#math-and-meta
1 messages ยท Page 414 of 1
thats not bad?
yeah, it takes steel instead of iron
how do you split that evenly?
but it has two big advantages:
- it uses the same resources as stators do, so production lines for motors get super easy to build
- it can make use of steel alt and iron wire alt to be super efficient
and for the splitting - just use manifolds ๐
---+---+---+---+---+---+---+
A lot of recipes have ugly numbers, especially those involving wire.
- being a splitter, - is a belt
cant have perfection
you can, but the cost is so big, that you don't really want to have perfection
also, manifolds are perfect โค
@velvet basalt wires and screws are meant to strain your logistic system. You have to deal with it. The simplest solution is get mk3 belt.
The 2nd simplest way is to underclock whatever excess.
Or design a production line that uses double belts of screws.
Not Possible. No machines can output double belt. The input side is not an issue though....
Simplest solution is screw screws
I had to when making crystal oscillators for 45 quartz computer. 180 reinforced iron plates needs 1800 wire.
I didn't feel like introducing oil or caterium for the alt.
So I had to design a manifold that used 4 wire belts. I stacked assemblers at 12 with three belts going down the middle.
How would I setup the constructors
I could but merge them and manifold but that be to much on a mk3 conveyor
You can't possibly pull more than your highest belt size from a single pure node
anybody made a layout design for 1 terawatt of power yet ?
400 nucleur plants?
A powerful Satisfactory calculator, filled with features, supports overclocking, alternate recipes, live map, power calculations, etc.
You need lots of production. But good thing is, alts help a lot
Is it available only with two nodes?
3 nodes
Uraniun?
Y
I haven't checked it. Are there two uranium nodes on west side? The other is in cave at center.
There are 3 uranium nodes long time ago. Not sure what is the latest qty.
3
long time agoII think it used to be 2 nodes for stable EA, it's just upped to 3 since the experimental release. one is western of the cave area pretty central in the western part of the red forest and the other node is eastern of the cave at the edge from red forest to swamp area @mystic coral
1 Nuclear Fuel Rod a minute, is enough for 2.5 Power Plants?
Just checked, that was with 250%, apparently
yeah
So 2.5 at 250%, can be made into 2 at 250% and 1 at 100%?
you will still get the same amount of MW
Yeah, but 0.5 generator, will probably not work 100% of the time (50% idle time)
you should not overclock generators anyway
I was looking at what it would take to make full use of the 1800 uranium on the map. 420 nuclear power plants, is not a small amount of space. Hence why seeing what overclock that do instead
it's a waste of shards
we have nearly infinite space
also, 420 generators provide 1010 GW
even with using all the mats on the map, you probably won't go over 400 GW
unless you fill your base with jumppads or something
Not with current power draw, no. But we have more tiers on the way
or trains
Also, going full nuclear, means more oil and coal, for other productions
I am going full nuclear
yeah, but you don't have to usee all nodes
you are probably fine with one overclocked node
140 gens
I have seen I only need 1 node for 20 Generators
that's without alternate recipes probably
I chose alts for the rod production
Yeah, without alts, so is there just enough uranium for 16 rods a minute. With alts, 84 rods a minute
also, alts decrease the power needed to produce the rods
Is 600 the max ore, from an mk3 miner on normal?
and increase the general efficiency of nuclear rods to be better than coal
yeah, 600 is max
Because apperently, the calculator allows 621.4 from 1.0 mk3 miner on normal
I'd say that's like 1.048xxx
and it's just rounded value
try increaseing the precision of numbers in settings
or use consumption tool with exact input
Thats better: 1.03571 x Miner MK3
yeah, just as I thought
if you are looking to produce stuff from limited amount of resources, consumption tool is generally better
hey greeny i feel like you live in this channel
I do
That's what happens when you're both dev and tech support
blah, these alternate reinforced plates are messing up my ratios
what site are you using for those?
also why aren't you exploiting the 3rd iron node in the middle there
hello math people ๐ what will get me further, full tank of turbofuel or full stack of batteries?
depends if a stack of batteries has more energy than a stack of turbofuel
Turbofuel if I recall
Batteries might be better than regular fuel?
Haven't checked lately
wiki says they are likely to be
Satisfactory helper to calculate your production needs.
i suppose they are better, they are higher tier... thanks for info
is it better to over-dimension stuff (12 smelters for 300 ore/min) or juggle splitters a lot to get perfect efficiency (10 smelters)
can't see a good way to split 300 ore into perfect 30/min feeds
hm, if I split a 300 feed into one lv2 belt and one lv3 belt, will the lv3 one get a consistent 180 rate since the lv2 only takes 120?
Perfect splits (balancers) are overrated
manifolds > balancers
Literally just make two rows of five machines and run splitters down the middle. Easy to build, will balance itself over time.
It's actually worse to build too many machines. They'll starve no matter what.
but is starving an issue
Yes, the machines run at lower efficiency forever.
Uses less power to have the right number of machines.
And less resources to build
I'm more concerned with the final parts of the chain having full efficiency rather than the earlier parts
They will over time as long as there's sufficient supply
They'll never be saturated but as earlier machines saturate they will eventually get theirs
selfregulation ftw
but isn't that manifolds?
so the train station is unidirectional, can the train turn around or do i need to loop?
The earlier parts cant "eat" more than 30/min. So the final are going to fill up after a few mins.
Lower efficiency is only a few minutes long
ok loop it is then
oh wow there's a switch too
splitting evenly into 5 equal outputs is a real headache
I try to use balancers as much as I can, but not for outputs of 5, screw that
Yeah if it's a number whose only prime factors are 2 or 3 it's easy
Because it's easy to split inputs by 2 or 3
Otherwise you have to do some shenanigans
Oh I heard they made it so if you have two engines on either end of a train facing opposite directions you can program a train that will go back and forth?
But loops would be easier
It's never best to balance imo. Manifold everything. Multiple belts of material? Each gets its own branch.
but then, things need to be at a certain order, i suppose
biggest issue with having too many smelters/production machines on too little resources/components, is the swings in power when you first turn them on (belts would still fill them up a few items after a surge), and you'll have more variance (swinging highs and lows) in power usage. Stand bying the last few is one way around those issues. as for materials you'll fill the idle machines eventually, so raw materials are fine, higher tier components that you make at less than a dozen per minute you'll run into shortage issues.
you could also underclock all the machines, 20 machines going at 50% use less power than 10 going 100% for same throughput.
MAYBE a skybridge for the train wasn't the brightest idea...
@shy mason at this point power is probably less of an concern than CPU cycles
@royal cloak here is an easy 1 to 5 balancer that work with saturated input if you truly need it
@pseudo jay or you can use a trimmed down version
That one doesn't work with saturated input
The belt from merger to splitter needs to carry 120% of the input belt
I'm planning for a restart once trains get to stable version, so power usage is still on back burner of my mind atm, less oil generators and more nuc this time with trains encircling the island.
With nuclear you'll be hard-pressed to use all that power
At least until hadron colliders and teleporters come in.
now I'm thinking of having trains on ground floor of all future factories for receiving and organization, so I'm not going to tear down my current turbo motor 5 floor factory(that turned into a mess recently for the motors) and just restart instead.
It'll still take a while to get hundreds of nuc plants, but the 2.5 GW each will help with my idea of train colonies, and power being carried over rails makes it even better.
Btw: how many new alt recipes does the end of June update bring?
experimental got 6 new recipes for new items, wire heat sinks, radios, turbo motors, fuel rods, U cells, and control stators. Still no alts on super computer yet.
I really do hope we will se some smart logics at some point where we could set splitters to close the gate if one connected specific container runs full. and more things to actually to do, I really loved the if then functions that you could implement there.
uff, gimme a sec asd5a
there will be at least one normal blueprint for quartz and I think silica which have been only alternate before if I remember correctly
So they can still add over 30 alt recipes before we run out of harddrives
yeah, splitter -> (smart splitter + merger) is my go to plan for now for the mass smart splitter organization system to avoid lock ups.
and for the new ones I need to look up on my sheet
I counted 8 fresh alternate BP's with the update
dunno if I missed one, if you wanna cross check use the sheet https://drive.google.com/file/d/1a8ueOah9-87HHFT25ztRwOZaSdlMeHQy/view
not counting silica
fml
I think you might be right. I'm kinda tired and how ever counted the normal super computer BP as alternate fresh one
I've got no clue what two it is, even looking at the sheet.
yeah I am counting 6 as well. As mentioned, I probably counted older BP's in the list as well, I need to take a nap
nap sounds awesome atm, though it's still 11 am here
Ty
I think they also added some more crash pods in the middle regions as well.
I wanted to nap at 12 but business meeting didn't let me. now it's 5pm and I am not sure if it's wise to nap now^^
yeah holding out for dinner and early night slumber may be best
What's the screw it index?
No time for a nap you have to keep building ๐
I can link you my reddit post about that @pseudo jay there is exactly detailed what this is
it would me maybe take too much time right now
the speadsheet's name is for how to avoid screws in general that in general just fill up /max out all conveyor belt systems they are a part of.
just produce them where they are needed. Don't build screws for mainbus production
consider producing screws on demand
same for cable
now my issue is plastic that I still have to bust out for super computers after replacing them all with rubber
Ah, in the spreadsheet it is a value representing ressource cost also regarding rarity of each ressource
it's an issue as I refine the rubber and plastic from crude oil at the pump (north coast) then load it into a truck to take to the cpu factory, plastic is filling the truck route I set for it and would have to weave the two routes together if I wanted to change the refineries.
yes, it does. It relies on overall resource rarity with also taking purity into account
If i try to use smart splitters, the plastic / rubber imbalance eventually locks production down.
on the second site of the sheet you will see the resource cooeficients more detailed
nvm that wasn't a question. But yes, that's what it is, just giving some sort of logical comparable value to the blueprints
anyway, trains will help with that, as well as foresight to just dedicate an impure oil node to plastic as well.
yeah, indeed. Smart sorters with logic funtions could solve such issues easily how ever, then you could smart split items from belts without caring for overflow stops
the only non alt recipe I may stick with is possible crystal oscillators and beacons. alt for everything ftw.
plastic
i think smart mergers should be a thing
plastic alternate is one of the most useless recipes
more refineries and more crude oil
I'd like to see splitters/mergers reacting to set conditions, like a full container @fluid ether
I'll play around more with trains, industrial containers and smart splitters, at some point I'll be happy with a system then start building the mega factories I can stick with for future updates.
you would need wiring for that,too. I would like that
I consider plastic to be obviuosly excluded, same as screws, as you eventually don't need it.
indeed
still using rods for building materials, and may have a container of plastic / screws for building equipment at main hub, but saving trains/main bus for rubber / steel pipes / some form of quartz / and quickwire to make everything else from.
yeah. I mean you still will need plastic for super computers and stuff, like you can't live without that completely tho
The planned location for super computer / turbo motors is sitting on a local node of oil, and importing additional rubber in, so 1-4 refineries of plastic should be fine.
@arctic nova a great start, but it ignores how many products need a certain ressource, i.e. Iron ore is vastly more used than copper ore
*ps talking about the west end of desert canyon below north oil coast if you are looking for it on the map, next to the four inland oil nodes.
But judging that gets iffy, it depends on planned production and chosen alt recipes
thank you @pseudo jay I do agree, but it's more like a guideline then actually saying "this is the best bp". It just says basically that if you would make wire from copper you could not build as much wire if you'd use the entire map then when using the alternate with iron for example.
that way it stays comparable, there is basically no way to couple that to actual real demands
@pseudo jay only 28 spare left.
and the copper + iron ore to ingots can be useful in some places, but doesn't help if you are also using copper ingots for quickwire and aluminum. just be flexible enough when planning and it'll work out
Yeah
Considering all coppers first go to aluminium then go to caterium as well, you left none extra copper for alt irons.
Exactly, I am moving away from alt iron because aluminium
at some point you just need to start considering individual sites and how best to make with what you got and can import.
I even tried to state that out as good as possible in the description of the sheet, yet I guess most random ppl that will see it will just look at the sheet and take it too serious.
Hopefully there will be alt aluminium ingots
yeah, I recently prepared my main bus productions to be ready for 1.200 for all my base materials....that was hell of a work
doubt it, there's no alt caterium ingots
I did also switch from alternate iron back to normal
Other then bayer i remember there is another common process for aluminium
unless they start using SAM or something
There is no alt caterium, as you already have alt quckwire.
they could have alternate buildings that have different recipes that have a higher electrical input.
or mods could do that
Lovely
I do need to tear down my current ghetto base and expand my tower, eventually I want all higher tier production inside it
The tower is 21 by 21 with 20 floors or so
I need to better plan my tower conveyor walls, inside belts are fine for fitting rows of assemblers and manufactors on as few foundations, the outside belts feeding between floors are spaghetti.
@arctic nova you should paint a creeper face on the center wall
there's like one level where I have 160 smelters to process 1,200 iron, 1,200 copper and 1,200 caterium. and that thing currently has like 5 levels if I remember correct
I plan to expand to 27 by 27 and add an atrium where transport between floors happens
I actually plan on doing some mario pixel art at the long side of the building @glacial hemlock
I would make a train sprial elevator just to transport goods between floors
Smelters and foundries are banished from my tower
make it all black and call it the cube @pseudo jay that was actually also an idea I had in mind ๐
I'm restarting my game so I can add a train station and ind storage sorting on the ground level of future mega factory towers, 5 wall high for trains should be fine for sorting the storage.
if any1 remembers that horror movie
That was a good one
But for that to fit I'd need to reduce the amounts of floors drastically
well, this is resource refining there, before it was all on open platforms. Main production side is somewhere else
I hate to have smelters in normal buildings.
used to have smelters and miners in my buildings, the hole in 2nd floor for top of miner didn't help at all. So belts + conv walls it'll be from now on.
That tower needs to be wider, because I am like 5 foundations away from a lod barrier
it'll be a while, or more foundation and curved wall types, before we start to do anything besides square towers,
well, ppl allready use awesome circular buildings
If I am looking at the plateau with waterfilled craters the water doesn't render properly from the top of the tower. I need to fly 5 foundations closer for the higher detail to load and render
but it's super annoying and complicated to do
that's super unfortunate ๐ so will you at some point move the thing?
nothing you can do graphical settings wise?
nah, the different biomes have different loading/streaming zones. if you are at the edge from on to another you might have those issues
makes sense, so take screen shots from center of the biome looking at the mess from now on.
I am super lucky, I build on the edge from northern forrest to canyons and at least I have not experienced much texture issues there
It is super noticeable with water
okay, I do have this when I look towards you into the desert as well
that is an awesome spot, did that last time. Going to start on bottom right corner of map / grasslands next time to work on trains, as the space will help as I get familiar with track layouts.
how ever, there's not much to complain beside some trees that look like nuke-explosions (mainly because it's a desert with not much in it)
du you know that pure quartz node in the desert inside a cave?
I am exploiting that one
was planning on trucking it out to gold coast a month back before aluminum
Belts
fml
I read the other definition of exploiting
I really should sleep
metro confirmed^^
that cave quartz spot is a place of nightmares without a jetpack or explosive weapons, alpha spitters guarding it and back cave for nest of stingers.
tbf...alpha spitters won't hit you if you just strafe around them. They need to improve on their AI
or basically every spider
oh, that one, was talking about the middle one under the center region
wasn't planning anything for that corner after they moved the bauxite node, at least until they tell us what the SAM does .
South of it is a limestone node with is nice for on premise silica
ive got an idea
for my tw plant
ill build 3 factories off each uranium node
reducing travel lengths and stuff
then converging the nuclear waste to one central site
Why not go hard core mode and wall the spawn in with nuclear waste?
where you make the steel pipes and encased beams for the millions of ind storage you'll need
the bot hates me and my steel....
Regarding more shapes for buildings: you could use train tracks for curved decorations
yeah, going to work on 3 station basements, with railway switches / lanes to each of them to then return to main lines, as the base and get that worked out before building any manufacturing foundation above.
It'll be a mess along with all the truck stations/routes going to and fro each miner / smelter site
I guess you even can centralize it around the most central uranium node, the other 2 nodes are not that far away
I don't know how many other resources you need as well, so ofc I might be totally wrong on that^^
Satisfactory battle royal/survival? @shy mason ๐ interesting idea
just have 2 radiation spokes / belts of all the uranium. as long as you have enough concrete, you don't need much else that can't be just belted in.
you just type my name into the search bar? that was months ago.
I ment your idea of the nuclear waste around
oh fml
that was @pseudo jay
fuck it, my spelling and grammar is shitty right now, I can't count, I can't read the correct lines, I will have a nap now
oh, all good, I did have an idea of having two teams face/dual against each other to build quickest to space elevator tiers with separate tech trees, thought you were reffering to that.
still a great idea. Probably on custom map which might be better balanced and or 2 different games
anyways, I'm out, later o/
later
very messy but heres rough idea
blue = train line
red = main base
yellow = poweer
you'll need to export the aluminum/bauxite before it gets too rad infested, but would work.
I need to start making maps of my ideas at some point.
@tacit bison have you got a nasa supercomputer my man
my pc hurts loading that pic.
if that is all the belts, can't be too bad, if that's 10% of all belts/ bus you have, yeah you need some liquid nitrogen overclocking.
at least until they start using more multi cpu core threads for the game, it's using close to 1.5 threads now.
but that's an optimized build problem, and not something you do with early access to test gameplay.
It's the majority of the belts. Everything goes to and comes off of that bus. Currently 8 lanes wide, 7 high. Not much is optimal in regards to balancing, but that wasn't really the goal of the build. More to just have everything easily accessible. I'd be lying if I said it doesn't get framey sometimes when I've got it all in view
wow i hate these train tracks a lot
but unless i build it on columns, it will block my oil supply, which will block my power
who scales down to seconds? rather scale up to a full minute since you allways get the items/min ^^
@unborn parrot that screenshot looks like you are still dictated by a factorio mindset. Still looks great tho
Yeah, call me an ass, but just "paving" over large portions of the map to create a flat factory feels.... cheaty. That's just personal though, I know better then to tell people how to play the game ๐
for a while I could only build floating platforms if 1 edge of it was on a natural wall or supported by walls already there, so you're not the only one.
@hexed jungle well, it's what happens in the real world, so I'd call it realistic :)
Oh totally
I don't really enjoy these 2d bases with bus. We have 3d space to build and we can build nice factories, don't build like in factorio
I agree, though, that it's not the most beautiful way of building. which, sadly, also applies to real-world industry
@wind spade yes, those large bused feel strange in satisfactory, IMHO
I like building "single-concern" buildings in a large complex. I try to make it endlessly repeatable vertically, so I can just add new floors when I want to expand output
busses should not be in Satisfactory. They are not really valid here
@wind spade i like bus-like structures for getting raw materials from the mines to the factory complex
I make a floor for each stage of a process, like 1st floor for circuit boards, 2nd for crystals, 3rd for cpus, 4th for radios.
people put everything on the bus
@shy mason same here
My friend always does that, and it has such a neat charm to it. Exploring his factories feels so much more organic
here's hoping the mark 2 constructors and assembliers stay the same size.
or I'm going to have to start over again....
I also try to make the outside of each building look a little different.
I'm hoping they make them better than mk1
my buildings all start as squares until the spaghetti feed belts come in
like if mk1 assembler eats 4 MW and mk2 assembler would eat 20 MW and have x2 speed, then there would be almost no reason to use mk2s
I think that I'll nuke my entire factory except the mines and resource delivery busses as soon as the experimental hits ea
Would be nice if there was a bulk-deconstruct
yes
I god damn KNEW IT
might delete everything, not worrying about drop chests, and then save editor away the crates.
nah, searching all the hard drives once was fun, doing it again would be painful
but tbh, bulk deconstruct mostly means you haven't planned your factory well, so it's kinda your fault ๐
yeah busting the chart and jetpack for that again.
I don't mind making a new save, but I have a habit of doing that with most games. Then I end up knowing the first 20% like the back of my hand, and only reach the end once or twice
(keep in mind the word mostly)
and I don't want to rebuild my slightly insane oil delivery bus that spans 3/4 of the map from east to west...
means bum rushing tier 6 to play the game though
I don't think anyone nukes a whole building without knowing they done fucked up
@wind spade my problem would be that my entire factory is laid out for belts that do 450 items/sec
erm minute
@shut arch well, manifolds make things easier
I have lots of manifolds and balanced storage outputs, but the belt throughput has lots of consequences for things like the number of constructors on a floor.
now they are going to go 480 without any change on your end. new max is 780, and that is going to make my floors huge.
I mean, nothing really bothers you with that. If they decreased belt speed, sure, that's an issue. But they increased the speed, so you have no reason to rebuild the part that is working nicely
went from 15 smelters to 26 for the rows per maxed out miner. going to start doing the double manifolds for similar factory floor plans.
even if it's not using fastest belts
but I really don't want to wine. deconstruction and rebuilding in satisfactory is quite fun, almost as with lego
the only complaint is that my RSI hates the click-hold-release action for deconstructing. ๐ญ
I remember seeing someone say they made a macro for it
yeah, I'll also do that for the next big rebuild
Honestly, deconstructing the factory machines and belts isnt too much of an issue, but if you put a lot of effort building walls, it can definitely get real tedius to deconstruct everything
it's still sad that the game is such a tendon wrecker by default
took me a painful 2 weeks before I realized the sprint toggle was a thing
ouch
yeah, the "big rebuild" is something, that can easily be avoided if you plan for the future
Live and learn. Most of my mistakes are/were simply due to ignorance of the games systems
@shut arch if you have autohotkey I can send you my script
It does the left click holding on a loop
How long do you time it for?
It's exactly 1s
ahk. Good to know
Yep, or if you're epic you can just make a script for it yourself
I tried to make one for Autowalk and Handcrafting but it failed pretty spectacularly
Where's the "lazy bastard" achievement for satisfactory
I'm guessing the lazy acheivement is only make 100 items or less of each item by hand until space elevator "I win" button.
leaving 100 so you can make them by hand to make the assembler.
@crude girder hold E, open inv, release E, close inventory
I know that one, but scripting
Someone tell coffeestain to make everything "on click down" so we can automate the game about automation
There are a lot of QoL and stats I want in the game but will probably never exist for another year or so
I wish the space elevators required like 10x the materials to bump to the next tier. It seems just low enough that it's tempting to try and rush it by idling some hand crafting
There will be further balancing in the future. But be reminded that AAA titles have an average play time of 40-80 hours
True, I guess its a boon that it's so low while we're getting so many updates so quickly. It would be a huge pain to rush to end-game to try out new features on a new save
tier 6->7 requirements are pretty huge with 1K items that come out at 1-2 a minute per manufacture at full production,
Ah, I still haven't really touched experimental/next update (If I'm getting my tiers right here)
you're right
Is ir more energy efficient to underclock for 7.5 as 7 whole 1 half or 8 @ 0.94
Here comes Greeny!
The calculator site leads me to do the latter. That is why I ask.
smelters as example:
8 @ 94% = 29 MW
7 @ 100% + 1 @ 50% = 28 + 1.3 = 29.3 MW
around 1% difference
So I'd say pick your favorite and go with it
How about manufacturers? Difference might matter.
calculator shows the latest, but it's mostly because of how it's coded internally, that it was easier to do. Also the other one can be easily calculated (8.42 machines = 8@100% + 1@42%), not like the other one
@stark lichen it will still be 1% no matter the base energy (at least I think. Let me verify that)
Thanks. Although the half is less of a hassle to upgrade if you increase production by less than double.
manufacturer:
8 @ 94% = 398.5 MW
7 @ 100% + 1 @ 50% = 385 + 18.1 = 403.1 MW
around 1.15%, so not a big difference
Wow, it's really better to underclock 8 by 6%, rather than 1 by 50%?
and it's even 2% more than needed
I mean off the top of my head I'm pretty sure underclocking 2x constructors to 75% is worse than having 2 but cutting one down to 50%?
so it's even slightly more better
2x75% constructors = 5 MW
1x100% + 1x50% = 5.3 MW
@barren elm
Fair enough, good to know
is there a chat or graph for clockspeed to power use
@jovial obsidian hang on, I'll make one for ya in wolframalpha
@wind spade thank you
@jovial obsidian https://www.wolframalpha.com/input/?i=graph+y+%3D+4+*+(x%2F100)^1.6+for+x+from+1+to+250
it's for a constructor, if you want anything else, replace the "4" in the query with the power consumption of the thing
okay thank you, it really goes up there
it's really terrible how the bird throbbs after you kill it
Don't kill the bird
train loop complete.
this channel is getting more and more offtopic
@tough fiber can you keep those posts out of this room as they are not #math-and-meta related
pple are abusing this channel for screenshots lol
but we already have a place for screenshots
i know
they even dropped the cool down period after I asked
it's still there
1 minute instead of 2 now
lol sorry actually didn't notice where i was
(also when i talked about the bird lol)
the 24hrs in the streams and videos is a bit insane - need a separate one for videos - I personally don't care about people's streams and get the 24hr wait for posting those
Wrong channel?
Yeah Wrong channel @wild spade that comes under #off-topic-general
How many fuel gens can you run off one pure oil node
overclocked or normal?
Normal
I am incorrect its Pure oil at 240, 4 oil refineries, 12.5 generators giving 1875 MW
at 100% usage
im at 5,900 on oil at 100% use
1 node?
2 nodes and im still growning the gentors, i can see if i can run it off 1 node
100% consumption?
you only using 20% of the power available
right now iam because i overloud on my supply chain so shut it off to impove the gens hense why i asked about powedr to clockspeed eailer
Does anyone know if there is a program or anything that helps map out the most efficient way to create items autonomously?
@lucid nexus https://satisfactory.tash.fyi/ this might help
that is great, thanks
your welcom
Alright thanks to @wind spade im almost done with the layout of my quarter of terawatt factory which im gonna have to make 4 of them ๐
wish i could read it lol
I cant even take screen capture of it ๐
all is good, not even the PSC key?
i dont want to start a production line just to make pcs ill stick to those aha
made it on an excel sheet, took me long enough ๐
your insane
Youre a mad man
Mad man is too light he is just insane
Indeeed
At what point do trains beat conveyor lines?
In terms of throughput.
Rail is cheap too!
A 2-cargo train already beats a mk5 belt.
On a flat circuit i can pull 40 cargos from just 1 locomotive
Iโm wondering what the best ratio will be. Iโm building stations assuming a conservative 2:3...
I have hills...
is 800m is far enough to justify the huge buildings?
I have a sky base roughly 150 plates/ 3 conveyor lifts in the air.
How fast do trains travel?
120kmh
So transport belts require 1 item per meter to build?
yes
#math-and-meta pins for production planners but as for the amount of foundations you will need that is up to you
@unborn parrot
@frigid sluice 2 meters per item
how much beter in % is turbo fule?
Phew, only 2,5k beams
@upper crater I just answered on Reddit to somebody, so I'll just copy it:
Leaves, Fabric: 15 MJ
Mycelia: 20 MJ
Wood: 100 MJ
Biomass: 120 MJ
Alien Carpace, Alien Organs: 250 MJ
Coal: 270 MJ
Biofuel: 300 MJ
Compacted Coal: 600 MJ
Fuel: 750 MJ
Battery: 1000 MJ
Turbofuel: 2000 MJ
Nuclear Fuel Rod: 750 GJ (750000 MJ)
thx
Is there a formula to keep production at 100% whilst using trucks/trains without overloading the belts/other infrastructure
I know I can use storage bins as a buffer if necessary but I want to keep everything moving
for trains: no, it won't work
for trucks: super hard, probably won't get it correct, since trucks are kinda random
That's a shame, I like using the transport options but after an indefinite time eventually the materials will get backed up
I generally run a truck or two first while I'm getting production set up for something and then replace it with belts when I've got nothing else to do.
Can someone demonstrate to me how you can load balance 3 MK3 miners producing 300 items/min to freight stations for trains? Each station has 2 inputs so it gets tricky when you try to get 1800 items/min into them
*600 items per min
why do you need to load balance them? Just connect each miner to one platform
I figure Iโll have to condense the belts at some point once they reach my production base though
Without priority mergers and splitters, it gets complicated having to dedicate a whole belt to one production line instead of using that same belt over multiple production lines (waterfall merging concept with a stacked bus)
what on earth are these terms, who comes with them lol
"waterfall merging concept" and "stacked bus"
Like if my production line only need 1000 items per min but I had 2 belts of 600, how would I be able to balance it efficiently and still have some left over to use down the line
I assume with the first one, you are talking about manifold
you don't. if you supply 1200 items into a 1000 consumption chain you'll overflow with 200/minute. thats how consumption works
Yes, my machines use the manifold but I want to be able to still use leftover backup resources to other production lines
well after the last split, you are left with 200/min leftover
If I split it, it will only go 1:1 and not backup.
Where might I split off to use the excess if the splitters only split 1:1 and donโt fill up one side and then go to the other output once the other is full?
just do it like this (ignore the numbers)
you will be left with 200/min at the end and you can use it for anything you like
Ah I see. Even though the splitters alternate where the item goes every other item, it will still be enough to cause a backlog and get that 200/min output?
it works the same as a manifold
the last machine will fill and when that is done, rest of the resources continue to the next splitter
What I mean by priority splitting is once one output is backed up, it will then go to the other instead of trying to split it evenly and siphoning off resources from the other side. If you have priority splitters, you remove the time needed to fill the machines.
the time will still be there
it would be cool if we could limit/disable the buffer on machines
I suppose the streamer I had been watching didnโt know fully how manifolds could work. He would try splitting off a belt that supplied more than the manifold could handle, but the dilemma was that it kept splitting evenly so the manifold would never get backed up.
He would try โwaterfallingโ resources down from higher belts in the stack, but that too wouldnโt work since the mergers also evenly go 1:1
yeah, that's not how to do it. If you have more than a manifold can handle, just use the other end of a manifold
Thank you greeny. That basically just solved the existential crisis I was having in game. Also, do you have a list of the best combo of alt recipes without having to go calculate it on your site? I suppose someone has already worked out which recipes are the best.
recipes aren't best or worst
Now that trains are out, time and efficiency are almost zero-factors so based on whatโs available on the map, the ratios of resources available would point to which recipes are going to work the best. (I.E. alt screws arenโt necessary, and alt computers arenโt as good as crystal or caterium)
Like screws would make the B team while the caterium/crystal computers would make the A team list. I know thereโs a good combo out there that would be the most efficient for the existing nodes out there
yeah, basically it's about availability on the map. But also some people have different end goals, some want to make as much SCs as possible, some do HMFs or a mix between those two, etc. etc. So it's hard to say which one is good or bad. What I'd suggest is use the analyzer tool I have to figure out what suits you the best
Say for instance, I might save more resources doing crystal computers if I use all the alts for the production chain, compared to caterium. But in order to calculate all that I would need like 10 production lines to calculate which would make more sense
Oh hey which tool generates that little chart greeny?
A powerful Satisfactory calculator, filled with features, supports overclocking, alternate recipes, live map, power calculations, etc.
@open plinth that one?
like the green-red tables?
Thank you greeny. I also notice youโre very active on this channel and I appreciate the work you do here. Your calculator has been a tremendous help
for example, 1 reinforced iron plates raw costs based on recipe combinations:
you can also sort it by clicking on the ore icons or "total" text
This one
Will there be a guide for train efficiencies soon? (Best loco to freight car ratio, etc.)
@open plinth that's a manifold fill tool https://satisfactory.greeny.dev/machine-fill
A powerful Satisfactory calculator, filled with features, supports overclocking, alternate recipes, live map, power calculations, etc.
Ah the manifold fill time tool
Thanks ๐ It was cool
I liked the time till efficient display
@haughty quail trains are the topic of today's SSS, but I am still writing it, I hope it will be out in a few hours. For the efficiencies, put simply: if you are on a flat land, you shouldn't have issues with 1 loco for X freight cars (any normal number will do, unless you go crazy over 30). With hills and stuff, you shouldn't go over 20 cargos, safe bet is 1:5-1:10, but needs testing, since it depends on speed it starts with and length of the hill. But if you are able to stop&start the train on the biggest hill, then you are fine
I see. The AI likes to take it easier on speeds than I do, and Iโm sure thereโs an equation it uses to calculate all that
tbh we weren't able to figure anything out. 1:5 seems safe even for AI though
@open plinth yeah, it is used to estimate the time manifold will start working for 100%. I made it to counter all those pple that say "I don't want to wait hours until my manifold starts working, I'm gonna make balancer"
If we can find what equation it uses, we could really crunch the numbers and figure out exactly how many locos you need based on the 3D path of the track that you could visualize with one of the save visualizers out there. I imagine the spline stores numbers on the slope and curve
we may be lucky and devs will give us more info on that. But until that, I'd stick with 1:5
I usually just solved my impatience by letting the source fill up a container while I finish building the rest of the machines, and then I just manually fill them up to max ๐
It probably uses the same methods that the other vehicles use when climbing up hills. Like some gravity coefficient that amplifies over the slope. We could probably ask and theyโd give it to us.
either that or let the manifold run while you build the next part
the next part will probably take you longer than the time manifold requires to fill
I kinda hate buffers
they hide the problems your factory has
well, rather postpone them to later
Maybe, I'm hopeful that maybe one day we'll get something like Factorio's combinator system and alert poles, so I can detect if a container drops below a certain level and fire an alarm ๐
Mods can do that once they add in full support.
I'll just paste some math from my Reddit posts here:
Total fuel efficiency for vehicles (max energy = energy from one stack):
Biofuel: 300 MJ, 200 Stack size, max energy 60 GJ
Coal: 270 MJ, 100 Stack size, max energy 27 GJ
Compacted Coal: 600 MJ, 100 Stack size, max energy 60 GJ
Fuel: 750 MJ, 100 Stack size, max energy 75 GJ
Turbofuel: 2000 MJ, 100 Stack size, max energy 200 GJ
Battery: 1000 MJ, 100 Stack size, max energy 100 GJ
Nuclear Fuel Rod: 750GJ, 50 stack size, max energy 37500 G
can someone help me out getting my 1 to 10 splitter?
@daring surge sure, but maybe consider manifold instead
---+---+---+---+---+---+---+---+---+---+
Unless 10 constructors need more of the ware than your current belt capacity (in which case a 1:10 splitter wouldn't work either)
in that case a manifold is still better, because you have more machines running at 100% so less power spikes
manifold?
i got 450 coal and 450 iron i gotta divide over 10 foundrys
yeah, you can do the manifold for each resource separatedly
aand i just crashed my game twice cus i forgot to make my last freight cart stay on a flat surface while docking
btw does that also make sure every line has equal amounts of resource comin through?
yeah, it balances itself after a while
the first machine will be eventually full and will let the other items flow to 2nd machine
if you have enough input, then you should be fine
this is your setup
will be efficient in a little over 10 minutes
but first 8 machines are efficient in 4 minutes already
so it's pretty good
ok thx
What are these batteriea you are talking about?
item from experimental
No batteries involved
some people reported that they can be used as a vehicle's power source
idk if that's correct, but if I assume it is, this is their fuel value
Oh, how do you make/use them
what's the math on tilted jump pads? Like how many foundations forward and up would you go?
last time I used them, they weren't really consistent, throwing me in a range of ~2-3 foundations from the center point. I am not sure if they fixed them, but if they didn't, I wouldn't recommend them
Just manifold
Your machines can only receive and output a certain amount, once it starts clogging up, the rest of the line will receive the resources
It'll take awhile to saturate, but it should work at near 100% efficiency after a while
@inland shadow Because of their inconsistency, varied upon angle, speed, jumping or walking, pressing forward or letting go of keys, they're very annoying to use. Plus in early game trying to place a gel pad for landing is a nightmare. I used a system of three at certain angles to get up a small cliff for a time, but ended up just building a ramp of foundations that worked far smoother and didn't eat my biomass (pre coal)
@safe halo @inland shadow however, the inconsistency can be removed with by eliminating player input by using conveyor belts
Ie. Player gets onto conveyor belt and gets dropped off on jumppad without doing anything
@fierce ruin so i'd guess : if an item has 100 amount each stack > you'd need a wagon for every 3km of rail track to match an mk4 belt? < and a wagon every 2.5km for mk5?
@pseudo jay True, though without flight key (pressing w) added, you don't go impressively far.
that's assuming loading/unloading + acceleration is near instant ofcourse
Sounds about right plus track is so much cheaper than belts
it might be cheaper, but still: considering the reports on trains getting 'bugged' if you have multiple trains using the same splitter at the same time > it might still be more efficient to use belts
so many reasons not to use trains right now, so little benefits to them
1 steel pipe and 1 beam per 4m of track
you're forgetting i have unlimited everything anyway
(well except alu sheeds, only got like 6 mk2 containers of that atm :p)
I need to get ready for work bbs I still want to discuss tho
i want trains, but having a seperate track for every material type seems super oversized
@safe halo frankly you can guarantee consistent behaviour without eliminating player input. I have gone for stackable conveyor poles for vertical movement because of that
so it'd be better to wait till the train splitting problem is fixed
also (random example) > right now i have 3 long distance mk4 oil belts, i would need a train with at least 6 cargo containers just to replace those , means i also need 6 platforms for every station on that train etc (since you cant have a train stopping at a station that's shorter then it's full length)
fairly sure that'd make trains more resource expencive then belts
More expensive to setup but more performance friendly from what I understand
You could reduce the length of stations by running several trains on the same line
@crude girder I cant manifold
I need too split the plastic
For the manufactor and for circuit board production
S/can't/don't want to
What ratios do you need?
Oh 37.5% and 62.5% is easy: split input belt into 8 belts, then merge 3 of those for 37.5% and the rest for 62.5%
@warm root you could have more mini trains on the same track to make up the difference, for eg only 1 loco and 1 car and then just make more of them to make up the missing
@tough shale you can manifold. Just run if near the first set of machines, then run the rest to the 2nd set of machines
you don't need to manifold to ratio
you have 2 computer machines and 4 circuit board machines, all need plastic. So a simple 6 machine manifold
I trust you hope it works
sure it does. It's the same as when you split to multiple machines
@pseudo jay For the angled one? Can you grab a screenshot?
lemme build that real quick
@safe halo something like that
you can tweak trajectory by adjusting the end of the conveyor
Oh, yeah. I get that. I meant it's difficult without using things like conveyors, fences, or in your case poles to block a player into only going that one direction
Yeah. Ladders are pretty great
damn, does anybody know how to convert a number of seconds in google sheets to hh:mm:ss?
@wind spade https://www.youtube.com/watch?v=0zADtaH-8yI
How to convert decimal time to hh:mm time in Google Sheets. Sample Google Sheet: https://docs.google.com/spreadsheets/d/1DUyBmjppDcPdFJr-snxF9fE6leuZDz9jl9os...
@random fog thanks. I found this formatting option, but couldn't figure out that it has to be in days, I was assuming it needs minutes or hours.
for me I was using "Format > Number > Elapsed time"
but the cell has to be in days
that's why they do the /24/60/60 for seconds
and I had my cells in seconds and was trying minutes or hours, for some reason, I didn't thought it'll need days
yeah. It shows elapsed time in hh:mm:ss and needs days as input. who could've known
well its typical to spreadsheets to have additional cells just to get numbers be converted in another format accepted by a function
oh yeah. But I didn't need them. Since the seconds are calculated by a formula, I can easily add the division as well to the formula
thats when you don't need the number somewhere else ๐
nah, it's just a small table of Freight Platform fill/empty times
lol small tables sometimes grow
make sure not to end up with a table kilometer in diameter
it's only 6 columns and 11 rows
... they said ๐
unless they publish a patch with new belt speeds or new item stack sizes, I am ok
yeah, I am including that
since trains are in experimental only, I don't see the need to calculate with old belt speeds
what would be great is a new splitter with adjustable throughput for any output hole
I already talked about this a few times.
Basically if you want to set exact number (like 40 ipm on the left side), then it won't work even if they figured out some magic way to do the splitting (they won't, since it's impossible to do exact split)
if you want a ratio (like 1:4), that is doable
Both options are useless anyway, since we have manifolds and they are way better than any ratio splitter
manifolds?
well, gimme 10 minutes and i'll write piece of code capable of converting throughput values into ratio plus input throughput
ok, but don't forget these assumptions:
- belt doesn't have to be full
- items on the belt doesn't need to have equal spacing between them
- there may be a more than 60 seconds gap between batches of items (some alts are producing 64 secs)
---+---+---+---+---+
that setup would be so brilliant if only machinery would accept not a whole stack of materials but only needed for 2-3 products
also:
- the throughput for last minute doesn't have to be the same as the throughput for the previous minute
yeah, that's like the only manifold disadvantage. I actually have a tool, that pretty accuratedly predicts the time a manifold needs to start operating at 100% of capacity
the times are mostly around 5-10 minutes, with the last two machines often being the slowest (around 20-30 minutes). But that differs based on item stack size, throughput, number of machines, etc.
just let me choose throughtput for 2 outputs and let the rest go into last one
so, a priority splitter?
oh, please keep it /min
yeah, that's exactly what manifolds do (kinda)
yeah
but you can't make this happen far from prod line
you have to pass it all through
and they don't need any setup for that (no setting up ratios in splitter), are easily extendable (again, without setting any splitters)
and wait for ages for it to saturate
it will always work in a finite amount of time
and considering we have no permanent output, where we can put our resources, our production lines will get stuck at some point anyway
:lurk:
well, i think about it like that.
if its so complicated to make what i'm talking about, okay
but let's assume you really want the splitter and you don't want to use manifolds. Sure, that's a valid assumption.
But there is no way to make a splitter split exactly 25/min
I am open to discussing this and I can provide examples that will break any piece of code you will pass to me
yeah, but it won't even pass 60 ipm
@feral dew did a research on this
because of how splitter works, it may actually pass a little less than 60 ipm
gimme means to slow down belts to values below that and i'm actually okay
I can try to find his answer on a slightly different setup
but the same case basically
well, i just could make a fast belt buffer
then "slow" piece
of say 20 / min
and vuala
what's the matter?
theoretically that would work better
@feral dew another case of a "splitter to do X per minute on a side belt"
and I mentioned you because we got into the "split 60 ipm using mk1 belt" issue
btw you thought about internal buffer inside splitters in this case?
they don't have it
yeah, fair.
wouldn't work anyways
round robin i guess
you only know that you need to put 20 ipm on side belt. How do you that from the splitter's point of view
I'm talking about the splitter you mentioned, with rate option
actually no
think of it like this
we got outpust 1, 2 and 3
you decide where priority goes
not necesserily where biggest number is
then you got the algorithm
all outpust have internal buffers
btw, @feral dew we need that post of yours pinned here (the one where you explain why low mk belt split doesn't work). I'm tired of searching for it for conversations
buffers are fine, but they only delay the issue, not remove it completely
if the top 1 priority's buffer have got 2 items in it, pass
then top 2 priority, same
then the rest
if all full, input stops
outputs just draw from internal buffers with given speed
first of all, I would hate my splitters if they keep some items inside. I don't want my rare 1 SC/min to be stuck in some kind of a buffer just because
buffer capacity discissable
it could stuck only if some output full
of not, no storage
second, imagine this example:
you have input of 30ipm on a fast belt (e.g. mk4). The items come in batches of 30 items in ~4 seconds. Then there is 56 seconds gap.
this breaks almost any split you want
well, it doesn't
it does
you got nice supply of things yesterday, doesn't mean that today you have to return all the low priority stuff because supply stopped
e.g. if you have 5:5:rest split, with 1,2,3 priorities
first item goes to first lane, 2nd item goes to 2nd lane, maybe 2-4 items will fill some internal buffers, but rest will go to the low priority
so you will get at the best 3 ipm
but not the 5 ipm you wanted
(and yeah, you can argue, that you can increase stack size of the buffer, but I can increase the numbers and we are back with the same issue)
gimme a minute
okay, sorry, now is one of teh few times that I'm actually playing the game
but I'll try and figure out what's going on. @random fog What in the world are you trying to say?
keep in mind, that I'm not trying to sink your idea or anything. I think output rate splits would be nice. But I'm just pointing out that it's impossible to do an algorithm that won't break on special cases
@feral dew he is rooting for output rate splits, e.g. 50 ipm left, 30 ipm top, rest right, with priorities in this order
ah
@random fog So, the reason why output splits wouldn't work
is because if you decide the ppm on that output, there's a good chance that it won't be an integer ratio of the input rate
Mk1 belts and Mk2 belts work flawlessly, since 120/60 is just 2
now he's stuck with trying to figure the algorithm of the output rate splitting, he tried internal buffer in the splitter. I must say that is actually a good idea, but it just dealys the issue
yea, internal buffer resolves that
oh geez
๐
no, it just pospones the issue for a bit
I don't know what greeny's example is, but ultimately I'ma sking you what you think of buffers
because, that's an issue in of itself
once the buffer fills up, it's the exact same thing as if the buffer didn't exist to begin with
my example was touching a bit different issue than @feral dew is trying to point out
explain it thoroughly then, I'm building a manifold
@feral dew my example was 30ipm on a fast belt (like mk4), in batches of 30 items every minute
or basically any number of items in batches, so there is a big gap between some items and the splitter has no way to deal with that (since it doesn't know there's the gap)
its a question of a goal there
for me its okay if a batch arrives and 5 ipm goes on a priority side only while batch is still there
i mean, 60 items arrives, they pass away in, say, 10 seconds
and 5 ipm ports sucked only 10 of it
well, okay, thats the goal
wait what. Each of the 5 ipm outputs should only take 5 items from the batch. Remaining 50 items should go to the last output
ok. So, you are saying, that it should suck the items even though it's not their "time" yet?
and keep it in an internal buffer?
how big is the buffer?
not at all
๐
you now confused me more than before. So what happens if a batch of 60 items arrive?
explain then
i guess internal buffer could be one for all, not per output
so, while batch is still there, 5 ipm output takes one item as soon as batch arrives
and a new one once in 12 secs
from the common buffer
like it was a leak in a pipe, through which only 5 ipm could go
what if the buffer is emptied before the next batch arrives?
what if there isn't immediately an item after 12 seconds?
then no item is delivered into 5 ipm gate
buffer size lets say 3 items, relitively small
and your 5 ipm isn't 5 ipm but less
imagine you had chains of those splitters
like one doing 200 ipm split, then another one doing 75 ipm split from that, another one doing 20 ipm split
because you didn't have an item to put there
but there are no items in the first place
they passed away
totally confused right now sorry
Okay, let me ask something
what if your input is in packet form
where, there isn't an item ever 24 seconds
so, every 12 seconds, there's a part, or isn't
- one batch comes in
- buffer gets filled
- buffer starts sending 5 ipm (1 item every 12 seconds)
- buffer sends everything it can
- buffer needs to send an item now, but there's no item
- buffer doesn't send item, so the output isn't 5 ipm
- next batch of items arrive
basically, distributions like 5 ppm, 5ppm, excess ppm, doesn't work
because there's a good chance that the input would be inhommogeneous
you say as if there was a periodic tick for that 12 secs
i meant something different
my bad
yeah we are just trying to tell you that your setup won't work.
once in 12 seconds, i mean, if the buffer is not empty, and the output did not send an item in last 12 secs, it does
yeah, but that can mean, that you send less than 5 ipm
when does it put items in the buffer?
when it arrives
and if there is a chance that 5 ipm split won't do 5 ipm (even though the input is obviously more than 5 ipm), then the splitter is unreliable
and unreliable splitter will be replaced with manifolds
because that's what we do with unreliable splitters
well, that's different
5 ppm divides any belt
so that won't be a problem, the problem is when the input is homogeneous
and when does the splitter take in buffer. Taking in buffer by definition reduces the throughput on other belts
not necesserily
how not necessarily
give one example where taking in buffer doesn't take an item away from another belt
you got an item input, same moment you got it in priority gate
it was inside buffer during the very same game tick as it exited
oh geez, then when does the buffer go into play?
buffer stack up only when splitter is unable to fit items into priority gates
so, you got the rest output
So, let's say that we have a split from a 270 belt into a 60 belt and a 270 belt, or 270 -> 60/270
with buffers the 60 belt will receive an item and won't be able to recieve another after 4.5 ticks of the 270 input belt
after recieving the item, the 270 belt will receive an item, and every other item will enter the 270 belt, with every other entering the 60 belt
after 4.5 ticks, the 60 belt's buffer should have 2 items in it
at the 4.5, the 60 belt opens again, and 1 item from the buffer enters the 60 belt
which means we have a repeat of the initial case, except the buffer now has one item in it
naturally, after 5 repeats of this cycle, the buffer will be full
and it will act as if the buffer never existed, with every item entering the 270 belt, instead of every other item entering the 270 belt
in other words, the buffer achieved nothing instead of lengthening some period of time
the whole point of a buffer is to give an item to a prioruty gate when it needs it, but input is not stable
in the ideal situation where you got reliable 270 input it does not play any role
think of it like if there was a simple splitter
270 in, 270 out, 270 out, but on last 270 after 5 meters starts 60 belt
what
no..
why not
This si what I was going to explain in the very beginning
so, the problem is the non-integer ratio between 270 ppm and 60 ppm
270 breaks down into 135 + 135
Just listen
The 60 belt can reopen after 1/60 minutes, since it has a belt speed of 60 parts per minute, and 1 part length is *one part. Similarly, the input enters an item ever 1/270 minutes. That means that the 60 belt can recieve an item after 4.5 ticks of the input belt
Well, obviously it can't take in an item after 4 ticks, so it has to take in an item at the 5th tick
Which means that during a period of 5 ticks, 0.5 ticks are spent not accepting items, which reduces the throughput
thats the reason for a buffer
in fact, the output will be 45 ppm.
to put away round-robin mechanics
I literally just explained why the buffer is a temporary solution
wait a sec
you mean that under no circumstances never ever can 60 belt be a fair 60 belt?
No.
I said that when you split 270 -> 60/270, the 60 belt won't accept 60 ppm
how does it suck items from constructor, why can't it suck items from buffer just like that?
โ โ ๐
^
think of a priority output gate as if it would be output of a constructor
and constructor's inventory was the buffer
and the belt's speed was adjustable
i just don't get why it would differ from such a situation (where 60 belt is actually a 60 belt, not 45)
@feral dew I think he is right with this one tho. Still doesn't mean the splitter will work, but in this example you and he provided, it works.
Since the internal buffer is not related to belt tick speed, but game tick speed. So it can output an item every Xth game tick as needed for the 60 IPM to work.
I've done my part ๐คท
I don't believe it would work.
that is of course only in the case, that the buffer has an item in it
if you let only a priority output to take items from buffer, it would be full
and non'priority output get items only when buffer overflows
@feral dew imagine a chest there instead, directly connected to one output of a splitter (if we could do it). If you connect a 60 ipm belt to the chest, you will get 60 ipm, if the input is more or same as 60 ipm
you're saying to smoothen out an non-homogeneous input?
or you could imagine a storage container capable only of one stack of items, but when it fills up, input items get moved on another "low-priority" belt instead of stopping the input conveyor
@feral dew depending on the size of the storage, it could work. But at some point the storage will be too big to be useful
and so, input is 270, output is 60
still breaks on special cases tho, so not really useful anyway
and the "low-priority" is 270 too, but it gets its items once in a while
So that's what you've been talking about.
If you want something like that in the game, go upvote something on question site, people've been asking for that since alpha, yet we have been getting by perfectly fine without
just like you are perfectly fine with assembler, still prefer things like c# and python
๐
no, I think those things are on completely different orders of magnitude.
well, but in this case manifolds are C# as a superior language
they take much time to get filled up
fun fact: even if we had perfect splitters that will be capable of splitting X ipm like you said, manifolds will still be better
reason why: because I don't need to set up any ratios in a splitter and just put a row of those and don't ever change it again
the same if I underclock/overclock the machines, manifold will automatically balance stuff, with the splitters I would need to change the ratio
i would place a manifold behind the "magic splitter"
the same if I upgrade the machines (MK2s are comming), some of them will stop working (output filled), etc. ect.
the "magic splitter" is required to govern what assembly line gets items, not what assembler
you can manifold manifolds as well
red is manifolding in series
blue is "magic splitter"
when you have this assembly lines in different parts of the map, you don't want to build few extra kilometers of belt
and wait overnight for all this to fill up
I doubt it's overnight
it would be still the case if only you could setup a priority (even without ratio) for the splitters
most manifolds have efficient times of around 20 minutes
also, you could build the priority assembly, connect it to the miner and start building "not a priority assembly", while the first one is filled
500 for a stack
and you can also overrride the wait time by prefilling machines
and lastly - get rid of screws. They are useless
eventually made me build large splitting system
They maybe useless, but the mechanics remain to be like that
In fact, 2 things coult make me happy in this story:
- production machinery input would not accept item if it is already enough for 3 crafts
- priority on simple splitters without ratio
priority on splitters mostly means, that you don't produce enough raw materials so you need to prefer some part of a factory over another
or i want top speed production of a part of my factory today
then connect a belt with enough resources to it
what if i don't have 2k of new resource in my pocket
then take it from the first few machines as they fill
what if I only have one hand and I'm blind?
what if?
there is always a case it doesn't work
after that, they would start to fill all over again
I mean, just connect it and let it fill
it'll probably fill before you build the next part anyway
thats the difference of whether you have quality of life or not
I don't see how priority splitters would help you with that issue
screw story did not fill up even when i finished all the big assembly line it was meant to feed
not even close
what's your setup? How many machines, what's the recipe you are making and how many screws are you inputting?
and then i build some other factories, new oil refinment station, still not full
standart recipe, 5 assemblers of reinforced plates
how many screws input?
sooo, how exactly does your setup look like? 3 belts into 5 machines?
let me launch the game
just fyi, if you had mk5 belts, the fill time will be 14 mins, 35 secs
which is super acceptable
there were no mk5 belts released
considering first 3 machines are already efficient in 2 mins 47 secs
so it's most likely something wrong with your setup
it began to work when i evenlysplit these belts
wait a min, i'll make a screenshot
it was two splitters on 1 and 2 belt feeding merger on belt 3
plan was to wait until belt 3 gets loaded full
then belts 1 and 2 would send less screws and everything would line up
belt 3 never filled up
I'll wait for the screenshot
its 6 assemblers
i managed to even out this line only when i unlocked mk4 belts
and yea, you can't use manifold when there is no belt to handle it
so:
- you need to be 100% sure that you have enough input
- all the input belts need to be moving at all times
- I may be wrong, but those input belts on the right side look like MK2s
- and of course you can use manifold instead of this ๐
the 3 inputs from right side are not evened, so, pain
to say what @wind spade wants to say: "SCREW SCREWS!"
and yeah, screw screws, reinforced iron plates are like 100 times better
all moves smoothly
there are enough screws, otherwise it would not begin to work after evening the belts
wait, so where's the problem, it works?
it didn't until i unlocked mk4 belt