math-and-meta

[royal bramble]

Sooner or later you're going to want to automate just about everything
What I do (other people do differently) is basically rush coal so that I don't have to worry about power, then automate each of the basic parts and send them into a container/dimensional depot, then go and explore the map

[languid laurel]

Do you explore the map for drop sites or is there a specific reason

[royal bramble]

I'm not sure if you've seen them yet, but there's a collectable that gives you alternate recipes which can help a lot as you build more
And the alien artifacts are also big

[royal bramble]

Yeah, the crash sites have hard drives, which give you alt recipes

[languid laurel]

||hard drives?||

Oh ye

Yh the alien technology in the Mam I can’t find to find the one in the middle

I found a lot a the Mercer sphere things and the rainbow s things but not the purple thing

[worthy shuttle]

Lol why did u censor hard drives? They aren't something locked completely 😭 most people who stsrt would just ignore em lol

[royal bramble]

Yeah, SAM is annoying to find
There's one close to the dune desert you can find once you have explosives
Otherwise you have to wander a bit
I would hold off until you have your base set up

[languid laurel]

I’m not gonna lie I thought he was trying to not say them in case I hadn’t found them (im also a lil bit of an idiot so idk why I did)

[languid laurel]

Ok thx

[worthy shuttle]

Ngl I didnt look through the convo as im bored atm😭

[languid laurel]

Fair lol

[vapid gorge]

my advice is find coal nodes next to water - most of them are

[languid laurel]

Thx for your help Craig I’ll try take your advice

[languid laurel]

K ty

[worthy shuttle]

That is true its pretty uncommon for coal to not be next or close to water

[vapid gorge]

I would say 80-90% of coal nodes are next to water

[languid laurel]

Last question is the map fixed or is it random

[royal bramble]

Fixed

[worthy shuttle]

Its fixed so u can watch yt vides withiut worrying

[royal bramble]

And there's only one map - each of the starting locations is just a separate area within it

[languid laurel]

K thx I’ll try explore the map on my own tho

[languid laurel]

Ah that makes sense

[vapid gorge]

there is an online map , but imo the exploration and map is excellent. I wouldn't use it until you feel like you've thoroughly gone over it and gotten what you can out of it

I kinda wish sometimes I could forget it all and explore it again xD

[languid laurel]

The map is so well designed the terrain is so cool and I love using my parachute against a slope and flying to places I probably shouldn’t be yet

[royal bramble]

Just wait til you unlock ladders 😄

[vapid gorge]

get the zip line and blade runners too 😉

[languid laurel]

Oow blade runners sound cool 😄

[vapid gorge]

they are the best item in the game, once you put them on you'll neve rtake them off

partially because nothing else fits into that equipment slot 😛

[languid laurel]

Are they far int the game?

[vapid gorge]

you can get blade runners very early

I believe it's quartz tree in the MAM

[languid laurel]

K ty for your help

[vapid gorge]

mostly my main advice to new people is treat the milestones like a tutorial, once you unlock everything you can treat it like a sand box

[vagrant lynx]

My entire fuel generator workspace. Gonna be utilizing 1200 crudeoil/min which should power my entire building till alot later

Just gotta figure out the exact math to have it planned out

[unique cypress]

If my math is right, this should be exactly enough space for 160 gens

[vagrant lynx]

Like the entire are be fuel gens? Or like I can have my HOR refineries and the whole setup there on the platform

[mint dew]

can someone help me i have 2 120 belts and i need to split them into 1 line to 135 and one to 105 but i only hjave 120 belts any heklp

[dusky dust]

With the added complication of "strict" belt-speed limits like you've got, IMO the best solution for that kind of thing is just to use overclocking/underclocking to produce exactly what you need, rather than trying to split things

So like have one set of machines producing exactly 105/min, and then split up the other one by whatever's convenient (maybe one set of machines doing 60/min and another doing 75/min, or whatever -- set the ratios so that it's convenient for whatever recipe is consuming that total 135/min)

[mint dew]

its beacuse i only have 120 beltsa

[dusky dust]

When you've got faster belts, one of the other main ways to do a split like that is to just do a "naive" split where you just split (for instance) a 240/min belt right in half; the half which only needs 105/min will be getting "too much," but it'll eventually back up, and once it does, the other side will be getting the overflow

[dusky dust]

Right, that's why I recommended what I did. :)

If you take my initial suggestion to the extreme you could go so far as to use clocking to send the resource to each machine in exactly the amount that it needs. For instance the vanilla Reinforced Iron Plates recipe needs 60/min screws; if you're setting up a bunch of RIP assemblers you could literally have a little screw-production setup right in front of each one which produces exactly 60/min screws, without "combining" the outputs any further than that

[jaunty stone]

Bro you have 2721 water p/min as a bio product

what are you going to use that for like

[brisk smelt]

satisfactory modeler is retarded

it just goes back, uranium is a closed loop

[dusky dust]

That word's considered something of a slur nowadays, btw

I'm pretty sure that Modeller can be convinced to handle loopbacks properly, btw, though since I don't use it myself I'm not sure exactly how

[jaunty stone]

I use modeler and I love it and I have had no problems so far

i am actually doing my 100 fiscionium rods with it

[vapid gorge]

did you solve it?

often times things like that you can jsut run 2x120 and have different machines on each manifold

for example have 105 feeding 1 group of constructors , and the 15 extra feeding another machine doing another thing

[scarlet sky]

Got a headlift question. I want the three fluid buffers here to hold a short reserve. The refineries feeding those generators produce just enough extra fuel to eventually fill those buffers, after which point I want the excess to be sunk, by a packager to the right. The vertical rise in the pipe to the right side of the pipe feeding the generators should take care of that. But how do I deal with the headlift creted by the three fluid buffers to make sure that the generators get priority before the packager?

Aside from not stacking the buffers, which wouldn't align with future pklans

[vapid gorge]

if you absolutely must have fluid in reserve , that won't help since once the breaker flips generators stop burning it, have the buffers on the side of the pipe with a valve set to zero

also - use power storage.

[scarlet sky]

It's not a reserve power question but I need to be able to shut off the refinereies briefly for troubleshooting purposes, and without letting the generators run dry. Yeah, building plenty of power storages

[vapid gorge]

power storage would also solve the generators being a bit dry and be simpler

to fil the buffers you're going to be starving it anyway. Just shut the system down while you do stuff. Restart with power storage

[scarlet sky]

ohh the use case isn't that the generators stop burning it, it's that the refineries stop producing it for a short time, but only by me manually shutting them off

[vapid gorge]

I don't understand the difference sorry

[scarlet sky]

Basically it's just a headlift question. Would the headlift created by those three buffers be enough to cause the fuel to flow over the pipe rise

The generators need to be fed before the packager on the right side

[vapid gorge]

buffers have variable head lift so depends how full they are

[scarlet sky]

(assuming they're full)

[vapid gorge]

just add a pump after and don't worry about it

but you really don't want a bunch of buffers in series with how you feed machines

[scarlet sky]

Yeah but wouldn't a pump just set headlift to zero, thus the fuel would never go over the rise?

[vapid gorge]

power it?

[scarlet sky]

Oh shit

I may be overthinking this

[vapid gorge]

you absolutely are

[scarlet sky]

Put a powered pump on the output of those buffers. Bingo bango bongo

This is all coming from my "all machines must run at 100% all the time" challenge

And all the troubleshooting measures you need to build in to make it realistically achievable

[vapid gorge]

trouble shooting requires turning them off they won't run.
filling those buffers means generators won't be running at 100%

[scarlet sky]

Eventually they'll fill, because the generators use less fuel than is produced by the refineries supplying them

[vapid gorge]

so the refineries are going to stall because it'll be clogged?

[scarlet sky]

Nah you're not understanding what I'm trying to accomplish I think. The generators should pull fuel first, then excess is first used to fill the buffers, and only then is any leftover excess sunk

The use case is I need the ability to shut down the refineries to remove them from the power grid, but without emptying the generators. The buffers give me a 20 minute or so window to do that

This is a small scale start to something that will eventually be much larger

And also that the refineries must run at 100% otherwise

So therefore, they need excess sunk

I think it'll work like this though, it'll just take an hour or two for this all to settle

[agile junco]

Ooof, my train math didn't work out. I wonder what I did wrong.

I'm doubting myself setting them to wait until full and 0 sec, they are stacking up on each other and then it's overfilling while they are all going to drop stuff off.

Actually, maybe it's not the settings... still seems bad.

Ohhhh it's becuase the "whole train" is waiting to be fully unloaded/loaded... including a car I use to carry a small amount of something... damnit.

[mint dew]

yea

[agile junco]

I wonder if it would have been better to package nitrogen gas instead of making nitric acid offsite...

But then I would probably have to have shipped water too.

[mint sky]

any tools to create a 14 to 9 load balancer?

[wind spade]

well, any reason to do so? 😄

[mint sky]

14 blenders making silica, 10800 total, so 9 mk6 belts

[wind spade]

hook each belt to amount of machines that will process the belt fully

[mint sky]

I dont want 14 belts tho

I only have space for 9 as the 9 are already preplaced

[wind spade]

you don't need 14, you can do 9

[hard locust]

well i didnt handled it but i would make feed it back in obviously

[wind spade]

just flip it around - merge machines so that the machines you merged will fill one of your 9 belts to the amount you need

[mint sky]

everything is already completely finished building

i just had to rip this part of the factory out completely

[wind spade]

I mean, then you're on your own, but you're making it hard to yourself for no reason

always consider logistics before you build anything

you won't run into these issues

[mint sky]

I did consider it, what I didnt consider was beeing sleep deprived and not looking at numbers correctly

[wind spade]

happens to the best of us

[mint sky]

so I missed one blender but build everything else and now I gotta rip that part out

and need a 14 to 9 balancer

[wind spade]

or merge in groups of 9

that's just changing a few belt connections here and there

[vapid gorge]

and possibly a bit of clocking

[mint sky]

im afraid all that isnt possible

[vapid gorge]

clock your machines and selectively merge them, that's your most powerful logistical tool for the whole game

ok well then you're making things hard on yourself :\

[mint sky]

Guess I am. The whole quartz purification and distilled silica ratios are so fucking annoying

I couldnt think of any ratios I hate more

[vapid gorge]

ratios in this game are absolutely meaningless

you have full control over the machine outputs from 1% to 250% base output

clocking is youre most powerful tool
ignore it to your pain and suffering

[mint sky]

Ratios can still suck ass

[vapid gorge]

only if you fail to use your most basic tools to manage outpouts

and then that's on you

[mint sky]

Guess it it, can you just shut up then?

[vapid gorge]

nope

[mint sky]

If you wont help me with my issue then just piss off

i dont have the time for this shit

[vapid gorge]

I just did - now next time you set up distilled crystal you know that your ratios are not a problem

this applies to every single system you make from now on

[mint sky]

Alright, youre set to ignored now. ty.

[vapid gorge]

What will I ever do. I hope you remember to clock your systems so you don't wreck yourself before you checked yourself

[hushed silo]

[orchid brook]

before i scale this, is there any mistakes here?

the only thing i didnt do is looping the fluids cuz its not worth it

[vapid gorge]

get rid of the buffer and the hump

also - whether looping the inputs is 'worth' it will depend on how tidy your set up is and how much oil you're extracting

[orchid brook]

the big one is 600 and the 2nd one is 300 oil

[orchid brook]

whats the "humb"?

[vapid gorge]

hump*

oh, I did write hump.

you'll likely need a loop on the 600 pipe

[orchid brook]

hmm ok i will do it

[vapid gorge]

but also - get rid of hte buffer and hump

doesn't help and could easily murder flow

[orchid brook]

what does hump mean?

like the tower?

[vapid gorge]

yes

you might not need the loop on the 600 oil - you could try it w/o.

I thought you meant the HOR was 600 flow

crude oil and fresh water tend to be more stable and often don't need a loop

[orchid brook]

better?

[vapid gorge]

fluid towers don't help feed or flow, all they do is provide headlift, which is what that pump did before it.
and buffers can cause wonky flow issues

[orchid brook]

well if that is 900 oil and it was staright forword hopfully are easy AND i have a lot of knowlage from my RF plant

[orchid brook]

yay done

[orchid brook]

at wich point should i add a pump?

[vapid gorge]

generally when it goes above 10 from the initial point

prob aim for slightly lower

[orchid brook]

ok i am gone add one at the starting platform, good?

[vapid gorge]

sure

[vagrant lynx]

Isn't that 400x400 foundation?

[unique cypress]

1 gen takes up 2.5x2.5 foundations
160 in a 16x10 arrangement take up 40x25 foundations

Unless I miscounted, that's exactly how much space you have

[vagrant lynx]

Each fuel gen is 20mx20m so 160 of them is at minimum 3200mx3200m each foundation is 8m so isnt that 400x400 foundations?

[unique cypress]

No, that's 160x160 gens

Not 160

[vagrant lynx]

Also 16x10 arrangement?

I counted and its 25x41 which leaves some room

[unique cypress]

It fits perfectly in a 25x40

If it's 25x41, then you've got 25 foundations spare

And that ofc assumed one layer of gens

You could double stack them

[shadow sinew]

satisfactory tools wasn´t doing the plutonium rod processing correctly so I´ve had to do it by hand xd

[vagrant lynx]

I could. They're what? Under 32m in height? Which my floors are gonna be 32 high.

[unique cypress]

It does. You just have to input the waste you have

Because waste doesn't officially have a recipe

[shadow sinew]

even then it was having trouble calculating it, alone it´s fine, but when trying to do plutonium and uranium rods at once it completely freaks out, it ended up using only 840 uranium on maximize both without converters for some reason, it´s not like it ran out of other resources either

[unique cypress]

That's because maximize makes the same amount of both items when you maximize 2 at once

[shadow sinew]

I felt it easier to just write it down and do it myself at that point

oh ok, ig that explains it

I´m gonna be using the standard plutonium rod recipe because I´m just sinking them

I´ve also been told if at any point I need more power, I should use the converter to make more uranium instead of doing Ficsonium

I will use the plutonium for drones, and probably will keep an emergency stockpile of rods

[unique cypress]

I doubt you'll need more than 630 GW, but yeah that's cheaper and easier than doing Ficsonium

You can also set up augmenters and matrix them

That'll be up to a 4x multiplier

[shadow sinew]

Probably not, checked max ballistic warp drive and thats 730 GW pre sloops, I´ll do 100+ thermal rockets, and I´ll convert however many I feel like to ballistic warp, but prob max half the theoretical max

although it is possible that slugs and sloops push it over, so I´ll make some slack capacity on the nuclear setup, good for redundancy too

[unique cypress]

Yeah, well, max anything assumes you have nothing else in the world. Even if it includes power generation, that's always (mostly) RF, not nuclear

[shadow sinew]

Well, I plan to do 100% green energy by the end of the year xd

and I´m making fairly liberal use of the plastic and rubber alternate recipes as such

[pastel canyon]

what's a good ratio for reinforced iron plating/motor automation? im currently at 5 plates per minute (about to try to expand to 10 per min) but i dont know if thats too slow or acceptable

[unique cypress]

It's acceptable as long as you don't keep running out of them

[pastel canyon]

thats fair yeah.. im just not used to having to build big shit so i ended up unintentionally making it small and now im stuck with 5/min

[limpid knot]

How long would the longest drone flight take? Want to ballpark some throughput calculations. Assuming batteries/rocket fuel speed.

[unique cypress]

You're not stuck. You can always make more somewhere else

[pastel canyon]

fair enough

[unique cypress]

I'd estimate it at about 0.5 stack/min across the map diagonally and about 1 stack across horizontally/vertically

But it'd be best if you actually set up a drone and tested it

Note that the throughput the drone reports is often a bit off, so if you want a more accurate number, you'll need to actually measure it

Though even that'll give you a better estimate than I did

[limpid knot]

Ok, thats helpful, thanks!

Would the awesome shop throughput monitor be more accurate?

[unique cypress]

The throughput monitor is perfectly accurate, but you'd need to gather data from it for like half an hour to an hour to actually see that accuracy

The lazy man's version would be to time how long it takes between drones

Because they should always drop off exactly 9 stacks, at least if you oversupply the loading side

So it'd be 9 stacks every x minutes

Time like 5-10 drones and you have a decent estimate

[limpid knot]

Ok perfect. I suspect my plan will be sufficient for my needs, but now I can better gauge things. Thanks!

[scenic gale]

I am currently in the process of trying to kill two birds with one stone and increase my power capacity while also producing an adequate amount of rubber and plastic, but from what I can see from Satisfactory tools I have noticed there's a split in the fuel requirements for the plastic/rubber and the turbo fuel that I'm using for fuel generators.

I have looked at maybe implementing a VOP Junction for this to give the fuel in the turbo fuel production line priority but just wanted to make sure that, that is a suitable approach and there isn't something better I should be doing. Naturally I don't want my rubber/plastic refineries to potentially cause any issues with the fuel being supplied for the generators.

[pastel canyon]

after a bit of tinkering and a lot of math mistakes i have managed to fit the reinforced plate factory in the small fuckin space i gave myself and have successfully upgraded it from 5/min to 10/min and honestly... that was... weirdly satisfying, i think im gonna build another

and the screws arent bottlenecked!

[scenic gale]

Something like this maybe? 2400 fuel in divided into four lines and then 530, 430 and the three fuel lines? Provided I did this properly, not quite sure if the inputs were done right since I could quite make out how to design it for multiple inputs from the guide.

[zenith sinew]

Depending on input pressure, you might want to have initial pumps too.

Not sure.

wait a second...

disregards, wrong direction.

[scenic gale]

just wanna make sure this is a perfectly viable solution before I go to actual implementation of it, or if there is an alternative

[unique cypress]

The best idea would be to make fuel in modules that produce exactly as much as the next step needs

[scenic gale]

I guess I could shuffle the blenders around a bit to make that work, just means i'll end up with an extra blender but thats no biggy

provided the ratios work out for the heavy oil residue

I'd imagine I'm gonna have to split one of my HOR lines to get that to work, which does sound easier than implementing a VOP here, alright, keeping it simple it is, thanks

[unique cypress]

I'm a fan of putting every machine making one item all in one place and splitting it to every place that needs it. But it doesn't really work with liquids. Especially if you want to give priority to some of them.

You could package it, use whatever janky item management method you want, and then unpackage it, but that's more work than splitting up the fuel production

[shadow sinew]

how many items per minute can a drone port transport excluding travel time?

[unique cypress]

The port itself? Full belt speed

[shadow sinew]

i´m getting 283 uranium per minute on the drone port, but it´s not coming full from the impure node by the rocky desert

well, I hope it is 300

[unique cypress]

The transfer rate depends on the drone(s) too, not just the port

In fact, it doesn't depend on the port at all

The port can take a full belt and it instantly shoves that into a drone, if one is docked

[shadow sinew]

in any case even if it is actually 283 per min, surely it will improve to at least 300 when I use plutonium rods instead of turbofuel

[unique cypress]

You could add a second drone to the route if you haven't already

[shadow sinew]

it´s only 1000m from where I´m making my mega reactor, it feels kinda bad xd

prob will do one drone per 150-200 per min at least for the rest

[slender wind]

What I have seen, a drone can carry 9 stacks per run. If there is less material availabe, it will take what is in the station.
When the drone and the stations are set up, you can see what the drone will supply. Including travel time.

[shadow sinew]

so yeah, it´s prob work out with the plutonium as fuel instead of turbofuel if it´s 283 per min now

[nova vortex]

is there a way to plug in how much iron ingots i have to figure out how many motors i can make?

[unique cypress]

http://satisfactorytools.com/1.0/production and maximize mode

But make sure that you switch back to items/min mode afterwards

[nova vortex]

how do i find maximize

[unique cypress]

Click on items/min next to an item

[nova vortex]

how add the amount of iron ingots i make

[unique cypress]

In items, input

Make sure to disable iron ore while you're there if you don't want it to make more ingots

And SAM

[nova vortex]

im so confused i cant change any numbers

im making 1500 ingots and i have the alt recipie so i dont need coal and i wanna know how many i can make with that

[unique cypress]

27/min https://www.satisfactorytools.com/1.0/production?share=LNM5LTbDlKM2LHGkcKx7

[nova vortex]

mm ok

bc i realized this is only enough for my pipes (1500) ingots

[vapid gorge]

jesus, use Steel at that point. So much less work

[nova vortex]

NO my whole point is to cause me pain and agony

[unique cypress]

It's not really less work

[nova vortex]

iv done to much work to restart

[unique cypress]

Cobalt is just allergic to iron pipe for no reason

It's neither expensive nor a lot of work

It's just a lot of iron

But imo better 10 iron than 4 iron and 4 coal

Or at least sometimes

[nova vortex]

i was so happy when i realized i didnt need coal i havent had motors automated since like EARRLYY phase 3 im now phase 4 teir 8

[vapid gorge]

4 coal + 4 iron = 4 steel pipes
(solid steel)
vs

16 iron = 4 steel pipes

you are shockingly bad at maths

possibly teh root of why your split waste water system didn't work

[unique cypress]

I didn't do any math this time so I'm not surprised it's wrong

[pastel canyon]

should i aim to cut out screws from as many recipes as possible? because i just found uhh this and i dont know which one i should go for

[wind spade]

no, you should aim for recipes that you like

[unique cypress]

Stitched plate is a great recipe, especially with iron wire

[vapid gorge]

eh, up to you. Screw management is easy though
just make them right before you use them
in a lot of cases the Steel Screw recipe can simplify that

[vapid gorge]

perfect for advice giving xD alternative facts

[pastel canyon]

oh

yueah fair enough

[vapid gorge]

misrepresenting reality for yoru benefit

[pastel canyon]

is it just any wire can be used in that recipe? or only copper?

[vapid gorge]

there's basically no 'meta', it all depends on what resources you want to use and have available

[unique cypress]

Alt recipes don't make a different item. They make the same item in a different way

[vapid gorge]

wire is wire. Doesn't matter the source as long as the item name is 'wire'

the game doesn't remember how an item is made

[pastel canyon]

thats what i did with my current reinforced plate factory, i have like 3 constructors making screws that immediatly get fed in 2 assemblers

[vapid gorge]

and thats the core part of belt management 🙂
I mentioned steel screw before because you can make a lot of screws from 1 machien and just send steel beams out

makes any screw recipe easily done

[pastel canyon]

that'd be nice to find, i still have 1/2 hard drives i still need to crack open so im hoping on that

because im about to start building my steel foundry

[vapid gorge]

It's a solid recipe.
I often use the Bolted Plate and Bolted Frame recipes combined with it. They use TONS of screws but output items fast

[pastel canyon]

though im a little hesitant to start 'cause i wanted to automate rotors before that but holy shit they have some weird ratios so i dont know how to do it

[vapid gorge]

the real power of most recipes is what other recipes you use them with 🙂

[vapid gorge]

have you unlocked 'clocking' in the MAM?

[pastel canyon]

yeah i have, why?

[vapid gorge]

then ratios are meaningless

just clock your machines to what you need

[pastel canyon]

....yeah im a little slow, fair point

[vapid gorge]

no stress 🙂 never forget clocking - it's the single most powerful tool you have for logistical management

and this is a logistical management problem solving game 😄

[pastel canyon]

yeah, its just my brain saw the 100 screws/min and was really confused, so instead of dealing with that i expanded my powergrid to 8 coalgens

[vapid gorge]

it makes ratios and decimals a total non issue

[vapid gorge]

on that point, I recommend you find a spot with 3-4 coal nodes next to water, and just make it all into power. You'll need it
every new machine you unlock takes significantly more power

[pastel canyon]

yyeah i've been told im probably gonna end up with like 16-32 coalgens by phase 3 so i've been kinda preparing myself mentally for that, though while exploring i think i found a pretty good spot, its like a big ass lake with a bunch of coal veins

it also has like a natural slope i use to go up and down even though i mostly just jump into the lake from the top of the cliff to get down

getting the power to my main factory was a bit of a slog but not that bad

its like a kilometer away so thats about the farthest ive gone from homebse

[vapid gorge]

I would set up like 64 coal gens by the time you have mk3 belts

[vapid gorge]

why was it a slog? 1 cable right?

[pastel canyon]

i was told that the number mightve been 64 but i was ignoring that hoping i wouldnt actually have to set up that many 😭

[vapid gorge]

just make multiple groups of 8 🙂 https://cdn.discordapp.com/attachments/558721941410807812/1439217940141441107/image.png?ex=694bd0a8&is=694a7f28&hm=92978b540a29356e4114a68de28714ec4e1b4b9cd034db982cea6516b17f2e9e&

[pastel canyon]

yeah but i had to walk my ass back and forth to set down the poles and wires 'cause i have 0 movement

[pastel canyon]

..that is so much more efficient than whatever the fuck i have going on 😭

[vapid gorge]

!wikisearch CG

[glad apex]

https://satisfactory.wiki.gg/index.php?search=CG

[vapid gorge]

I basically just modified one of these layouts a little and kept making chunks of them

mk3 belt, 4 nodes, that'll get you to phase 4 pretty easily

without having to think about power again

[vapid gorge]

as for the walking - have you unlocked Blade runners yet?

[pastel canyon]

yeah i did whatever the fuck this is 💀

[pastel canyon]

yeah i have 'em now, its just when i was doing this i didnt have them yet so it was just a little painful putting down the wires

[vapid gorge]

ah no sweat, and do you know about sprint slide jumping?

that's a very reasonble coal set up - could be a bit cleaner but solid.

[pastel canyon]

oh yeah thats a little weird, i have no idea why this factory building game has like, movement shooter slide canceling and bhopping but i've been abusing the fuck out of it to explore

[vapid gorge]

but yeah, that coal set up isn't bad I'm sure you could make the belts and pipes a bit cleaner but ... yeah far far from the worst I've seen

[pastel canyon]

one thing that stood out to me though was that when i finished it and was hooking everything up the power was.. like.. fluctuating alot, almost like it had to "warm up" and it eventually stabilized out at a very comfy 700 MWs of production, it kept falling up and down going from 460MWs to 700, no idea why it did that though

i thought i fucked up while building it or something at first

but its stable now so i dont dare touch it

[vapid gorge]

manifolds take time to fill up, very normal

with fluids it's often good to set like 1 or 2 machines to 50% clocking so that the whole thing floods before cranking them back up

full pipes are happy pipes

[pastel canyon]

well i kinda did that accidentally because i first made the pipes, then the coalgens and then hooked up the conveyors, so they had time to fill up

[pastel canyon]

i also have no clue what manifolds are

[vapid gorge]

you've been using them this whole time 🙂

[pastel canyon]

is it like the conveyor setup

[vapid gorge]

a belt or pipe with junctions that feed a series of machines, the first machines will fill up faster and then over flow to the next

[pastel canyon]

oh

yeah that makes sense, are they bad? is there a "better" alternative for them? or are they fine?

[nova vortex]

@unique cypress i decided to go with my original 60 motors a minute the first set on the left is js for my pipes the next set is for my screws and the last set is for my rods (this is js my iron) doesnt inclue my copper

[wanton hawk]

Idk if this is a question for here but here goes. I'm working on a pretty big save and I've been using trains a bunch (16-18 across the map so far) but was wondering for the long term what was more efficient performance wise (cpu/gpu) between more trains and mapwide belts?

[vapid gorge]

almost certainly trains

[limpid knot]

They're good, so are the other options (load balancers). Manifolds have the best space advantage and let you easily dump excess. Load balancers are pretty, take a lot of space, and can be complicated to get just right. Most use cases, a manifold is just the simplest and most fool proof way of getting solid performance. They just take time to saturate, thats all.

[versed stone]

First of all i was reading monitors a minutes not motors lol, second of all 60 a minutes jesus christ, what are you going to need all the motors for anyway?

[uneven rampart]

First time using satisfactorytools.com and im messing around with some plastic and rubber production. Looking over this setup, I am very confused how this would even work if its sending the rubber and plastic in a loop between the recycled rubber and plastic refineries. Anyone understand how this works?

[unique cypress]

You just send all of the plastic to recycled rubber, all of the rubber to recycled plastic, put smart splitters set to overflow on both of those belts and the overflow is your output

[uneven rampart]

ohhhhhhh that makes so much sense now

thank you so much

[worthy shuttle]

Oh I thought u dont need to use smart splitters?

[versed mango]

I make my rubber and my plastic with the rest of the fuel. Like that, I can make some power and with little bit of water and the rest of the raffinery, I have my plastic and my rubber

That less pain since a normal need 2 refinery no overclock, that will power up 4 generator of electricity with the fuel and you can use 2 other refinery to make the rubber and the plastic with the residu of the 2 first refinery

[worthy shuttle]

But wouldn't ur setups for ur other stuff produce enough hor to burn off in gens?

[full dawn]

Is it worth it to loop a 300m3/min mk1 pipeline on itself? Feeding 10 refineries from a normal Oil node

There was also an image that is in this Discord somewhere. It had a message on it "Shamelessly stolen from CobaltofDoom""

[wind spade]

You don't

[orchid brook]

ok the first part is done is there any mistakes here, i triple checked everything

[unique cypress]

You don't need need them in that case, but for example a 900 plastic/300 rubber would actually need one, so it's just easier to always use them and not worry about it.

Plus the system starts up a lot faster with them than without them

[wind spade]

Why would you need them for 900/300?

[versed mango]

No, other set ups kinda worst, like if you make the plastic with the crude oil, you just make 10 or 20 m3 of heavy oil, that can be changed in fuel with some water, but that is not enough for the gens

[unique cypress]

Because with a regular splitter that setup would have to rely on the rubber output backing up to supply everything recycled plastic needs

I'd rather not rely on that if the fix is as simple as a single smart splitter

Plus, like I said earlier, it starts up faster if the loopback fills first

[pastel canyon]

alright, thank you

[pastel canyon]

question, would something like this be possible? because im looking to automate modular frames but i dont want to build another reinforced plate automation factory as im running kinda short on iron nodes and dont know how many more iron things im gonna need to automate, so i would like to know if it could be possible to create an "overflow release" splitter that, for example, when my modular frame storage is full would pull the reinforced iron plates into a separate storage container

[frosty owl]

It is very much possible. You might want to take a look at the Caterium branch of the MAM research ^^

[pastel canyon]

well, firstly, very glad to know its possible, secondly, i guess i should stop putting off making a caterium factory 💀

i've been meaning to make one for like, 2 days now but kept putting it off to do other stuff (im starting my steel foundry instead of making caterium 😭)

but thanks

[unique cypress]

You don't really need a "factory" for it, though it'll certainly help. But even handmining and handcrafting can get you there

Smart splitters are pretty cheap to unlock and build and are very early in their tree

[pastel canyon]

oh

thats actually very helpful because i think my closest caterium vein is like... 800m away from my main factory 💀

[unique cypress]

Distance is irrelevant if you have depots

[pastel canyon]

depots as in...?

[unique cypress]

Dimensional depots

Another MAM unlock

[pastel canyon]

oh the purple storages

yeah i have them unlocked, though ive just put a single one down near my storage tower and i hand feed stuff into it because i dont wanna waste mercer spheres

[unique cypress]

It's not really wasting because you can just dismantle them if you want to

Plus, there's nothing else to spend them on

Only depots and depot upgrades

[pastel canyon]

oh is that their only use?

okay that eases my mind a bit

will be making alot more of those then

[nova vortex]

So i havent had motors automated since like early phase 3 and im now phase 4 tier 8. So i dont wanna EVER run out of them

[agile junco]

Anyone have experience with chaining path signals through a complex intersection? I'm having a bit of trouble.

Oh nevermind... it was because a train was just barely in a segment and it didn't look like it.

I need way more staging space in this temporary station I've built for all these trains.

Nothing like deleting and rebuilding block/path signals in the same places to fix "signal loops back on itself". Definitely some bugs with how segments get indexed internally.

Totally a mess for now, but it works while I work out the exact numbers and design the final thing.

[lusty summit]

Hey guys if I've got 2 miners pushing out 120/min materials and only 90 of that is being consumed, How can i utilize the rest of the 30 unused material without making the 2 factories using the 90 become delayed

[unique cypress]

Make the other factory only consume 30 and it won't take any ore from the first

[lusty summit]

oh with the under clocking systems?

[unique cypress]

I mean an iron/copper smelters takes 30 normally I think?

So just connect one of those to the line you already have

[lusty summit]

its because im using a foundry now with coal so the numbers are wonky. 2 foundrys are using 90

[unique cypress]

Ah, yeah, then underclocking will help

Alternatively, you can unlock smart splitters in the MAM so you can use their overflow feature

Then the ore that isn't used by the first factory will overflow to the second

And only the ore that isn't used

[lusty summit]

is that this?

[unique cypress]

No, that's for merging

You might be able to use it for ingots with the same effect?

But a smart splitter is, well, a splitter, not a merger

It's in the caterium research tree

[pastel canyon]

these were the most painful 2 and a half hours of my life, i will never work with decimals again

all this shit for 10 modular frames / min 😭

[unique cypress]

Good luck with that 🤣

[pastel canyon]

im cooked arent i

im gonna have to work with decimals again

fuck my stupid pioneer life

[lusty summit]

thank you for your help 🙂

[unique cypress]

The game is all decimal as soon as you get to like steel lmao

Wait till you encounter repeating decimals or some shit

Or pure iron with its 7:13 ratio

Or silicon CB with 11:5

Or HEF producing 2.8125/min

[versed mango]

I make a factory for 8 Modular frame per minute with 2 iron ore pure

😭

[agile junco]

I guess I need to figure out how to expand this to multiple trains: https://satisfactory.wiki.gg/wiki/Tutorial:Train_throughput

It's trivial if you use the wait until fully loaded... but I cannot always do that.

[unique cypress]

You can always do that :)

All of my trains are set to only depart when empty/full

Every single one on every single station

[agile junco]

But I have a train which carries Iron Plate and returns with Nitric Acid. If I tell it to wait to be fully loaded/unloaded I cannot say "just fully loaded", so it stalls forever waiting to unload plates.

I could always make the station larger and do two trains, but that's not what I want to do.

[unique cypress]

The math for multiple trains is simple, though. Replace round trip time with time between trains

I'm pretty sure that just works

[agile junco]

So I'm trying to work out the averages to see if my way is feasible.

Not if one train gets stuck directly behind another train and ends up just collecting nearly empty loads.

At least that's what I'm concerned about.

[unique cypress]

Then the time between trains is nearly 0

The math will still tell you what's the throughput

[agile junco]

Sure. I guess what I'm trying to think through is if there's a way to game the wait duration setting to ensure that number stays in the proper spacing.

[unique cypress]

And if you put in close to 0 RtD into those equations, you get nearly 0 throughput

[agile junco]

I think it should be possible, although not super nice.

[agile junco]

You actually get nearly infinite, then divide by zero.

RtD is the denominator.

Throughput ratio = TtF/RtD.

[unique cypress]

If you set the delay to 1/n th of the RtD, where n is the number of trains) I think it might work

[agile junco]

Yea exactly, that was my thought as well.

Maybe 1/n - .4513/n or somethnig like that.

[unique cypress]

Actual throughput is a minimum of 2 values. Train throughput and station throughput. With near 0 RtD, train throughput does go to infinity, but station throughput goes to 0. And the minimum of 0 and infinity is 0

[agile junco]

Oh no RtD includes the delay nvm

[agile junco]

Hmmm OK, I think I see what you mean.

[pastel canyon]

i mean thats 2 normal iron veins and 1 copper vein because im using stitched plates

not like im using 50 veins, i was complaining just 'cause i had to fuck around using weird deicmals and overclocks/underclocks

[pastel canyon]

great!

[agile junco]

I'm doing something wrong... I measured my RtD to be 1.68 min and my TtF is based on 100 stacks in a 32 slot car with 2 Mk.6 belts loading... so that's 1.33 min. Yet the car is partially loaded. Maybe my stopwatch time was off.

Nope stopwatch was correct.

Oh, I think I forgot the docking time in TtF.

[versed stone]

ok yea you gonna need them

[unique cypress]

@agile junco here's a screenshot of my formula for default stop settings
x is RtD, n is the number of trains, l is train length (wagons), b is belt (/pipe) speed, s is stack size (50 for fluids) and tp is pause time (~0.45 min)

This assumes the trains are equally spaced, though.

If they're all brunched up, you might as well assume you only have one

But it's probably even worse than just having one due to the extra pauses

[agile junco]

Yea I was just working on this idea. Thanks!

[agile junco]

wait, why isn't belt speed in the second argument to min?

Oh nvm, I get it.

This is the TtF < RtD case

[agile junco]

I'm going to save this:

I'm wondering if it's always valid to set the train delay to OR d/n

I feel like it should be.

[unique cypress]

Then it'll always depart as soon as it loads

That'll not space the trains out

[agile junco]

Yea, I was assuming you have enough trains...

Oh... I meant AND

derp

[jagged brook]

so i have a question im on phase 4 right now trying to get some infistructure up but ive got a hard drive here and both seem pretty good which are better in the long run

[unique cypress]

And tip: you can use \left( and \right) to get brackets that match their contents in height (assuming I'm right and this is written in latex)

[agile junco]

heh, I definitely knew that.

[unique cypress]

I basically never use rigor motor and extensively use iron wire

I feel like rigor motor adds too much complexity for something as simple as motors. Plus it replaces a cheap resource (iron) with a moderately rare one (quartz)

[jagged brook]

i mean yeah id need a manufacuter right? i dont have them unlocked
but i do have a somewhat close quartz node

i mean iron wire does just sound good

i mean i was looking at it like its a 3+1 to 6 instead of a 2 to 1

[unique cypress]

Rigor motor does need a manufacturer. And so do crystal oscillators

[jagged brook]

ight so iron wire then

so i have a througput quetion is this the place or no

[unique cypress]

You don't need to take either fyi

And you should do that if you don't want either

[jagged brook]

i mean both seem great

[unique cypress]

I'm saying in general

I don't take half the recipes I get

Though I'm not saying you should do that. But you can, if you want

[jagged brook]

cool so this may seem like a dumb question but is it better to under clock a foundery to make steel at 30/m and have 4 or is it better to over clock 2

power shards are not a problem nor is space in all honesty

[unique cypress]

Depends what you mean by "better". More machines making the same output take up more space but also consume less power

[jagged brook]

well wouldnt two over clocked a bit take less than 4 underclocked a bit
idk

[unique cypress]

If you want 120 steel, 2 foundries at 133.3333% take 47 MW (23.5 each) while 4 at 66.6666% consume 37.5 MW (9 and a bit each)

[jagged brook]

so 4 underclocked is technicaly better

[sly canyon]

oh thanks this is reassuring ||(cries in early phase 2)||

[unique cypress]

More space, less power. Neither is strictly "better". Depends what you care about more

[jagged brook]

right

much appreciated thanks for the insight

[sly canyon]

why not just do 2 100% and 1 66%?

[jagged brook]

i do have one last minor question are like adaptive framework and automated wiring only used for the space elevator?

like should i just make this setup with it being removed in mind or are they needed later

im on space elevator phase 2 so

[unique cypress]

They're only used for the space elevator, but they're used for every next stage of the elevator

[jagged brook]

fun

okay cool thanks

[unique cypress]

Then it's 3 machines and 41 MW

There are basically infinite options

[jagged brook]

some one say infinite

[sly canyon]

tbf, if you ask me thats the perfect balance for space and power usage

[unique cypress]

Especially if you also include the ones that don't have all machines running at 100% uptime

[unique cypress]

Underclocking all of them equally (to 88.8888%) would save you 0.3 MW while still being 3 machines

Imma be honest, all of my machines are at 100% until tier 9, and all at 250% from tier 9

I would overclock all of them earlier than T9 if I had the shards, but I usually don't

[oblique hollow]

never had that issue with insufficient power shards, then again i dont scale production as much

[unique cypress]

The power differences are too small for me to care as soon as I get to coal

Yeah, well, my tier 8 factory has like 800 machines at 100%.

That'd be 1k shards to OC them all

I have maybe 800

And so far I only used them for miners, oil extractors and some gens

I could probably scrounge together 1k if I pulled them out of the gens

But even then, that's 1 factory I can fully OC after collecting slugs for 8 tiers

Maybe I should've OCd the iron wire constructors or some shit but whatever

Placing the machines takes like no time compared to dragging the belts between them

[versed violet]

Spoiler: ||In tier 9 you unlock ability to manufacture power shards en-masse, no slugs required||

[oblique hollow]

we were talking about pre-tier 9 for the other stuff

[agile junco]

@unique cypress you ever use anything like Z3?

[unique cypress]

Considering I don't know what that is, probably not

[agile junco]

It's a constraint solver. I've been meaning to learn it a bit. This might be a nice task for it because it's pretty simple.

Basically tell it that the RtT is fixed (assuming it's the same regardless of the number of trains), and the other constriants and let it optimize the solution for cars and number of trains.

[wind spade]

You don't need to loop at all, you can chain it

[olive void]

noob here, what will a basic splitter do to an input of 5? will outputs always be integer? so like 2 & 3 (if only using two outputs) or 2&2&1 if using all three outputs?

[unique cypress]

Didn't say you do

[unique cypress]

Individual items can only be integers (you can't split an item in half). But item per min rates are split exactly in half. After all, 2.5 items/min is the same as 5 items every 2 minutes (when averaged)

[olive void]

ah ok I see. I'm trying to go from 5 items per minute to 2 items per minute

[vapid gorge]

what you're describing is a 'manifold'
it'll self balance on it's own as it overflows places

[olive void]

Oh yeah actually duh that makes sense.

[vapid gorge]

it's the simplest method for managing items like that 🙂

if the different sections aren't convenient to feed in a simple manifold, you can use clocking to have machines outputs onto different belts

[wind spade]

If you chain it, you don't really need any splitter/merger thing, and definitely not prio ones

[unique cypress]

it's hardly an issue to use one though

[wind spade]

No, but I was replying to your message about "needing" a loop

[unique cypress]

no, I said that if they use my design (with loops), they need a smart splitter

not that they need a loop

[vapid gorge]

@worthy shuttle so what are you trying to do exactly

[worthy shuttle]

The aluminium loop thing someone gave me a reference one but idk how to continue its sequence past the 3rd ref (aka the recycled one(

[vapid gorge]

ok how much bauxite total are you processing and what recipes?

sorry I was away from my pc

[worthy shuttle]

I have 600 buax to work with but idk how much aluminium im planning tk make yey

[vapid gorge]

basic recipes?

https://media.discordapp.net/attachments/1287987061361610793/1288029095216545792/alum_options.png?ex=694d86d2&is=694c3552&hm=4292b81c91a7aef4c87accc98b3322347ee8aeacc6f4ff95f7e620a5541df4fb&=&format=webp&quality=lossless&width=860&height=1216

if so you need 2/3 of the solution refineries run on fresh, 1/3 on waste

as you can see, the top left diagram has 2x 100% refineries doing fresh water and 1x 100% refineries runnign off waste

so if you have 600 bauxite, runn 400 bauxite pm with fresh refineries, 200pm on waste 🙂

clocked however you like

[worthy shuttle]

So if i want to upscale i just keep it as everyone third to under clock it?

[vapid gorge]

ok so just as a thing - if you want to 'upscale' make a new system

really really recommend you just process each bauxite belt on it's own and don't interlink them in anyway.

it's possible to do it but it's a giant headache and doesn't really give you any benefits

[worthy shuttle]

Ohh i meant like i haven't started making the aluminium yet and was thinking if making it big as possible at the beginning so I dont have to worry for later

[vapid gorge]

oooh yeah then just make small groups.
process each belt of bauxite independently and expand as much as you like

[vapid gorge]

though if you want to do a big system I'd probably use Sloppy Solution + Electrode Scrap.
I find them very convenient for large systems

[worthy shuttle]

Whats do they trade out for in the recipes?

[vapid gorge]

well Sloppy cuts out silica production and produces more solution

Electrode trades coal for oil
but you need a lot less oil pm so you don't need to move much.
you do need to process it into HOR then Coke though, but you also get more scrap

worth it imo

[worthy shuttle]

That is true might do some save scumming to get it

[vapid gorge]

Get all the recipes, go on safari. All useful

[worthy shuttle]

😭 its the power ones that keeps me from wanna do it

[vapid gorge]

Bring materials to make bio burners

The vast majority don’t need power though

[worthy shuttle]

Ik but the ones I have left are massive power amount (above 200) or late game items like heat sinks

[vapid gorge]

oh that's only like 2 or 3, you hsould have most recipes then

[dusky dust]

AFAIK by that point you should be able to have everything you need to construct Alien Power Augmenters, btw -- one of those will open any pod on the map

Useful to make sure you've got building materials for one tucked away in your Depot storage

[worthy shuttle]

I have a slight library of ones but they are horrible alts like alt wires and alt ingots

[pastel obsidian]

You can also make a bp of biomass burners filled with your favorite biomass type

[worthy shuttle]

I never used a single blueprint yet 😭 I make everything a 1 off design

[vapid gorge]

those are all fantastic xD

alts are just options and shine in their situation 🙂

[worthy shuttle]

But these alts are ingots mixed with water or just use caterium instead of copper

[vapid gorge]

yeah? that's great. Gives you options to not need copper or iron in a location - changes up posible factory location lots

pure ingots gives you more ingots! but I like the alloy recipes better usually

fewer refineries

[worthy shuttle]

Fair

[agile junco]

@unique cypress well, after a couple hours learning Z3, I think I have a working optimizer based on the equation you sent me.

It's consistent with my couple test trains and the tables in https://satisfactory.wiki.gg/wiki/Tutorial:Train_throughput

If you want I can share the code. It's python (unfortunately), otherwise maybe I can package it up somehow for people.

I'm thinking it shouldn't be based on belt speed but rather input speed in general, since you might have a system with output of some non-belt speed number.

[vapid gorge]

for example, super computers entirely made from caterium and oil using caterium wire 🙂

https://www.satisfactorytools.com/1.0/production?share=Z6tknI7esHM2v0YNt2Vd

[worthy shuttle]

But this aluminium one i just missing rhe sloopy solution so I mma reroll till I get it

[vapid gorge]

fair xD

[worthy shuttle]

LETS GOO I GOT EM

[cunning sorrel]

how to design a balanced coal powerplant

[vapid gorge]

feed the right number of coal and water into it 🙂

!wikisearch CG

[glad apex]

https://satisfactory.wiki.gg/index.php?search=CG

[royal yarrow]

Is that with Mk.2 pipelines?

[cunning sorrel]

same question

[wind spade]

Not necessarily

[royal yarrow]

isnt mk.1 300m3? so it cant be 360m3 unless its mk.2 pipelines right"?

im still new to the game so i could be reading it wrong

[wind spade]

But no pipe section is required to carry more than 300 per minute

[unique cypress]

If you want to add input/output rate into the equation, then that'd go as a third input for the min function

The platform's maximum throughput is determined by belt speed, not really input rate

If you connect 2 full belts to a container, then that container to a platform, you won't get those 2 full belts out, ever.

Platforms pause their belts for 27.08 seconds whenever a train docks. So the transfer rate in/out of a platform is 0/min for 27.08 s every time a train docks and 2b whenever one doesn't

Trains dock once to each station per round trip (ig they could dock twice or more, but basically nobody does that), so that first component of the formula is a weighted average of 0 and 2b over x with weights n*tp and x-n*tp respectively. Which simplifies to what I put in there

[plucky badger]

You maths and metas people is it worth load balancing in and out of trains

[unique cypress]

It's basically required if you want to get the maximum theoretical throughput out of them

And you also need ISCs as buffers and the trains need to be set to depart only when empty/full

[vapid gorge]

just clock what you want put on a belt in the first place

definitely not needed

[lyric aurora]

Hey yall, could someone please rq look into screenshots cause i need help, or yk, whenever you feel like it, no rush :)

[unique cypress]

Could've just pasted the image here lol

I'd use a balancer (obviously) but this can be done with a single splitter and 2 mergers

[lyric aurora]

Ohhh, thought it had to be screenchots xD

[unique cypress]

Images are only disallowed in #satisfactory and I think #off-topic-general

[lyric aurora]

Hmm, soo i dont need a smart splitter? Also, idk about balancers, theire preety bulky and i dont enjoy working with them that much.

[lyric aurora]

Oooooh, thankss, mate

[unique cypress]

If you have 3 belts of 300, then you split one in half, and you have 300, 300, 150, 150. Merge 300 and 150, you have 450. Repeat for the other one

A balancer would give you equal belts regardless of inputs, which is useful sometimes (mostly trains)

Balancers and mixers can also turn a random mix of belts into any other random mix of belts, as long as the belts can fill up

[lyric aurora]

Okay, ill start to consider them. Could be useful since im starting my train network

[agile junco]

Yea I don't think it really makes sense to model train stops like that. The biggest limitation of all this is still assuming perfectly spaced trains, but I don't see a great way around that without making the modeling much more expensive.

And I decided that adding input throughput was pointless, since as you said, you can just ensure that your train has >= that you need.

I added a --throughput flag which makes the program solve for the minimum throughput which is >= to that value. So it can determine number of trains/car/rtd. or let you set them and have it work out the others.

usage: trains.py [-h] [--stack STACK] [--belt BELT] [--rtd RTD] [--max-trains MAX_TRAINS] [--trains TRAINS] [--max-cars MAX_CARS] [--cars CARS]
                 [--throughput THROUGHPUT] [--full]

For pipes, use --stack 50 and --belt=<flowrate>

options:
  -h, --help            show this help message and exit
  --stack STACK         Item stack quantity (default: 100)
  --belt BELT           Belt speed (default: 1200)
  --rtd RTD             Round trip time, otherwise optimized for (default: None)
  --max-trains MAX_TRAINS
                        Max number of trains (default: None)
  --trains TRAINS       Number of trains (default: None)
  --max-cars MAX_CARS   Max number of cars (default: None)
  --cars CARS           Number of cars (default: None)
  --throughput THROUGHPUT
                        Min throughput needed (default: None)

$ python3 trains.py --stack 500 --belt 480

minimize cars, minimize trains, minimize rtd, solving optimal throughput
Stack Size: 500
Belt Speed: 480
Trains: 1
Cars: 1
Loaded: full
Round Trip Time: 17.118 min (1027.08 sec)
Throughput: 934.6886 items/min

Matches output from the graph here: https://satisfactory.wiki.gg/wiki/Tutorial:Train_throughput#/media/File:Train_throughput_as_function_of_trip_time.jpg

Also most useful is that I can solve for a given round trip time and needed throughput:

$ python3 trains.py --stack 100 --belt 1200 --rtd 9 --throughput 3000

minimize cars, minimize trains, minimize throughput >= 3000.0
Stack Size: 100
Belt Speed: 1200
Trains: 5
Cars: 2
Loaded: full
Round Trip Time: 9.0 min (540.0 sec)
Throughput: 3555.5556 items/min

I'll try to post the code today or tomorrow. But it's my first time writing Z3 and I really don't know what I'm doing that well with it still.

[agile junco]

https://github.com/nixpulvis/sat_is_factory

[full dawn]

I have a Pure overclocked oil node filling 20 refineries. Should I feed them with 2x 300m3/min pipes instead of 1x 600m3/min pipeline?

I'll loop them of course

[agile junco]

The 600 coming out of the extractor is fine, why would you need to split it?

[full dawn]

I might have mistaken it for when people have problems supplying fuel generators with a 600m3/min Mk2 pipeline. Some people have suggested to split it into 2 mk1 pipelines

[hushed silo]

quick question for the daring

assume air resistance, friction dont matter, how much 1 engine 4 cargo trains would more or less take to transport all that bauxite into pit?

[unique cypress]

I'd probably use 4?

One per node "group"

One per should probably be enough

[agile junco]

I've also heard this, but honestly haven't had that issue myself.

[agile junco]

What's the RtD?

I'm also starting to think about unifying my Bauxite processing a bit along these lines, but there's no way a single train is going to do it.

[dusky dust]

That is a common setup which can cause problems for folks, though there are piping arrangements which can work fine. That does tend to be a finnicky system though which can rely a lot on exactly how you build

[hushed silo]

4 trains per all of that?

[hushed silo]

what does this acronym mean

[agile junco]

To answer your question though. You basically want to minimize the loop to the smallest size a single 4 car train will utilize fully, Which is 32*100*4 items per round trip.

[agile junco]

Round Trip Duration, sorry.

[hushed silo]

oh, i dont know if train will be round, i just wantes to highlight bauxite

[agile junco]

wdym? It has to go in a loop, no?

[unique cypress]

you can probably assume a single 1-4 train will handle like 2000-2500/min

if you have mk6 belts, you might need to add an extra train to one or 2 of the groups

[agile junco]

I mean, if the RtD is low enough it can be much more than that.

[hushed silo]

so 6 trains more or less

that will need to be big station then

[agile junco]

he means 1 train 4 car.

not 1-4 trains.

[unique cypress]

ah, right, how much bauxite is that total?

[hushed silo]

12k

[unique cypress]

4 platforms can only handle 7100

you probably want to use 2-8 trains

then 1 per node group will be plenty

if you only want to have one unloading station

[agile junco]

Belt Speed: 1200
Trains: 9
Cars: 7
Loaded: partial
Round Trip Time: 15.0 min (900.0 sec)
Throughput: 12250.56 items/min

[unique cypress]

1-4 will work if you have 2 station on the unloading side

[agile junco]

asssuming a 15min RtD you'll need 9 trains and 7 cars each to process 12K.

[unique cypress]

it's pretty obviously not 15 min RtD lol

[agile junco]

Yea, IDK why I put that in lol

maybe like 6

[unique cypress]

unless you want all trains to stop at all stations, then it might be

but that's dumb

each train can have only stations on schedule

[agile junco]

Stack Size: 100
Belt Speed: 1200
Trains: 3
Cars: 8
Loaded: full
Round Trip Time: 6.0 min (360.0 sec)
Throughput: 12800.0 items/min

Yea, that's the thing. Your first point about breaking it into 4 groups is best.

[hushed silo]

so how many unloading stations do you think ill need

[unique cypress]

I mean 1 is enough if you use 2-8 trains

2 if you use 1-4 trains

[hushed silo]

realy?

[agile junco]

Wait, why would you use more than one unloading station?

[hushed silo]

why not

[unique cypress]

Either way, you need 8 platforms

[agile junco]

For the same train?

[unique cypress]

Well, 7, but that doesn't divide well

[hushed silo]

gotcha

[unique cypress]

No, not the same train. One for 2 or 3 trains, and another for the rest

[agile junco]

I mean, if you break them up into 4 groups and each one is only like 3-4K buaxite each, then you can do something like this:

Stack Size: 100
Belt Speed: 1200
Trains: 3
Cars: 3
Loaded: full
Round Trip Time: 6.0 min (360.0 sec)
Throughput: 4800.0 items/min

3*4 is 12 trains though, so you might get some congestion.

[hushed silo]

congestion is what im worried about honestly

[agile junco]

There are different kinds of congestion.

There's number of trains, which impacts crossings and stuff... then there's backups waiting to unload, and station size, which number of cars also impacts greatly.

If you have 3 trains with 6 cars, you need to be careful to have staging space for the cars when they're waiting to dock.

[unique cypress]

Set all trains to only depart when empty/full, add a waiting bay before the station (s) and you'll be fine

[agile junco]

That works for 3 trains, but if you have more than 3 in a single route it can be iffy.

@unique cypress btw, any interest in testing out my optimizer? I'm pretty happy with it, but it's not perfect.

I'd be curious what you think could make it more useful.

[hushed silo]

allright guys

ive read all info

but im too tired to respond 😅

so thank you and good night, see ya

[unique cypress]

Oh, I was planning to take a look at it. I just don't have the time or energy to do it today.

After all, it's Christmas and I'm busy visiting family

[agile junco]

Yea, no pressure. I just got pretty excited about this last night.

message me if you want help getting it to run.

I think I'd like to support train route with multiple stops as well... but I haven't thought about that much, or needed it.

[fierce ruin]

is this the most best way set up if i want to prioritise plates for reinforced and excess overfill goes aside? for example 135 for all 5 (27 per 5) and 15 is left over

and yes thats a splitter next to my player and merger in mid air

would be nice if i got tips for improving efficiency to max everything

[primal shadow]

So it’s a python code to calculate amount of trains needed to transport a certain amount of items right?

[limpid knot]

That manifold looks a-ok.

[primal shadow]

Sorry Ngl been lurking and im curious af

[fierce ruin]

thanks, its bit compact which i like too haha

[agile junco]

Basically. Or to solve for how many items you could transport.

[primal shadow]

How’s it work? Does it ask now many or do yiu write into the code

[agile junco]

Also can solve maximum throughput.

[agile junco]

It uses a solver called Z3 to optimize an equation for the throughput.

[agile junco]

This is the main equation part: https://github.com/nixpulvis/sat_is_factory/blob/5233754965bac93cc933814195bd1b4d6cee5720/src/sat_is_factory/train_solver/train_solver.py#L34

[primal shadow]

So you’re maximizing it?

[agile junco]

Well it depends.

[primal shadow]

Wait no I mean Z3 uses like math to maximize basically a graph or something

[agile junco]

I'm actually generally minimizing things. For example, minimizing number of trains, and cars. And minimizing throughput >= the desired throughput.

If you just give it a round trip time and no throughput, then it maximizes the throughput tho

[limpid knot]

You can nudge structural pieces in instead of the unused splitters you've got in your splitter towers, like concrete pillars. Different aesthetic. Yellow-hologram clipping doesnt impact performance of anything.

If you turned it 90deg you could have two rows, three and two machines respectively, fed by the same inputs. It would reduce your part count a little.

You can try using the stacking belt towers to run a column of two belts between, and splitters can be applied to belts after the fact, at which point you can use angled belts or even belt lifts to feed the machines.

Then its a two-sided manifold, and that's all I've got for you. You can run six machines instead of five, and underclock their outputs to match current output, and it is a marginal power saving for same output with a slightly larger--but symmetrical--footprint.

[primal shadow]

Ok that’s sick

[fierce ruin]

they already underclocked

[primal shadow]

Oh holy it’s not like in python

You run it with sh stuff right?

[fierce ruin]

but thanks for the tip

[primal shadow]

Like it’s a program that’s in your computer with integrated commands?

[agile junco]

Read the README. You'll need to install Z3 though.

[limpid knot]

Your setup is perfectly functional, no need to change it. My suggestions are cosmetic, and where they're not, the gains are so marginal or lateral, that it makes no functional difference, only aesthetic. (Which, at the end of the day, its a game. Nothing matters except user experience.)

[primal shadow]

Oh ok I get it

[fierce ruin]

i much prefer very minimal but performance over looks for machines, for build and shelter i like the cosmetic tbh

[agile junco]

I want to make a windows app, but it's a bit of work.

[primal shadow]

Ya that’s like a lot of effort

[fierce ruin]

still looks nice looking at the machinary

2 conversations at once love that

[agile junco]

If you have a mac or linux machine it's easy to install. On window it should be easy too with NuGet.

[primal shadow]

When it says 5 trains two cars it means 5 trains 2 cars each right

[agile junco]

yes

and it assumes they are spaced evenly.

[primal shadow]

Im not at my computer rn but will 100% try when I get the chance

[agile junco]

Which is something I'm still trying to work out exactly in practice, but should be possible I think.

[primal shadow]

Like exactly to the second?

[agile junco]

It's a nice little generalization from the wiki and it solves things pretty consistently.

[primal shadow]

That’s sick

Must’ve been fun to make too

[agile junco]

Yea. I'm still new to optimizers like this, but now I kinda want to implement a factory solver too.

Since I love SFtools, but I often want to optimize for other things.

Like # of machines.

or footprint size

[primal shadow]

Doesn’t satisfactory manager kinda do that

[primal shadow]

Oh I see

[agile junco]

satisfactory manager?

[primal shadow]

It’s the one on steam I don’t remember exactly what the name is

[agile junco]

modeler?

[primal shadow]

It’s literally sftools but a steam game

[primal shadow]

Ya that one

[agile junco]

I use modeler a lot, but it's not like SF tools really.

It's not an optimizer.

it will help you build things out and keep track of ratios.

But it cannot tell you want to do to get to an end result.

you still have to choose all the recipes and flows yourself.

[primal shadow]

I like that better

Sftools plays the game for you

[agile junco]

I partially agree

But sometimes I like to double check my builts or find new recipe combinations.

[primal shadow]

Fair

[agile junco]

And it's not playing the game for you if you build the optimizer yourself 😉

[primal shadow]

Ya I’ve been getting into trains in my recent play through and it’s my first time using them (however I have been watching satisfactory videos for years, just never had a pc to run the game), and I was wondering like what the optimal amount of items per minute compared to max belt is per cart

Then I realized it’s more complex than that lol

[primal shadow]

Ya exactly that I find so cool and I’ve been thinking of trying to make one with python but I don’t think I know enough python

[agile junco]

It's cool playing with this program, because it makes it so obvious that just adding another train can actually reduce throughput.

E.g.

$ python3 -m sat_is_factory.train_solver --rtd 5 --trains 3

minimize cars, maximizing throughput
Stack Size: 100
Belt Speed: 1200
Trains: 3
Cars: 1
Loaded: partial
Round Trip Time: 5.0 min (300.0 sec)
Throughput: 1750.08 items/min

$ python3 -m sat_is_factory.train_solver --rtd 5 --trains 4

minimize cars, maximizing throughput
Stack Size: 100
Belt Speed: 1200
Trains: 4
Cars: 1
Loaded: partial
Round Trip Time: 5.0 min (300.0 sec)
Throughput: 1533.44 items/min

At some point, the trains are just wasting time docking.

[primal shadow]

Woah wth

Are you calculating for traffic?

[agile junco]

No

[primal shadow]

Or is it loading times

[agile junco]

yea

[primal shadow]

Ok wow

[agile junco]

Traffic would need a complete model of the railway and signals... way too much.

[primal shadow]

So why wouldn’t I just do a huge train with like 10 cars vs having 3 trains with 3 cars

[primal shadow]

Ya that’s what I thought I thought for 2 seconds and it’s require a whole simulation or something at that point

[agile junco]

I like to minimize cars, because my whole build philosophy is to minimize build area. And train stations with a lot of cars are big.

Not to mention, they go slower, so there are limits.

[primal shadow]

Ya but you can add engines

[agile junco]

Yea, you're not really optimizing an equation at that point, but simulating a system and taking an average.

[primal shadow]

Ya lol

[agile junco]

But yea, by default I prioritize minimizing cars, to save space.

[primal shadow]

So what’s better for throuout efficiency: more cars per train, more trains with less car, or a mix

[agile junco]

I added a --minimize trains option to change that though.

[agile junco]

Depends. Both can work.

[primal shadow]

Ooo interesting probably gonna use that

[agile junco]

Adding cars is definitely simpler though.

[primal shadow]

For sure but there’s always the situation where you can’t

[agile junco]

For example:

$ python3 -m sat_is_factory.train_solver --throughput 3500 --rtd 6

minimize cars, minimize trains, minimize throughput >= 3500.0
Stack Size: 100
Belt Speed: 1200
Trains: 3
Cars: 3
Loaded: full
Round Trip Time: 6.0 min (360.0 sec)
Throughput: 4800.0 items/min

$ python3 -m sat_is_factory.train_solver --throughput 3500 --rtd 6 --minimize trains

minimize trains, minimize cars, minimize throughput >= 3500.0
Stack Size: 100
Belt Speed: 1200
Trains: 1
Cars: 7
Loaded: full
Round Trip Time: 6.0 min (360.0 sec)
Throughput: 3733.3333 items/min

I might choose 3 trains of 3 cars, but you might prefer the one of 7. Both satisfy >= 3500/min

[primal shadow]

Ok I see

But 2 trains with two cars will almost always be less efficient than one train with 4 cars

For example

Ya this program is really cool I’ll give it a shot when I get the chance

[agile junco]

Yes, that sounds right.

[primal shadow]

When it says loaded full what does that mean

[agile junco]

That the train is full.

[primal shadow]

Fair lol

[agile junco]

https://satisfactory.wiki.gg/wiki/Tutorial:Train_throughput

[primal shadow]

One thing I don’t get; do we assume we’re putting one maxed out belt into each freight station?

[agile junco]

the --belt argument assume 2x of those belts.

That's something I'm on the fence about maybe making a bit more clear. But I don't want to confuse things. And right now I'm following the wiki a bit.

[primal shadow]

Ya no that’s fair I probably didn’t read the readme fully enough lol

[agile junco]

But I could see using the total input throughput instead, so 2400 instead of 1200 for Mk.6 belts.

The readme is woefully incomplete on actually explaining all this.

I wrote this last night and need to actually document things in the --help output.

[primal shadow]

Ok

[primal shadow]

Isn’t that bad because of loading not letting anything in?

[agile junco]

But if you play with it a little and compare the numbers to the formulas, it should make sense.

[agile junco]

?

You can't get around the docking time being down.

That's why nothing you can do will ever get you 2400/min throughput.

[primal shadow]

I misunderstood, I thought you were saying you should load two full belts per cart

[agile junco]

python3 -m sat_is_factory.train_solver --stack 500 --belt 1200

minimize cars, minimize trains, minimize rtd, solving optimal throughput
Stack Size: 500
Belt Speed: 1200
Trains: 1
Cars: 1
Loaded: full
Round Trip Time: 7.118 min (427.08 sec)
Throughput: 2247.8224 items/min

This is the most you can ever get from a single car, since there's nothing that stacks higher than 500. And even if there was you wouldn't get 2400.

[agile junco]

You should.

[primal shadow]

But not all items will go through?

[agile junco]

Well techically, you can only get 2247.8. But that's almost 2 belts.

Correct.

If you need > 2247, you need more cars or trains.

[primal shadow]

So you’re losing ~100 items for the sake of having perfect throughput

[agile junco]

No, you just cannot do it on a single bay.

so best to send those 100 items to another bay and let them deal with it if you need them.

This is actually where balancers can be helpful, since if you want to make sure your trains are following the schedule, you want to set them to depart when fully loaded/unloaded. So you want the bays to fill at the same rate.

I'm dealing with a sushi train rn, which carries nitric acid and iron plates... which means I cannot fully unload/load, so I'm still testing wait times to solve the issue that way.

[primal shadow]

Ok I think I get it, correct me if I’m wrong; this program calculates theoretical max throughput, and can minimize cars and stuff depending what your max belt is. It always assume 2 of the max belt into every car, but that doesn’t mean at all those belts should be full (unless coming out of a container)

[agile junco]

Here's a better example... let's say you're carrying ore (stacks to 100). This is the max you can do with a single train/car:

python3 -m sat_is_factory.train_solver --stack 100 --belt 1200

minimize cars, minimize trains, minimize rtd, solving optimal throughput
Stack Size: 100
Belt Speed: 1200
Trains: 1
Cars: 1
Loaded: full
Round Trip Time: 1.7847 min (107.08 sec)
Throughput: 1793.0519 items/min

So if you ask how can I transfer two full pure nodes a distance of 2min:

python3 -m sat_is_factory.train_solver --stack 100 --belt 1200 --throughput 2400 --rtd 2

minimize cars, minimize trains, minimize throughput >= 2400.0
Stack Size: 100
Belt Speed: 1200
Trains: 1
Cars: 2
Loaded: full
Round Trip Time: 2.0 min (120.0 sec)
Throughput: 3200.0 items/min

You can see you need more cars/trains.

[agile junco]

The best way to think about it would be that the belts are split evenly to each car.

[primal shadow]

Ok hence balancers

[agile junco]

though technically that only matters for spacing out the cars and having them leave loaded when they should.

Like this program might tell you the optimal solution involves 2 fully loaded trains. But they in reality get backed up right behind each other, so the second one docks and doesn't get fully loaded unless you get the setting to make sure it waits.

That setting being optimized is dependent on the belts being balanced.

If the optimal solution has you running 1 partially loaded train, then this part is irrelevent, so you could skip the balancer, so you might still need > 1 docking bay if you need to carry more than what a single bay can provide.

A single bay, despite having two inputs, can never reach 2 * belt speed, because the docking time they get paused.

Storage units, despite what some people believe, do not address this issue. They just may help smooth out the draw from bursty building rates on the production side.

[pastel obsidian]

The use of storage units is to make sure you are able to to export one compressed belt from a station

[agile junco]

But you could have a lot more than one belt generally.

[pastel obsidian]

you can have 99% to 19% gains you are leaving on the table

depends on belt and item stack count

[agile junco]

I'm just saying that storage units don't really solve any actual problems.

[unique cypress]

They help with a different issue though. When the belts pause, and you don't have buffers, items start backing up in machines connected to the platform. If it's something like a miner, it could easily fill its internal buffer in 27 seconds and start idling

With a buffer, those items are kept inside it, and then loaded into the platform when the belts are unpaused. No idling

[agile junco]

Yea, I guess that falls into my comment about bursty builds. But yea, anything where the internal buffers will backup before the dock is done.

[primal shadow]

Ok I get it

Thank you im looking forward to using trains again

[agile junco]

Yea exaclty, and as the belts get faster the more you notice the efficency lose.

[pastel obsidian]

that is due to the station not being able to accept new items for 27 sec , honestly play with the mod that sets it to zero for some QOL

[agile junco]

haha, or just allows the storage to continue to fill... but that's not realistic.

would make the whole point of my little coding project irrelevant though lol.

The mod I'm more interested in would allow waiting for fully loaded/unloaded per car.

[pastel obsidian]

you can fully load and unload

[agile junco]

Or a general purpose way to control the maximum number of a spesific item in a train car.

[pastel obsidian]

its an option

[agile junco]

Read what I said more closely.

[pastel obsidian]

I think it would be nice to just specify how much we want each train / wagon to load / unload

[agile junco]

My situation is that I really don't want to have to build another station just to transport a few iron plates when I have a perfectly good train carrying the nitric acid. But I need to maximize the acid and it's a bit clunky.

tomorrow when I get some time I'm going to test just adding manual delays to keep the train separated though.

This is my first large train build in general, though I've decided 8,200 quickwire is just getting belted a medium distance to the nuclear plant.

Honestly it's not too bad, I think the highest throughput line is either something like 2500 aluminum, or 3000 quartz.

[primal shadow]

Hey I do have one more question; if all I want to do is have a consistent throughput on each end, as in 1200 go into one station, 1200 come out of the other, do I need to do all this? Or is there like a quick and dirty way you could recommended? Like 1 max belt per freight, assuming you put it into a industrial container

[pastel obsidian]

2 belts into a ISC and one belt out

[agile junco]

And do the quick math to make sure the round trip time is not greater than the time to fill the 32 stacks a train can carry.

Otherwise add cars/trains.

Or do what some people I've seen do and just run 8 car trains everywhere lol.

[primal shadow]

Ok I think I’ll try that but there’s already spots in my world where there’s a limit to the amount of cars I can have so I’ll try the program in that case

[agile junco]

Optimize all the things!

[hushed silo]

this sounds like some fight anime analyzer

but the program looks sick ngl

[fierce ruin]

am i better of having depots when producing parts or have them where i have all my storage which means they will be transported too to the storage

[hushed silo]

about the storage, imo if u want one its better to have it in one place, however in this game u dont realy need one because dimensional depot exists

[fierce ruin]

yea

true

[hushed silo]

and then the temptation of box feeding and giga factory in one place kicks in

the urge

[unique cypress]

The point of dimensional depots is that it doesn't matter where in the world they are

I always build mine right where production is, because why would I build them anywhere else

But if you want central storage for whatever reason, you can put all your depots there instead. I suspect it won't make a difference

[fierce ruin]

also with the depot, its quite annyoing because my rate is 30 per minute but i dont alwways use raw materials so the lane becomes 270 not 240 and that annoys my assemblers too

[unique cypress]

Overflow to sink

[fierce ruin]

thats what i got with my plates

[unique cypress]

Oh, and also, you should always have a container between your production and the depot uploader. As a buffer

[fierce ruin]

idk how that would be useful tho

[unique cypress]

So the depot can be refilled at full belt speed instead of just production speed

[hushed silo]

just so there is storage if ur production rate of said item is lower than depo upload and ur usage is higher

[fierce ruin]

and i got another issue since i dont have super high belts, this issue is with my reinforced and modular frames + rotors, they are all finished product of 22.5 or left and depot is 30 what do i do

[fierce ruin]

my production is same as my depot tbh 30 for 30

[hushed silo]

it will change later on dw

[fierce ruin]

that depot is 30 and that constructor 30 too

[hushed silo]

so lets say there is an item that u are making 5/min but u need like a couple of hundred of it right now

storage is prefect for that

dont need to wait for them to be made

[fierce ruin]

alr i get your point

ill add one then

[unique cypress]

Depots go up to 240 with upgrades. And you can have several per item. They get fast.

[fierce ruin]

but what about this issue

[unique cypress]

And it's still not enough for my concrete needs

[fierce ruin]

the reinforced plates 22.5 vs depot 30

eh

[hushed silo]

why is this an issue tho

u dont need to be making each part as much as depo upload speed

[fierce ruin]

yeah but hold on

[hushed silo]

okay i fastened my seatbelt

[fierce ruin]

for modular frames i need reinforced plates, but because i need depot for instant place mats with reinforced, its an issue

because i got low numbers on reinforced

22.5 reinforced merged 1 belt -> depot which is 30 -> modular frames

the depot eats it up if i use reinforced lots

ill have barley any modular

[hushed silo]

place depo after modular frames are fed

ore use smart splitter and set overflow to depo

[fierce ruin]

i wish i can set number on smart splitter

[unique cypress]

Make more reinforced plates

[hushed silo]

thats what u can do now, eventualy u will be making more

[unique cypress]

I always make separate production specifically for the depot and nothing else

!wikisearch independency

[glad apex]

https://satisfactory.wiki.gg/index.php?search=independency

[fierce ruin]

my brains splatted on the ground, too many choices to make

alr whats the best pick here then:

A) place depot after prduction is made which will go into depot and if more excess then into sink

B) make seperate production which might be bit complex

C) place depots before asembleers

i have option c atm

[hushed silo]

b is the best but long term
a is short term for rn

[fierce ruin]

Alright then thank you

with B, im thinking of splitting 2 paths for production from the miner i think

rather going for other nodes

[hushed silo]

dont be afraid of other nodes, they dont bite

[fierce ruin]

i have another fear of predicting how much i need resourcers as well and if i aint gonna have excess after the production

i like everything perfect but it hurts veins to think

[hushed silo]

if u will be looking for perfect all the time u wont move much u know

[fierce ruin]

in progress?

[hushed silo]

related to time yes

[fierce ruin]

how does everyone else cope with these things

keep in mind im like tier 4

[hushed silo]

personaly i make an idea, lets say i want to produce this and this here and here in X ammount for this

then i make measurement, sketch building shape and think of a couple different options

[fierce ruin]

this game requires engineering freaking hell 😭

i need a degree

[hushed silo]

u dont

just experience

[fierce ruin]

fair i guess ye

[hushed silo]

if u just want to play without much consideration floating platforms always work the best bcs u dont need to plan that much and make it up as you go

[fierce ruin]

ill scrap my current factory build and do everything again and go for B option

[hushed silo]

u dont need to scrap

if its is there it can serve you

[fierce ruin]

do you want to see my world

[hushed silo]

u can send pictures to #screenshots sure

[fierce ruin]

alr

ill send all the logistic notes too

[hushed silo]

however what i am trying to say is that u dont need to stress that much about it

come up with idea you wanna do and make it happen

[fierce ruin]

i think i lack imagination

[hushed silo]

to gain satisfactory imagination u need some experience first 😃

we perceive environment in relation to our own ability

meaning to get some ability u need practice and try new solutions from time to time

[plucky badger]

If u want i can get a ss of my load jn and out put but i have 2 ICSs per freight platform but i have 12 long trains with a belt jumbler on one side

[fierce ruin]

i guess i like being perfect and clean so its hard

its in screenshots

[hushed silo]

its very nice and organized

[plucky badger]

I have gone with a belt jumbler on both sides like a belt balancer 3 to 3 but 5 to 5 back to back

[unique cypress]

Figure out how much of thing x I want to make, plug that into SFTools, play around with the recipes until I get something I want, then just build that

[fierce ruin]

thanks

[plucky badger]

So it slightly evens out the inputs but not balanced and more compact

[fierce ruin]

how do i find this SFTools

[hushed silo]

just google

[unique cypress]

http://satisfactorytools.com/1.0/production

[fierce ruin]

ah

[hushed silo]

bro has sati tools under ctrl + b command or smth like that xd

[fierce ruin]

flipping hell this useful

thank you

this my current build atm

[unique cypress]

I have it pinned in my clipboard lmao

[fierce ruin]

is this visual thing calcuates machines and power in the most efficient way possible?

[vapid gorge]

it just tells you how much clocking % you need in total

[unique cypress]

It calculates for the lowest resource consumption possible given the constraints you set

It doesn't minimize power consumption though

[vapid gorge]

so 4.5 constructors is just 450% clocking however you like it

[fierce ruin]

ohhh

i want power efficiency

[vapid gorge]

also - the difference in power consumption through clocking is so little, and making power is so easy, power efficiency in the game is basically meaningless

[vapid gorge]

clock 450 constructors at 1% for the 4.5 constructors then

[unique cypress]

Then better start setting all machines to 1% speed

[fierce ruin]

xd

[unique cypress]

1% is like 4.5 times more power efficient than 100%

[vapid gorge]

but seriously, power is incredibly easy and clocking doesn't make a huge difference . I wouldn't bother

[fierce ruin]

so you guys build with overclockers right?

very minimal

[vapid gorge]

I'll OC things as close to 250% as my designs allow

[unique cypress]

Yeah, I overclock everything to 250% as soon as I have the shards

[vapid gorge]

when I have infinite shards anyway

[fierce ruin]

oh dang, when do i unlock this resourcer which allows me to

which tier

[unique cypress]

Tier 9 💀

[hushed silo]

personaly i OC for the ratios i want but also trying to save as much space as possible

[fierce ruin]

dang, at least i got somersloops and i can get 2x XD

[unique cypress]

With sloops, there's like 5000 shards to be found + you can get slugs from doggos

[vapid gorge]

treat the milestones like a good tutorial - you don't actually have to design 'good' systems to finish all the phases. You can create the sloppiest factories imaginable and still finish everything

use the tiers to learn and then build your own projects after

[unique cypress]

If you don't build on a huge scale, that's usually enough

[fierce ruin]

idk how you even have them somehow get that

[vapid gorge]

there's lots of slugs onthe map

[unique cypress]

Only like 1000 slugs to collect

[fierce ruin]

do i put a slug into their slot?

[vapid gorge]

you make shards

[fierce ruin]

you got a point tho, cuz max stuff is better to use, i hate thinking about expanding constanrtly and deleting my factorieis that i spent so much time to make it pretty

i might just start building more bigger for purpose

[unique cypress]

Generally, the bigger you build, the bigger you need to build

[vapid gorge]

I really wouldn’t bother before you’ve unlocked everything, your needs and style will change

[unique cypress]

Because the bigger you build, the more building resources you need. Which means the factories for those materials

[fierce ruin]

true

ale then thanks for all your tips

i appreciate it

[hushed silo]

dont forget to turn slugs into shards in constructor with sommersloop inside for x2

[fierce ruin]

oh yeah i always do that, i even did it with all the drops from animals and it gave me so much dna points

tons of tickets

[hushed silo]

good

[fierce ruin]

is there a way to edit the iron ore to set limit so it can all calaculate

[shadow sinew]

in the resource tab

[fierce ruin]

thanks

[fierce ruin]

what about belts

mk3 limit

is this the best set up for depot @hushed silo