#math-and-meta
1 messages ยท Page 224 of 1
i take a practice drive of the path, then record it on second loop if i don't have some landscaping to do
like just open the record menu and drive the path
i try to spend as little time on it all as i can get away with
like here's the route i have one driving just to pick up a miner's output and get it to a train station:
I mean, after hitting record what do you do? Load at the station and drive to the other?
Where do you start the recording?
i start at one of the stations and just drive the loop after hitting record
afterward, i'll save the route with a meaningful name in case i need to set up another tractor on it
sometimes after i'm done recording, i'll fix up a spinout i had on the route by deleting some path arrows
what would a 5 way splitter look like (a .2/.2/.2/.2/.2) (i dont mean to interrupt)
i really don't use them much
5-way is a 6way split either 2-way then 3-way or 3-way then 2-way, where you loop one of the 6 outputs back into the input line
when you have a situation where you need such a construct, consider overclocking machines to 125% or 250% to change the 5 into a 4 or a 2
well, if you are wanting to consume 20, an OC to 250% changes it to 8
I wonder if anyone uses trucks much
their capacity pre-1.0 was pretty much never needed (like i mean you just didn't need it. Ever.), with miners doing 1200/min, there's a use case for them
I never used them, was either doing long belts or trains
But since they're in the game, I wanna give them a chance
Especially with the nicer fuels
you're better off with belts or trains, but sometimes they can save you a cross-country belt project
Some terrain is just a pain for trains
For remote nodes like that, recording a path for trucks can often be just as much of a pain.
Also I hate building belts on bare ground. They just look so ugly
hah, yeah
in any event, if you use them for anything, i always opt for the tractor until it proves it gets too full per load
recording routes for the trucks is an exercise in patience that even I tend to run out of with them
much better in u8 & 1.0 than they were before u8, but they're still awful
i think personally, it is that way because CSS wants to try to give us 'muricans some common sense ๐
have you seen the DoshDoshington video where he tries to make a Factorio megabase built around cars transporting everything instead of trains and it goes to shit
no, but i've tried myself. have the t-shirt.
In this video, I drive myself crazy
Consider supporting what I do: https://www.patreon.com/Zyllius
0:00 Intro
1:25 Starter Starter Baseโข
6:40 Starter Baseโข
11:41 All Sciences
15:07 LogiBrainโข
19:56 Building the City
29:40 Extreme Measures
31:51 Mostly Legit
36:52 Rocket Launch
39:01 Busy City
Music Used:
Runescape - Pheasant Peasant
Command &...
How many resource nodes were you usign when y'all were tier 3 - 4? i feel like i am "under pace" for somewhat fast production
I got up to tier 9 without building major bases outside my starting area tbh
before you get trains cooking, really it's whatever is nearby in your starting spot
But also I hate building stuff I know will be obsolete which makes it difficult to build optimal stuff before unlocking mk6 belts to actually fully utilize pure nodes
but then production would be slow wouldnt it? did you just like wait a ton of hours?
i've built a crazy long conveyor belt, but instead of transporting the item from A to B and being processed there, it makes more sense to take the component from B to A and process there ๐ฆ can conveyor's switch direction based on what is connected or I would have to lay all that again going the other way?
Yeah I'll set up production and then leave the game on overnight, I don't recommend it but it works
Im afraid you'd have to build the conveyor line agaiun
Like the rate itself is pretty decent, I aim for 400ish/min of most relevant materials, but setting them up takes forever
this is what i have tapped after having completed phase 5 - some of it has been added after i completed the game
there's no need to worry about your miner/belt speeds being obsolete, just build your factories with room to grow up to a 1200 ore input
the main anxiety is if you know you're gonna wanna use more complex recipes later
My strat is to get to tier 9 with the minimum footprint, unlock mk6 belts, and then exploit the entire map
Dang rn im just going for 120/min
Ah but this assumes I can build with foresight instead of vomiting out machines as they fit together
Though i must say i havent explored much
something of a trick in the early going is that the copper and iron alloy alt recipes multiply your ore->ingot very considerably
I tend to rush to mk3 miner and mk5 belt, but that's because this is like my 7th playthrough
I understand the value of the alloy alts but I'm a pure alt truther
Im 150 rotors away from unlocking foundrys i'll see what their up to but, How do i get hard drives?
Power is cheap and increased setup time and complexity is the reason I play this game in the first place
find crashed drop pods
I like the alloy recipes whenever I'm sick and tired of building refineries and dealing with fluids
there's 5-6 in your starting biome if you look around
the main thing is that if you think you're going to upgrade a recipe choice later, keep those machines separate enough to change out
ie, if you're pre-oil doing smelted caterium, and you know you'll want to change to tempered later, keep it modular and leave lots of room to replace the smelters with foundries
tbh by the time I get around to upgrading existing builds instead of making new ones, I'm usually so disgusted by my early-game build that I want to burn it down and start over anyways
if you'tr looking to make 60 pasta/min, there's a place for pure copper. if that isn't your goal, pure copper is a fricking waste. Iron Alloy makes more ingots per iron ore than pure iron, so don't think pure maximizes your screw production capacity.
haha
yeah, I've been there. I've yet to have a playthrough where I looked at my early-game buildings and thought "yep, I nailed it, this is still lookin' great"
Is there a way to find crashed drop pods or i just have to meander
The thing about pure recipes, it's a pain to have a nice stackable blueprint for them.
You need pumps all the time
the quartz research tree lets you search for them with the scanner
the object scanner can find them after unlocking it in the mam, but that's kind of a later game unlock
I'm trying to come up with vertically stackable blueprints to simplify factory layouts, like I can just block off a 3x3 area and know it'll have anywhere from 4 to 40 constructors
But refineries are obnoxious to stack
best idea: climb something high and look around
Yeah the Pure/Alloy alts are when you either only have one type of ore nearby and dont want to do logistics or taking all of the resources nearby for one goal.
or alternatively when the ratios that pure make are too perfect not to use them
I'm unloading the train or unloading the station?
the train
tankiu tankiu tankiu
Unload the train into the station
Like one spot Im using south of the gold coast, by the pure sulphur node.
ALL of the copper is being alloyed, but I dont have any copper I want to pull from rn so im doing the rest of certain iron nodes as pure.
What's the math and numbers for a pure iron node 1200 ore if you wanna do pure ingot recipe?
Or let's say 600 iron ore, cuz for 1200 I can just double the design
a really gross number
Most of my builds are based on 600 ore input
Up on a tower on top of my factory and now i just see how close everything is dang
2,228.5714..../min of iron ingots per 1200 
yuck
its an irrational number too
honestly I literally only use default or alloy for iron (and alloy is mainly for solid steel)
the game has almost never left me in a position of wanting to stretch my iron ore further
That's why I typically don't do pure, except for pure caterium, that one plays out nicely
How can they all be impure
grassy fields
Yeah like, the area im in, used 520 iron ore for compact steel, the rest was divided up into pure... and one smelter 
tempered caterium gives the same ingots per ore as pure, and foundries > refineries
leached also gives the best yield now
eat it, pure caterium. get lost
Pure caterium is still good
Pure is good if you arent diving into more oil burning
petrol coke
the main use for pure caterium now is to avoid cutting into any other resource
but it's more useful than ever to produce a coke supply (coke steel is great too), so I quite like tempered
Yeah it would be great if you had extra when diving into coke production chains.
i meant im feeding 20 constructors by splitting one conveyor in to 5 sets of 4-way splitters, i just dont know how to evenly split up the 5 part
it's actually quite a bit of iron 8x300/min in one spot
oh that may solve my life at some point
I found a caterium node, why cant i mine it?
if you make the nearby coal node into compacted coal, that 2400 iron can be 4800 steel ingots/min (for a butt ton of power)
need to blow up the rock on top with 
oh i dont have that
the research you do in the MAM plays a huge part in making life easier
it unlocks quite a few of the cool toys in the game
I have 1000 compacted coal as by product from rocket fuel production. Currently just sinking it, but I'll need to turn it into steel at some point. That's gonna be a massive tower of foundries 
keep in mind you can also burn it in coal gens
With all that steel my rotor, stator and motor needs will be fully met.
Im using 260/min for steel, 1040/min 
gunna be a NICE chunk for you.
that's endgame hmf-land ๐
I have 45 HMF for now
yeah, i just finished up 90/min... that took i think 3600 ingots
Can a hard drive give me a recipe i cannot make
Wet concrete is one such
If i get iron screws ill be happy
speaking of
i used a reroll
Copper rotor ftw i can now make rotors comfortably
Congrats
Quartz Purification i believe
Pure recipes too.
You can get leached alts before you have the blender to make sulfuric acid
how to get rid of blueprint addiction in satisfactory -chegg -quora
Does this pump pump the water up the tree or does it need to be closer to the base of the tree for the uplift to come into effect?
Embrace it.
It would if it were powered.
awesome, just drafted it up as an example, thank you!
Yeah, head lift doesn't care about horizontal distance, only vertical.
So it'll push it up 20m from where it's at.
no, but pumps show the headlift they cover
Yeah ik
Yeah horizontal distance is infinite. Only vertical above the current point matters.
Could build down 100m and up 115 and the 20m pump will work fine
That's certainly the idea. I'm not entirely convinced it holds entirely true, what with junctions nibbling away at head lift.
They subtract headlift?
Makes me glad rocketfuel is a gas... 72 junctions
something i find useful in building is to run pipes either horizontally or vertically, but not diagonally up so that you can reason easier about where to place pumps
how many generators do you have?
Is there a way to mass manufacture ionized fuel for power production in a way that makes it appealing over rocket fuel? Every chain I can come up with uses way more resources and power for not that much overall power gain, if any. Seems like rocket fuel is better for power generation and ionized is mostly for messing around in the jetpack and vehicles.
@brazen scarab u mean like this
it does provide a boost in power, but costs a lot of power. the point of ionized fuel is that it allows you to generate dark matter residue from power shards for other processes and sink them into fuel gens
You can probably look into dark ion fuel, it doesnt need power shards
dark ion fuel is something else entirely as a recipe for eating DM residue and being made in a converter, it lends itself to slooping
Dark ion fuel eats 2.4 times the rocket fuel. And aluminum. It is by no means cheap either, I would argue it's one of the most poorly balanced recipes in the game.
It's like the developers forgot that packaged fuel is twice as much as unpackaged.
the dark ion fuel is pretty sucky of a recipe until you do this to it:
It does produce more compacted coal per unit of rocket fuel I think?
let me check on that
So maybe it fits some weird steel production meta?
It works in practice all the time for me
But compacted steel is a pain, too. Another questionably balanced recipe. It's pretty much worthless in tier 3/4.
with slooping, DIF, you get back more CC than regular recipe
Im wondering why dark ion fuel cant just be done in a refinery with the fluid rocket fuel
Or why time crystals cant be made in constructors
sloop cost is most likely the reason
Same reason as diamonds, they want to make the power cost higher
idk, dark ion is just a weirdly bad recipe
also, when you look at it, remember it takes 2:1 packaged RF
But why do that unnecessarily?
Then the solution is just to halve the requirement
probably sloop cost
Or ficsite in converters instead of foundries. Funny story, I totally tried the foundry first without looking closely at the recipe for that. And then when it didn't work I built an assembler.
i kind of think it's one of those 'lets add more recipes to the new machine' sort of thing
well, i mean, early game accessible machines vs ones that you need to tech up to build
I think why the ficsite, time crystals and diamonds need special machines is that the processes need conditions that can only be created inside those machines, or can only be done by such machines as they involve atomic/subatomic reconfiguration etc
Ive started in the desert and found 2 Pure Coal nodes, its a good distance away from water but wont be too bad to belt it over. Whats the ratio for water extractors to Pure nodes? havent worked with pure yet so im a little confused.
With mk1 miners, 1 pure node=3 extractors=8 generators
120 coal+360 water in 8 coal generators supplied by 3 water extractors at 100%
8+3 is the magic ratio
So something like this?
also, if you're looking at the pure nodes in the north DD, you might want to consider using normal coal nodes instead of the pure ones
Whys that? they're not suuuper far
I end up doing 4 gens on 2 extractors, literally
This was overclocked to feed from 780/min coal.
pure nodes will end up kicking out a lot of coal with mk3 miners later in the game, you might find that useful to use for stuff other than intro power
from the pipeline manual that is pinned in this channel:
Ah ok that makes sense, dont think there would be enough space in this area for all the water collectors i would need for pure mk3
yeah, usually coal power is used just to get through early game until you get to oil power which jumps your capacity by orders of magnitude
what tool is that?
SCIM in the pins
satisfactory-calculator.com's map view
when i start in DD, i prefer one of these areas (or both) for coal power:
the coal in the upper east corner is a pain to use because of the cliff, but the two spots circled are great
This is what ive got goin for my area. Started on a pure iron as i couldnt find a decent spot after a bit.
1 - Copper, making Cables and Sheets (was wires, changed it).
2 - Iron, Making screws, plates and rods.
3 - Smart plates or whatever there called
4 - Reinforced iron plates
5 - Motor
6 - Modular Frames
7 - 2 stone farms, not good but they work good enough
Definitely wish i chose a better spot, but overall not too bad where im at
I used the south one for temp power 
I was doing Geothermals and remembered I needed more for rocket fuel production
good start, you'll find things start mushrooming in size pretty quickly in phase 2&3
yea i plan on using the trucks a bit more, as last time i played this game i just used LOOOONG belts
its really easy to get doing that in the desert, especially if you build a big flat platform. (the belts
)
my main platform is just a giant concrete slap wedged into the dunes, really like the look of it but i wanna try and make it look nice when i get stable power and production going
i've been camping out in grass fields & blue crater and trying to use all the stuff there, but am slowly creeping my way into other biomes
trying to use up the last of the blue crater oil atm ๐
quick question. diluted package fuel and diluted fuel are 2 separate alternate recipes? i just got the blender
Diluted fuel is better if you have the blender.
The packaged variant is only so you can access the same thing pre-aluminum
i am wondering if they are 2 diferent recipes as in do i need to get a hard drive to have the diluted fuel?
Yep, 2 HDDs required to unlock them separately
So uh Im messing with filling my pipes and changing things around...
looks like I got a slosh loop?
Nothing is using the nitric acid and all I did was add a loop 
Its one section of pipe for machines, the other lines are showing zero flow
It will probably stop, but just mildly hilarious to me.
im not sure how this works...
the water byproduct management
from north we get fresh water
i have a central buffer elevated at 6m always 270/400 full
waste water from aluminium and U processing
wouldnt my byproduct machines need a higher headlift?
It's glorious
i dont think this design works if water extractor and water byproducts are on the same level ๐ค
Anyone know why the pipe segment going left and right showing 0 flow rate? Full 600 Pipe incoming
second segments to left and right showing 201 flow rate same as first going straight through
out of the turbo fuel recipes, which one is worth using and why
It depends
Turbo heavy is borderline useless, and is only worth considering if you don't have diluted fuel.
The other two aren't better or worse, just different.
so, i was planning on going the old fuel to turbo fuel after getting the diluted package fuel, but once i got the blender i got the last turbo fuel recipe
You could look at nitro rocket fuel
If rocket fuel is a consideration, default turbofuel + default rocket fuel are the easiest to balance the compacted coal byproduct of rocket fuel.
i haven't unlock rocket fuel yet. let me look it up
i going to have to pass on that one until i get another hard drive
so, why would one chose the blend turbo fuel over the base turbo fuel recipe then?
i see it out puts more, but what is the challenge to that recipe over the base?
ionized fuel worth it?
Turbo blend is easier, fundamentally it's just sulfur and oil. Base turbo is coal, sulfur, and oil, so you need to route a third resource.
ah.
i really dont think, so consuming power shards just to make extra power doesnt make sense as their is already way more than enough power available using rocket fuel and/or nuclear
default turbo fuel uses coal and more sulfur while turbo blend fuel uses more oil, no coal and less sulfur
ionized fuel in jetpack
make sense. in my case that just mean transport sulfur
so you trade 540 more oil (turbo blend) for 240 less sulfur and 1140 less coal (default turbo)
*for 3000 rocket fuel
alternatively use darkmatter
yeah but the darkmatter is just inferior than the power shard one, but if it is for jetpack it can be good
yet to unlock. just unlocked the blender
personally i dont bother with ionised fuel, i find rocket fuel powerfull enough for the jetpack
I'm gonna make it, maybe a small part of general setup, just so this production exists in my world.
and i am at a loss on how to use satisfactory tools and calculator...
It works, the buffer isnt needed though, tested it myself (although I did reorganise some of the parts)
So you could say the buffer is only useful there because they are on the same level.
Is there any layout to make some kind of priority merger?
idk
fresh water in: 300/s
random water from aluminium procession
random water from U waste procession
wouldnt fresh water hog all the headlift leaving nothing for the byproducts?
I recommend using the search function on the discord, you will see the plight of everyone asking.
if you watched the video, the input has NO headlift because of the pump.
unpowered pumps zero the headlift.
This setup also works if you feed from below, low enough for the headlift to be at or near zero.
So the answer is no ๐
unfortunately, its one of the things people always ask 
well, I guess I can work around it by inverting the approach and using the overflow function of a smart splitter before a merger and some belt speed bottlenecking
huh really? ๐ค
Its mentioned in the video where that image comes from 
There is a mod for it
maybe it didnt click for me the first time watching 
meanwhile the video:
and im thinking: couldnt you just elevate the smaller one?
Test it
Using the blend turbo fuel recipe, the dilute fuel recipe, and the alternate heavy oil recipe to make 585 to 600m of turbo fuel (with a mark 2 miner for sulfer), can someone help me create a layout on satisfactory calculator or tools?
dunno worked for the 2 minutes i ran it
imo it entirely depends on your footprint
honestly i am getting lost using the site and having a hard time proceeding from there
2 fuel = 1 HOR
3 coke = 1 HOR
what is hor?
heavy oil residue
https:// satisfactoryproductionplanner.com/
there are several tools you can use find what works for you
1 blender makes 45 TFuel, 13 blenders make 585 TFuel / min
then you can calculate backwards
This page lists links to external online tools that can help you in playing by providing e.g. map locations, crafting recipes, factory ratios.
thanks. lost using the site
600 TFuel /min means 300 sulfur / min
id recommend satisfactory tools
where are you getting lost?
one how do you change the tier of miners?
what tool is this and thanks it much simpler
in which tool?
Model your Satisfactory builds. Set up your machines however you like, adding optional limits to the number of any of those machines and the tool will calculate how the parts will flow and how many of each machine type are getting used.
this one
no idea about that one ๐คท
in the game or in satisfactorytools?
satisfactory tools
you dont, as thats not really necessary.
Satisfactory Tools doesn't have miner tiers, it just shows you amount of raw ore needed
its up to you to decide how you get the resources, a tool cant build the entire factory for you
satisfactory calculator sorry
same thing in that website
scim shows miners afaik
i mean, its still up to you to decide how u want to get the resources
changing the miner tier in the site doesnt really make a difference
starting from the top, i have 600 crude oil into 20 refinery set up to make 800 heavy oil with resin as a byproduct. the end goal is to maximize the crude oil for the blended turbo fuel recipe. i am lost as to what to do from that. I plan on using the diluted fuel recipe as well
then you use this
converting everything to HOR-equivalent, except sulfur
loop is stuck ๐ค
I can do this easily with Satisfactory Tools... set the limited Oil, activate all recipes, then maximize for Turbofuel... when I want to something more fancy I take the TF amount into a non-maximized calculation sheet
being honest, this make no sense to me. can you break it down?
oh, and deactivate SAM input too unless you want to convert stuff ๐
looks right...
this is how to manually calculate, to be precise
turbo blend fuel takes 15 fuel, 30 HOR, 22.5 coke, 22.5 sulfur per min
sulfur can't be converted to HOR
2 Fuel can be made from 1 HOR
3 Coke can be made from 1 HOR
so after converting, you are left with (15/2) + 30 + (22.5/3) HOR per minutes
45 HOR/minute per blender
like I said, switch to non-maximized before experimenting more... because "maximized" can have interested side-effects outside its primary use-case
you have 800 heavy oil
800 / 45 = ~17.778
you need ~18 blenders
I normally don't use maximize at all but just guess the output and then adjust it if it takes too much input
and calculate how much of the HOR needs to be converted to Fuel / Coke
luckily the ratio is easy, 1/6 each
yes. that is what i have been getting lost on. language rather than number is something i am better at. thank you
you need to process 133.333 HOR/min for fuel, and 133.333 HOR/min for coke
1 diluted fuel takes 50 HOR/min, so 133.333/50 = 2.667
1 coke machine takes 40HOR/min, so 133.333/40 = 3.333
number diagramms are very information dense... some people (like me) like it, others not...
but I think its easier to convert from numbers to "described steps" than the other way around
i get lost in the number and can't keep up with them the deeper i am in
how do i debug this system ๐ค
alu setup?
you can try to break them down one by one... start at the resource input and get an idea "how the hell do I mine/get such amount"... then move to the first machine... then to the next connection between machines...
sure
don't try to force the "big picture" unless you need it
do you absolutely need two water extractors
is it mathed out
Step 1: interact with machines showing yellow light to know why they have a yellow light.
Step 2: fix the reason(s) for the yellow light
Step 3: let the system run and be ready to go back to (1) if you see yellow lights
2 input is easier to understand, 3 or more with various needs becomes a bit much
2nd just emulates this
or would you rather work with spaghetti
how much does that blender produce
15 water
and how much is the first extractor producing
Q unless there is a loop involved you can take inputs one by one
so 15, 15
30
can I also see the UI pics of the first and second refinery
what 30
this is not the first water extractor? the one that is not emulating a blender
well i am going to destroy my factory, and i need a power grind that can take on all that i am able to do with my current phase. plus i want a factory that is always on which is another consideration while upscaling
The smoother the power draw, the higher the efficiency, the easier to spot issues 
first is 65 water
120+15+65=200 exactly the amount sloppy needs
gotcha, it is mathed out
and I assume you have tried emptying the pipe and emptying the water extractors
if it fills up later then idk
VIP junction in the pipe manual might work
and power is the last thing i want to think about. after i get that taken care of, next on the list is my person pain. trains, from building them to how to set up, just the route and logistics of it, while desciding to got modular or not for the factory.
:|
you can drag the water to trashcan
bottle necked again somehow
The pump is unneeded and I would suggest merging the byproduct water with the extracted water AFTER having already merged all the extracted water.
Having byproduct water come into the pipe like that rather than using just one junction feels weird to me.
Tbh, I'd just slap a limited valve on the input pipe coming from the Extractor, unlimited valves on the recycled water inputs and add one buffer for the recycled water. Then I'd leave it forever...
Do you think the desert is the best place to produce Cooling Systems?
which pump got 2
Both ๐
1 must stay
oh right the one after alu didnt get into buffer because of the lift reset
Why?
i dont trust valves
Do your thing then~
They must've made this post for you @vapid gorge 
https://www.reddit.com/r/SatisfactoryGame/s/VvqPWuDtJx
The comments should interest you
What's your rational
i like valves to limit flow instead of doing splitting math or limiting extractor but not as backflow preventer
they suck at that. Valves are mostly not recommended to be used
Tbh, I'm not invested enough to try adapt my solution (which I already explained) to other building preferences ๐คทโโ๏ธ
Better start a new factory than destroy and rebuild the old one
I was just curious what peoples opinions were. There's a nitrogen node out the way in the desert, and plenty of iron/copper/coal
Organisation would probably make fixing your problem a lot easier
There is no wrong answer it just depends on what you have available in your logistics network.
I know, I just wondered if there was a better location than others ๐
In terms of having most, if not all, resources close by.
Thanks!
how can i adjust the lift thing like this i was building 2 or 3 on top LMAO
doesn't seem very math or meta-y
especially when I answered you in #satisfactory already ๐
first click places one end, then just aim at where the other end needs to go before clicking again
oh yeah didnt see it, just tag me like this np
when using a fluid valve, does it stop the fluid amount from going in that direction? I have a pipe of 400 but i need to break it into 3 section. 2 section are 133 and i want the rest to flow... so do i put in 133 for 2 of the valves?
they stop a percentage of flow going through
it depends on pipe fullness, and also the values aren't that granular
so if i tell it to stop 130 out of 160, will the remaining 30 go thro?
can someone help me with this math???
I have 10 lines of 480 sulfur each that feed 4.8 blenders per line. and I have 4800 Fuel which is split into 12 lines of 400.
what's the best way to split up either the sulfur into 12 lines, or the fuel into 10 lines?
I guess how do i split 10 lines of 480 each, into 12 lines of 400?
what do you mean with "adjusting"?
your looking to load balance that?
yea
it will just distribute half/half... but then one of the belt blocks (because it doesn't consume half) and the remainder is redistributed to the other belts... its a self-balancing system as long as you provide enough input
you could divide each input by six into 80/min, and then merge 5 of them together to get 400...
smart splitter 10/12 lines, 80x10 leftover
then, merge 5 leftover lines each to create 400x2
so it will restrict the 130 but allow the remaining 30 to go thro?
what happens to the 30?
do you have a machine that consumes 30?
what do u mean smart splitter 10/12 lines.
feed 480 to a machine with overflow
you can do it 10 times as there's 10 incoming belts
you feed each of the 480 line into a 400 one with a smart-splitter and collect the remainder with an "overflow" belt
ahh okay I got it thanks
what is a typical goal for elevator parts per minute?
well i have a pipe of 400m. i need to break it in to 3 segment which will all be use. 2 of the segment on the same pipe will be broken in to 130 each with the remaining going to the next part.
number of items you need divided by time you are willing
ive haerd people talk about using all path signals before, does anyone know hwo something like that would work?
as long as ALL attached outputs of your cascade of splitters consume together less or equal the amount you put it, it will will out... some machines will get too much, but then block the input belt, which will redistribute the part...
the only scenario where this will NOT work is if you can consume more than you provide, e.g. by having a container (consume as much as the belt delivers) at the end of a line... in this case your need a "smart splitter" (MAM, Caterium research)
all path signals would not work... all path signals except the signal before a train station could work
i am not talking about belt. pipe with valves is what my question are about
pipes have no splitters... you just connect them and as long as no machine is too high it works
forget valves... short story, they don't work as you think... longer story: they are unnecessary
you put 133 for all of the valves
at least that's what The Manual says
but would it be a good idea? it sounds like itd end in disasters. i thought path signals are best for small areas of rail where many rails intersect
so they stop that 133 from flowing in that direction or they stop only up to 133 from flowing in that direction?
they stop when the machine is full
so what I believe would happen if the three branches are (133, 133, unlimited):
I don't hink its a good idea, but I am unsure about the details
if so, the remaining is allowed to flow through?
there is no such thing as "remander liquid goes there"... that is the way to a disaster ๐
yea same. maybe i misremember, but i thought i heard someone say that its actually better than blocks and paths, somehow
the valves somehow "communicate" to each other its flow limit. so it is effectively (133, 133, 600)
then, each valve only allows portion of fluid proportional to its limit: (133/(133+133+600), 133/(133+133+600), 600/(133+133+600)) * (incoming flow)
so, the unlimited pipe ends up taking most of the fluid, and 133 pipes end up starving
I don't know the exact mechanism
but The Manual implies such a mechanism
so what I believe would happen:
train "books" the entire path to the next destination, because Path signal.
Other path signals know what path this train will take (because the entire network is one large path block) and approve or disapprove depending on what block a train is passing
u can think of valves as "pushing back" against the fluid, and if theres another way for it to go it will go there instead, because it cant build up enough "pressure"
at laest thats how i view it
yea thats what i thought, but that would cause tons of issues when you have 6 trains doing the same route with a 10 minute round trip time.
there could only ever be 1 train on the route but i have 2 or 3
idk. if the path crosses a junction then there is a high probability that another train won't be able to leave station until said junction is cleared
yeah
I stay with my advice... if you think you need a valve, you should consider redesigning your factory
well my goal is having the valves block only 2 segement of 133 from going through. what remains can flow to the next segment of pipes
usually my advice is: if you think you need a valve, no you dont. itll just work. if it doesnt, add a pump
im convinced that pumps are actually sacrificial alters wihch can be used to sacrifice energy to the pipe gods
how else do you explain a single pump using 8000000 watts
by spamming pumps, you are dividing up your pipeline manifold and therefore making it simpler
well you see, it also works when you put lots of pumps infront of the manifold
putting in a valve is worse than putting in a pump ^^
because "restrict to x/min" is difficult with a valve, especially if you are NOT overproducing
i used 8 mk2 pumps infront of my manifold to get it to work correctly
going up, to get 370 meters of headlift. not on flat ground ofc
this is my issue
whats the issue?
one pump per pipe should do the same effect than two pumps... unless you need more headlift
forget the total numbers... look at what groups of machines need/produce
exactly. you need more headlift. obviously. 50 meters of headlift could neeeveer be enough to move water up 8 meters. 370 is enough though
then just connect X outputting machines to Y inputting machines (which consume it all)
the calculator just shows you the ratios and amounts, it doesnt tell you which machines to place where
you can split them up however you want
e.g. Look at the "Dilluted Fuel"... you have "2.66" Blenders... so two blenders at 100% and one at 66%... or other similar distributions...
what does one HOR producing machine give you?
i am not going for sphagetti, i am now just trying to find how and where i want pipes and why.
That looks simple enough, split the HOR refineries into two groups of 400.
One group does the Diluted Fuel, Petroleum Coke, and 133.333 of the Turbo Blend.
The other just all goes into Turbo Blend
(because 133+133+133 = ~400)
(133.33.... + 133.333... + 133.333... == 400 exactly)
underclocking to modify the input amounts is incredibly useful at times
if a can take a pipe of 400 and use valve to divide that into 3 segment 2 of which are stopping only 133 from going through i'd like too. one of those 2 will be for fuel, the other for coke, the remainder get mixed in with the other 400 hor going to turbo fuel
pipes don't work this way
they'll stop themselves, by virtue of being full
basically what i am aiming for
then just do it lol
pipes balance themselves out, just like belts
you don't need weird valve balancer attempts
you dont need to limit pipe flow.
you just need a group of consumers (e.g. all of Diluted Fuel, all of Coke and "some" Turbo Blend) that consumes 400/min
try to think about how many Turbo-Blend (some of them over- or underclocked) you need to consume EXACTLY 400/min
don't solve the problem at the pipes, solve it at the consumer and provider
build groups of consumers that fit exactly to their group of providers... then just build a common pipe between them
you also potentially have the option of direct feeding, though I don't know the production vs consumption numbers offhand
pipes aren't belts, but you can apply some similar principles to how you connect things up
I fucking swear I went through all five stages of grief solely because of mk2 pipes
Blended Turbofuel ratings are not easily to "smooth out" with hole numbers
(but there are a lot of cases where they don't work the same)
yeah and i just now realized my plans slightly need to change.
i'll have to much hor for a MK2 pipe if i combine the hor i am not using in the coke and diluted fuel.
good thing we said to use two pipes of 400 then eh
you should NOT combine the output of all HOR production
that a given
then just listen what Meindratheal said... try to think about two groups of 400/min HOR
i just though that i could even out the 533 that was going to the turbo fuel part with what remain of that 400 that i divided into 3
producing exactly 533 from factories that produce 40 each is possible with over and underclocking
you don't need to make it exactly 20 refineries that make HOR... you can make it 18 that run full (100%) and a 4-6 that are underclocked so you get the right distribution without thinking about valves
In the satisfactorytools.com calculator, where it tells you a power requirement of "X (up to Y)", is X the minimum or the average?
you could even build special HOR refinery groups for each desired consumer group...
in short: fluid dynamics in satisfactory is unintuitive and valves are especially so
i think i am starting to understand
make a HOR production group that outputs 133.333/min... that would be normally 4-5 machines (depending on OC/UC)...
then build two of these groups... connect them
then build a group that produces the 533/min... connect that one too
just don't mix too much... this way you prevent the headache on splitting or designing "mixed consumer" groups
actually this will work out
i keep forgeting that i'll have 2 line of 400. that was my issue
you can distribute the stuff in lines anyhow you find useful... 2x400 is just an easy example
e.g. FOUR Blenders for Turbofuel consume exactly HOR at 120/min... as much as THREE Refineries produce.
so you could connect all "full groups of 4" TF Blenders with 3 Refineries...
just find some round numbers that reduce your total problem ^^
2x400 is definitely where I'd first go, but yeah there are many options for how you set this all up
so you can do nice round numbers, nice
at least for HOR ^^
as soon as you reduce the rest of the problem below 600, its easier
Honestly i have to apologize. something clicked which made me realize i wasted your time, but i might not have if realized it if it wasn't for all of you, so thank you very much.
no, you did not waste our time... at least if it clicked now
its often a matter of finding the right perspective... afterwards things get easier
thanks for being respectful. Numbers aren't my thing so i am a bit slower on it than others. also, i took me a while to look at what i already had to understand what i needed. often i get too focused on one thing, because at that time i thought i had a connection.
and Turbo-Blended Fuel is a wild network of interconnected numbers ^^
but at least it doesn't contain a loop... these can be wild
don't show 'em recycled rubber/plastic
without a loop its easy to understand... ๐
yeah sometimes these things just take a moment to click, no worries
i have those recipe. not ready to deal with it yet
if you want to experiment with them, keep the numbers smaller and just build more of the solution later ๐
i have a list of thing to tackle, and that isn't on there at the moment. but when i am ready i will
but having enough power is a good start
I ended up doing pure iron into iron pipes
Itโs pretty nice. While not perfect 1:1, when I clock the refineries just one time it puts out 97.5 per refinery
Which is close enough to the 100 each on the constructors
The steel rotors idea ended up being pretty great
Glad to hear it
I ended up giving my copper supply a break via using a few alt recipes in my supercomputer line
machine consumption restricts the flow better than valves?
they say valve are useless
because they are.
in my case my number adds up without valves
pretty much
there's some wild and potential set ups that do some odd things that have valves involved
they are however, much less reliable than the non valve methods
so it's better to just go 'don't use valves' unless you're specifically into finding those weird set ups and using them
do machine cycle times get reset on load?
curious
my machines were stalling, i thought it might be because cycle times get reset on load
I'm pretty damn sure they don't, otherwise you'd notice all your machines in clumps get syncronised and I've never seen that
is there a way to flip an entire belt line? i did an oopsie
I'm currently making 1600 fuel and feeding 80 gens with it. I want to upgrade it to turbo or rocket, but without redoing the setup (i.e. I want to use the fuel as an input, not HOR). If I use the nitro rocket fuel alt, I can do that but end up with Compacted coal as a byproduct. Does it make sense to use that plus a portion of the fuel to make turbofuel, which then goes through the vanilla recipe to make rocket fuel (feeding back a bit more CC), or should I just keep it simple and ship off the CC to use for steel or something?
teh alt is definitely simpler for your purposes. But I'd just go nuclear so you don't have to deal with awful gas in mass manifolds
what do you mean by the last part? just dealing with lots of pipes?
gas misbehaves a lot, and you'll need to build LOTS of gens
you can manage it but it can be a real pita
especially if you're someone who wants to build fuel towers
like the gas mechanic causes issues in pipes?
besides, they increased fuel and turbo fuel consumption in the gens specifically to help people avoid building hundreds of fuel gens before going nuclear
depending on how you build it yes. It's very easy to get issues if you don't plan it out with informed design choices
What's the main gotcha there? I have a decent feel for the limitations of liquids
He probably means liquids and not gases
ehh.... a lot of people build the towers I mentioned, which... bad idea
both. Gasses are kinda worse as they have no grav priority
this was never an issue pre 1.0 because you just didn't use nitrogen like that
so if you insist on doing the RF, keep it in smaller manifolds, under the limit by a good margin, if you have multiple floors? dedicated pipe for each floor. Best chances of it working first time
Main motivations were that I hadn't done anything with rocket/turbo fuel before (I built nukes my last save), and I designed the fuel burning plant to be able to splice in a turbofuel plant between the fuel production and gens
I'm getting close to the point where power will be an issue again, so time to expand. Then time for nukes
fair enough! just keep the two main points in mind. Honestly those are good points for liquid pipes too but maybe even more important for gas
thanks, wasn't planning on fuel towers, just a flat build with multiple pipes
I got my fuel to fuel gen system running perfectly so at least that part is ok
cool! gl and gn!
sounds like if I go rocket, it would be best to just ship off the CC for other use or sink it if i'm lazy'
why the hell is pipe meter all the way in the beginning of the pipe and not on the middle?
I bet if you target that for deconstruction that's two pipes.
same for others - it should be enough space for pipe meter but it's not here
If there's anything in the way when the pipe generates, it won't generate the indicator.
This includes things like power cables.
just unlocked the mk3 miner, what a beast
I did similar setup but for RF. Each blender outputs 150 RF uses a mk1 pipe. 14 oil gens at 250% and 1 around 98.8 (idr exactly). I did the loop back method for the pipe and everything worked out great
I did use the alt recipe for RF cause itโs way better
What do you mean by oil gen?
Oil generator
Fuel, probably.
Oh ya fuel my bad
I was looking at it more and I may need to use the loopback to make some turbo as otherwise it's going to take 1600 sulfur and I only have mk5 belts, so that's at least 3 nodes to tap
or I leave half of it on fuel. Going from fuel to rocket is a huge power boost anyway
Is it worth making Ficsit ingots from iron or is it a waste of SAM?
saving SAM is generally the best approach unless you're using very very little of it
I also actually tend to find Aluminum Ficsite most convenient, because I tend to produce aluminum in high volumes before allocating it (due to the finicky production process and all)
so I actually end up having a big supply ready to convert
i just calculated how much i need for my fuel power plant:encased beam:17,940
motor:3,900
sheet:26,520
rubber:13,000
quickwire:13,000
cable:3,900
concrete:123,890!!!!
quartz:15,600
oscillator:4,030
iron plate:10,400
wire:79,820
rod:5,200
steel beam:12,090
reinplates:2,080
plastic:5,720
mod frames7,800
Have fun lol
thats uhh, modded, right?
absolutly NOT
oh wow i really underestimated how much of these things you can do in vanilla
Heeheehoohoo 240 fuel generators go
the reason its so expensive is because the blueprint im using has batteries and lighting (W signs) built in
OOOH
i thought per minute lmao
wow i should really get some sleep
yea that makes wayy more sense
wait you thought i was saying 123 890 concrete/min?
is it like 3 am there?
yea the max is 90450/min
the motors are totally doable tho
and the rubber and plastic too. i thought maybe you included parts needed for other parts further down the line as well
in that case i think itd be possible besides the concrete(?)
well the beams and wire might be difficult? im not sure
fun fact, you can make 27632.9 motors per minute technically
the oscilators also might not be possible
i didnt even look at the oscillators yet
screws are even more insane imo, you can make over 5 million per minute
the max screw bulid uses up basically all the resources
besides nitrogen, uranium and the tiniest bit of caterium
HOLY SHIT
3430 oscillators/min is max
eXACTLY. when im done with this world ill do a max screw run i think
itll be difficult to figure out how much to sacrifice to power, it needs like 500gw and it uses basically all the resources
it mostly uses aluminum to steel beam alt and steelbeam to screw alt
it uses up all the concrete to make aluminum
sulfur gets turned to compacted coal for steel
crude oil to make more concrete and petro coke
caterium gets converted to bauxite and quartz to silica for aluminum
Man the blueprint UI gets so confusing sometimes
I just wasted an hour of work because I assumed opening up the window to pick what subdirectory to save it into actually saved the blueprint, because you click a big plus sign and get a message saying "Blueprint Added!" with an exclamation mark and all
But then if you don't close that window and click the Save button on the main window that doesn't mean anything
By the time I noticed it never actually saved I was already far enough into the next design that the autosaves weren't helpful
Yeah they really need to fix it. I've lost stuff like that too.
I sat down to restart the blueprint design, paused for 5 seconds, quit to desktop, and bought + installed the Factorio expansion
Instrumental convergence is the hypothetical tendency for most sufficiently intelligent, goal directed beings (human and non-human) to pursue similar sub-goals, even if their ultimate goals are quite different. More precisely, agents (beings with agency) may pursue instrumental goalsโgoals which are made in pursuit of some particular end, but ar...
Since steel is more efficient for the screw making process than iron, and steel can be made with coal, and you can make coal out of biomass, technically it is also optimal to convert all the wildlife into screws
How many extra screws could you get if you add in some extra source of coal to represent you running around like a madman throwing all the flora and fauna into a grinder?
I see you're also a fan of Robert miles?
no actually, who's that?
he does videos on ai safety
Twchnically it's not the most screws since you also gotta put sloops into max-per-cycle screw constructors, and maybe also add biocoal and charcoal on top of that
it's really cool
Also this has the issue where it uses up a bunch of sloops that im sure go elsewhere in this supply chain
I know this is the wrong channel, but I cannot post pics in the main. Is there a known bug that trucks just stop driving? As you can see in the shot, the map says it's driving, andthere is no exclamation mark in the compass. But the thing is just standing there.
Wait I just did the math and a maxed out + slooped constructor running the Steel Screw alt singlehandedly backs up a mk6 belt
260/min x 2.5 x 2=1300
he also did stuff in instrumental convergence
tho I don't really see how that's relevant to the calculator thingy
Ohh okay I see the relevance
You've made the paperclip maximizer lol
I linked to a subsection of the page
"That's a logistical challenge for a player to solve." - Snutt in one of the videos, sorta
The challenge for the player to solve is the soul-searching to find out why they would even try that
they're kind of buggy and sometimes just go off the rails (pun intended), you can either hop in and disable/renable the autopilot and mess w/ it a bit, getting the truck in a better spot or realize like the rest of us that the truck just isn't very good
Don't ask "why do that", ask "how do that" 
They should add an easter egg where if you do that the extra screws get flung out off to the side of the conveyor as grabbable items
Also make them physics enabled so it crashes the game in seconds
unless you really need the truck's capacity (which is like nearly never), you're better off using its more fuel efficient and manueverable little brother in the tractor
A punishment for being the most efficient screw producer, wow
Make it output nuclear pasta lol
Overloading a mk6 belt off one constructor is so funny to me
If you need the truck's capacity you actually need a train
Yeah, I realized that a long time ago, but in this playthrough I just wanted to use trucks for the fun of it and the aesthethics. It's something I haven't done in my other savegames. :/
ehhh, maybe if your moving a miners 1200/min to a train
if you need capacity, you need a train. im a trainlover and truckhater
America playthrough where you fully exploit every oil node and use only cars
aw, im doing that but trains
in the past the usa was really good with trains, maybe im just past usa
all vehicles suffer the same derpage, its just the truck is more pronounced with it all
trains are derpfree, thats why i love them
not quite
I'm in Boston, our trains catch fire sometimes
i've caught them doing some funny things, but much less hassle overall
im curious
i mean like, wayyy back when.
Only trucks*. With tractors it will probably be Russia or Belarus
when you load a save sometimes they're in a spot where they can't renegotiate signalling properly
O.o
im gonna become one of those RETVRN twitter guys but for a comprehensive rail network across the US
whats a RETVRN
maybe it is fixed now, but the signal state wasn't saved in your save file so it could end up being bizarre when you reload
there are fascist dudes on twitter who talk about how great everything used to be in the past (i.e. when they could be bigots more openly) and say we should "return" to those times, except they stylize it as "RETVRN" to make it look more Classical Roman or whatever
that's one of those things that doesn't really start becoming a manifested issue until you have A LOT of trains
we should RETVRN to the days when twitter didnt exist ๐
im going to have tons of them i think, i have 6 for just one train station
well, two, but ykwim
it's called X now, if you didn't get the memo ๐
elon deadnames his kid, im gonna deadname his website
why do we discuss twatter in satisfactory chat
"and-meta" covers a lot
because we're all silly in the head and convos easily drift off course
unlike trains, because they're on rails. ๐
i miss the days when the world wasn't on the internet
i hope my tinybouts can service all the trains ill have if my plan comes to fruition
tinybout = tiny roundabout?
i think the fact that they're tiny should help with that
yes
they're smaller than the 3x3 turn and use some bugs to work correctly
maybe, a lot of stuff with train signalling, just like concurrent programming is completely opposite of what you'd intuite
well, not the whole roundabout is smaller obvsly, the turns are smaller
isn't a roundabout just one big turn?
or do you mean the entrances and exits from it
like the circle is the same size but the routing into and out of it is more compact? im intrigued
a roundabout is just a N-way intersection with the special property that it can only let a single train pass through, even if there is no rail contention between trains
also @edgy leaf have you played Factorio? their trains are really robust but do some truly cursed stuff
no i couldnt live without 3d
They have a demo if you want to try it to see
usually you cant place 4 rails on the same connection but with the H glitch you can, its what i do to build the roundabout
what why did my brain read orange
lmfao
no i just plopped the blueprint down ontop of the other rails to demonstrate
Ohh okay lol
wait ill show it in use
It looks like it should make sense but imagining trains trying to bend that hard is funny
its divided into 4 sections. the two colors on that one side are similar but different
caught a train driving through!
you could divide it into 8 sections to get it even more efficient, but i think the signals look a bit silly..
with blocks so small i think blocks are actually more efficeint than paths, maybe
i ended up going with a roundabout because normal intersections are really annoying with vertical rail
also i think the smallabouts are smaller than normal intersections
roundabouts are crap in terms of throughput though
i'm not 100% sure of that assertion if you can break the circle up into blocks that allow multiple trains to pass
best case scenario you have same throughput, but mostly not
classic X junction allows two left-turning trains at the same time, roundabout doesn't
@edgy leaf btw, i was fussing with nuclear numbers and it looks like about the max that can be made is around 3tw with full ficsonium recycling. it's kind of hard to pinpoint the exact number, but it looks like somewhere btw 2.5-3tw you run out of just about everything
without sloops?
hm
i get 2.54 with not much utilization doing matrices and base uranium
my staistics 101 knowledgei s a bit old so maybe im wrong, but wouldnt linear regression be good for this?
yeah, but that isn't as cool
i'm having to do a lot of back tracking through the calculations because all the tools still don't have a 'use sloop' button
Hey just want to make sure I'm not crazy because I'm thinking about this at work. Working on a power plant and I can't implement this until I get home.
I have three 600mยณ oil pipes, two 300mยณ, and one 150mยณ for a total of six pipes. I want to split them evenly into six pipes of 425mยณ but I have found myself frustrated with understanding fluid mechanics. From what I understand though, what I've drawn out below should work right? I am trying to avoid valves if I can (other than using them to prevent backflow). Sorry for the extremely rudimentary drawing lol
The system will even out if you have no bottlenecks where it must go over 600 to equalize
unless you build vertically ๐
cause gravity
if you have higher up components they get their liquid later
I don't think I have a bottleneck. After this balancer there would be a vertical climb but there would also be pumps.
then you just prime the system by letting it fill before the consumption is turned on, and if each branch does not consume more than 425 it should work
or you use valves and limit each branch to 425
At the top of Turbo Towerโข there are six rows of 13.75 heavy oil residue machines. Technically each row only needs 412.5mยณ per minute but there shouldn't be a problem over filling it right?
My limiting factor in this build is sulfur and I could reduce one of the extractors by 75mยณ but other than saving power there's no compelling reason to do that right?
so I used satisfactorytools to calculate my factory I want to make but how do you guys actualkly make these kind of numbers? Do you juist take the closest number you can get or am I missing some setting which gives these crazy .xx digits
Did you hit maximize by chance?
Focus on the numbers of buildings. So the 3.232x Computer manufacturers, I would just build 4 manufacturers at 80.8% (3.232/4). Then they woudl consume the right amount per minute.
But actually, I just round up. ๐คฃ
I also don't use the numbers from that tool.
I always do the math by hand, working backwards from the last machine, and only use the tools to see which recipes I want to use.
I did cause I wanted to give the tool my input and then let it calculate what my maximum output would be
this would be a Versatile framework, modular engine, adapte control unit (guess which stage of the space elevator im at xD )
Oh. Then take that number, and put it back in. Maximize doesn't optimize anything.
Also never use maximize with multiple products.
It maxmimizes the first product, then sets the rest of them to the same amount IIRC.
aha
yea this is the one I made
Yeah, so you know you can do 4.04 of each. Pick the actual number you want to make of your final products and switch to that.
aha
yea the limits I have is more in the items, input tab
where I gave it X coal, iron ore etc
Yeah, if you switch it to 4/min for each it's much saner numbers.
Still get some adventerous ones though.
haha yea but thats fine I mean I dont need to be super optimized
I jsut dont want those weird split numbers that are just almost impossible to get to without 60 splitters xD
!wikisearch manifold
You don't need to worry about the splitters. As long as the total on a belt is less than its capacity, it'll balance itself out.
alright, thanks for the input both ๐ I will try this out
And for this one, I would do 5 assemblers at 99.05%, and 2 assemblers at 99.04%. ๐ If I cared.
hahaha
Just remember, clock speeds are only accurate to 4 decimal places. XX.XXXX% is the most precise you can get.
So 88.45673% gets rounded to 88.4567%.
Easier to do it evenly IMO.
And valve settings only have 254 distinct values and will round down ๐
Definitely relevant. ๐
Set the first one to 97.8412%, then copy paste to all 17 foundries. Saves you a bit of power, too.
Yea I also read something that pipes technically also only flow like 597 instead of 600 or something along the lines of that?
I didn't say it was relevant I just thought it was funny and never get a chance to mention it otherwise, doesn't detract from your point
That was a bug prior to 1.0 that should be resolved I believe
ah ok ๐
Yeah. "bug". "resolved". ๐คฃ
Totally not "intended behavior" and "skill issue". Nope. Not at all.
doing 600 is fiddly, you need special techniques bc
It can be fiddly.
Sometimes it works super great, actually. ๐
Sometimes it requires a substantial amount of goat sacrifice.
it is fiddly because the fluid dynamics in the game is "realistic-when-it-wants-to"
See, I don't try to explain why it is the way it is. Because I can't access the code, and so far I've yet to be able to actually make any provable statements to cause.
Every explanation I've come up with has some edge case that is absolute nonsense.
I want to look at the code just to figure out why they could only allocate a single byte to valve precision
I mean, that's a choice, not a limitation.
It's a bizarre choice for an application where you definitely would expect more than 256 discrete values
One of many such things in Fluids, I promise.
Using one more byte for a 16-bit float would let them get step sizes of 0.009 instead of 2.3622
Right. They clearly don't want to give people that level of precision.
Well if they didn't want it they'd make the UI make that a bit clearer I think? Also the fact that the step sizes are seemingly random increments determined by the data format and not, like, 2.5 or 5, makes me think this is a technical limitation
It has to have something to do with how it scales when you have hundreds of valves in the world, but I'm at a loss how exactly
I mean, I've so far only found one actually valid use for valves at all, and the amount of precision they need for that application is, like, "within 15 or 20 of the target". ๐
Also, don't forget, there's a reason there aren't good ways to artificially constrain belts the same way.
They've very intentionally designed the limitations of these things.
I don't disagree that they should instead use like increments of 2.5 or 5, you're absolutely right, but that's not a technical limitation of any kind I'm aware of.
I get that, it's just extremely weird from a game design perspective to give players a tool like the valve, tell them it works a certain way, and then design it to not actually work that way
It does work that way.
The "technical limitation" is the fact that it's 8 bits instead of 16
I mean, if you drag the slider instead of typing it in, it can only be set to a valid value.
That's bad UI design then
If they wanted to be consistent it should round to the actual value it's using when you type in a target and hit enter
I agree with that, too.
Otherwise you get reddit threads like this
It also isn't "severely messing up" their fuel factory. ๐คฃ
That's sort of the other side of the coin: valves just aren't useful in most scenarios people think they're useful in.
And my main gripe here is just that it's downright bizarre to only allocate a single byte to a variable nominally allowed to be between 0 and 600 unless there are HUGE performance issues I'm missing somewhere
That creates issues even without decimals! You can't even do whole numbers!
You're gonna really love this. It's a double precision float under the hood. ๐คฃ
(my main gripe is that players has to refer to a user-made document to kinda sorta understand the pipes)
/** Set the limited flow through this pump. Set this to -1 to use the max limit, i.e. valve is fully opened. [m3/s] */
UFUNCTION( BlueprintCallable, Category = "FactoryGame|Pipes|Pump" )
void SetUserFlowLimit( float rate );
/** Get the limited flow through this pump. -1 if max is used, i.e. valve is fully opened. [m3/s] */
UFUNCTION( BlueprintPure, Category = "FactoryGame|Pipes|Pump" )
float GetUserFlowLimit() const { return mUserFlowLimit; }
/** Get the limited flow through this pump. [m3/s] */
UFUNCTION( BlueprintPure, Category = "FactoryGame|Pipes|Pump" )
float GetFlowLimit() const;
/** Get the limited flow through this pump in percent of the max. [0,1] */
UFUNCTION( BlueprintPure, Category = "FactoryGame|Pipes|Pump" )
float GetFlowLimitPct() const;
/** Get the set maximum flow rate through this pump. [m3/s] */
UFUNCTION( BlueprintPure, Category = "FactoryGame|Pipes|Pump" )
float GetDefaultFlowLimit() const { return mDefaultFlowLimit; }
/** Get the replicated flow rate through this pump from 0 to MAX in the range. [0,1] */
UFUNCTION( BlueprintPure, Category = "FactoryGame|Pipes|Pump" )
float GetIndicatorFlowPct() const;
/** Get the replicated flow rate through this pump. [m^3/s] */
UFUNCTION( BlueprintPure, Category = "FactoryGame|Pipes|Pump" )
float GetIndicatorFlow() const;
So the variable we can set is one byte, but it's 8 bytes under the hood
how to merge vertically (yes it clips in some spots, but hardly noticeable). This one can merge up to 18 belts. xD
why? cuz I'm tired of pulling a heckload of belts from my low-output manufacturers to my storage.
Although that code might just be for pumps. Who knows.
It looks valve related, not sure what else "UserFlowLimit" could be referring to
from front \o/
if it is valve related, that would suggest valves implement AFGBuildablePipelinePump which is a whole fun thing by itself. I'm gonna go see if I can validate that.
Or rather, those functions are for pumps, but the Getter and Setter for UserFlowLimit look like they involve a valve
There's no headers for Valves, otherwise.
My guess here is that every pipe has a variable tracking its flow limit, which is set to -1 (unlimited) by default. Adding a valve just splits the pipe at that point and creates a new pipe segment after the valve with the flow limit variable set to whatever
And, in fact, that block of code contains the only two references to "valve" in the entire set of headers we have.
I'd be interested in seeing where in the code it accounts for flow directionality with valves
Because they would need to be referenced/checked there too, no?
It would probably be somewhere in functions related to passing fluids from one pipe segment to another
PipeJunction has a directional concept.
It has both a previous and current, and a positive/negative flow.
is it better for through put to have block signal at beginning and end of each train station or just beginning
Wild. Valves literally are pumps. Wow.
the train station should be a single block
but wouldn't affect throughput if two trains catch up to each other
in an open system two trains doing the same loop will always end up tailgating each other no matter what
law of entropy ๐คทโโ๏ธ
that doesn't make sense to me tho, like they both spend time stopping at the same stations right?
like obviously you're right but i don't get it
You can force them to wait at a station and they won't tailgate.
yeahh but traffic always exists and will manage to fuck it up
That, in fact, is the way to achieve optimal throughput.
that is if the loop exists in a closed system
It.. doesn't necessarily?
Is PipeJunction referring to just the union of two pipe segments, the +-shaped placeable Junction object, or both?
So valves are pumps that cannot pump? 
pump cosplay
wait so if a train is full, it just goes skips the station I see
eh no
So they're just implemented as pumps that don't create headlift? Is that why they reset headlift (or used to, idk)?
it will still do the load/unload cycle
That's my guess.
except absolutely nothing will get loaded
That's hilarious
Wait maybe that's why it's 1 byte
Pumps don't need more than 1 byte for flow limit because the player can't set it
I like how you just refuse to accept the answer of "they intentionally limited it that way".
But then they expanded that class into valves and suddenly the players can set it but you don't have the dev time to unfuck everything that would happen if you made it 2 bytes so you fudge the UI
Yes, I do refuse that answer
Releasing a mechanic even though you don't actually want it used and making it broken/unintuitive in subtle ways to discourage its use is awful game design on several levels, they aren't that dumb
So, should I use valves or no?
Still no.
What makes you think they won't want it used?
And what makes you think it's broken? Honestly, as far as I can tell, valves do exactly what valves are supposed to do.
It's just that most people expect them to.. do other things?
no
if you need to use valves your system is designed like shit
People expect them to let through 60m^3/min when they type in 60 and hit Enter, that is not at all unreasonable or their fault
Or dealing with recycled fluid from a well head.
Okay? You do realize that's not actually the same as "this thing has a specific level of precision"?
CSS, I love them, but UX has never been their strong suit.
Honestly, the bigger crime is why I can have a system that should be doing 600 in Mk2 pipes and just doesn't, and doesn't give me any indication why whatsoever, aside from some pipes not able to stay at 600.
Like, you're picking nits over something that's never off by more than 2, when the UX crimes they've committed are far, far, far worse than that.
It is the same thing when that level of precision is 1. not conveyed anywhere (no, the slider sticking at weird-looking decimal places does not count) and 2. is LESS THAN HALF of what you'd expect for something ranging from 0 to 600
I mean, even if you double the precision, it still doesn't land on many, if any, round numbers.
Being off by any amount will eventually cause stuff to run dry if it rounds down and you did your math assuming it, you know, works as advertised
ive NEVER had the MK2 pipe bug
Good for you! ๐
it hits 600 no problem
You don't double the precision. You double the size of the variable from one byte to two, which increases the precision by 256x. You go from step sizes of 2.3622 to 0.009.
Yay!
When I heard about valves being a byre limit I'm like wtf this is 2024
Yes this is exactly why I'm still harping on it lol it's so bizarre
Was it that way in early access?
The only variable that small in a modern game should be, like, a boolean /s
Or enums if for whatever reasons they don't want to use actual enums and even then
Oh my god imagine it's an enum table with 256 distinct values
Stop
๐ญ
Funny enough. ๐ Even at 16-bits of precision.
the loading bug would be enough to cancel that out i think (?)
I mean, valves don't need that much precision to begin with.
how hard would it be to represent values as fractions instead of floats?
I would like it if their step size was smaller than 2.3622 or if that was conveyed to the player at all
yea it should at least be a slider that can only move in 2.3622 increments
having it be smooth is super misleading
Floats exist specifically because they're the best way for a computer to represent fractions unfortunately
It can.
The slider won't let you pick invalid numbers.
The UI fail is that typing the number and hitting enter doesnt show the rounding it does behind the scenes
best general purpose use case for a computer to represent decimal numbers.
That's just... not true. ๐คฃ
the real world device it represents is analog, so it looks really jank in game to have float size limitations appear to the player
It would have to be used in a calculation that produces a double cause I'm sure the flow rate is that
Closer, with "general purpose". But Decimals are better in a lot of general purpose usecases.
im not talking in the valve, i mean for machine cycle times
It's already a double.
๐
The valve flow rate?
using -1 as a special number
the rust user in my is crying
According to this header, the flow rate limit (that has 1 byte of resoltion) gets cast to an 8-byte double lol
Treated as, not cast
void SetUserFlowLimit( float rate );
Funny how your decision to refuse to accept that it might be intentional makes everything else look wrong to you.
I can't find the byte declaration
but the getter function just returns a generic "const"
Char, I guess
Looks like it returns a float to me?
Oh duh I missed the part at the beginning. That makes this even more baffling.
No, it doesn't.
Speaking of precision, has anyone figured out where to clock machines for theoretical best precision? Even product number? Even production time? Or is it based off some internal tick figure that's not visually accessible?
Clock speed is the only number that's always the right precision.
So that's the number you want to do math off.
Well, I guess cycle time probably too. ๐
I've always figured production time is what everything hinges off of, huh
Yes I literally refuse to accept that they intentionally set the step size for valves to 2.3622 for game balance reasons if a smaller resolution would have worked
@slender cradle what do you think?
It's sort of both? You divide the cycle time by the clock speed and you get the new time?
So the conclusion is that a smaller step size either doesn't work for some reason, or there's some technical debt that's not worth dealing with
does the valve flow rate thing have more than 8 bits of precision? idk that language
Meanwhile: they're passing around doubles internally.
Yeah that's too braindead an engineering decision even if it was somehow for a balance decision. They're better programmers than that
they just want to troll us i think
It has to be unintentional
to get us to sacrifice moar goats
I can't see how it'd even exist if it was unintentional.
It's not easy to accidentally get 8-bits of integer precision out of a double. ๐
Software bugs can be f***ed up
yea but ive never heard of it still working but only with 8 bits
Not in my experience
That... seems surprisingly uncomplicated
My theory is that since valves are coded internally as pumps, that means pumps always had a flow limit variable, but it was just 1 byte and that worked okay since the user couldn't directly set it. But then they expanded the Pump class to include Valves, and then that variable became user-editable, and there were too many functions built assuming the variable was only 1 byte to be worth rewriting.
i mean, power shards are just time crystals and dark matter
You can't sink power shards. That adds some complexity back.
The real question is.. why can't we restrict flow in pumps. ๐ฆ
wait wait wait wiat, valves are internally coded as pumps?
having some personal ionized fuel production for the jetpack whihc u can sink is how i plan to deal with it
Or they forgot it was a char
Yes.
i think many recipes are dark matter positive (?) so you wouldnt even need the sam. just coal and quartz
How so? This entire thing is self contained.
What if it was some jank-ass solution like "0 if unpowered, 1 if mk1 pump, 2 if mk2 pump"
thats. interesting.
McGalleon doesn't seem to think so in the other server. ๐คฃ
yea
It could be done like that, But I wanted to avoid having to figure that ratio out
I believe that
I mean. We have the header files.
if you use dark matter trap it can also be dark matter positive
When were valves added?
0.3.6.
tug is deep in the lore
I'd be interested in the header files right before that version and right after
They were added alongside Mk2 pumps.
i couldnt just pull out a version number like that
and then what do I do with it?
I had that question about 2 minutes ago. ๐คฃ
use clocking and two recipes so its dark matter neutral
or use the dark matter for other things that need it
i love making factories depend on eachother, i know its technically not "good" but it makes everything feel so connected and alive
Yeah, the only good way to deal with DMR is to balance it in and out.
You can't really overflow it.
That and I am unsure if I'll run low on SAM in my world
DMR is horrible, you cant package, train, or VIP it so its literally impossible to VIP or VOP
Once you're running, you don't need any SAM for DMR.
same with EPM
If you move EPM, I'm going to come find you, and I'm going to fix you.
maybe i want an EPM train, who knows. its easier than moving electricity
I fully expect Josh to build a full map EPM train network.
I try to keep it reasonable. I have two next to one another. A makes Quickwire for B as a byproduct, and B coke for A as a byproduct again
Is it, though?
I am hella proud of that
not possible sadly! epm and DMR cant go into trains
Since rails literally move electricity? ๐
shhhh dont make my perfect plan look bad
