid of the overflow that i will have in 50 years XD
50y? when update 4 is coming out?
Yo guys, for you, how many foundations is necessary (width and length) to have a plan that can go to the most complicated recipe ?
I would like to do an only place where i can make produce everything like smelters, constructors, assemblers etc.. (power is gonna be on a other place). I'm afraid of getting to short coming to uranium recipes.
I was thinking to do 60 width and 100 length.
What do you think ?
I know, but i still want to stay close to the ground. I know hyper tubes exists, but i want my base, to not roll over gravity rules too much ^^
than you need a lot more foundations
i have 81x160 foundations just for nuclear fuel rod production
and 80x111x4 for every other production :D
So what would be the right choice if i wanna separate my base, like uranium in a place, oil in another one etc...
its up to you
Im finished with automating all things i have. only rotors need to be automated anymore. How to you like it?
Any tips for improvement?
Or do you see somewhere a misstake?
Interesting. It seems that with all the new recipes, sending a 60/m dedicated belt to storage, of everything, seems possible
There is even head room
takes 2 seconds to test it...
yeah I just did
its been my experience that things not doing anything, dont use power... so my assumption is no...
I needed to know before connecting this
ngl that looks like a massive waste of materials on pumps
are you spacing them every 18-19m?
looks like every 18-19cm
but it looks super trippy
also I just did a test... pumps are ALWAYS powered regardless of if water is being moved. their animation is going, or if water has stopped moving or if they are unable to punp (due to too large a elevation change) or if all storages are full and water requirements are met on the other side (for generators or refineries)
as far as I can tell there is no situation where pumps are not taking power...
@vast copper definitely might wanna change your layout up.....
Where did you find the need to move that much liquid?
I mean late game youll need a few k/m of some resources to acomedate high level materials or power requirmenets...
True true
60 super computers / minute uses almost 10k oil/min worth of plastic.
Fluid flows through pumps even if they aren't powered, so they work as one way valves, but they won't add any height without power.
Makes sense
looks like every 18-19cm
@nimble knoll every 20m, you can zoom in screenshot taken from far away and it's has this visual effect
is a 48 -> 21 | 27 splitter even possible? without breaking it down into like 3 -> 1 | 1 | 1 splitters and re-merging it all
use manifold splitting?
So in my preliminary work before updating my calculator, I've made a visualization of the entire recipe graph.
The boxes are recipes, the ovals are items.
The red ovals are resources.
Also I left out the recipes for unpackaging fluids.
The next step is to pick out just the subgraph for the oil stuff.
Okay, here's just the stuff made in the refinery.
oh yeah i can understand that to a degree
Hmm, but I think it's missing something.
Ah, it's missing the petroleum coke recipe.
Oh, it's missing a bunch of stuff.
hah, one sec.
@worn moat
Take the 48 and remove 6
Using 3 consecutive 2:1 splitters the final result will give you a pair of 6s, re merge everything but a single 6 and then split the rest in half and add the 6 to the side you want to be 27. As far as Iโm aware there is no other way to split resources like that
The other option would be to just flood both the 27 and 21 just by splitting to both and waiting for it to backlog
Okay, there we go:
That's the recipe graph of everything that uses the refinery.
I could maybe throw in some colors to clarify things, but for a quick and dirty first pass, it will do.
Ah! So I am.
@reef turtle you make the graph seems like you are making turbomotor until i click it open, lol
Added it:
Also HOR alt
Ah, yes, I missed that, too.
Do you want me to export you list of all refinery recipes to check?
Well I don't see anything essential missing, but it's hard to check on phone :)
Here's the updated whole-graph image:
ah my eyes
Looks pretty simple
Let me see if I can hack in some colors quick-like.
(There's a sophisticated way to do it which minimizes shared colors on touching nodes, but I think I'll skip that for now...)
Would be nice to have each recipe combine colors of it's ingredients and each ingredient would have average color from all the recipes (and we start by giving some colors to raw items)
Or it would look like a crap xD
Oh and also loops
Nvm
Well, that... sort of helps.
debatable
Colorized refinery graph:
So I hadn't looked too closely at the aluminum stuff until now.
Making alumina solution also produces a small amount of silica as a byproduct.
And then you need a larger amount of silica when using the base aluminum ingot recipe.
Math on (If mining every node on maximum and turn all mats into products) Energy consumption ๐ ?
I'm think I'm coming to the conclusion that you basically need a sink hooked up to your polymer resin output.
Why not just spam make fabrics
The way I'm thinking of it is: Pick an item. Suppose every other item is backed up. Can I make more of that item?
So suppose plastic and rubber are full.
Now make fuel.
Priority splitters would help a lot, but c'est la vie.
Maybe I will make that version with the many splitters/mergers after all.
Or train station
Takes a whole lotta space to do that.
In any case, I'm thinking that with some manner of priority splitters set up to bin excess polymer resin and petroleum coke, you can have a totally resilient oil processing system.
But I still want to run some more numbers and actually build it.
I know we don't normally speak mods here, but I have a math question about a mod to see if I got this right. Is that ok?
just pretend it's not a mod: D
๐คท I don't care but I think the motivation is that it would get confusing to be talking about modded mechanics, and misleading to new people who'd have a hard time discerning what's in the game and what isn't. So in that spirit, probably not okay, not that I have the authority to say no.
But if it's really quick...
well shoot. Now that I'm trying to explain it I'm confusing myself even more! lol
It hurts itself in confusion!
it was very effective
ya, I'm trying to write it so I don't confuse everyone even more.
We can be confused together
we can be fusion together
np ^^
whats the best alt for heavy modulare frames? ๐ฎ
Heavy Encased Frame for me. Heavy Flexible Frame, can make more/min but you add rubber to the mix and that is why I prefer the other one. The Encased Frame one, uses the same resources as the original, but you don't need screws, they are replaced with concrete, witch you need to make the Encased Industrial Beam.
??? the original ingredients (Mod Frame, Steel Pipe, Encased I-Beam, Screw). Heavy Encased Frame (Mod Frame, Steel Pipe, Encased I-Beam, Concrete). Heavy Flexible Frame (Mod Frame, Rubber, Encased I-Beam, Screw).
So Flexible needs more Modular frame, more Encased I-Beam, More Screws.
Encased needs less Modular Frame, slightly more Steel Pipe, slightly less Encased I-Beam -> output goes up by about 1/2 from 2 -> 2.8125/min
Concrete I find very easy to make, and I almost always have an excess of it. So the fact that I can remove screws from the recipe and use my concrete and keep everything else the same and gain more in output is a win for me.
yes i saw this all in the wiki...
BUT i need the double amount for Beams, more of double of Pipes AND Concrete only for the production. but okay i will thinking what is a good alt for my, thanks
Find the alt for Encased Industrial Pipe. That should really make your Encased Industrial Beam production much better.
I say easier in this way. Encased I-Beam (24/min Steel Beam, 30/min Concrete) -> 6/min
Encased I-Pipe (28/min Steel Pipe, 20/min Concrete) -> 4/min
So at first it looks like it's worse output.
But to make steel Pipe it's (30/min Steel Ingot) -> 20/min Steel Pipe. Steel Beam is (60/min Steel Ingot) -> 15/min Steel Beam. So I can use the same amount of steel ingot and have 40/min steel pipe vs 15/min Steel Beam. So I can take the same input for Steel Ingot, and have almost enough to make 2 Encased I-Pipe (It's really 1 1/2 ish).
okay, this recipe i us it rn
so im looking for it ^^
thx to greeny too ๐ cant wait you finished your update 3 c:
how many frames/m makes you with that rn?
From the #faq check out (https://daniel2013.github.io/satisfactory/calculator) and use it to plan out the recipe. You can then see what fits best for you given your set up.
That would be easier to compare the results.
I say that b/c I'm not at my computer right now. Technically I'm "working" so I can't play at the moment.
okay^^
is turbofuel worth or same like in update 2?
its super worth
Fuel gen use:
fuel 15mยณ/min
turbo fuel 4,5mยณ/min
so turbo fuel can provide more than 3 times as much as fuel
okay ๐ so with heavy turbo fuel or normal turbo fuel?
both
depending on if fuel is a byproduct or a main is what type of turbo fuel you will choose
yea okay!
Here is my final Plan for the basic Production. Do you like it? Any mistakes or ideas for improvment you see?
Manifold instead of balancers
manifolds are just splitters in a line instead of trying to split everything evenly
Ok, I will look for a tutorial or something
overlock machines is a waste of shards, power and so xD
200% machine is using like 400% power
it's really a preference, if you like splitting that way, go ahead. It only gets out of hand when people are trying to split 7 lines into 3 or something odd
that's a bad example now that I think of it :p
manifold is just easier and u can easily expand it
Ok, I will use the tips in my next setup, ty
np :)
ned help spliting got 1200 resorces and net split tham in 240 resorces for 5constructors
why not manifold/overflow?
?
i think got mk4 only
How many conveyer belts are you bringing it in on
non i got 1200 outcom. ned to split it so i can put in 240 in 5 constructs
isnt there som program who helps ?
yes
So what you're going to want to do is take the three, put a splitter on each of them
480 each? so three conveyors?
Place three different mergers
ou wait
Then connect each splitters output to a different merger
That gives you equal lines
Now, it's time to create the 3-5 split
So here, you're going to take the initial input and split it by two
The one going to the left will go into a splitter to turn it into three outputs
The one going to the right does the same, however, the third input goes back to the merger that you place at the beggining
Replicate that on each 3 outputs
And now you have 15 outputs
Merge 3 of them
Each
Now you have 5 lines of rp
240
Here, I'll even draw upa blue print for rhis
i got 8 refinerys whic ech gives out 150 so together its 1200
where is @wind spade when we need him xD
basicly 8 ref. 1 producing 150 all 1200 ned that amount in 5 asemblers bu 240 each
if i merge tham its 300 or 450
wut
Here
what do you demand mortal
S not the most effective thing
But it works
Greeny, they want a 8-5 splitter
Im sending them one rit now
We have the topic again between manifold and splitting
I'm not as good at explaining what manifold is like you
manifold is great, i just wish i had faster belts for when i max out the mk5
quickwire smh
manifold has slower start up time, but much faster build time
And fewer headaches
manifold is easier expanded, can work with different speeds etc
@upper crater just use manifold and not splitting items :D
i end up doing stacked manifolds with the uppers feeding to the lowers later on when belt capacity is maxed, is that the normal way to handle it?
whats the termin manifold?
overflow method
Basically just a bunch of splitter in a row
@harsh fractal I do stacked, or reload from the side, depending on which works better in the surroundings
cool, good to know
im trying to understant this blueprint u send me. basicly its fil 1 asembler and so its full it go to 2 and so on?
ok. time to build andsee how its work in game. an i was trying to figut it out for 4 h whit spliting...
spliting is just Loss of time and takes up too much space
Manifold may take longer for everything to run at 100%, but you can also do other things in time
what do you prefer, bolted iron plate (with screws) or stitched iron plate (with wires)? bolted are time efficient but need a lot of screws
screws are cheap af so i would go for bolted bois
I am missing something with Greeny's consumption calculator? https://satisfactory.greeny.dev/consumption Been working great for everything, up to steel beams/pipes. For steel beams, showing a 1 to 1 ratio of foundries to constructors. But a foundry pushes out 45 steel ignots per minute, and constructor using 60 per minute (For steel beams that is)
A powerful Satisfactory calculator, filled with features, supports overclocking, alternate recipes, live map, power calculations, etc.
Greeny didnt updatet it to u3
Ah, didn't realize that was affected.
Glad I caught it early, because my steel beam factory was just finished, didn't take much adjusting to make it 100% efficient
Alright. I think today I am going to be crafting some linear programs to start playing with some actual oil numbers.
Hopefully I can find where I stashed the code from when I did this same preliminary work for Factorio...
(Ah! Found it.)
... Huh. That wasn't a result I expected.
I realize this is probably incomprehensible on its own, but I got this for making 60 plastic/minute:
[('plastic', 0),
('recycled plastic', 5),
('residual plastic', 0),
('rubber', 0),
('recycled rubber', 0),
('residual rubber', 15),
('fuel', 0),
('residual fuel', 15/2),
('alt heavy oil residue', 155/16),
('alt polymer resin', 25/8),
('water', 60),
('crude oil', 765/16),
('surplus plastic', 0),
('surplus rubber', 0),
('surplus fuel', 0),
('surplus heavy oil residue', 0),
('surplus polymer resin', 0),
('surplus water', 0),
('surplus crude oil', 0),
('cost', -765/16)]```
The numbers are recipe-rates.
So apparently the most oil-efficient way to make plastic is: Turn oil directly into HOR and polymer resin; turn HOR into fuel; turn polymer resin into rubber; turn fuel + rubber into plastic.
Net effect, it makes 60 plastic/minute using about 47.8 oil/minute and 60 water/minute.
And no surplus production.
Though this may be predicated on running with a perfect ratio, which is probably impractical. I should work out how this translates to numbers of refineries...
You can make is so that you get 3 rubber/plastic per 1 cruid oil. With the right setup
I'd be interested in seeing what that setup is.
Ahh, the dilute fuel thing.
I didn't include that in this linear program I was testing with.
Which tool is that? I'd recognize dagre-d3 anywhere.
Adding the diluted fuel recipe does indeed arrive at that solution as well:
[('plastic', 0),
('recycled plastic', 170/27),
('residual plastic', 0),
('rubber', 0),
('recycled rubber', 70/27),
('residual rubber', 10/3),
('fuel', 0),
('residual fuel', 0),
('alt heavy oil residue', 20/3),
('alt polymer resin', 0),
('packaged fuel', 0),
('diluted fuel', 80/3),
('unpackage fuel', 80/3),
('packaged water', 80/3),
('water', 200/3),
('crude oil', 20),
('surplus plastic', 0),
('surplus rubber', 0),
('surplus fuel', 0),
('surplus heavy oil residue', 0),
('surplus polymer resin', 0),
('surplus packaged fuel', 0),
('surplus packaged water', 0),
('surplus empty canister', 0),
('surplus water', 0),
('surplus crude oil', 0),
('cost', -20)]```
That must eat a whole bunch of electricity.
the only cost of power is manhours invested in building infrastructure ^^
And also the finite global supply of resources which can be turned into power.
Sure.
Bauxite and Caterium are more limited than power resources, you'd run out of these long before running out of power if you're optimizing for top tier item production
Yah
Uranium is only goo for nukes
And I'll have to see the max you can pull off of uranium atm
I mean assuming that it's all still the same
Each rod would last five minutes
And produce 2500
The most efficent uranium cells
Produce 17.5 per minute for 20 per minute of uranium pellets
And that 17.5
Becomes 1.05 per minute for each 20 uranium per minute
We can produce in total
119.7 nuclear units per minute
Which would supply 598.5 nuclear plants
Giving us a grand total of
1,496,250 MWs
MWh is a quantity. A Watt is a Joule/second.
A MWh is a MW times an hour, or a constant multiple of a Joule.
When you get your power bill, it's usually denominated in kWh or such, or the amount of energy you consumed.
can someone please help me with math? i dont get it
so i have 1 nuc. fuel rod per minute and i have 2 nuc. power plants already. both are runnting at 100% and 2500 MW peak power. how many nuclear power plants can i build? (included the other 2 plants)
I admit, I don't know off-hand the rate at which nuclear plants consume fuel rods.
I knew it at one point during Update 2, but I've totally forgotten it, and for all I know it changed in Update 3 anyway.
Hmm. So I've got this idea that simply solving the entire recipe graph in a single linear program could be fast enough for the purposes of my calculator that I could do so in place of a graph traversal.
(The alternative is the hybrid approach I take in my Factorio calculator, where I extract a number of subgraphs which are then each solved as a linear program.)
The constraint is that it needs to be fast enough that I can compute the solution every time the user clicks on part of the UI.
The complexity is superlinear, I think something like O(n^3), so it's fairly sensitive to the size of the graph.
Right now the calculator requires the user to manually choose the alt recipe for each item. Being able to just throw the whole graph into one linear program would give it more of an ability to pick which recipes are most optimal.
And, I suppose, there are probably opportunities for optimizing the linear program solver that I wrote.
But there's some inherent slowness in there from the fact that I don't just use regular ol' floats for the calculations.
is there a design somewhere for an optimal screw/reinforced plate/rotar setup?
I've read through this channel for the past day or so and checked the pins, haven't found anything yet
Have anyone made a calculator to help with item/s when transporting with trains?
No, but the math there depends on the distance being traveled.
Also on the stack size of the item.
You can talk in terms of the number of cargo wagons arriving per unit time.
What I'm really trying to find out is a way to calculate if trains are better/worse then running a long line of belts
the math should be (items * stacksize * xInCart * carts / timeFromAtoB)
There is also the limiting factor of being able to load and unload cars with two belts per.
A freight car has 32 slots.
station has 48 slots
Also: It's 780 items/minute on a belt.
even better
So 1560 item/min
Or about two minutes to load/unload a car for an item with a stack size of 100.
So if you can complete a round trip in less time than that, you're beating the belts.
basicly, from a to b in under 2 mins including loading+offloading ๐
a to b and back.
Or, more to the point, it's the rate at which you can get freight cars.
So conceivably you could run multiple trains on a loop.
But the game isn't very good at that, I think.
no train loops are wonky at best, currently doing double-headed trains
And of course, there's the factor that running a big long track is less of a pain than running two belts and power poles over the same distance.
Or more than two belts.
You can scale the train up by adding cars way more easily than adding more parallel belts over that distance would be.
Do we know how long the unloading / loading animation locks the station?
I don't.
25-30s iirc
(Satisfactory's trains really do just make me miss Factorio's vastly more sophisticated train system.)
so even with the fasts belts you would not beat the loading animation
The un/loading animation takes 25 seconds.
https://satisfactory.gamepedia.com/Train_Station
2HD trains is wonky for multiple stations / complex network
I want better wait conditions.
conditions in general please
I want a loader station to wait until the train is full, and I want an unloader to wait until the train is empty.
it might come eventually
yeah
i want proper train collisions too.
Yeah, part of the issue is that you would basically need a whole train signalling system.
A la Factorio. And that can be a lot.
it would be interesting to see how they're gonna implement it and how it will differ from factorio.
go on
lmao, if anything, 3D is more complex than 2D
3D means you can have tracks cross without actually intersecting, but it doesn't really simplify the implementation of junctions.
Although impractical, we can have cloverleaves or even stacker interchange in SF. That means only y-junction will be involved
and how do that makes it simpler? a junction is still a junction.
and train collisions needs to be checked everywhere, not only junctions. you could've 2 trains on the same track going at different speeds. You could've tracks at different Z levels, and if the track above is not high enough, it should collide as well.
you could even have trains going in the same tracks but different directions, crashing into each other. Something that's not possible in factorio due to signals.
assuming you set it up
enlighten me, how it's simpler.
๐ train signalling delayed by 1 year
if anything signals in this would be much more complex
"how about keeping trains noclip''
''good idea''
i feel like collision check wouldn't be that hard to program
anything in 3D will always be more complex than 2D, generally, cause you have 1 more axis to worry about.
also angles are weird
best case scenario, it's equal, but it won't be simpler.
collision check wouldn't be hard, you're correct. What's hard is the system preventing that collision.
you can't just give player train collision w/o any system to make the players prevent that
imagine factorio w/ train collisions but w/o signals to manage it.
Can. All train tracks to be 1-to-1 system. All junctions handled by transfer station
If we get collision on trains, tracks would have to have a boundry to assure your trains can pass landscape and player build things at any given time
If they keep the tracks as they are right now, you would have to have a train checking everytime you build a new line, then incase something is impassable you would then need to move said tracks without any way to angle the tracks manually
#hellOnWheels
Whenever I play with trains in Factorio, I think to myself, oh man, if only we could build vertically and build like complex clover leaf intersections. Having proper signalling in SF for trains would be glorious
There's always this:
they could probably add it so collision only detects player-built stuff, so terrain collision doesnt matter
I would be content with detecting only collisions between trains.
^
Cloverleaf intersections aren't really efficient in terms of throughput
Versus what? A roundabout?
Nope, that's even worse
Issue in cloverleaf and roundabout junctions is merge before split
greeny knows trains. haha
If I use the diluted fuel with water and that I transform it in turbofuel how many electricty can I get with 300 oil/ min?
@fierce ruin like 22,000 Mw
@fossil thicket ok thankyou I think that's enough ^^
@reef turtle what is the purpose of this design? isnt it always more efficient to do something where each rail is multi-directional? ignore the missing top section here.
A rail supporting traffic in either direction greatly limits the number of trains which can run on that rail at the same time.
that would only apply if trains have collision
Well, yes.
But if trains can ghost through each other, then you only ever need the one rail.
not necessarily when you have different directions
That is to say, junctions and signaling are non-problems.
Perhaps I did not understand what you meant by, "each rail is multi-directional."
the turns
in your screenshot i see that the middle top rail is unable to go up or down
so like, youre limiting the directions a side can turn to for what reason
oh, it goes down and right but not up
shit nvm
i see now
thats a really wierd way to make a junction
was this created to eliminate slow down?
i see now with the signaling
so instead of crossing 2 other rails to go down, it does it before the merge
Stack interchange has the highest throughput imo. But they won't allow any u_turn
That intersection is designed for high throughput with no roundabout and a relatively compact use of space.
And it shld be made large enough that any segment is longer than the longest train
(Roundabouts can sometimes screw with the pathfinding in Factorio.)
There's also always: https://en.wikipedia.org/wiki/Diverging_diamond_interchange
Diamonds are in the first place, the cost effective nerfed version of cloverleaf. So i doubt it would have the throughput.
But it is safer than normal diamond
but isnt all of this currently irrelevant with how trains work
i dont play factorio but I've played a lot of openttd
Yeah, Factorio draws a great deal from OpenTTD.
Eew..... diverging diamonds? You need the patented Markus Reese square-a-bouts
DD Disgustings
Trying to link to my pic on steam, but on mobile it seems to do something wierd where it wants to download pic when I try to go to picture. So instead, the picture hub?
https://steamcommunity.com/sharedfiles/filedetails/?id=1799853967
what is the tractor/truck/train speed comparing to conveyors?
depends on how fast it goes ๐ but speed is pretty much irrelevant, since you care about throughput
thats mine question. about throughput an average tractor/truck/train route
depends on distance between stations and number of trucks
can't be really generalized
distance isn't important; average speed is probably ๐
early it is hardly limited with stations conveyors; but i wonder if at conv mk. 3 or higher there is still a limit
assuming relevant routes (0.5-5km)
and in the end you're still limited by belts
yeah, so 0.5km has way way way more throughput than 5km
about 10+ times the throughput
and about this i plan making multiple stations on both ends. Like 4 loading and 4 unloading for each truck route. So need math to match trucks to stations ratio
tbh if you need high throughput like that, I'd just go for trains. Trucks are pain when they are multiple on a single route
collisions?
and setup
i make elevated highways 200 meters above ground so setup isn't too hard ๐
just pres W button and wait
well the collisions are still there
also the need to provide fuel
and if you are doing highways, then you can just go for trains anyway
need unlock trains first. and using trucks mostly for coal; so don't need fuel
also "setup isn't too hard" > "I make elevated highways"
๐
well, it's your call in the end
but there's no math to cover this, you need to figure it out by calculating your lap time, truck inventory size, item stack size, etc.
*no precalculated math
it's hard to generalize something as random as trucks
For coal isn't probably. 24 stacks x100=2.400 coal with average 30-45km/h speed (500-750 meters per minute).
so it looks like about 1.000 items/min supply
for 1km road
it takes about 7 hours to unlock train in my playthrough. Not sure if a proper truck setup will use less time.
There's probably a more scientific way to do it, but I just keep an eye on the how long the container is sitting full / empty on each side to know if I can add more trucks to draw from a pick-up stop or if a drop-off stop is starved for resources.
What is that?
uh lol this is for Satisfactory related math
@clear gulch Use Thales' Theorem.
AB = 4, now any ideas?
Not sure what you're asking.
But, as Ondar said, this isn't really the forum for general geometry homework help.
i need some project for 8 to 8 conveyor MK2 equalizer (can use MK3 at short range)
Wait for mk6 belt, then your project ends
nope; i'm just doing iron mill for 960 iron/min for midgame
i don't want wait infinity of time to get higher tier conveyors
@sudden pumice 270 is mk3 belt 960/270 = 3.53333. 4 belts (120 + 120 = 240 in 1 belt X4)
its unreal dude)
I'm trying to set up an efficient modular framework factory and am already confused
@sacred blade whats wrong?
I have no alt recipes so I need 3 reinforced iron plates per minute, but they get produced at 5/min/ So maximum efficiency would be 3 plate constructors and 5 assemblers, right?
better do 10 per minute right away
and that assumes that I have enough rod production to fill 5 assemblers
@clear gulch nah; i found a solution. 32 smelters to 3x32 constructors (ingots->plates+ingots+screws) so 8 belts; each one connected to 4 the same type constructors
probably it will dry out sometimes but i'm too lazy for exact math
okay. that helps
Just through it in one of the calculators and it got me a full number of each machine except for the assemblers at the very end
I just need 14 miners now
@sudden pumice
we probably did not understand each other
what your normal ratio between iron plates; rods and screws?
I think I have three green power slugs?
use shards only in miners
you can scavenge for them more ๐
so if I put one slug in a miner, how much does that overclock it?
@sudden pumice
we probably did not understand each other
@clear gulch this is mine project. i wanted to equalize at few point conveyors from 32 smelters.
@sudden pumice try to make a modular production, but in my base 90 plates ~600 scres (i don't remeber) 90 rods
so I can cut my miners in half if I get 2 shards per miner
@boreal cypress agree about shards in mines only
how long do you think it would take to find that many shards?
so I can cut my miners in half if I get 2 shards per miner
@sacred blade jeah
@sudden pumice try to make a modular production, but in my base 90 plates ~600 scres (i don't remeber) 90 rods
@clear gulch now i'm planning to overkill production with 32 constructors for each thing. At least some unwanted will be idle ๐
4 hours to find 11 slugs?
@sacred blade you can use this to find slugs
https://satisfactory-calculator.com/de/interactive-map
Satisfactory Helfer zur Berechnung des Produktionsbedarfs
it feels like spoiling the game experience though
you can unlock scanner for slugs in mam
@sudden pumice and then make everything from steel
@clear gulch next build will be steel mill; also planned for about 1000 iron input
@sacred blade like said up top. SCANNER IS ALL FOR YOU)
got it. I will use that. thank you for your help
np)
@sacred blade also to find slugs elevated highways are nice thing
@sudden pumice r u using coal generators for this?)
@clear gulch currently 16. But can easily double coal gens asap
I'm for big productions. But they are useless at the initial stage, that is, before oil.
1 pure and 4 normal mines tapped; next untouched mines in about 1km from base
oh are you serious? my coal set up from last night makes just not enough power
its make 400 MW and I need 402 MW
@sacred blade omg) 2400 from 2 mk2 miners
Checkmate - make 600MW coal plants (8 gens each) ๐
that will be much nicer for power math
I have an 8 generator setup with one mk1 miner and 3 water pumps
1 gen = 15 coal per min. x8 = 120. really better dude
@boreal cypress nuclear energy is good?
@boreal cypress it was a shit on release
i dont have nuclear now, 50GW max is just fuel right now
but soon (when i finish my water supply) i have 144 Nuclear Power Plants
wdym? :D
wait a second
i wait 2 seconds just for you
cool, but i'm lagging
oh man; 3 oil places in less than 1km to mine base. how to use trains when everything is so freaking close ๐
thats not a desert xD thats water
waaaaait
real Map and so aren u3
@boreal cypress
was there oil before?
@oh god
Iโll go look for messages from last year
@boreal cypress it was my favorite place for generatord
@clear gulch that map is outdated, dont use it
Satisfactory helper to calculate your production needs.
@sacred blade add one coal gen to maximize efficiency, you can have up to 3 fully overclocked coal gens for every 1 water extractor
The two hours spent designing this better be worth it
now I need to go find enough power slugs and get enough resources to build this
whats a .svg?
similar to a pdf
I didn't download it as an image cause the text is small and unreadbale
pdf work; svg no ๐
pdf downloaded on 8 pages
than use png or so :D
easy to read
the design got messy at the end when I had to merger everything together
just open original and zoom in
thx
why do you use balancer and not just manifold?
what?
108 votes and 75 comments so far on Reddit
I only have mk2 conveyors
and not enough iron plates - spent those on my coal generator setup
@sacred blade why not feed all constructors by single conveyor bus?
here conveyor mk2 main bus (960 inglots/min). feeding iron plates, inglots; and screws (direct recipe from hard drive) in each row.
32 constructors for each factory (plates; inglots; screws). Self balancing via overflow
How much Coal do a Coal generator need?
15 fully operating; and 45 water
so 8 gens for conveyor mk2. Best set is 4 gens per pipe (and 2 water extractor) to keep math power safe
Ty
You can do 6 generators on a pipe, with 3 water extractors.
but you will need use 24 gen block to make math good ๐
but not symmetric ๐ฆ
i had 36 coal gens before i had fuel
Like, if you have 240/min coal, that's 16 coal gens, with three pipes, each supplying (6, 6, 4) generators.
And eight water extractors.
usually people do 8 gens per 3 extractors tho
greeny when will you update calculator? xD
when my fucking client gives me a break with useless requests
jk, I'm working on it, it's hard and I don't have much free time anyway
clients are the worst
this is the worst client of bad clients tho
IT illiterate who thinks they know way more than they do and tell me how to do my job
did you tried to turn it off and on?
I literally got a request to change some spacing between elements on the side, that said "take out a ruller and measure it on your computer, it should be X cm"
uff ...
oh I've been there, I did software consulting for 20 years, I've known that type.
best part is that even after two years of being my client, they still don't understand that not everybody has a screen of the size like they do
(even though half of their requests is to make some X feature mobile-friendly)
I've seen that same lack of comprehension while at the same time resenting my very existence because our software replaced their old job. But at least they pay you ๐ (I hope, they better be).
Anyway better keep it on math-and-meta.
continuing in #off-topic-general if anyone wants
For my part, I'm working on updating my calculator.
I've got an updated data file, I've got updated images, I've got a rough idea about what I want the math to look like, and I've got a rough idea about what I want the UI to be.
Next I just want to make sure that "put the whole recipe graph in one linear program" will actually be fast enough.
whats the base throughput of a single rail car? say single car, flat level track exactly 60 seconds round trip. how much material can it transfer
A car has 32 slots.
So it also depends on the stack size of whatever item you're moving.
so a caterium wire car would be 16,000/m ?
But you'll likely be gated on the belts which load/unload the train station.
yes but you also are very unlikely to have exactly 60 second round trip. just a base number to do the maths off of
If you reckon tier 5 belts and a stack size of 100, it takes about two minutes to load or unload a car.
Once I set out to actually observe the container with a timer in hand to see if I could just roughly measure the number you're trying to calculate now, and I very quickly realized the belts were going to be the bottleneck.
so a caterium car with an 8 minute round trip could maintain 2000/m worth of belts on the unloading end. with the assumption of infite input for simplicity
freigth station have 2 inputs.. so 2x 780 would be 1560/min
1560
thx
I usually look at trains from the other side
its cap by the belts :D
this table shows how often needs a train come to keep the throughput of the connected belts
ya but if i have those base numbers i can other calcs
but without the cap depends on stack size ... stack size 100 than it would be 32x100 per minute
keep in mind tho that when loading/unloading, the platform can't accept any input, so you can't really max 2 MK5 belts
ahhh perfect. so for a double tapped caterium wire...10 min 15 sec. awesome! thanks
there's around 20ish seconds of animation iirc
25 seconds.
in which the container is inaccessible (I think only for input)
naw output too
ah alright
well output doesn't really matter much, since it's already capped by input
but can't the train hold essentially infinite resources? so the time it takes isn't really important, it just matters that the freight platforms don't completely fill up
i was concerned i would have to add a second caterium car. but that saves my bacon. sweet
well you can just add more carts though
so you divide those times by the number of cars you have
The advantage of a train is that you can scale the length of the train without needing to do anything to the track (just the stations on either end).
So if the comparison is to a bunch of belts over the same distance, the train is way cheaper.
@nimble knoll no. you check how many belts are coming out of a single platform and use that for calculations how often does the train need to arrive
if your lap time is longer, then you need to add more freight cars
or reduce your throughput per freight car
(And in Factorio, the enormous advantage of trains is that you can run multiple trains, usually carrying different items from different places to different places, over the same tracks, which means that you build this one piece of infrastructure and it connects everything.)
@nimble knoll double not divide. if u add another car
but i mean if you have one freight car, and the time is 5 mins for example, if you add another freight car and a platform on each end, can't the train take twice as long and still get the job done?
oh yeah i meant double
yeah, but the table assumes that those belts are per car
@wind spade but if u add another car you are adding another unloader...which doubles thte belts. so the math still checks out
@warm elk well not if you merge them ๐
e.g. if you have 2mk 5 belts coming out of two platforms
then yes, you can double the numbers in the table to get the numbers for your situation
I think the point is that you can always add more cars in order to shove the bottleneck somewhere else.
tru. so theoritically in a vacuum, it will double all the numbers.
aren't these times the maximum that your train can take though?
max a single CAR can take. single car + single freight station.
and if you add more cars what happens to the time?
does it stay the same or multiply?
double
ah
assuming you have the same amount of input
and now it's just split
otherwise it stays the same?
1 locomotive, with 1 car and 1 freight station on either end with 2 input belts. thats the assumption
1 loco, 2cars, 2 freight station and 4 belts....double the chart
IF you are NOT merging any of those 4 output belts
@warm elk uhh, no
so is this table essentially how long it takes to fill the freight station?
if you double both cars and belts, it stays the same
so caterium wire (stacks to 500) with 2x MK5 belts unloading/loading it...means your train can take 10 minutes and 15 seconds round trip before the unloading bin would be empty from the last dropoff
@nimble knoll no, since freight station is a bit larger than freight car's size
but if you double the cars but split the input into both cars you get double the max time?
essentially you are doing this:
- take the number of belts that's coming in one platform (max 2)
- use the numbers from that row of the table based on those belts
uhhh
confused myself, edited the original post
if you have 2 mk5s coming into the STATION and then split it 50/50 to two platforms, then each platform has only 1mk5 coming in and you use the numbers for 1mk5
surely if you have two belts going into one freight station, and one car, you will have less time to get the train there and back than if you put one of those conveyors into a seperate car?
yeah, but then again, the table is used for 1 platform, not 1 station
it also assumes you are inputting with 2 belts as well as outputting
nope, there are rows for 1 belt
if you inputting with a single MK5 and draining with 2 MK5 its gonna half it
well... only depends on input
so this is like belts input per freight station
its both
wtf
output doesn't matter (as long as it's the same or more)
it's the lower of the two numbers essentially
same or less...
okay i think i get it now
or itll drain faster. and the train has to arrive sooner to keep it topped off
it's not about keeping output full. It's about keeping input empty
i think were saying the same thing haha. im calling the draining side -> the END of the line the output. the input side is where the production is..and inputting into the car
based off the table, my coal delivery for my heavy frame factory has to be about 11mins or less since I have 1 mk3 belt, and it's about a 2min trip so i should be fine
but i don't even think it's fully saturated so yeah
are you putting coal in one end with a single MK3 and taking it out the other end with a single MK3?
yes but i think it's only carrying 120/m so i should use the time for 1 mk2?
I mean there's no reason to use different amount of belts for input than for output
cause that would be 26mins
so let's assume those are the same
there isnt, but assuming can be dangerous +D
i just use mk3 for literally everything
ya...so belt size is actually a bad variable for you. it should be production rate
MK5 is gonna give you bad data if you are outputting 100/m haha
so if it said 60 items/m, 120 items/m, 120 items/m, 240 items/m, etc.
270* but yes
well actually no cause that's 2 mk2 belts
the next one would be 270 yeah
really weird how they made mk3 270 not 240
and mk4 480
it's tailor made for the vanilla concrete recipe ๐
wait so 2 mk3 is faster than 1 mk4? lol what
mk5/4/2 are devidable by 60 but not mk3
they all are but 3 is special
@nimble knoll yes by 60 :D
well at least they made it more not less
two mk4 are much faster than 1mk5
yeah that's odd
by 180
i wonder why mk5 isnt 960
maybe mk6 will be, or it could be completely different again
it really just comes down to your time. time the trip, see what your train takes. and use that as your base for what you NEED in # of cars
i wonder what mk6 will be when 2mk3 are faster than mk4, 2mk4 faster thank mk5... 2mk5 are 1560 so mk6 must be below 1560
or maybe it would be 1920 because 2x960?
why couldn't they have just kept it as each one 2x the last lol
2 mk3 is 520 = 40 more than mk4. 2 mk4 is 960. = 160 more thank mk5. 2 mk5 is 1560....soo if they keep going with x4 itd be 920 ?
i went the rwong way with my maths
i doubt that tho. 920 after 780 would be shite
also 540 not 520. brain why are you failing me
my math is so broken today xD
well 4x160 is 640 so maybe 780+640 = 1420? for mk6
which means 60 more than mk4.... and its 180 not 160 for 2 mk4 vs mk5. god damn brain.
so its TRIPLE not quadruple scaling. s0 540. off 1560. 1020 belt.
still seems low tho. for mk6
scratch that. missed your previosu comment
when 2 previous mk are faster than the next mk it must be <2x780
x1, x2, x3.5, x8, x13. no real pattern there either. multiply by 60 for each belt tier
2mk3 are 540 right?
ya
ya i did that math up top
but that doesnt check out either. cause 2 MK2 is 240. MK3 is 270.
they have no rhyme or reason to their madness!
2mk4 are 180 faster than 1mk5
so 180/60 is 3
2mk5 are 1560
180x3 are 540
1560-540 is 1020
maybe thats the solution?
i bet the devs were just like "hmmm yeah that seems like a reasonable speed, lets go with that"
so its TRIPLE not quadruple scaling. s0 540. off 1560. 1020 belt.
@warm elk oh ... you did it already xD thanks brain
there's no formula for belts guys ๐
i alrdy said that @boreal cypress XD ahhaha. but 1020 seems so low
pfftt. this @wind spade guy here bein a buzzkill
so mk7 is 420 i guess? xD
knowing coffee stain...its probly everyone puts a reasonablish number in a hat and they draw it for the next belt speed
yeah exactly
so maybe there is an infinite number in it :D
oh that would be interesting
I mean the speeds are fine as they are right now
the items just teleport
no complaints about speed....was just goofin off trying to find a pattern instead of building caterium factories
yeah there's no pattern ๐
oh what if there was a teleporter belt but it had a low max items
all we now know is, that the speed is devidable by 30
i wish theyd add an interface to train stations that show the average round trip for a train
every run it makes would dial in that average
jeah would be nice
thatd be stupid simple to code too ๐
how do you lane balance overclocked coke for mk5 belts?
its very difficult
screenshots?
nvm i think i have an easier way, overclock only to 108% and it should go into the belts easily
at the cost of space
underclock*
over....petroleum coke is 120/m
unless power isnt an issue at all
50% / 60 per min then?
well you could use the resources for something other than power ๐
once u get to nuclear power is meh. XD
why do you need to lane balance anyway? use manifolds and don't balance at all ๐
eventually youre going to run out of belt to put things on, no?
what exactly is a manifold
make sure the input is more than or equal to the amount you need
if it is less it won't work properly
well you don't need that
X being supply?
X being a machine
if its what it looks like to me..ive heard it refer to as bus line setup
https://www.reddit.com/r/SatisfactoryGame/comments/btexf7/satisfactory_saturdays_2_balancers_vs_manifolds/ @fierce ruin
108 votes and 75 comments so far on Reddit
S is Splitter
well for merging it works the same way
welll...depends if X is input or output ๐
well you would put mergers on the output
@warm elk bus setup is something different
x is neither
@boreal cypress i dont get it, this is literally what im doing
ive always been bad at terminology : /
i use manifold in almost everything. and bus lines (if they are what i think they are) for my input lines
stacked like 6 high or so belts. fastest belt i have. the split off left and right into production lines
underclock and build more machines, i think im going with what greeny said
what's your issue anyway?
if you have a belt that you want to split to X machines, use manifold
if you have X machines outputting to a single belt, use manifold to merge
if your machines output more than 1 belt, use more manifolds ๐
i have an awkward output of 300 per minute which wouldnt fit evenly on 780 per min belts, it would be a balancing nightmare
or just more belts ๐
or don't use full 780 belt and just put 600 on the belt
or underclock 300 to 260 everywhere and have 3 machines per belt
possibilities are endless
the goal is to squeeze as much shit in the smallest space possible while maximizing throughput, i must use the full 780
then do the 3x260
welp; you can squeeze 4+2+4 belts on 1 foundation
with only 4m heigh
walled in conveyors take pretty much no space
where should i be putting my pumps?
i occassionally get below 75 flowrate on the last 2 overclocked refineries
not sure if thats normal
are all other machines full?
now its 75 to 76, maybe its just balancing out?
yes
right now i have a pump right before each refinery just to make sure
you dont need a pump :D only for headlift
why are they losing flow rate well after the lift then
because pipes keep some fluid for themself to fill up
could be ^^
I am a bit late to the party, but it is wise to use vehicle transport for compressed item, then only de-compress it locally. Such as caterium ingot vs quickwire, or iron ingot vs screws
screw is better for transport than iron ingot with 1.25 compression (bigger stack)
5 times biger stack with 4 times decompression
It makes me happy to see that i've managed to build a fully efficient heavy frame factory, zero downtime. Left my pc on for 4 hours and it's supposed to make 2 heavy frames per minute and when i came back at the 4 hour mark it had made 480 heavy frames.
I had even made it so every machine before that was only making the bare minimum of materials needed to get the heavy frames made
@forest schooner how much ore as input?
i have two miners, between the two it's 515ish iron ore
and for the coal it's 205/sec if i recal
for the first try as a noob with only 140hrs in the game, i'm very pleased at my first attempt at something like this, the rest of my save is garbo and not efficient at all
meh; it looks i need much more ore i thought for mine project...
What are you working on?
i wanted make heavy modular frames at rate 6-10 per minutes
if i had to make a rough guess thats going to cost you arond 4000mw/hr
at the high end
just done part for modular frames (with 20 unless some lower tier stops) per minute and 1k ore input
Just for frames? hmm
just for frames; and some rotor/plates leftower
i was wondering why that felt high
i didn't do exact math; i just make tier 1 iron things for 1k inputs. (32 iron plates; 32 screws; 32 rods -> 16 reinf plates and 8 rotors->12 modular frames.
big number of plates/screws/rods for autobalancing
you need about 500 iron ores for 2 HMF, and that figure is correct. But you might want to look at the alternates.
"the alt recipe" refers to which one? isn't there more than 1 alt for that item?
you need a whole chain of alternates to really enjoy the HMF production.
I need some help actually
Just finally built the Space Elevator but my factory composition is too fucked up.
Perhaps someone can give an advice of how to make an optimal and good factory?
Or should I just like reset everything and start from 0?
(Pretty sure this ain't the channel to ask these things)
Starting from zero is pretty much bad idea. You can always abadon your factory and make another next to it
Do you will be free from early waiting game
That sounds like a good idea hmm. I could just scrap out all machinery and then rebuild them.
The position I left my factory at was pretty fucky tho. Barely any ground and I'm having to settle things down.
But also close to plenty of iron resources.
Not scrap until your next better modular factory will be operating. Good point for rebuild starting base is after reinforced concrete
Because coal miners mk2
And youll stop being limited by power do hard
True. Right now biogens are what are fucking me up the most. Cuz I can't even run two things good without things going down.
Any reccomends regarding assembly machines? My factory started going downhill after I built them
Chainsaw is your friend. You should be easily get about 800 mw od power from biomass
You can build more biofuel generators to keep things producing while you work towards coal.
So keep wood plant with 20-30 burners. And leaves +wood biofuel factories (two lines)
800mw from biomass? you'd spend more time feeding your generators than playing the game lol
6 burners along with the 2 from the hub should be able to last you into coal
Tom - i always start with 20 burners and go chopping. Mostly use IT at about 20-30%
I have enough biomass to build as much fuel as I want actually. Just the place where my hub and my generators are at fucks me up.
it's super overkill lol
And use as auxillary when coal fail
I don't generally build more than about 8 biofuel generators actually. I don't use the ones from the hub because I'm ocd about the fact that they burn at a different rate than the standalone ones ๐
yeah its silly they're worse than regular biomass gens
More burners = more time between refuel
10 to 15 biomass burners is pretty good.
So 20 is just for convince not overkill
the problem is that max capacity increasing also makes people less likely to go for coal
because why change if your capacity isnt being reached?
well, I agree with the benefit of reloading less often, but I wouldn't actually consume that much power from my biofuel. Or I don't anyway, by that point I'm into coal.
I like rushing coal because trucks
If you do consume all the power of 15-20 gens, then you lose the reloading less often benefit anyway.
Bryan - i mostly use 6-7 burners. Have triple supply to rarier refuel only
20 burners at 30%>7 at 95%
Burners are too cheap to not to overkill
https://discordapp.com/channels/370472939054956546/553550313533997057/695985894498238464
I made thing... A crazy thing... But it splits 4 x 600 belts with screws and 1 x 720 belt with rods into rotors
I still think that biomass burners should have a belt input for automation.
one of the only things you can't automate and it makes the trek towards coal power that much more tedious.
would be nice for beginner
That's exactly why they don't have input
as you progress towards coal, you have a pretty good sync between increasingly better biofuel types and factory energy demands until the only concern with bio is that you have to fill them up every half hour or so and then you have to learn piping to progress into higher tier power
you still need to harvest biomass
which you have to do anyway to make room for your factory
you harvest very little at the start simply for your factory :p
I tend to go on deliberate journies for biomass
yes as i said, you progress from having to scrounge up for very little and then progress towards biomass which massively increases energy value to just having tons of biofuel as you just clear vast swathes with the chainsaw and then you get to coal
But that's really neither here nor there; it's a game of automation, so having a machine that requires manual input every x minutes is pretty much an outlier
you also have to handcraft iron plates to progress through Tier 0
there is also a machine to automatically make power
for crafting turbofuel, will i yield more of it by diluting the heavy fuel first? or by putting the heavy fuel directly into turbofuel?
you'll get the most turbofuel if you use the heavy oil alt to turn crude oil directly into heavy oil residue and then turn the heavy oil residue into diluted fuel, unpackage fuel, convert to turbofuel
turn crude oil directly into heavy oil
what recipie is that?
each 1m^3 of crude oil will net you 1.777 m^3 turbofuel
it's called Alternative Heavy Oil Residue
it takes 3 crude oil and gives you 4 HOR
let me get this straight, dilute the alt heavy oil, then use the fuel to make non-alt turbofuel?
yep
well hang on
take the heavy oil and turn it into diluted packaged fuel, then unpackage the fuel, then use the regular turbofuel recipe
ahh
so you could get a bit of rubber/plastic I guess or just sink it
i wonder what i will do, when i dont need the HOR anymore
but that means i have to craft empty canisters
yes but once there's enough canisters they'll all be recycled in the loop and you don't have to keep crafting them
because the diluted fuel recipe needs packaged water
so it'll use the empty water canister and put the diluted fuel into it, then you unpackage that fuel and send the empty canisters to get filled back up with water and the cycle repeats
bruh thats alot of work
all I know is with this method you can take a 300m^3 pipe of crude oil and turn it into 666.667 turbofuel powering 148 fuel gens for 22,222.222MW
yeah it's a bit of work but it's worth it imo
holy shit
you asked how to get the most turbofuel lol
thats what is "imKibitz" doing :D
that's why I got bored in my save and stopped working on nuclear lol. I simply don't need the power ๐
lol nice
what's the best chain for heavy mod?