#math-and-meta
1 messages Β· Page 118 of 1
but if you're realising that every time you build a factory there's 10 things you could have done different or better? you're playing right
just don't let it stress you out
alright, good talk, this helped me out alot. appreciate it
How bad can throughput losses be on belt to belt Mk5 connections? I quickly upgraded an aluminum setup to 600ppm bauxite from a single Mk3 miner, and it ran fine for dozens of hours. Now after making a new and totally separate line of batteries and packaged nitrogen that utilizes 150pp bauxite, the original aluminum facility occasionally halts due to water buildup. I have no idea why, but bauxite running dry causing machine start/stops is one of my first guesses.
wdym 'throughput loss'?
as in you're trying to move 600ppm on a mk5 belt and your machines are starving of bauxite?
In theory the belt is now moving 780ppm. But in practice Mk5 belts can't do full throughput if there are belt to belt connections, IIRC
old data - that issue is essentially solved
even on a massively long mk5 belt, with lots of belt to belt connections and low fps I doubt you'd ever get anywhere close to 600 throuhgput with that bug in any case
either math or build error
or both
In the last test I saw, the maximum loss we were able to achieve was ~1% if memory serves
in non mp without terrible fps I saw results with like 99.9998% throughput
I'm on my way back to the miner in case it turns out I was being stingy with power and only clocked it to produce 600ppm or something
and with margins of error who knows
well if you only need it to move 600ppm why not clock it to 600?
I need it to move 750ppm
The 600ppm line was my first aluminum setup for sheets and cases, I added a new line that needs 150ppm for batteries and packaged nitrogen
ah must hvae mis read sorry
well if it was a fluid issue stuttering it if anything should make it more full of bauxite ready to be processed
it happens π
Does anyone know a mod that can set drones to only take off when full? one two way drone can take around 300 IPM, So for a 780 nodes, I need to use 3 drone ports. But lose out because of the 3rd drone taking off less then half full.
whats wrong with it taking off with it being less than full?
at best you'd only save a couple batteries
A 780 should only need 3 ports if you are crossing the whole map... 2 is more than enough for most :/
Esp when you have double drones on each port
curious how ppl set up massive water arrays, I'm having trouble with getting consistent flow using a water tower and key pipe system to feed roughly 6 pipes a 600 flow rate. The issue comes when one or two of the pipes I'm trying to feed just simply wont get any flow rate from the key pipe even though all the other ones run fine. if anyone could explain better i would appreciate it β€οΈ
its been a while, but last time i tested, 2 ports couldn't carry fast enough and back up. But that was a while ago.
don't use water towers for the most of it. keep it simple
keep things simple - don't merge pipes and link systems like that
If there's no flow ever, rebuild your connections where the flow stops. Try removing floor holes, valves. Reconnect observing the flow.
you donβt use water towers in general? I always thought pumps consumed to much power on a large scale.
been staring at pipes and redoing connections for a few hours now, itβs hell π
nah, pumps are super cheap if you're getting to the point of tons of water needed for large production>
just don't build silly sky factory hundreds of m in the air
pipe rules to go by #1190597884681396296 message
and honestly if whatever you've build is 'good enough' just keep going, don't burn out redoing a whole thing, jsut learn for next time
sounds good iβve been playing off an on since update 3 so iβve got a decent grasp of most of the game just trying to learn the intricacies atp
appreciate the help π
no stress - the rules in the link are flexible mind you - you can bend or break them in certain ways if you know what you're doing - but as listed they work for any situation.
~1% of the loss in U7 (for belts of the same segment count ofc) 
@fringe pawn feel free to @ me for more details on single input sushi systems 
I don't mean to brag, but they do be my area of expertise π (I also just really like to discuss the subject
)
I'm currently DJing, and I don't chance running my factory since I didn't bring a second laptop π
That's not a Satisfactory DJ 
idk what a water tower is but personally I just keep it as simple and organized as possible so that when there is issues I can identify them easier
@woven fossil see #math-and-meta message for alt recipe choices
yeah i was up literally all night working it out. i think i was just overcomplicating things by connecting multiple pipes together creating back flow issues
i slapped 80 pumps down and itβs working perfectly now
alright
just drive by the station, a node that looks like the pause icon should appear, if it does then your truck will load
0, that should make it wait automatically
@obsidian creek see #math-and-meta message for alt recipe choices
will 200 water flow into the pipe at the front? and 100 to the left
@frosty owl If I build a single input factory with intentional excess that feeds to storage ISCs and then a sink once those are full, does that work? For instance, an HMF factory that also produces excess encased beams and so forth.
its easier if you just use a valve.
Or if you dont have one - to not try to adjust flow like this
ah true
if you use a valve you must use 2 however
one set to about 200, one set to about 100
do they need power?
no
its better to set them to a bit more than the needed limit so they can compensate a bit for their inaccuracy
underestood, thanks
I really cannot wait for them to fix those arms clipping
Wait. Hold up. They said they're going to?
Refineries too?
iirc both were mentioned in regards to 1.0 improvements on stream at one point, but I don't know if they ever confirmed that they were definitely coming.
That would make me very happy. Because otherwise the temptation to pack them at their hard clearance boxes is simply too much.
Anyway, I'm fuzzy on how smart splitters work. If I put caterium and copper ingots on that same line, am I good to set each smart splitter to the appropriate ingot to feed the machine, then overflow out the center output?
Should be
Because overflow is also "any" I assume?
Yup
this should work as long as the head lift of the 2 outputs is the same. You can put a pump at both outputs to ensure this. Note that if one output is backed up, the fluid will redirect into the other unlike when using valves.
Perfect. There will be some extra production at the end of that line for discretionary quickwire and sheets anyway, but the numbers work nicely if you set he assemblers to 2 AI limiters, needing 10 sheets and 40 quickwire per minute. Effectively 8 cat ingots and 20 copper ingots each.
Though now I'm going to try a version with fused quickwire.
Bah
Do I tolerate the potential imprecision, or go down to 2.7 AI limiters per minute and ditch the repeating decimals for all input and output numbers?
(fused quickwire version of the AI limiter blueprint)
well unless the ai limiters are then going to another process it's not going to matter since if they only eventually go to another process it'll probably not be the nubmer you need anyway
It's an early game ticket farm strategy
But eventually they're great for electro rods and oscillators.
so once every 10,000 hrs it might stutter since you rounded the 4th decimal up? oh no
The other upside is that throttling those clocks gives some power efficiency benefits. Though did underclocking power efficiency get nerfed at some point with blueprints?
no? but if you're worried about using power you're playing wrong. there is more power available on the map than you can ever use
and it really doesn't save you much. OVerclocking a whole system to 250% only uses 33% more power.
Huh. I could swear overclocking used to have a bigger penalty than that?
BUt if 250% is only a 33% per item penalty, that's solid, especially in lower tier machines where that's just a blip
if you need to produce X parts per min, and instead of having all the machines at 100% you clock them at 250% the overall power cost is only8 33% more
cause you need fewer machines
similarly you need more machines underclocking so it doesn't save you that much
I get enough idle time that build costs tend to be a nonconcern
Changed production building and extractor exponent from 1.6 to 1.321928
From Update 7 patch notes, is that what I'm thinking of with power use changes for clocks? It's at the bottom of the clock speed page
Oh, huh, Google brought up the Fandom wiki page first.
And now Fandom links are crushed π
Any belt leading to a machine input can allow for overflow only if it is single-item. In other words: sushi inputs (going into a machine with no splitting) can't allow for overflow
Yeah, they reduced the power penalty for overclocking a bit iirc
Also linear scaling for generators' clocking/power-production
Generator scaling just inclines me to stick with coal for longer. I got this most recent gold cup with 100 coal plants only. In retrospect I should have just used shards to go from 48 to 96 or 124 (effectively).
The individual machine will use more power than that. But per item its o ly 33% more
if i have three things that output 120 per minute and my pipes can only transport 300 per minute and i need to split them into 8 things of 45 per minute how would i do that?
hey guys im still kinda new and I want to create a "not speggiti" factory but i'd like to make each building as efficient as possible, is there any good tutorials/websites that will help me get my biggest bang for my buck?
sastifactory calculator
!wikisearch coal_generator
Efficiency is just "machine running at all time. That can be achieved by basic math π or there are indeed online tools listed in #welcome that can do the math for you
Multiple pipes - or just feed it directly into the next machine
@gloomy dawn see #math-and-meta message for alt recipe picking
is there an estimated date for 1.0?
Sometime this year
thanks π
Later
Not now
now that I have mk4 belts I took my petroleum coke and boosted my power grid for now instead of just sinking it
that should last me until I get fuel generators up and running
Does anyone know if using faster belts has bigger impact on performance than using slower ones? Of course I mean the impact with huge amounts of them, all transporting items, going through splitters and mergers.
probably not more so than just the fact you're using tons of belts in one spot
not having a belt freeway is prob the best bet
not in one spot
No, I am just slapping mk 5 in all of my factories, no matter what the actual need is, and I started wondering if I wouldn't save some fps in the long run if I used lowest needed mk of belts in a given spot
probably not to any noticeable effect
Speed is just a number to multiply by, so most likely not
@quiet panther see #math-and-meta message for alt recipe choices
I haven't noticed any lag differences with mk3. They just don't render individual items anymore. And my game freezes up at every autosave. lol
I do like how my manifolds fill when the belts to individual machines off the manifold are mk 1, though
tbh if you're new, I would recommend evading sushi belts
but basically it depends on what kind of sushi you want
In this short tutorial I show how to load balance a mixed belt using a single splitter, something that can currently be done only by programmable splitters. The resulting outputs will carry exactly 1/2 or 1/3 of EACH of the items carried by the original belt, just like when using smart splitter to split each item off the belt and load-balancing ...
sushi load balancing is a bit of different beast
what issue are you having with it specifically?
Mb, I had that link on hand
Was about to say, person is confused-> so you link them the most complicated form of sushi π
don't make it a loop for one
big loop
Problem located.
are you just putting random stuff on this line?
sushi in this game works best in small isolated sections
e.g. you have set of manufacturers needing 4 different ingredients, you put those in the exact ratio on one belt and feed it to manufacturers
decide on the series of machines you are feeding with sushi
put on that parts per min on the line
set each splitter to feed the item into a slot and set main path to 'overflow'
have a sink at the very end at teh start in case you need to send some extra out to stop clogging until it self balances
absolutely don't make a loop
this isn't factorio
should look something like this , but with a sink to start it up
My entire world disagrees with this statement.
compared to Factorio it's true
same as Cobalt I assumed person came from Factorio due to making gigantic loop for sushi
You know my automated response is "Fuck Factorio" so... π€·ββοΈ
You can make gigantic sushi just fine.
Just don't loop it.
in Factorio you usually just build a giant loop of belt and use circuit network (basically programming/redstone) to put items on belt if there's not enough
And I will continue to say that people should not bring anything from that game over to this one.
Because this is yet another example of it causing confusion when they don't treat Satis like its own game.
you know very well that I'm with you on that
I'm just explaining why I thought the person came from Factorio and why I said they should think smaller
Size is fine.
Loop is not.
well size is limited by belt speed
@median heath yeah the looped sushi description made me think of people's factorio busses
Injection manifold π
those should burn in hell, same as hypertube canons
Hypercannons are a bug.
Injection is a perfectly valid method of play.
Feature
it took most of the evening, but my Heavy Modular Frame setup is done
you can also see my Quickwire/AI Limiter setup in the background
Just have a smart splitter set to "overflow". Regardless of wether you make a loop or not, that's all you need to control all excess/unwanted items.
Note: if you don't loop, the overflow splitter must be at the end of the line being split. In a loop, it can be placed wherever items have made a complete loop already (eg: not between merging of items X and consumers of items X, AFTER the consumers)
Loops can work surprisingly well, actually.
Though they do usually end up bringing the worse throughput-wise situations...
Loops are the sole thing that can break any of my systems.
Skill-issue-git-gud 
Just a few days ago I encountered a post on Reddit detailing an extensive idea exactly on this: user wanted to make a central production area based on one big (~16 belts?) bus of sushi belts; user ALSO wanted all items to be distributed along all belts equally (so each belt has the same items/min) and that ALL items made a full loop before being consumed/overflowed.
So, eg: Smelters make ingots; output is collected and split in 16 to merge on bus belts; after the bus loops around the whole factory, ingots are collected (smart splitters) from the bus and fed to constructors right next to the Smelter.
So each bus belt would actually be carrying (at at least one point on them) THE ENTIRE PRODUCTION CHAIN of items involved in the factory, though "just" 1/16 of it...
I haven't played in a while and it seems overclocking changed at some point so a bunch of my old power stuff no longer ads up.
.
Is there any up to date guides on coal and fuel power generation that doesn't use alt recipes?
(Got a friend to buy the game the other day so I'm helping them out)
The wiki is up to date. The changes in power made its clocking simpler as it's now linear: 200% clock = 200% fuel intake and power output (before one had to clock to ~250% for the same output effect)
I assume both .gg and .fandom are both updated.
Fandom is not being updated
Thank you for that, my friend has been using it a fair bit. I will let the know.
also consider breezewiki for those who haven't left fandom yet
fandom doing SEO horseshit doesn't help at all
just start without fandom and you're fine π
@brisk shoreHi
yeah bot won't answer to you
how many m tall is it in all
bc that things that gets u water can bring it up 10m it self
yeah just one
ok thx
I think next on my to do list is a computer factory, might need some help figuring that out
I wanted to do the crystal computer/silicon circuit board setup to eliminate oil from the production
I have no idea whatβs going on here, why is it looped?
it's an issue with trying to do sushi from a single delivered vehicle too. it'll probably balance out eventually but you made this into a buss I think? feeding machines with material and putting the finished material back on?
yeah I'm gonna stop looking at this
Why....
if I wanted to do Crystal Computers, which quartz node would be easier to setup around? Also my current layout for reference
up to you really and what other recipes you're using with it and how many comps pm you want
I don't have mass production plans in mind at the moment, just enough of a facility to get 1 assembler's worth of crystal computers per minute, which is I think 2.8125
doesn't really matter then really right? jsut whatever seems more convenient to you rn
Fundamentally, you're putting quartz, iron, and copper together. Assuming silicon circuit board and crystal computer. You could throw cheap silica and pure quartz crystal into the mix if you want to get the most out of it, in which case you want to bring the solids to water.
Real basic question.
For coal power (no over or under clocking)
How many water pumps are needed for coal generators.
I've been using 7 coal generators and 3 water pumps for each section.
Is it possible to squeeze in another coal generator or two without the water being an issue from only 3 water pumps?
(I'm bad at math)
I'm just trying to get enough power to last until fuel generators.
open coal generator and see how much water it needs
open water extractor (not pump) and see how much water it gives
do math π
Math says 8 coal plants need 360/min and 3 extractors can pump 360/min
but when i tried it, they were pumping like 355/min for no observible reason
had to overclock one just to keep all 8 coal plants fed
you have to wait until the pipeline system is full
instead of overclocking an extractor, temporarily shut down a generator
i did
pipes got full. turned it back on. 5 min later im off doing something else and check the grid graph
ripples driving me nuts
leave all pipes the same. add one more extractor and set all to 75% for 360/min again. works fine.
Sounds like a volume problem not an extractor problem :/
were you having all 3 extractors on 1 pipe coming in?
no they were connected at 3 points along the pipe running across the plants
if it was a volume problem (max pipes) why would running one extractor overclocked fix it then hmmmm
that shouldn't be teh fix in any case
pipes often take a while to be 'fixed' after a solution, leads a lot of people to think something is corelated
likely something else that 'fixed' teh issue, even if that something was just time
it ran for an hour, no issues
all pipes showing full
clocked it back to 100% and within 5 minutes it was shorting one plant again
I couldn't begin to guess at the issue you had w/o having been there and seeing it myself. I can tell you that Overclocking would not have been teh actual solution
so, it was "A" fix
i can tell you that overclocking just one by just 25% absolutely allowed all plants to run full with no interruptions
No, it's a fix like people who do homeopathic medicine
they have a cold, take what is esssentially water, and then feel better after a few days
the only thing that actually helped was time but they think the water was the solution
no changes to any pipes
so youre saying, the fact that all the plants would run without interruption was just in my head
yeah, flooding a system and/or 'time' fixes a lot of pipe issues
that the power graph being PERFECTLY STRAIGHT was just coincidence
so why then, when clocked back to 100 did one plant start cutting out about 5 minutes later
was that in my head too?
Couldn't tell you - might be the way it was laid out, maybe it wasn't properly flooded first, maybe it hadn't quite balanced out?
All I'm doing is trying to get you away from homeopathic pipe maintenance.
i was never there in the first place so no thanks
Overproduction is a remedy to an imperfect pipe balancing algorithm.
No need for jibes about homoeopathy and placebo... just on your part some more understanding of computer science of calculation errors in fp math, and coding of a recursive balancing algorithm
Fluid loss from pipes does not happen. If you make a packaging loop with a perfectly flat 2km pipe, the loop will never stop
by 'over production' you mean initially flooding a system, usually by lowering the consumption until it's flooding? sure
in this case water extractors are an easy option to OC for a bit, but it's not the overclocking that does it, it's creating a flooded system.
it's also entirely possible they just needed to wait for a bit since coal systems are very easy going and have a high tolerance for weird pipe things
and there's TONS of people who develop bad pipe hygiene by doing a ton of changes and 'fixes' and landing on whatever they did last as being the 'fix'
happens constantly. Exactly how people wind up thinking homeopathy works
As anyone with an understanding of computer science would understand, floating point error is not consistently above or below the true value, so it does not account for a net loss or gain of any significant amount.
Now there used to be an actual fluid loss bug ages ago but it was stealth fixed. I still see people going on about it like it exists on reddit.
Fluid loss, and calculation errors/non infinite recursion are separate things, I did not mention fluid loss.
The guy with the coal gens running dry didn't mention whether imperfect flow caused an extractor to back up and momentarily stop producing - so that could be a possible explanation, and what I'd put my money on as an explanation for the observation of 360m3/min not making it to the generators 100% of the time.
Devs are a small team with limited personhours and more important things to fix, just accept the idiosyncrasy of pipe calcs and overproduce and be done with it, the small power loss from idle and inefficiency of stuff cycling off and on is less of a cost than the potential need to run around manually resetting things, IMO.
If you build your pipes properly, they will work reliably, and your machines wil have no detectable downtime
Overbuilding instead of building correctly is certainly an interesting approach.
see you're doing the homeopath thing too - Yes, flooding a system initially (often by underclocking the consumption) is often important to get a stable system. Just generally overfeeding can do the same thing sure - but its a poor solution
It also restarts and re floods itself if there's ever a problem, needing manual attention is what I call a poor solution.
It does not/should not need manual attention if built properly
some basic pipe rules to have stable and consistent pipe systems at all times #1190597884681396296 message
You don't need manual attention for pipes
And pipes work fine tbh, just people not understanding them leads to "bugs" which are not really bugs
Which leads back to people making correlations to actions they do with pipes rather than causations
Pipes are plenty forgiving, this weird jank works. 6 lines of 240 that are all interconnected so that one pump can affect everything.
that looks like a pile of issues waiting to happen
coal power?
gonna guess coal power
coal power is like a stress ball, you can do just about anything to it and it'll be fine. throw it against a wall, twist it whatever. Almost all the problems people seem to have with coal power is doing literally impossible things like more water pm in a pipe than it can handle or not having enough headlift.
after coal fluid systems start resembling a peach, or an egg, where it's very easy to wreck shit if you just slap it around
It's 100% stable after dozens of hours
Sure. I did state coal power is extremely resilient to bullshit
I know because until I expanded it, I kept exactly enough overhead for the hoverpack. Any flicker would have dropped me from the sky.
Still not seing anyone say it doesn't work..
doesn't make this statement true though 'Pipes are plenty forgiving, this weird jank works. 6 lines of 240 that are all interconnected so that one pump can affect everything.'
'coal generator systems are plenty forgiving' is accurate though
greeny seemed doubtful
Maw, what we're saying is that later fluid systems won't put up with bullshit like that
Sure they will. It's a miracle my aluminum works. I just fumble around until things are stable.
πΏ
I'm just saying that setups like these have way less chance of working properly. I'm not saying they will never work
yeah, almost every time people tell me a system is stable like that they send me their save and I see flickering yellow lights within 30 seconds.
I've noticed people can be very lax with their observations
especially if it's "the 8th time rebuilding that setup and it's finally working"
Usually I add some sort of loop and that fixes things.
sometimes slapping a system around enough doing random shit can work - almost every time people have done this and shared pics or sent me their save I see yellow lights in the mix of buildings.
I try and make it more obvious for myself, I want things to fail totally versus in a less obvious way.
Hi, idk if I can ask here, ill delete if not. Im building 16 coal generators with 6 water extractors. Should i divide those 6 extractors to use 3 for each 8 generators? Cause i made just one line from those 6 extractores to feed all of the 16 coal generators and somehow something is not working. Sorry if the explanation is inaccurate, english is not my first language. Thanks
Edit: Just found out pipes can "only" carry 300m3.
!wikisearch coal+generator
until mk2 pipes they can do 600m3
not sure why you have only in " " xD
Thanks, havent unlocked yet. Anyways, Ive seen people use 3 extractors for 8 generators, so i also need mk2 pipes for that? Cause its 360 water
no, just mk1 , click the link
I see, thanks
Does anyone know why this shows a bunch of 0/minute steps at the top of the screen? https://www.satisfactorytools.com/production?share=LapRjXRz92EUo0PLEDrl
@wind spade
thanks, I was too shy to ping
Has to do with the way the program optimizes, iirc. It finds any way to reach the goal using the available recipes and minimum resources, and sometimes that spits out -0- steps like that.
Rounding error, can safely ignore
Ok, thanks. I was mostly curious if I could change how I was using it to stop them from showing up.
I'd never seen them before, so I figured it was some kind of error on my end
You cab move them out of the way but that's about it xD
Disable the offending recipes to remove, but I just ignore when it happens.
What does the blue thingy on the left mean?
That's where current headlift stops and you attach the pump there.
If you attach downstream, it probably won't get flow.
Useful snapping point.
Alr, thank you
where to get a ratio of nuclear PP at max overclock to required fuel and produced waster and to how much plutonium rods i can recycle from it?
Anybody know a way to calculate a specific oil setup?
My head hurts from complex equations.
I need to achieve:
- 100/min Polymer Resin
- 1707/min Heavy Oil Residue
- 850/min Fuel
- 4393/min Plastic
- 2800/min Rubber
No leftovers. I don't wanna sink thousands of petroleum coke.
Oh and Satisfactory Calculator isn't very helpful with my specific demands.
What exactly are you trying to calculate? How much crude oil you need? Building counts?
What alts do you have?
What are you doing with 1707 residue? It's not an end product...
Use TOOLS instead. Be sure to check the alts you intend to use.
OMG how do i always hear about SF calculater? This is 10x better!
thats why...
tryna get stack/minute for every item that the player uses.
so far i figured that i'm interested in only 23 alt recipes.
You want stacks of packaged HOR? That's an interesting rabbit hole to go down, are you going for packaged sulfuric acid, nitric acid, alumina solution, and so forth as well?
There are 32 items worth storing for buildables. Plus consummables/ammo.
There's an argument for storing ores and ingots if you want to included prefilled buildings in blueprints or to otherwise fill up a manfiold. But I don't know where that sort of thing could ever make sense with liquids. You need to add a packager in somehow for that.
Unless 1.0 adds the ability for us to use packaged liquids directly and get empty containers back.
Or even just chuck the containers as part of the process
uh, no, for liquids, only fuels.
and generaly only items the player uses. Even the eccess ones... but i won't be storing stuff like Copper Powder
@pulsar atlas see #satisfactory message for alt recipe choices
anyone know how to get this
doesnt work
Skill issue.
π
Said skill being your ability to read the "
" in my reply.
And the inability to read the actual answer already given --> "
"
k well im tired long day lol my b
Depending on where you're at in the game I'd pick one of the first two
All 3 are good.
see #math-and-meta message for alt recipe choices
I would personally pick Bolted Frame. Now that enough time has passed and the person has probably picked already.
computer factory online. went ahead with my silicon circuit board + crystal computer plan
How many ppm? That's going to feed your sink nicely.
Probably somewhere around 4-6
3 assemblers, I guess 7.5!
I forget how significant being able to make computers with the oil free alt chain is.
For all that there's only so much oil there's also only so much quartz, but I suppose quartz has less in the way of needed uses
is it that important? lots of oil on the map
I forget how significant being able to make computers with the oil free alt chain is.
??
Oh, maybe you don't know about the recycle loop?
Oil is not a limited resource at all.
regarding it being that important, I have a recollection of playing a long time ago with people discussing plenty about how amazing it is to make computers without oil, heralding it as so much better
oh, maybe the recycle loop didn't exist then. I was only around u4 or so
but i have done nonsensical recyle looping shenanigans before
People say a lot of dumb shit π€·ββοΈ
Is the oil-free chain good? Yes.
It is OMG SO AMAZING MUCH BETTER? Hell no.
I had the looping in an ammo factory in a save that no longer exists
maybe you talked to some odd people π
Like the computer chain that is just oil and caterium is quite nice and convenient sometimes
I mean, Sev, it wasn't just like some random people in chat, it was everyone, even Greeny I think
I miss my ammo factory. One big cube generating every type of consumable I might want to use, from rebar to filters
I take it that was before all hte new ammos?
It had some special rebars and nobelisks I remember, not sure about bullets
It is strictly speaking liable to have been from a different time than when everybody was on a hype train about computers without oil
It's not liked I leave the game and play a lot again on a tracked schedule
I guess silica cb and crystal comps just need crystal, copper and iron with the right recipes. That's fairly convenient too
in my current, modern save I'm at the point of hand crafting oscillators to get radar tower, one machine for heavy frames, and oil only feeding one refinery each to get rubber/plastic. As it stands I'm not really remembering much Vanilla to use quartz for besides computer alts
entirely depends on what alts you go for. You can have nearly no silica use if you aim that way
in fact I think you only need a tiny bit of crystal if you really force it
do we not go for "get every alt"
get every alt isn't use every alt π
you can set up chains that use TONS of silica, or little to none is all I'm saying
these are all the recipes with silica
and the HSC CB and ingots which are the main ones have non silica options
same with the U cells too
and concrete
huh I guess you could make a whole game that doesn't use silica for anything
The main thing I like silica for is balancing out the raw quartz draw with crystals, because of their numbers
wdym?
Crystal takes 37.5. Silica takes 22.5. 60.
xD I guess, but what if you don't use crystal and cilica in that ratio in that spot? Like I said, yo ucould just never use silica
and you can make pure crystal and cheap silica
Bottom up factories using full output of nodes.
I could convert an entire node to either, or both, and there's always sinking if I didn't have a way to use the full node. Like if another input couldn't match it
but then does it matter what ratio their recipes are in?
could jsut split a quartz node 1/2 1/2
Remember when everyone was declaring that Fluid Trains sucked?
It's just nice that the return to 60 counts grants exact machine counts
I do have some vague memory of that being a thing people said. Again, everyone, even Greeny
And again, everyone was wrong.
Yea, nah, they definitely were not wrong to say "bringing the dries to the wets is better than bringing the wets to the dries".
Pumps are part of that, it's not just the trains.
Of course I trained around oil to my ammo factory and space elevator factory anyway.
Though I see we even have mk 2 pipes now. I don't remember those
"bringing the dries to the wets is better than bringing the wets to the dries".
- Better is subjective, so yes that is wrong to say.
- This was never said in reference to trains, this was always about coal power. Which is an entire phase before trains.
Yes, 2 was said in reference to trains. It was said in reference to anything that ever involved moving liquid.
People saw Fluid Cars can hold only 1600 an assumed that meant they were bad as Freight can hold 3200.
So they ran with it.
When the actual math proved the opposite to be true.
The math people are the ones I listen to, I never cared about the arbitrary populace.
And then I did what I wanted anyway because who cares if I just need to make a longer train or another power plant or whatever
Sev is a train math person π
I can imagine people not liking dealing with fluids saying something like that though
*Logistics math.
For the record, better is not subjective.
Preference is subjective and doesn't need to adhere to objectivity. I once made a huge screw factory and ridiculous screw train. Because I could.
But with this being such a math and construction game, "better" is objective.
The most output from the least input, located where it needs the least distance for the logistics
For the record, in Satisfactory, better is absolutely and without room for dispute subjective.
It is absolutely and without room for dispute objective. That only changes when there is infinite nodes.
Continue in ignorance, I am done with this conversation.
I suspect you might be misremembering things people have said
'better' is only objective when you make very specific requirements though.
and since everyone's requirements are different everyone is going to have a different end solution to 'better'
People play for different reasons, in different ways and pursuing different kinds of "fun". Saying they should all strive towards the same "objectively best" solution seems shallow and reductive, tbh...
This discussion seems to be about the word "better" having different meanings for different people - for Old Song it seems that "better" = "logistically more efficient", Ven Cum "better" = "more fun for a given person", Cobalt "better" = "more aligned with personal requirements" and so on.
logistically more efficient is also wobbly. On what metric? cause if it's just throughput just belts everywhere
And especially because "better" has different meanings for different people, you should specify what do you mean by better
Yeah, I couldn't find more fitting description, I aimed to mostly underline the problem here
Playing better is overrated. Maybe my next gold cup will cut out oil and copper (except powder).
Yeah. I constantly feel peer pressure seeing how much time I'm spending on trying to integrate my buildings into the environment without harming the plants and so on, thinking how almost everyone would see that as utterly stupid and pointless. And yet it gives me a lot of joy afterwards.
Imo (loosely speaking) for something to be "better" than something else, it needs to fulfill a player's preferences "better" than other solutions. Since the last "better" in that sentence depends on user preferences, it's not objective ^^
Eg: players who likes to rush through milestones are unlikely to find balancers "best" for their use, more like a waste of time; similar for players who don't care about ore efficiency or big numbers and Pure Ingot recipes and the Water complexity they add... And so on
More joy = Better imo
"Better" being subjective is very close to "better" being dependent on user preferences. But it misses the pretty common scenario when "better" for a given person means "better aligned with external rules" as for The Old Song.
In which pressing the case of "better = preferences" might be really hard to understand for that person, because without going a bit more abstract (into the semantics), external rules (as in this case "better = less materials + more output with minimal logistics", which is a common rule from the real world) will never be equal to personal preferences. And for a lot of people external, real world rules seem to be objective, hence "better" might also seem completely objective to them.
So while I agree with you, provided that "preferences" also contain "unconscious preference to follow external rules as objective truth", I want to state that "better" being seemingly objective is quite normal and there's nothing wrong with that.
Just my two cents on that.
While that can be true, this still applies ^
If you want to ask/talk about best/better, you need to specify what your best/better means
I definitely did not say that
At least not in current balance. Maybe in like U2 or something
Yes, specifying your assumptions / mental framework is the best strategy in communication.
Which is usually what we mean by "best is subjective". Not that everything is always subjective, but that we can't knkw preferences of the user, so from our POV it's subjective
I understand that, but a person might not know that their preferences are actually preferences, as I tried to outline in my bigger piece up there, so they won't understand your point of view
It's like religion or morality or a lot of stuff - most of people don't question them, don't see it as their personal choice and preference, they assume it's an objective truth
I usually just ask "what do you mean by best"
That's super valid
Just for rounding out context, this is what started the rabbit trail:
"bringing the dries to the wets is better than bringing the wets to the dries".
Someone declared that, I countered with the truth of it being subjective, and so on until now.
You seemed to be more or less agreeable to the point when Song said "better is not subjective" and you countered with "better is absolutely and without room for dispute subjective" which is always a nice way to talk to others and build mutual understanding
instead of "better" it is usually better in your own interest to just say what metric it improves (or you think it improves)
better and best are often thrown around when talking about recipe combinations and whatnot and thats just always such a dry topic
because its like "what metric do you think it improves, instead of just being 'better' "
Right? After all, the only thing that's better is sushi...
Seriously mixed up belts π
you know I disagree π
but since my fiancee is called "Sushi", I can just pretend you're talking about her
From a game performance standpoint, wouldn't bringing the wets to the dry be more efficient, based on pipes being less CPU/GPU intensive than an active belt with flowing material?
All sushi.
No smart/prog splitters.
I proposed this before.
Game has to calculate headlift, fill/flow of each segment, sloshing, etc... Might be worse.
Yeah, but there's no real rendering cost for the pipe versus a 780 belt of screws for example.
*I'm assuming
There's no rendering cost for anything outside of simulation radius.
From an item count standpoint however, if you're worried about hitting the item limit, wets to dry is better, because the items on the belts count towards the item limit
the true reason why bringing the wets to the dries is usually not very beneficial is simply cause it takes so much longer, without considering troubleshooting
what. no the items on belts dont count
unless you mean transport speed efficiency
Does each item thats on each belt in terms of inventory capacity not count against the item limit in the world? I assumed it did, similar to items inside cotainers, machines, vehicles, etc
Those aren't items. They're shaders.
they are all meshes written directly to the GPU, they are not real objects
And the item limit has to do with built things. Not inventory. (Unless I'm sadly mistaken)
the only item data is an array inside the belts, which is just stinky normal data
and doesnt contribute to the uobject limit
oh cool, thats good to know then
Exception: inventory UIs
UI elements count towards the limit
Weird... But interesting. And a constant, more or less.
right now no, its a major leak contributer
open an inventory and close it again a few times
Oh no...
surefire way to crash the game due to hitting the limit
so dont go around opening/closing containers too fast
Hello,How do I know how many trucks I need to transport a certain amount of material? Example I have 4 miners outputting 480 each, let's say I want to take all that to spot A, and the time to drive the truck would be 5 min. How many trucks I need to have the same output at destination? Thanks β€οΈ
What about the parts shown in dismantle mode?
how many items stacks do you accumulate in 5 minutes if its 480 items per minute?
if its ore: stack size is 100
stencils, not real objetcs. just shader magic
yea iron ore
100 is the stack size then
480 * 5 = 2400 items
divide by 100 = 24 stacks
a truck stations transfers 2 stacks per second
a tractor has 25 slots
assuming the tractor takes 5 minutes from A to B and back, thats 24 stacks every 5 minutes, so a single tractor could do it
when in doubt, add a second tractor to the same route
oh wait 480/min EACH
alright thanks so much, I couldn't get my head around it. Also dont they collide if on the same route?
that would be 4 trucks then, right?
2 stations and 4 tractors because a station only has 2 belt connections
if you have trucks, those have 48 inventory slots
my final guess would be 2 stations per pickup / dropoff point, with 4 trucks in total
Yep, they can block each other.
give them WIDE rooms so they dont have to face opposing traffic
its not an issue if they all run the same direction
but opposing traffic to close to each other is bad
and I guess I need buffers at both ends?
would probably be beneficial, but my guess is its optional
all right, thanks you very much β€οΈ
too bad there is not a voice channel in this server, it would be very helpfull
we used to have them but then lost them
what happened
cant say, too obscure and far back
Failure to be moderated afaik
I allow you to be wrong.
I am magnanimous 
Half-full belts "cost" the same performance as full belts - > the least belts (for the same items/min), the better - > sushi is bestest
Jokes aside, I tent to agree: I think pipes cost less performance than belts, so bringing the wet to the dry should cost less performance (fluids are also in smaller volume, overall, I think, so less items/min moved far away overall)
fluids are not smaller in volume, game counts in liters
Good point. I stand by the rest
also items need just "how far progressed along a belt", while fluids have the whole dynamic thing going on. My bet would be on pipes more resource intensive
I've always assumed the same thing
Yeah, I'd definitely assume that pipes would be more computationally intense, especially on the cpu side where you're going to run into issues anyways
I wonder if fluid trains slosh while driving on the tracks π
They probably have a good internal baffle system π
I've always wondered how much fluid simulation even adds to the game, versus just making pipes 'stupid' and having them automatically feed.
There's something to be said for the coolness of conveyor belts rendering individual items which you can pluck and immediately use if you want. The nuances of manifolds versus balancers and everything else that comes along with splitter behavior. But pipes don't get any of that, and don't have the same wow factor right now.
I never thought of that, but that's a pretty valid observation. So much work went into the pipes and the mechanics seem to be a good attempt to simulate fluids and yet in the end it doesn't seem to be worth it. It's kinda sad in my eyes. But I have no idea what could be done to make the pipes actually fun and engaging.
Now that 8 constructor and 10 smelter blueprints are a thing, that disincentivizes refinery recipes that much more, too. And you can even go multilevel blueprints if you want even more in one click.
I personally enjoy refineries and resolving belts+pipes placement issues on tight vertical builds. But I avoid blenders with all of their inputs so close to each other.
Endgame, you can just plop a 26 smelter blueprint right at a node, I just realized.
does individual pieces of train tracks count as separate objects for the object limit? Is it best to keep the pieces as long as possible?
They probably do but you shouldnt worry about those
As you wont need as many tracks as you will need other things
Overclocking machines would help more in that regard
My advice is don't worry about the object limit
Even if it becomes an issue it's not really an issue
How so? I know it can be changed by modifying some file, but what about crashes and performance?
Because you will almost never hit it
Oh, ok. I saw too many times "object limit is much lower/you'll hit the liimit much faster than you'd expect"
And started worrying
It's the opposite actually
It's just that some items count as more than one object, so you can't just count number of placed items
Ok then, that's a relief
You can load-balance pipes.
I mean, it takes a whole lot of space and adds geometric restrictions to the build, but... It's possible 
The more I play rocky desert with blueprints, the more I keep deciding to make things out of only iron whenever possible. There's just so much of iron, and not needing to route a second resource is nice. I even find myself reconsidering cast screws and an otherwise all default rotor factory.
4 blueprints, done. You might be able to squeeze the whole thing into a multifloor blueprint.
I'm definitely not going to do it, but that's how I'd end up sectioning rotors out in that build, because there's plenty of spare iron to be inefficient.
how much does 1 coal generator takes coal per minute?
build the gen, put coal in, it will tell you π
I know it but i dont have coal nearby and i need to know how much i have to place
well then you can either find coal or open wiki π
Bruh isn't this discord supposted to helpπ
give a man a fish, feed him for a day
teach him how to fish, feed him for life
basically, I'm showing you how to find info yourself, so next time you are gonna get stuck, you'll know where to find the answers
15/min
@wind spade that one I directly answer because it 100% should be in the Codex, but it isn't for some reason.
I'm pretty sure they have updates to that in 1.0. Codex is very lacking, but I assume they didn't give it much love because it isn't a major part of the game
All other machine information can be gained without building the actual machine via the codex.
Power could easily do this as well, as I have shown many times how to format it.
For some reason it remains the exception.
they don't show the most important metric - item energy π€·ββοΈ
Fair. But for more simplistic understanding they could just list available fuels and burn rates for coal, fuel, and nuclear generators.
This would also solve the problem of how the vast majority doesn't know you can put Coke in the coal gen.
this sounds like something that should be in the generator directly (and I know you don't like it, but Factorio does that and it works great)
(if you hover over fuel slot, it shows you accepted fuel items)
Does Fucktorio have an in-game codex?
in inventory menu, you basically have list of all recipes (non-handcraftable ones are red), so I'd say yea
it's not directly codex in terms of "how can I make X", but you can search in it, so it's pretty close
List of recipes is far less than what the codex provides.
what would you want from it?
So I would say their lack of codex is why the hover over slot works fine.
Our having a codex makes it better to have the information in there.
I want nothing from that game π
no I mean what would you want from codex in a factory building game
so that I can say whether or not the information is available in Factorio
But the codex has recipes, also has all the buildings and their descriptions, recipes tell you WHICH buildings make them, and so on.
in Factorio if you hover over a recipe, it tells you where it's made
Yes, so same information, just not centralized.
well it's "centralised" to the inventory menu (as handcrafting doesn't require a workbench and happens automatically on the background)
Satis has a centralized codex.
Hence my saying the generator info should just add to that system instead of randomly being a "hover for info" like nothing else in the game.
I didn't say "hover for info", I just said that the generator could/should provide that info as well
the same as it says how much water it needs, it should say how much of fuel it needs
That info is already available once you put either resource in the generator.
I'm saying that does not need to be altered (contrary to you).
And that they just need to put it in the codex.
which doesn't resolve the issue of "how do I know what can I put inside"
and yes, that can be done in codex, but also pretty much all the info you have in the codex, you also have in machines, and I'd rather have it in two places than in none
I think that's a fun discovery in itself ^^
The part of my brain that hates imprecision hopes Compacted's burn rate gets slightly adjusted in 1.0
You can do this recipe in 2 blueprints: 1 Blueprint to turn ore -> Ingots (9 smelters for 270 iron). 2nd blueprint can actually fit everything in your list for rotors, it'll require 3 floors: 1) 8 Rod constructors on floor 1 + all inbound/outbound logistics 2) Floor 2 with 12 constructors -- very tight but can be done 3) 6 assemblers -- also tight, but can be done.
It's 50/7 so doesn't the whole number ratio solve any imprecision?
you can still make it whole number π€·ββοΈ
i think its funny and should stay
do foundries end up taking as much space as smelters end game?
trying to decide if they need their own floor in my mega factory
The best way to find out is to place foundations then a smelter and a foundry to see first hand.
overall not per piece
like i think i will need 2000ish smelters, but only maybe 250-500ish foundries? Haven't played endgame since U6 and alot of it depends on the foundries alt recipes usefulness
Well, it's hard to compare the space requirements of a foundry and smelter due to the different beltwork needed. Is this also in terms of choosing between copper alloy and default copper ingot?
over 300 hours and I'm just now finding out you can put coke in the coal generators.
I remember just continuously building more ISCs to store petroleum coke to deal with it, instead of coal generators. I had run out of room on the platform I built, such that ISCs fit, but a sink didn't. And I just kept building the ISC column higher instead of making room for a sink once I saw coke wasn't worth much in the sink anyway.
why not sink? π
even if it's low value, it's still free points
I did eventually, at whatever point I got tired of going vertical with more ISCs
I was as green as grass at that point, so I think I might have had the idea that it might come in handy for something down the road.
Not if you sushi
π
Never stockpile any unlimited resource.
I mean, I know that now
So... never stockpile anything?
stockpile hard drives and artifacts π
HDDs and artifacts are a safe bet. Beyond that, for all we know fertile uranium for maximum plutonium rods will be the future meta because SAM + plutonium waste becomes the best sink item.
for that matter at some point where I wasn't around did they ever say anything about what those artifacts are actually going to do?
Or there will be new alts using SAM that change nuclear entirely.
Yeah, I don't think anyone is trusting the rotor switcheroo as being the only change to nuclear.
I'm still hoping for another recycling stage. Even if that amount of power is pointless.
I would not like to see that.
I am all for recycling stages, so long as they don't reach a point where I just go "I can't be bothered I'll just sink it instead of finishing the production line"
Mk2 and Mk3 nuclear nobelisks would be amusing, if they required plutonium cells and [third stage] cells, respectively.
Enrichment is fun!
I'm really hoping for a mk2 Awesome Sink in 1.0: Would be really nice to have one with a 2nd input (especially if you use sinks at truck/train stations)
just build two
That feels like a weird sort of edge case to have so many things close together than you can't manufacture into something to then put in the sink, that you'd need a second belt.
Maybe we'll get the ability to snap mergers directly to any input or output and that'll work out regardless.
i thought they said that this wouldn't be a thing (in reference to mk3 oc miner output) but i may be misremembering
remind me, one pure coal can feed 6 gens?
Yes
one pure coal with a mk1 miner produces 120 coal/min
one coal gen consumes 15 coal/min
120/15 = 8
so it can feed 8 generators with a mk1 miner and mk2 belts
cool. been a while since i played so it's slowly coming back. I want to get to nuclear before i call it quits again
i think i found a uramium spot overworld where there's a node underground. I didn't think the radioactive effect extended above ground
Might be a deposit nearby
Does anybody know the file location for the scanner's dong sound?
Usually I don't spend my time looking into specifics about anyone or anything's dong.
But that's just me.

@deep ingot, i would recommend option 2
(from mcgalleon's pipe guide, can be found in the pins)
That's actually exactly how I had it
I had exactly the amount of output water being consumed directly by alumina solution machines which were not connected to any other water source
But small perturbations in any part of the whole process would have one machine or another go idle for a second here or there and eventually the water kept backing up
i also like to make sure the system runs at 100% even when using option 2, which you can ensure by having overflow splitters going into awesome sinks
if the system is running at 100% and you've done a full system flush and let it get going from scratch, and the pipes are still clogging it's most likely something to do with your pipe network, you can send screenshots in a #1038092680493801533 post and someone should be able to help
Glad to see all of my ideas present on that sheet though. Coal seemed the most stable option to me since it's entirely independent and guaranteed always on
you can always resort to that if you can't get other options to work
but it's usually ideal to deal with the waste water at the plant
Yeah that's how I've got it now, will probably just leave it. Putting the water back into the input feels elegant but doesn't really offer any real benefits besides the effort spent on setting up the coal plant haha
Thanks for the resources
if it's already up and running and don't need the coal for something else then its probably fine
alu can always be a pain
why isn't this setup working? I put smart splitters to feed the specific material, but they're not getting fed into the manufacturers
What are the settings on the smarts?
left material, center any undefined
Center = Overflow.
Always.
that's not the problem, the problem is they're not feeding into the manufacturers
Ah right, I forgot I don't know how sushi works.
My mistake. I will see myself out and try not to look at my entire portfolio that's sitting by the door.
Once you have done that to all of them, make sure THIS one actually has somewhere to send the overflow.
My recommendation, given this looks like a temporary setup, is loop it back into the ISCs. NOT the line itself.
Should be able to just loop it back to 1 smart that has each side set to 1 of the 3 items, going into the ISCs.
If you want to clean it up even further, you can swap the output belts from the ISC's that are going into the mergers to lower grades.
Mk1 for the Frames, mk1 or 2 for the Beams, and mk2 for the Rubber.
Let me know if this did NOT fix it.
now it's working
If the total amount need by the 3 machines is less than 480/min that should be fine.
High potential that the Modular Frames just start looping and block the other 2 products though.
If the loop part does jam, just route it back to the ISCs instead.
225 per minute, and yea it is jamming
Looping train = rollercoaster
Looping Factory Cart = ? π
@opal locust regarding Any Undef - has very rare use-cases that only like 3% of the playerbase will ever build to.
And is almost never used on smarts.
When you get into the optimal use-cases of poggers, that's where you see Any Undef.
I'm making 28 versatile frameworks per minute, and I need 2,500. This is going to take a bit.
I guess I'm in 3%
Tho I wouldn't say "building a sorting system" is only 3% of players
- So you have actually played and built on recently?
- Pics?
- Where/why are you using Any Undef? (only if 1 and 2 are completed)
sort container has belt with any undef going to the right, so that things can overflow if not defined for sorting
"PL" = ?
I do not understand this drawing.
"SORT" is a container for "dump your stuff in here"
mix of personal sorting and train sorting I guess?
train does not sort, train is not mixed
He also didn't provide 1 or 2, so π€·ββοΈ
it's a central storage system
ISC is the storage, things come by train (or any other option), get stored in the container and the top line of smart splitters is set to "overflow" for sink
Not one that I have seen π€·ββοΈ
bottom line of smart splitters is for sorting your personal inventory and it hooks to the same belting so that you return your stuff to the storage when done
at the end of the sort line (right edge) is processing of things like biomass, slugs, etc.
well there's not a singular way to build sorting + storage systems π€·ββοΈ
Is it possible to bring the throughput of a mk1 belt down to 30 pm with sushi
So I have 60 Batteries merged with 60 Limestone on a mk1 belt would that make 30 Batteries and 30 Limestone travel on this belt?
yes it would
ok
you can also just loop half of a mk 1 belt back on itself to make it 30/min with no other resources
is there an intuitive explanation someone can give me for why this works? does flow straight through a junction simply supercede flow at right angles?
though i guess then it wouldnt need to be vertical
No.
We can explain the how.
But we cannot explain the why.
It does need to be vertical.
We simply accept that it does work.
dang ok thats what i didnt want to hear haha. I was hoping it wasnt just like an emergent property of the game
ty though
It works.
It works reliably.
Do you need more?
yea i like that it works, but its more satisfying to me if i could have logic'ed my way to building it myself instead of just observing quirks and saying "guess thats how it is"
There are many advanced things in the game that cannot be logic'd to.
They were found via expirementation and testing.
Some of them were then reverse-engineered to discover the why, but that doesn't work on all of them.
makes sense, just wanted to make sure that this was indeed one of those
Build fun rooms like this and see what happens when you do different things.
There's also some very random fun headlift stuff that occurs only at "sea level" because the Z position of 0 is actually higher than that.
So "sea level" is technically negative elevation.
no clue if this is right channel but which one?
All of them are good.
Pick whichever you, personally like.
You will unlock all alts eventually.
pretty much always you want to pick yourself rather than asking others
alr ty both of you
the linked message has tons of info
oh alr ty!
Is this System fail safe for Batteries
what even is the point of this
Batteries Distribution in Nuclear pp
what are you trying to achieve?
what you've shown is basically "how to make 60/min into 60/min, but more convoluted"
Is there something useful I'm missing about Iron Wire being 22.5/minute output?
if you need that much as input π€·ββοΈ
The only recipe where the ratio matches appears to be the Beacon
Ah, Batteries apparently match
any recipes match if you clock
Unless you scale things up. 4 constructors put out 90 wire, 8 constructors will put out 180 wire. And of course you can play with clocks.
Ok, so you can do 8:3 to make cable or something. That's not awful
only if the divisibility works with the limited precision of clock speeds π
You can make stitched iron plate work
that doesn't matter
if you clock machine higher but only provide enough resources for the smaller amount, it will make the smaller amount
That's pretty cool, actually. Looks like 5 iron wire to 3 stitched plate. Now I just gotta figure out where that iron plate rate fits in, lol
Down clock the stitched plate assembler to need 20 wire and 10 plate per minute. That makes the ratios very easy.
I'm going from memory, but I think it's 80%?
Math looks like that would be 53.3 repeating %
clock to 53.3334% and easy
My goal here is exactness, lol
One part in ten thousand is still an amount of error
the machine will stop for a second once in a few hours
Don't get me wrong, I do usually just build my factories by using your wonderful calculator and plugging in the clock rates it gives me. This is just something I think is fun for its own sake, not because the inefficiency will matter in practical terms
What's the clock speed for 30 wire and 15 plates per minute?
Cool, that's 80%
That pulls things down to a point where it works well with default wire and iron plate recipes
8 wire constructors at 100%, 6 plate constructors at 75%, 6 assemblers at 80% works? 3 blueprints, less if you go multi-level.
That's cool. So it's got synergies with both iron wire and normal wire
It's a bit less elegant, but you could also do 5 iron wire constructors at 100%, 3 plate constructors at 93.75%, and 3 assemblers at 100%.
I like how it reflects on input. I often clock it so that it takes 30 ingots/min and often end up making sets of two, for a nice 60/min
That whole thing could fit in a double-decker blueprint, so you could just tile it endlessly.
And with just one input, one output. Nice
That concept increases the appeal of iron only products. I just went back to looking at the default rotor recipe with cast screw, in the context of blueprints recently. Modular frames also work nicely if you go iron only.
your loopbacks will just gradually clog until all 60/min continuously flows out of the only exit belt left
to do this kind of fractional split, both the 15 and the leftover 45 eventually have to exit the loop and get used/sunk elsewhere.
merging back into the source is a trick that CAN help with achieving unusual ratios, but both portions still have to have an exit path
and it's not needed here anyway because 15/45 is an easy split
Are you building an 15p/m throughput limiter? Looks like it should work.
I use similar thing to limit packaged fuel deliveries, so it doesnt fill the whole drone port on receiving side.
any situation where input exceeds output = something will back up
and there's nothing stopping most/all batteries travelling straight out the "15" exit once the loops get full
mk1 belts
although yeah, once they fill, it's 60/min through exit
No, merger will always take from all inputs in turn, so it limits the actual input belt to max it can fir.
so it's 20/min output
the particular dynamics of clogging might give you a bit less than 60/min out, and distribute the clogging back toward the source
but more than 15/min will get through
so need a second merger after the first one and loop the 15ppm there
so i have to make an other Idea because I want a throughput limiter of 15/min no matter what is going to happen.
look, no matter what you do here
if you try to put 60 in and get 15 out, some combination of the following will happen:
- the system will eventually output more than 15 ppm
- the system's input will back up toward its source
different arrangments of mergers/splitters may hit different equilibriums, but that extra 45 doesn't vanish. some amount will get added to the output belt, and some amount will back up the input belt
Yeah, you need to sink or otherwise divert the 45 extra batteries to a different line.
you need to consume or sink the remaining 45
the idea here is to recycle 45 pieces, so they MUST go through the middle belt again. this leaves only 15 ppm usable space.
I do not remember how to recycle excalty 45 into belt tho, I know loping back two belts from splitter to merger limits the belt to 20ppm
but those recycled 45 get added to the 60 input
so on the first iteration of the loop, you now have 105 coming in instead of 60
and since the belts are all Mk1 and can't carry that much, that's when the clogging begins
I get that you can play with clogging. But why not just sink instead?
I think im going with 20 and a merger with susi
why sink useful things?
By all appearances they are not being used? So it's questionable whether they are useful?
that is the purpose of the exercise
Diverting to ISCs also works if you want to stockpile them for whatever reason.
20 ppm limiter
plenty of reasons to sink "useful" parts
- the factory that consumes the part isn't built yet
- it's going to personal construction storage, which will back up eventually if not spent at the same rate
- it's going toward vehicle fuel, which may be consumed unevenly resulting in excess
- want points
resons to limit belt:
- it goes into very long route that will take eternity to back up
- items are needed elsewhere, but actual number is not know.
Why not just down clock the battery machines instead? You're going to clog them which will functionally do the same thing?
that assumes manufacturing happens at the same area. Batteries may well be shipped by train and divided here.
The only difference is the time it would take to jam the machines. So again you're just effectively stockpiling batteries in a different way. So why not just split off to ISCs?
I'm not asking why it is needed. I don't care. I'm solving the problem presented.
thats my solution i came up for
QW is 20 PM
The reason is i use a sort of manifold with drones that are bidirectional. If you want to hate me because of that, i have no problem with it.
I think this should limit to 15ppm, but not sure. there is no priority merger ingame
Obsolete, look here: #math-and-meta message
Ok, I'm pretty sure this will work.
We saturate the middle belt to 60, by looping back half of it.
The bottom merger can now accept max 30 items, because it will mix 1:1 from both belts.
The upper merger is now limited to 30 output, so it can accept the 15 we split below, plus extra 15 from the input line.
Input line is limited to 15 items, so whole throughput is 15 at the end.
This actually looks just like OP original drawing, so yeah, looks you were right, and it seems we can stack the 50% speed limiters by wrapping them around previous stages.
7,5 limiter
It won't. The math is wrong - the first merger, if the two inputs are each 30, will give 60 not 45.
Fixed version just below βοΈ
Not sure why you're doing the stuff with the mergers and splitters. A 15 ppm in will give 15 ppm with just a belt, similarly with the second example.
The blue color signifies the belt is being slowed by merger.
Because the merger cannot accept more items.
guys i need your mathematic skills
take a math class
so im using a 240 ingot node
ingot node? π€
whats the best approach to get 15 ingots into 1 smelter now
will i rlly have to split the nod 16 times?
manifold
isnt there like a smart splitter that splits exactly 15 ingot
note: the rest of the smelters need normally 30 ingots
no, there is a splitter that give priority to one of the outputs, and lets rest through when that belt backs up.
its just this specific smelter that needs 15 ingots
is it underclocked?
yep
just manifold it, and as long as input matches consumption, it will work
im building it in modules
so when the overall consumption is accurate to the input it wont overflow?
no matter what the values are, belts can be built in pretty much any way
--S--S--S--S--S
| | | | |
the 15x will back up tho right?
yeah things back up in manifolds
put the 15pm smelter at end of manifold
but it doesn't matter really
The clog would just backlog to the start (whatever machine the items originate from)...
You can achieve limiting conveyers to certain number of items so much easier
Examples welcome.
Not even sure what you try to achieve, 3/4 examples have the same input as output and are just trivial
I'm still curious about all that, but I didn't spot details on what exactly the OP is doing.
Without knowing that, π€·
Limit the throughput of a belt to 15ppm
Like you want to half an input? going from 30 to 15?
Limit [unknown amount] to 15ppm
RIght, but I'm unclear on the underlying reasons. THere are lots of ways to get a 15ppm belt, and lots of scenarios that might call for it.
"unknown but certain" or "any" thats a big difference
I'm the guy who just got a gold cup by using tickets instead of oil extractors. SO I'm all for solving weird problems with weird solutions. But the scenario needs to be clear.
Unknown, but OP mentioned drones, and I know drone ports will not back up, if you send other items along.
SF does not have unknown amounts
Load mixed good into a train and try again.
Your example just makes the input machine clog up and produce at 50%
So you can literally just underclock it to 50% and save all the mergers and splitters ^.^
If you are willing to build a sink you can just do that
don't mix goods
if you want to go from an unknown amount to 15*, make the input a mk1 band and do the picture twice
sometimes, you have a box.
Yes, but why are you introducing standing loops like that, when a single belt from the 15 ppm input to the 15 ppm output gives the same results.
yeah thats what i mean
Again, inputs are marked with blue because the mk1 belt is clogged. The source tries to supply more, but it can't.
Why clog the input when you can clog the output or build a sink?
And even if it's necessary to clog input, just limit the machine and put the desired number in
Probably will save* you a frame per second :)
Not a part of the exercise, but if you must have an example:
I have a train that needs to ship 12 caterium ingots per min to another base.
The wagon ships several other items.
The input belt comes from a box and supplies whole factory, so it is at all times in clogged state.
The train platform will not back up. I don't want it flooded with caterium ingots.
So I have 12/min throughput limiter on there.
The same wagon ships different materials? Oh boy I'd start a new save file
would it be a issue if i transport crude oil like 300m far?
heard transporting fluids is suboptimal
You can transport fluids unlimited horizontal distances via pipe.
I'm not adding five wagons to ship total of 300p/m
Just limit the smelter to 80% in the first place
which one of the 20 that ship by truck?
Again, not the problem we are solving here.
Transporting solids is easier. But not by much. The volume of solid versus volume of liquid comes into play. For instance, bring coal to water, and not the other way around, because water consumption is 3x higher, and you're likely still limited to MK1 pipes.
I'm all for just doing long pipes when it comes to nitrogen, because then you get rid of head lift issues too
I give up
If you have fun, you are in the right π
Gases in this game don't require head pressure?
*headlift and no
This appears to be attempting to limit a throughput based on what a belt will carry. The "slowest" belt allows up to 60 ppm so you're not going to be able to limit anything less than that. You could achieve a throughput less than 60 ppm throughput using splitters/mergers but whatever is extra needs to go somewhere outside the assembly being used, like looped back to storage or sink as has been said. It's not going to work being feed back in.
Hello angels and demons of math and meta. I come to you with an inquiry regarding pipeline functions.
The tl;dr version of it is this: I have four pipes totaling to 400 fuel each. I need to balance them into 400, 400, 366.67, 433.33, would this "theoretically" do it if the pipes to the left add up to 366.67 and the right add up to 433.33? Would I achieve same thing with valves, and even better, would it be better to limit it with the valves? Any and all thoughts on this topic will be appreciated. Also, breaking up the machines (as in adding more) is not somethin I would like to do.
I think an easy solution would be adding overflow junctions (a U bend going upward) to both sides, to make sure the blender gets fed the remaining fluid only after all others are fed. The pumps and extra pipe loop could be removed then
Could you help me understand it mroe?
https://www.reddit.com/r/SatisfactoryGame/comments/obqjrd/ficsit_pipeline_plumbing_manual_first_edition/ pipe line manual - there are junctions that can help with over flow
I have that saved
don't do whatever you're doing in that picture though
Well, what do you suggest?
Basic pipe rule - keep it as simple as possible.
hey do you guys have any good websites for planning the stuff out before you build it ive tried a couple but not many have been as helpful as i want
so clock machines to make that much down each pipe in the first place is probably your easiest solution
the actual machines? or the numbers needed in resources?
No can do, already tight enough on space.
overclock?
like flowcharts to make sure you know what youre building ahead of time
and how to clock them
best one
That might actually work
and user friendly
the other simplest method is alter the consumption machines to use what you have in the pipes in the first place
I can't see your whole set up or how it's built so I can't tell you what would be easier for you
The 30/min loopback would back up as 60/min exit that merger to try reaching the next but get backed-up by the 15/min.
You would need to "priority merge" some more the loopbacks (splitting them more before merging, to lower the priority of the original 60/min input)
You want one machine to be fed the overflow from 2 sets of machines, right?
Yea, this would be the simplest, that way the only "balancing" needed would be water
I don't think you need to balance anything. You can have all the fluid all in one linked network and it should work just fine?
I would recommend against that too tbh though water extractors have a higher tolerance for merging
nah but like i want something that will show me that i need to build the things connecting to each other and the only website ive seen that does that automatically doesnt allow constraints (i.e., i dont have a miner mk. 3 availible but it tells me to build one)
Maybe yes. I'd add overflow junctions just to be sure
idk i think i just suck at doing tech stuff
as in belt and splitter layout?
I really recommend you come up with that on your own. There's only 1 web page I've seen do that and they give you stupid lay outs
I've got a coal power plant that has a total of 1440 water flowing within a single network, using only MK1 pipes. But because no single point exceeds the pipe throughput limit, it's fine.
It's also not super hard, just make it down into chunks by your max belt speed.
you moving 1000 pm and have mk5 belts? break it up into 2 groups of nothing else is better for you
ok
So if you just inject those four 400 fuel lines at different places in the line, it should just self balance.
the other reason for doing it yourself is you can develop good logisitc layout skills and then create fun weird things
cool
The miner part is considered trivial as one just has to provide the ore needed, however they want/can
wait ive found something i can apply simple math to and figure it out kinda
calculatory
it doesnt let me post links
its calculatory. ovh (theres a space bc i cant post links)
also i might be stupid but im a noob is there actually no way to have a precise split between two outputs of a splitter (like two outputs, 33% of items go out one end, 67% go out the other)
there are ways but they are generally not needed
also Not sure what you're getting from something like this ?
aw that sucks
idk im really dumb man
ok so first off - your example of 1/3 going one way and 2/3 going another
a splitter evenly splits and it has three out puts so you can have 3x 1/3 paths right?
they could possibly add an even further upgraded version of the programmable splitter that allows items/minute output caps
whatever im just gonna suck at this
I think they have said "never" to that sort of splitter.
no no no don't stress.
why do you feel you need to evenly split things like that?
: (
autism idk
SF is a design and logistics game - a ratio splitter, what you described, would kill 90% of the problem solving
true
but it's ok you never need to split things like that. Do you know what a manifold is?
Already fixed in next message: #math-and-meta message
My bad, I thought that was a solution using sushi and filler items, kinda skipped it
i was making a mountain out of a molehill this really isnt that bad the only planning software im using is stickey notes so i can remember the required stuff i need for the final assemblers and stuff
i learned and am now using one for the factory i made thanks
what does the tractor use for fuel?
Anything you can fathom burning.
Leaves. Wood. Biomass.
Coal. Coke. Fuel.
Fabric. Paint.
Uranium Fuel Rods.
Etc.
record it and drive
do i stop the recording or just set on auto pilot and get off
stop recording I believe.
you might find a tutorial on youtube if you're having issues with it
gort it thank
It stops recording when you cross the start line again.
yeah found that out lol didnt work the first few times but got it now. thanks!
question, why are my smart splitters getting random materials in them that aren't being passed forward along the sushi line?
what are the settings?
and splitters have small internal buffers so might just be that
Define "random"
the plastic and steel beam
In the context of what you're building, how is that random?
they were materials from things I was previously building
and they're still there, not clogging up the line, just in imaginary space inside the splitter
I personally wouldn't define that as random, given you know exactly where they came from and how they go there.
But this is inherent to how overflow works.
Previously you were making Versatile Framework.
yes but it has cause me to be a few materials short multiple times because of this
Temporary setups can have this occurrence, yes. That's why I don't necessarily advise doing temporary sushi manifolds.
Their operation is based on infinite continuance.
If you're switching what the line is for, you have to flush it.
But most people don't know to flush it because that's a more advanced sushi thing.
If you want a sushi line that you can continually swap the settings of without needing to flush it, you have to use poggers instead of smarts.
whats the difference between all the different medical inhalers
Same difference between all the different Reinforced Iron Plates.
im guessing the recipe
Yes.
As long as you're also understanding that the item itself doesn't care what recipe it came from, it is the same item.
Fuel, Residual Fuel, and Diluted Fuel all produce just Fuel. Exact same item. Regardless of recipe.
gotcha anyway it can be used inf times likea mod that you know of or something
If you want to talk mods, that is the place βοΈ
ok. what do the bird things do in the dessert
be birds
so no use
Their job is to block you from building things occasionally. There are solutions to this. 
Ingredients. They make the same inhaler. (No indications this might change )
There's only one inhaler recipe that's infinite, using nuts.
all of them are infinite
I thought the bacon agaric is not renewable?
everything except hard drives and artifacts is renewable
Woah, mind blown. All this time I thought itβs not renewable.
lizard doggos can bring pretty much everything that seems non-renewable
Ah, you mean renewable through doggos. Thatβs not reliable enough for me.
What if you don't count that?
Mycelia is not renewable, right?
This is a table detailing raw resource renewability. A raw resource is any resource that can be obtained from the world without being processed (Crash Site wreckage is not considered).
Of all the items listed below, only Hard Drives, Somersloops, and Mercer Spheres are non-renewable resources.
Sans doggos, no foliage item is renewable.
if you don't count world updates
Hi, I have a question about trains. Im setting up a train with 3 wagons, one locomotive on each side. How do I make the train go back? So hes doin A - B but i want him to load and go back B - A.
trains can only drive forward
if you have loco on each side, it should work normally
so you either need a front-facing locomotive on both ends or just build loops
but for stations - train needs to arrive in the direction the station is facing
I do have locomotives on both side
make a #1038092680493801533 thread with screenshots of the train and stations
Oh ok! Thanks!
Wait does this locomotive also need a station?
for the arriving station? yes, it'll need a station facing the right way
The back facing locomotive doesn't
people make useless things sometimes
so there is absolutely no point to this?
Don't watch YT videos of other people playing the game.
Just play the game yourself. π
he showed like 2 inputs to 3 outputs etc. which is understandable but why would i want such a complicate 3 to 3
2 to 3 isn't really understandable either.
who knows, maybe it's a balancer, or a condenser?
It's a balancer.
yeah thats why im asking u guys lol idk anything about balancers etc
I don't use them.
thought u could expand my knowledge about balancers
I jsut put the number of items on the belt that I want or use the number that is on the belt
!wikisearch Balancer
Balancer may refer to Load Balancer or Belt Balancer. Splitters are built in a nested way, such that all downstream belts or buildings receive an equal amount of material, regardless if the supply belt is providing sufficient input. A factory that is built this way tends to start up faster, as there is no need to wait for the internal storage to...
I consider those generally useless, apart from radiation control.
so its just to start up the machine faster?
not really, you can skip the spin up time of manifolds by pre feeding
and it takes longer to build and more space so 'sooner' is not guaranteed
well that sounds pretty useless
it's mainly aesthetic
i see, thanks
or for the challenge of it
See my initial gif reply.
Note, it's a very satisfying aesthetic. However from a difficulty/progression standpoint, its a complexity you don't have to opt into until you build nuclear (and even then, its a soft opt-in)
Your original post picture is of a "belt balancer" as discussed in the wiki. It splits a number of inputs, which may be non equal rates, and merges to a number of equal rate outputs. I think it's an arrangement used in Factorio and brought in.
As opposed to a "load balancer", which tends to be an assembly of splitters and mergers that deliver the exact input rates required by buildings, rather than the overflow method which has building inputs delivers by a line of splitters and an input being the overflow from the previous splitter.
There's some math involved in load balancing, so some players see it as a challenge, whereas for some it's the norm.
Watching tutorials is okay if you need help or inspiration.
if you need inspiration, sure.
if you need help, copying someone else's build without understanding is like worst you can do
IDK... Depends on attitude. I've copied stuff for Minecraft redstone and used that to learn how it works.
that's a bit different, you don't need that knowledge to beat the game
in SF you learn in every stage and then apply what you learned in next stage
subjective
Manifolds are pretty meh. Your first machine will endlessly be full. The last on the line will almost always be short on material. Balancers create perfect efficiency.
The hting about this game is it is as simple or as complicated as you want. Balancers offr another layer and challenge to things. You don't have to math this game at all. Add a belt and when it is done, its done.
Not everyone finds that fun.
that's just not true
once manifold fills, it works the same as a balancer
and you can prefill it to get rid of the spin up time
'first machine full, last machine very low on materials' sounds like a system working at 100%
I'm less familiar with setting and hooking up balancers now that blueprints are a thing, but before blueprints, I always found large balancers more time consuming than just prefilling machines.
I frequently balance small machine count scenarios, though.
direct input is probably even easier
Yeah, direct input is great with blueprints.
Edit: Nvm, fixed it. Had a bad output from one of the final pieces going onto a mk1 instead of mk2, thought I'd fixed it, but I needed to flush the pipes to force a reset
I have 2x Petroleum Coke refineries using 40 Heavy Oil Residue per minute, for 80 total. I have another 6 refineries that are outputting a total of 80 Heavy Oil Residue. The additional conveyers are all working at 100%, and yet i'm finding that the 6x refineries outputting heavy oil (plus rubber/plastic) are often working at 50% or less efficiency, as it seems like the pipes get backed up.
Am I missing something?
do you have a plan you're working to? like from satisfactorytools? it'd be a lot easier to follow
So I'm finally trying my building skills at making a Nuclear Power Plant. My configuration will be 4 floors of 27, each overclocked to 250%. Water will be brought in by packages, to the tune of at least 65k per minute. I haven't found a solid number on how much waste is made nor how many fuel rods will be needed to keep the plant running. I do plan on having all the waste sunk after being turned into plutonium rods.
What is the best way to make a multi train car relay system? I'm not looking at making a push pull like I have in the past with fuel plants. One train wagon for each floor, every ?? minutes.
Before you go all in like that consider making a plan in satisfactorytools.com
You also can find the exact numbers for fuel rod consumption and waste production.
So you aren't guessing.
Also - if you build on top of water itβs easier
Also also - if shipping the water in, Fluid Trains will be better.