#math-and-meta
1 messages · Page 583 of 1
Muhahaha
How much 16 coal generators consumes coal?
All 16 of them will.
How much per min?
15 each.
So 16 x 15.
isn't there a chart somewhere on the wiki that shows the consumption rate for every fuel type? been looking for a bit now and can't find the thing
This the table you're looking for? https://satisfactory.fandom.com/wiki/Power#Resource_consumption
Is this your Factory? If so, this is amazing and I might just follow your design here...
consumption rate depends on clock speed and power production of a gen, it's better to use energy values to compare fuels
Would this be the correct place to share a splitter calculator that I made?
Yes
Sure, might have to ask mods for link whitelist though.
Well, may as well try:
A splitter calculator. It's goal is to either evenly split a line into N equal out lines or multiple out lines following a ratio.
https://github.com/IceMoonMagic/Satisfactory-Splitter-Calculator
I made it for myself and felt like sharing.
It's actually good
Well, it's good that someone besides me thinks that 😅
would be nice if I hadn't to install tons of stuff to run it 🤔 😄
Well, Python should be simple, but I don't really know much about installing Graphviz as it was apparently already installed with some other package on my system...
Um, oh, no idea how to use python.
yeah, though having it as a web app would probably be way cooler and more useful for non-tech-savvy people 😛
That's the plan for it yea, I'm making 2 because I like pain, one to get a container of all the other items, one for the stars. Gotta upgrade my power first though. Probably get all that done this weekend
I would make it a web app if I had any experience whatsoever in them, but if someone who does wants to make it one that'd be pretty neat
one of two reasons why I didn't include balancer planner on my site is that I don't think they are useful at all in satisfactory. They may be useful in some special cases, but pretty much always there's a (subjectively) better way to go around that.
(second reason being that I was lazy and didn't want to research how to make the code that would generate it in optimal way)
I wonder tho, the numbers are in weird places, is that some kind of bug or?
Looks like some kind of electric logic graph to me.
From what I've read online, that's related to line engine being dumb with labels
the rhombuses are splitters and the squares are mergers
the warnings suggest using xlabels instead, but when I tried that, they didn't show up at all, so....
btw, do I see correctly that you try to split using a full belt?
I mean try the possibility in the code
Yeah, the idea is using a smart splitter set to send all down that one belt, then overflow the rest.
the result being a fully saturate belt and a belt with less items on it. The items siphoned off can then be re-added to an output, giving more possibilities for simpler setups
I see. Just so that you're aware - that doesn't always end with the exact ratio you're looking for
people reportedly pushed +1-2 items/min on mk1/mk2 belts
anyway, if the tool would be able to output data (e.g. json format) instead of image, that would make it a bit more useful imo
satisfactory tools is not making the math very easy here NGL lol
(at least for me playing with it 😄 )
I really want to resolve this and even have idea on how to do it, just wasn't able to make it work 😦
That's probably due to the max throughput bug when trying to use said MK1/mk2 belt for more than 1 segment 
I think I've vaguely aware of that, being from the splitter's inventory filling up before overflowing, then when the flow stops it can fit onto the other belt. (Or at least that's what I think it is...)
I had a thought about maybe splitting water to each entry, but then, the player might want to combine some of those.
what I wanted to do for my website (but probably won't unless there's enough requests for it) is "balancer analyser", you make a balancer and it tells you how it splits resources
Id rather water as an output of Aluminum scrap not be used as the input for any recipes, even sloppy alumina
So, splitting might actually put you back on square one.
And instead just be listed as an output
It probably works best as a loop between each other?
It also outputs a .gv file that is the raw graph. I'm pretty sure there if you find another Graphviz engine, you can just shove that in.
In terms of saving the output otherwise, I currently have an option to dump the nodes into a YAML which is a superset of JSON
Well yeah but to keep the UI readable, Id prefer the arrows be "forward only" and I can figure that simple loop back on my own
does the planned Item overview that'll be out probably in 10 hours help you with this?
the loop back mostly just clogs it up visually and makes it harder to read, and then also introduces these bugs
rofl, greeny flexin on suggestions now
Oh interesting, thats pretty cool
What bugs?
if you just care about the number, then in the main row at the top there's total production and total consumption, so you can just take the total production and produce that much (and burn the extra byproduct water somewhere else)
Well less bug and more that it connects water outputs to inputs of whatever it can, so you get weird stuff like it suggesting this
If you mean splitting off to other stuff? Yea, I had it split off to residual rubber for some reason
(note the water output of aluminum scrap is being suggested to pipe into the residual rubber recipe)
Heh, it did the exact same thing for me.
yeah that's just because it goes through list of recipes and connects inputs to outputs in no particular order (well, in some order, probably the order of recipes defined in data file). I'm not sure if it can be made any smarter, but if you have ideas, I'm all ears 🙂
On a small scale it’s probably fine, but on that scale?
@wind spade So I think this is going to be my "stage 1" of my sky base build, thoughts on these item balances for being piped into storage?
https://gist.github.com/SteffenBlake/5aa4272ad505c9cc8f47803b1f67fcd5
lol no idea, remember that I have 30 hours ingame and last played over two years back? 😛 I have no idea which items are used in which quantities to what buildings 😄
ah lol, kk
Although for me it didn’t send all of the water to residual rubber, just some of it.
yeah, referencing you to this ^ as well 😛
Offhand I would say, perhaps have two "types" of water. Water "output" and water "raw" so to say
yeah I kinda already know if the source node is "miner" or "recipe" (gray or orange node)
And then modify the... two I think it is? I guess three, recipes which output water to instead output "output water"
I would straight up prefer if it never suggested to consume "output" water and instead it just displayed it as an "output" box, I can figure it out from there
ah like that... not really sure, as that would mean I'd have to hardcode stuff in otherwise automated data and also a lot of people like the feedback loops 🤔
would make the graph a lot easier to follow I think
No#t everybody is as mathy as you are though.
Maybe a "enable feedback loops" toggle?
That might be better actually.
I'm not really sure how would that work on low-level (the linear solver). I'd have to think about it
technically for the water it's possible, but I assume you mean any feedback loops?
Interested in checking that out.
what about a water disposal that pumps water into river so that I don't have to package+sink or wet concrete+sink the excess water
Honestly just water is fine, since so so many recipes use water
Also another feature I would love is the ability to "hide" (grey out and make way less visible) a node
Via just a checkmark on its corner
Why not just dump x fluid into the river?
it's pretty much ready at this point. Only thing missing is power consumption for particle accelerators (the variable one, not yet implemented, working on it now). I could probably even release it now, but I kinda want to do it afternoon-evening EU time
Cause water connects to everything so, it adds a lot of clutter Id like to hide
so in... 12-16 hours, roughly
Well, take your time on it. :)
would option "split resource nodes for each input" work for that? 🤔
I have to convert the fluid into solid so I can sink it
That might yeah, is that a thing?
no, but would be probably way easier to implement 😄
That sounds like my idea, but then I figured that people might want to combine some of it and not sure how that’d work.
since I can keep the solver as it is and just change how it displays the visualisation
Ideally it would work such that, if I check off a node, it also greys out all its children nodes "before" it that dont have any other parent nodes still checked
Water is the one big violator here with the lines at the start really.
So if I have Iron Ore -> Ingots -> Sheets, and I check off Ingots, then Ores+Ingots grey out
yeah, I know I've seen it in the chat above, but honestly I'm keeping like three conversations and completely lost track of who said what 15 minutes back. but yeah, credits to you 😛
But if I also had Iron Ore going into, say, Steel ingots, then Iron Ore would not grey out yet til both Steel Ingots and Iron Ingots were disabled
open U4 version and doubleclick on any node
The grey out function just needs to work again.
Oh gotcha, that was a thing?
was
it will probably eventually come back, just was lost while doing big changes to code
It’s fine. :)
or rather, I'm sure it'll come back in one way or another, just not in today's update
in U4 tool it doesn't exactly work like this, you can only do it on per-node basis. but I think it's already pretty close to what you would want
just gotta throw a lil recursion in there :3
I could, it's just that this way it enables the feature to be used in more ways
e.g. people can "mark" which parts of the production they already built
Currently I just kind of shove them in the back or something.
Edit: stupid autocorrect.
yeah, I really want the feature to come back, it just has to be re-implemented back since the code to generate the graph changed so much
(and I also want to do it in slightly different way)
So, looking though the output format types, JSON is one of them. I assume it should have everything related to the graph. It seems to have a list of all the nodes and another for the edges / connections between them. Most of it looks like how to make the specifics of the graph, but that can just be ignored if you want.
yeah, I didn't look at the code too much (also I don't know python), I just read readme 😛
Well, one of the options I added to the YAMLs is format, which sets the format the graph uses.
What is YAMLs?
The YAML Files the I'm using to simplify input settings.
this:
is:
- yaml
- it's
- a:
- configuration
- format
YAML Ain't Markup Language™
Lol
YAML is like JSON, but more human readable at the cost of being slower to load
and not being interpretable directly in javascript
Well, YAML being a superset, you could probably through the same data at my calculator in JSON, tell it to out JSON, and not actually have to worry about YAML
I just like json since it's so much easier to work with it (machine processing) and pretty much all languages support it natively
I had some issues with some more exotic languages that they didn't have yaml parser natively
(also I don't like that it relies on indentation)
That's fair.
Though apparently Java also doesn't have a built in JSON reader yet... somehow... how do people do Java... (yes I know Java and JS have nothing to do with eachother)
and finally, doesn't like tabs for indentation 😄
Yeah, well, the first language I really got into was python... which has very similar rules to for whitespace...
yeah, I don't like that 😄
Still better than XML, so much unnecessary stuff
can JSON support multiple "documents" in a single file?
oh don't talk about xml to me 🤢
no, json always evaluates to a single value
you can kinda do multiple by just setting the root value as array
I figured as much once PyCharm started yelling that JSON standard only allows one top-level value.
tbf I've never encountered a problem where I would need to have multiple top-level values
That should be easy to shove in though. Just gotta check if it's a JSON, then read the first item as a list. I'll do that later though...
since usually top level value is object anyway, so I can put all needed "top level" values as parameters in that object
(which also makes it good for things like APIs and such, since newly added "top level" values don't change much for existing software that uses it)
I mainly just liked the --- to separate the sections.
most of the JSONs I deal with are generated automatically anyway
(and processed automatically)
I agree that for manual configuration yaml could be better (or ini file), but for anything else it's always json for me
ᴰᵒⁿ'ᵗ ʸᵒᵘ ʲᵘˢᵗ ˡᵒᵛᵉ ʷʰᵉⁿ ᵈᵒᶦⁿᵍ ˢᵒᵐᵉᵗʰᶦⁿᵍ ᶦⁿ ᵃ ˢᶜʳᶦᵖᵗ ᶦˢ ᶠᵒʳ ˢᵒᵐᵉ ʳᵉᵃˢᵒⁿ ᵍᶦᵛᶦⁿᵍ ᵃ ᵈᶦᶠᶠᵉʳᵉⁿᵗ ʳᵉˢᵘˡᵗ ᵗʰᵃⁿ ᵈᵒᶦⁿᵍ ᶦᵗ ᶦⁿ ᵃ ᶜᵒⁿˢᵒˡᵉ
I was using the wrong variable...
oh yeah the classic...
WHY WOULDN'T THIS CODE WORK, I'VE DEBUGGED EVERYTHING FIVE TIMES AND oh there's a typo here
I lied, I did it now. It should also read jsons now, though I only tested one file...
One of my worst times was when I spent like two days trying to figure out why stuff was broken.
I had a string variable we will call foo.
I had accidently called a method like:
myMethod("foo")
instead of
myMethod(foo)
Which since it was expecting a string and got a string, it didnt throw exceptions, it just behaved weird and not how you would expect.
I cant figure out why this specific intersection is broken
8/8 of the signals on the intersection behind, left, and right of this intersection are all good and working
its litt
And then forward, 7/8 signals working, with the one block signal broken
Oh sorry, I screenshotted when they flickered off, the 4 path signals in the first pic are all flashing error
the 4 path signals in pic 1, and the block signal in pic 2, all give "Signal Loops back on itself"
Ive deleted the signals at these intersections and remade them, no dice
No serious, that is a litt train intersection (when it works)
yeah I have dozens of em now
only this one specific of them is broke, rest are working fine
I presume its some dumb problem several intersecctions away and the chain of fucking up shows up actually here or something
Im flying around inspecting all nearby intersections for errors but nadda
Rebuilt the intersection will be good choice.
Tried that, no dice
i got a problem that when i make a straight track, when i create it from the existing track its all nice, but when i start from a couple foundations further and conect it to that track it gets wobbely. These little things could be going on for you Soralin
nah its not that, tracks are straight and fine
Hello, I know this is probably a stupid question but how can I calculate how many rotors I can make without any alternate recipes and making 240 Iron Rods/ Min? I know I need to convert some rods into screws but not sure the ratio.
we could talk about it in #web-tools-dev
can't link threads by name hm
use greeny's tools:
https://www.satisfactorytools.com/production
oh wow, I'm a bit late
I have a computer factory making 30 computers p/m. I would like to use 8 p/m somewhere else, whats the best way to split it from the output belt ?
belt splitting is an inexact science, it'd probably be simpler to rejigger your computer factory to make a specific number of machines produce the 8 in total, and split them off as a dedicated output belt from the other 22, instead of merging them all to 30
thx for the advice
sadly the design is just made to throw out 30 on a belt hehe, machine wise and placement wise
yea, i was hoping you wouldnt say that 😛
splitter wise, probably do something like 30 -> 10 x3, and then 10 -> 3.333 x3, then 3.333 -> 1.111x3, and merge a 3.33x2 and a 1.111
gets you around 7.8ish without much fus
Send all 30 to that place. Eventually it will be filled and the remaining 22 can be sent on to wherever.
but also yea, manifold and overflow is the simplest overall option
think long term production times, everything will balance out
this is what i was thinking, just ship it, also i was not sure if this would maybe make trains ect more dense?
inexact science he says pff
rofl, forgot that someone posted that splitter program last night
Wdym when saying "belt splitting is an inexact science?" 
so yea, not inexact, just annoying
they posted the program to make all splitters 🙂
damn thx
but ye as usual just manifold xD
manifolding is the belt version of "you'll eat this dinner and you'll like it, and you're not leaving the table until the plate is clean"
gotta love it
credits to @timid cradle for that awesome tool btw
would be even more awesome if we had a web interface 🙂
Or you can just plop down a splitter and it will self-balance eventually
Nvm discord didn't scroll
Anyone have any good video sources on the math of the game I can watch listen too?
yo, is there anybody that could run a test for me ingame (or knows this)? I'm looking for the info how do particle accelerators behave when overclocked. For example, plutonium pellet recipe - they have between 250 MW and 750 MW consumption. What range of consumption do they have when overclocked? 🤔
hey guys - so I'm working out some setup plans and I'm seeing some difference in numbers for output in blended turbo fuel
the official wiki states that 300 oil/min should output to 800 blended turbo fuel/min (assuming I use diluted fuel and the heavy oil recipe)
but all my math is working out to half that number in the planning tools
which tools are you using?
can you share the production line so that we can have a look?
ignore the byproduct for now
that's just autofiller - I'll probably keep it as plastic and feed it into my systems
I'm only missing the turbo blend fuel recipe at this point so I'm getting close to a weekend of mega power plant building
seems like max consumption is way off
yeah it probably doesn't account for the changing power
both ends are * 2.5^1.6 so as expected
yeah I was expecting that, just wanted to confirm before I put that into tools and show to people 😄
had to hook this up to 3 nuclear power plants btw 
sorry 😦
i like how it starts under the max cons.
no idea where the 2295.3 comes from
like wtf is this number
is the PA the only thing hooked to the grid?
yes
then no idea lol
anyway, thanks 🙂 last piece of today's update should be ready in an hour or so
ok wtf is this number
its 1500 without oc so max consumption of the particle accelerator
oooh i think it gets the first number that it was started on and makes that the max consumption
wtf
yeah that would've been my guess as well 😄
I assume the wiki says "300 oil to 800 fuel", not turbofuel
is that fuel page or turbofuel page? 🤔
ah
let me plug that in and see what I get
yup - that was it, misread what the wiki was saying
there a link to this program? can't find it searching back through
whats this website called?
^
I like this one over the satisfactory-calculator.com tool because at large scale it sucks to use
the flow chart just takes the math work out for me, but leaves the building/planning to me
it's my happy medium
thank you 🙂
So i have a input of 720 iron and need to make a effective smart plating farm. (by effective i mean everything is always moving and constructors/assemblers never fill up)
How do i make one?
that's a weird definition of effective
Balancer is the only way to never make anything fill up..
But manifolds never stop moving even though they are based on things filling up.
guys its already getting answered on #old-questions-and-help no need but thanks
why post to multiple channels
Reasons.
makes it harder for everyone
cos they told me to put it in there but also told me the correct place to put these questions is here. my bad sorry
What tab is that going to be under? The old power tab?
there's more changes 🙂
yeah. Still working on the last part
sooooooooooo
where is it? xD
(which is making sure that particle accelerators display proper power)
update ready != update live
my disappointment is immeasurable and my day is ruined
(also, it isn't ready yet 😄 )
Oh hey, it decided to send water to the wet concrete plant, which is over a mile away, lol.
Firefox made it transparent apparently.
wow nice background 😄
That's firefox's fault.
if you copy the image using rightclick on visualisation, it will be transparent in all browsers
since it has transparent background 😄
Oh, sorry.
I can fix that by putting it as auto supplied though.
Oh come on now.....
https://www.satisfactorytools.com/production?share=n5C8eiHBJIDMO75MVeUD Here's the setup. I removed the concrete supply part so I could count what I need as it rearranged everything.
well, new update is live https://www.reddit.com/r/SatisfactoryGame/comments/rbzubx/satisfactorytools_qol_update/
@thorn bane you happy now? 😛
its beautiful ❤️
inb4 flood of bug reports of stuff I didn't manage or forgot to test
still hasnt loaded
hm 🤔
Oof. I'm not going to have as much time to play around with that as I'd like. Haven't even really touched Ficsmas yet either.
are you loading /production url?
can you check out what's in console?
yea its still /production
(F12 on chrome)
may just be slow internet or large production lines
i just found out i need more power 
I can revert the update so you won't know about that 

U5 tools loaded everything I had before today successfully. At a glance I like it!
just found out that the reason no one was responding to me on the Satisfactory reddit was because Reddit had Shadowbanned my account... 😦
@wind spade regarding the json file you gave me.
What is the deserialized structure of the recipe and item.
Cuz idk which are ints, uints and doubles
json doesn't have uints nor doubles
I know, but like the deserialized types do
that is language specific
The new one kind of makes things more spread out, but I can see that it tries to keep lines clear.
Yes but like for items there's "sinkPoints" and I assume that's not a floating point number.
And at least "stackSize" isn't either.
I'm mostly just curious what values they can be so I can set the correct deserialized type
https://www.satisfactorytools.com/production?share=xeLCiyfrdZ8mC8VKhoeQ This one kind of whooshes residual rubber way up.
usually all ints are without decimal part and all floats are with decimal part
https://www.satisfactorytools.com/production?share=3do2IPuenLmMxkjaYdYu This one though (which I'm working on) looks more organized now.
So "fluidColor" is an object comprising of integers or floating points?
The excess ingots is just me accounting for what I already had in the area, because I needed to set up signs for those.
Electrode scrap is still trying to give its water everywhere.
again, check the values. If they are x.x it's float, if they are x it's int
Well some of the values are both
then they are float 🤷♂️
hence why I am confused what they are supposed to be
in json it's number for both cases, so it doesn't show decimals if it's .0
Depends on the json library
🤷♂️ that's implementation detail, json supports both
You have an errant 'recipe' text at the bottom of the page btw greeny.
yes, JSON doesn't consider "integers" and "floats" to be a thing... its just numbers with optional fractional digits
and optional exponent 😉
lol WTF
I thought it was tab related, but it doesn't depend on the tab.
Didn't realize the json file contained more than just recipes and items, so thanks 😄
this kind of mismatch between serialization and internal storage can really be a pain for things like key-value maps... or comparison of function output with a testcase result
well that's fixed now 😄 https://github.com/greeny/SatisfactoryTools/commit/c1282105295d003bccc4b9d6c156eb7d2c3db7ce
technically you can just set everything to float lol
hahahaha nice name
Yeah but floating point boolean operators can be borked some times xD
How do you set the sink stuff? Or is the talk about floats and sink points for later?
it's just about which parameter in json is float or int
which I don't know because javascript 😄
so basically just "check the datasource if it contains float, if not, it's int"
not like I can guarantee that it contains float/int anyway
since it's automatically parsed from game data
and devs can change it at anytime
Well my parser deals with that.
But I have an automated deserialization function that converts the json object into a C++ struct.
The thing is that C++ struct has to be defined at compile time you know
It's not the worst is it?
how does your parser deal with the fact that devs may decide in some update that something will change from int to float? 😛
also that last > not being on new line pains me
(not even talking about naming 😄 )
Well it doesn't directly parse into the C++ struct.
It first parses the json file into a json object struct which can be all the types a json value can be.
Then the deserializer will convert that object into the C++ struct using the guidelines shown in using Variables = .
I know pains me too, but clang-format afaik doesn't allow me to do it properly
https://www.satisfactorytools.com/production?share=xAfw2CJ90sVl60Y3DwLv Net concrete reads a little odd because I set the items input for that to 200, but it reads as 0/m and net is technically correct as it's subtracting, but it's starting from 0. The consumption percentage seems correct though. Looks like it's working correctly, just displaying a bit oddly.
yeah... it's not considering input as a consumption 🤔 I'm not sure if it's the right call or not, but I remember doing this for a reason
now, if only I could remember the reason
fused wire
ah a man of culture
Ok guess what I just made deserializing an integer also allow coming from a floating point 😉 ¯_(ツ)_/¯
Bruh wth that shrug command removed a line
Would be nice if the scrap water stuff behaved better though..
point being that if you define something as int and then update later developers decide to change values to decimals, you would lose your precision because you have int instead of float.
Even though that truncates the decimal digits and might overflow if the number is too big
there's the removed line xD
ah I remembered. Input is not considered "production" because it's not produced inside of the production line. hence it doesn't count towards total production and for that reason the net is negative
That makes sense.
yeah, I'm not sure how to resolve this. I don't like the negative net, but I also don't like input being part of production 🤔 (what if you don't use 100% of input? is all production or just the used part?)
better generate the structs based on a JSON schema...
the small total production is fine imo maybe just change the big production number to supply that includes production + inputs
I know I can move it, just seems odd considering how it tries to keep things more together.
its 2021 use an existing json parser lul
ah the magic of black box node placement software
no idea honestly 😄
Ok.
my point is that a JSON parser cannot do this... because JSON doesn't tell you if "1" means to be float or integer
basically I have no control over where nodes are placed and I didn't write the software that I use to generate the placements 😄
Is there a downside to doing 1 train + at least 1 freight car per product made at large factories?
For medium to low volume items I was thinking of making dedicated trains that do not use overflow splitters to dump into Sinks but I feel like I'm missing something here.
For example modular frames could be made at one factory, go onto one freight that is dumped at multiple sites needing modular frame. As long as one destination factory isn't 100% the throughput I feel like this would work.
so if you want to transform JSON data into native structural data, use a JSON schema
no, this works quite fine... the downside maybe the number of train stations you need per factory
i think its actually better
i feel like the cost of mixed trains outweights the benefits
I like to use mixed trains to send low-volume stuff to central storage... but for high-volume items a dedicated train (or position within a larger train) has less headaches
those are actually the same variable lol... I'll see what I can do 🙂
Yeah, I currently have mixed trains and I'm not that happy with it. Feels like a lot of my time is just setting up the overflow splitters and then trying to calculate if I need to move it to a 2nd car because the 1st car may reach the max throughput of even 2 t5 belts.
mixed trains is like mixed belts imo
you can do it but i wouldnt recommend it
I wouldn't even do overflow splitters for trains, just hook each belt into one freight car
and on the other side use exactly the amount that was on the belt
Yeah I like that idea, use backpressure to stop production if it isn't needed. Overflow splitting everything means the factory is always running at 100% even if most of the low value items are just going straight to the sink.
doing high-throughput stuff with mixed train cars is just calling for trouble... because despite the "filters" of the train stations, there is not good way to make sure the proportions of the different items in the train car can get reset to the initial value without emptying the train car... which takes a long time to do
Yep that's what I have with Silica right now and I don't like it. So much just going to the sink.
Ok thanks for the info. I'll try building it all out. Haven't figured out if want a separate train for "storage & sink" but at least this a start.
I did a "compromise solution" today for my oil setup... I have a dedicated Rubber and a dedicated Plastic train station... and a third one that carries away a mix of "a bit plastic, a bit rubber, some fabric, some filters and some packaged fuel"... but the third is just to deliver stuff to central storage
better than to build 5-6 train stations
i even go further and only have 1 product per train so i can use "wait until full" but then again i dont have that many trains
What's gonna happen now when I deserialize the entire json file? 
Yeah I wish Programmable splitters had a "counter & reset" mode, that would make them worthwhile.
json schema also doesn't tell you that imo
json schema has a dedicated "integer" type
so you can say "if type is number, make it a float" and "if type is integer, make it an int"
But sadly I don't have the schema
They've said they aren't likely to add this since it's completely removes a core part of the gameplay loop.
there are a lot of interesting ideas for better Programmable Splitters... I use LOTs of Smart Splitters, but I have only 4 Prog. Splitters in use at my central storage
point being that the json schema that I have (and/or can generate) doesn't know it either, because the original game data don't have such schema
so the information is unavailable to me already
huh?
source
I guess it's possible to check out the source header files
I just don't find them more useful than a chain of Smart Splitters.
A Q&A stream where someone asked about it.
I use them to "pre split" the output of two train stations for my autosorter to increase the throughput... split both trainstations output first into groups, then buffer, then merge into the groups, then feed the group belt into the autosorters
my info is in regards to ratio splitters
"thats in my (mark the SF game director) backlog, because i think at a certain point you just go full manifold on everything, you dont do balancing, i think it would be fun to bring back some form of balancing"
"especially because the programmable splitters right now are kinda shit"
so definitely coming soon ™️
🤷♂️ Fair enough.
Idk what "counter and reset" means tbh.
And as much as they may change poggers to be ratio splitters, I'm hoping it is just ratios and not exact ppm amounts..
latest info is probably this https://www.youtube.com/watch?v=8CzaRyUtSog
oh awesome didnt see that one yet
Tokenizing the json file xD
Basically being able to buffer exact amounts. It would be the same as a ratio splitter in use but allows you to do it across train stations. For example if you said "Only load 100 Modular frames" (because you plan on loading 100 of something else at a different station). Ratio splitter would only work if everything loaded onto one Freight was done at a single locaiton.
I doubt these numbers are correct
they seem fine (except for items)
I dislike this personally.
Ratio splitters is fine.
Anything more precise than that I believe trivializes things.
Hey, thanks. That's exactly what I was thinking of. 😂
Oh shit I wrote "item" for the key of all the items, so it failed xD
but yeah except for items the other numbers are fine
Yeah it would make things very trivial. I just want some type of use for the Programmable Splitter that fits the word "Programmable".
(or rather, it's roughly what I think how many there are)
yeah I dislike even ratio splitters
we just got "load/unload" filters for train stations... and I am not even sure what to do with them... because they don't solve the issues I had with mixed trains 😉
just make your factories be able to input weird numbers bruv
They're just as programmable as a programmable thermostat. 😂
What about odd ones? 😉
this
imagine not making 4562.346 quickwire/min
like we can build pretty much everything from splitters and mergers, giving the splitter ability to split in exact ratio is pretty much the same as giving it ability to split exact amount and that just defeats any point of making nice balancers (balancers would turn into manifolds with ratio splitters)
@wind spade how fast does it read, parse and deserialize the json file in javascript for you?
which one? the one by devs? it's like 1.4 seconds
but dont manifolds already defeat the point of building balancers?
The one you gave me?
True.
But Mark's main point still stands in that poggers don't really have a function atm.
So either remove them or change them 🤷♂️
why would I parse and deserialise it? 🤔
to use for things?
I find past t7 it is almost discouraged to do even splitting because of trains. Back pressure and manifolds are just easier because you are discouraged from belting across the entire map.
yeah, I'm all in for any change or removal, just don't think giving it ratio is good way to do it
it's just compiled into the source code of the tool
Oh
yeah, but people may still want to build balancers 🤷♂️
discouraged? pfft, i spam too many conveyors
Fair.
It is opinion either way.
If you make them ratio remove the specific item selection.
Having both is too powerful imo.
See it takes 20 ms to read, parse and deserialize.
Don't know if that's good or bad xD
Genius Splitters that take quantum computers. 😄
bro youre like the 0.01% that uses sushi
nobody does that
it's somewhere in there 😛 https://www.satisfactorytools.com/assets/app.js?v=1638997480
That's just impossible to read ngl
honestly ive thought about making a new huub0 restart with 100 programmable splitters from the start
You aren't supposed to read it.
for you maybe, not for browser 🤷♂️
That's like cracking open some IL or Java Bytecode and saying that.
if you want to read code, just go to the github and read the typescript code before compilation 😛
I don't really see how this is relevant.
But that's easy to read tho, just like assembly
Then what's the issue? 😛
That js file is just impossible to read compared to assembly
Not if you know what you're looking at and for.
that's just mostly caused by removed whitespaces 🤷♂️
It's just as hard to read as assembly.
and yeah I can read it decently
And if you use an actual tool meant to read it, it's even easier. 😂
Well of course I can read it.
But it requires too much brain power to read it
But, again.. why would you bother trying to read compiled code?
Well it isn't really compiled per say
especially if the thing is opensource
True
Transpiled code. Fine.
But you gave me the link so
how would you call a process that converts typescript to javascript if not compilation 🤔
Transpiled and uglified and minified code.
It's technically transpilation.
Which apparently is actually a shortened form of "transcompiler", so it's absolutely still compiled. 😂
Compilation is the process of taking code and compiling it into an assembly language.
Then you assemble that into machine code for the target.
Transpiling is the process of taking code in one language and translating it into another language.
That's an utterly meaningless distinction.
transpiling is subset of compiling anyway 🤷♂️
though wiki on typescript mentions that it's "typescript compiler" and it "compiles" 🤷♂️
not like this is important or relevant to anything anyway
"tsc" literally stands for TypeScript Compiler. 😂
Yes they are wrong to a degree, but I guess call it what you want 😉
I also love how typescript compiler is written in typescript
Just like C++ compilers are mostly written in C++
what if someone made a backdoor in the typescript compiler
no one would ever find out 
Deno's written in JS, which a lot of people find to be an amazing contradiction.
Yes entirely possible
Bikeshedding much? 😛
¯_(ツ)_/¯
I thought we were done with the whole compiled-languages-better thing.
wait thats a thing????
Parkinson's Law of Triviality. Yeah. It comes up a lot. 😂
anyway yeah if you want to read code, feel free to go to github and read the typescript instead of trying to read the compiled/transpiled/transplanted/copied/whatever thing that comes out of webpack
Anyways I didn't so ¯_(ツ)_/¯
I, too, didn't.
I only wanted to know how fast JS would read, parse and deserialize that json file I got from you
But turns out you just put it in the source xD
The Svelte way.
Who needs dynamism when you can just make it the right way from the beginning?
js probably doesn't even need to do that, since it's mostly eval(json)
(since json is valid javascript object)
Not if it's JSON. And JSON.parse is actually pretty slow.
what a surprise since it's JavaScript Object Notation
Now you've got me curious if you can just eval it. Not that you ever, ever should..
yeah I'm not lol. But it should be possible imo 🤔
const result = eval(json)
Fun fact. You can't in the console.
or just eval('const result = ' + json);
also I'm not sure what even webpack uses to load the json
yeah probably this is needed
Yeah, that works but you can't actually access result outside of it. 😂 But yeah, point made.
eval('window.result = ' + json) wouldn't work? 
let result; eval('result = ' + json) should do it too. 😛
But even still. eval and with. Two keywords we don't ever need to use in JS-land.
Yeah, that doesn't surprise me. It's slow, but it's "slow" in the sense that it's plenty fast for most uses.
I don't even think there's a faster alternative anyway
Writing it as an actual raw object would be faster.
But probably not measurably noticeably. Before the pedants get me.
I guess, but I eventually want to put this thing on server anyway
instead of being served with each request
mostly for the purpose of multi-version support
And how long does it take?
it's pretty hard to measure in the middle of all the other code
sad
it's like... browser does all the things at the same time
Trying to parse it in js just doesn't seem to want to work xD
or you have invalid json
the file works fine with every parser 🤔
did you by accident insert a character in it or something?
how did you load it to js?
Copied into a `` string
maybe it has max length? or something?
Could be, but I don't think so
it's weird doing it this way anyway
Yeah I know but I didn't want to convert the file into a minimized version
Well I could host the file to test that way
Ok how do I read a file in JS?
depends if you're in node enviornment or not
node has fs
in browser you kinda have to request it
Just barebones html in chrome
You could import D3 and use that
Yes and I have totally forgotten as I don't do JS that often
There’s probably a cleaner way
<script src="jquery.js"></script>
<script> $.get('file.json'); </script>
😄
jquery?
If you have a url to the file there’s json.parse
oof
Can it take a path to the file?
Oh yeah jquery would do it lol
function httpGet(theUrl)
{
var xmlHttp = new XMLHttpRequest();
xmlHttp.open( "GET", theUrl, false ); // false for synchronous request
xmlHttp.send( null );
return xmlHttp.responseText;
}
If you’re just doing a local html / js file then yeah it should be able to read a local path
uhh no?
also this all kinda sounds like #off-topic-tech lol
Anyways it parses it in 16 ms...
Guess I've got to do a few things to improve mine
how did you get it to parse local file?
I ran python3 -m http.server first 😉
that doesn't change anything related to javascript
json.parse can't do local files, it only does strings
I'm guessing the `` string was probably changing the content in some way
yeah you should be using '' string
console.time("parse")
var xmlHttp = new XMLHttpRequest();
xmlHttp.open("GET", "/data.json", false);
xmlHttp.send(null);
const json = xmlHttp.responseText;
const obj = JSON.parse(json);
console.timeEnd("parse")
This is the source
But that doesn't support multiline
you know you timed it together with the http request?
I am working this out in my head correctly right? like this works?
then put it in as single line lol
Hard to do when the file I have is multiline
xD
Yes but you could always use a manifold
a who?
Manifold.
--S--S--S--S--S
| | | | |
X X X X X
oof that hurts my head a bit but i think i understand
Looks complicated but trust us, it isn't
If you have enough inflow, you can have one line that goes with individual branches to each machine and it’ll all work out (eventually)
A view from one end of a manifold.
Send all down the line and when it fills, the excess moves as the exact amount.
Instead of trying to make it exact beforehand.
I guess this puts things into perspective a bit more...
I tried the load balance thing and failed hard, I'll take another look in the future.
Oh i see, yah that makes sense for sure! For this particular instance i'm trying to make a truck route from my coal to my main power station which is currently belted, so i want to build and establish the network before switching fully to it, so i need some coal to go to the fuel for the tuck depot which wont give me quite the 120/min but at least i can test it for now
Just do a regular splitter into the fuel port of the truck station
It will fill eventually and will 99% of the time stay full
Power is one area I wouldn't use trucks personally, one flying or falling truck could crash the whole grid.
Just truck slightly more than you need and it's fine.
good point on both accounts
If the truck falls or flies it resets itself with the new system in U5.
People really need to stop hating them based on how they were.
I haven't tried them in a while, back when I last had a flying truck I think it made it to orbit and never came back.
The problem with trucks is that they are slow as heck, and mostly fail when trying to transfer multiple items in one
I don't hate on them I'm slow to trust things when they have gone bad in the past. Will I use them, sure, but in my own time.
The "problem" with trucks is they are capped at 1560 per station.
Which is coincidentally more than a single train car can do.
If the pathing is good they look to have brilliant throughput for not much infrastructure 😄
And I used mixed trucks plenty. They work perfectly fine.
@wind spade the new "Items" overview for tools is amazing, its exactly what I needed. This is tremendously easier to parse and makes everything "at a glance" now, amazing work!
Trying to make a 100% efficiency Rotor Assembler.
input 90ore, split 3 times to 30 in each furnace.
The rectangles are constructures. The top constructor is split into 3 and 1/3rd (5 of 15 rods) is moved down to bottom as r5.
Have you heard of manifolds?
New-People-Meta channel when? 😭
What is it? Just a series of splitters?
anyone that writes there would get: "use manifolds, get alt recipes, underclock to match ratios, 120 coal/8 gens/3 extractors", anything else?
Yeah if you have 1 input line with the throughput you need you just have it split off to each of the outputs. No need to balance individual belts manually
We have done some actual theory crafting and maths over the months...
In between the "let me ask a very basic question in the math-meta channel" phases.
I'm aware I'm being a dick atm.
So I could just completely scrap these two sections and use manifolds?
Yes.
Subtle insults are even more fun.
Yes you can always manifold instead of any other splitter configuration you can think of is always true.
So image is not required.
@soft scarab When you said to use manifolds, I can just use manifolds in those areas I just blocked off?
Yes you can always manifold instead of any other splitter configuration you can think of is always true.
So image is not required.
You can also have a manifold for the rod outputs. Anytime you have inputs or outputs you can have 1 belt with splitters coming off for each building
(Assuming the belt has capacity to serve / carry the throughput)
Ok.
And if you wanted you could put a manifold before the rod constructors too. That gives you more flexibility in base layout
I have the two stages there.
Rod production (remainder of 5)-> assembler.
rod production(adds the remainder of 5) -> screws -> assembler.
so If I run manfiolds along pmuch every line there inbetween stages. It will automatically distribute the appropriate inputs?
Yes.
Yes. It will not work immediately because it has to cascade down the line, but give it a few mins and it’ll be perfect
So it has to stack up and overflow first?
In order for the end of the line to reach 100% efficiency, yes
I was doing load balancing.
Yup
Ok so my setup still works. It's load balancing.
Manifolds are just a different method.
Yes.
I haven’t checked your math but presumably yes
Ah ok.
Thanks for the info but I'm going to keep playing with load balancing. I know its not easily scalable but it is in factors of 2 and I rather enjoy the precision of it.
Manifolds are generally easier to work with - less space required, less math/headache, and they work just as well. But load balancing can be a fun puzzle and if that’s how you like to play by all means go for it
Higher complexity ones also look cool as fuck too.
Balancer and Manifold have the same end result, just a separate method of getting there.
You can build arbitrary n:m balancers. The wiki has examples of prime balancers. Like 1:7 and 1:5.
Doesn't have to stay factors of 2. 😊
Yeah you're right, it doesn't have to be only in factors of 2, but its easily scaled up if it is.
If I went with manifolds it wouldn't have been as Satisfactory as the feeling of perfectly load balancing.
I still cant for the life of me figure out why this signal is broken. Ive torn down nearby intersections and rebuilt them, replaced signals, still no dice
Its just giving the "signal loops back on itself" message which is very unhelpful
Finally figured it out, not sure why but the trick is, one of the signals is bricked but simply rebuilding it wont work
You have to place a path signal before it, delete it, and then the intersection will start working, then you can re-place the path signal in the right spot and delete the temp on
@dusty cedar Like so? Moving to this channel cause we can post pics
well you would need a lot of pumps to move the water up 500 meters and the power draw
yet that's a example I figure
I don't have pics
just explaining the mechanics Scalti and TotalXclipse have demostrated in their content on youtube
Scalti's video on it is back in Jan. yet if CSS did not change the flow mechanics and they featured a water tower in their content involving U5 with the community managers I figure it is a permissible way to pump water and other fluids to higher levels without pumps on every single pipe
They haven't changed flow mechanics to my knowledge.
if so, then you could use a oil tower to raise the the oil to a raised refinery building for procesing
or water up to a refinery making solution for alumina
only using pumps on a pipe line to the tower then a key pipe to give the other pipelines the headlift of the fluid buffer tower's height
The Scalti video is Water Tower Mechanics guide
and he even uses the mechanic to supply his coal generators with water though the extractors are below with only pumps on the supply line to the fluid buffer giving headlift via the key pipe to all the pipelines
sure you can do that
but you can also just slap on 50 pumps since power is practically free in the late game
You could also transport the water with a train and build a train spiral
So this doesnt seem to be working. I have one way valves on the key pipe pointing downwards, and the key pipe is currently empty though
Do I need to have the key pipe filled perhaps?
Red being the key pipe of course
ah yeah okay, it seems the headlift from the key pipe works as if it is, itself, a giant fluid buffer
the headlift the key pipe puts on the others is equal to how full it is
As I watched my key pipe fill up, the pipes beside it started filling up and lifting to match along with it
As you can see here, the key pipe is only just filling up and the pipes it goes to are still not filling (the one full one is the "main" pipe with the actual pumps)
But the moment the key pipe in this section started filling, all the other pipes started filling right alongside it
This will mean, btw, going down from 750 Mk2 pumps (6000MW) to only... 30? 40? Mk3 pumps? (320 MW)
I think this should work then, gonna test this out on my dedicated water pillar #1 and hopefully, I should get 31 free pipes of lift without any pumps on em
Blue pipes are the normal ones, light blue is the Primary pipe (the one that will get pumps), and red is the key pipe
I think the main issue with this kind of trick is that it might stop working at any time when they revise/fix the fluid handling system
Maybe, I think though that might be a ways away tbh, this sort of stuff is actually kinda tricky to work on and finicky, so I can see it being a "save it for later" sort of deal
you need 19200 Water "up there"... do you need it for power (nuclear) or do you need it for production?
yes
And its actually 45,000 water I need :3
if its only for production I would consider a train... but for power I would go for pipes with pumps.
Its both and this is a self imposed constraint
My entire everything is 500m up in the sky, city block base, its quite fun
they maybe pipes with pumps for power and a train line with liquid containers for the rest? 😉
Nah, using the tower trick which I have hammered out, the power is quite cheap
I can comfortably do 32 pipes of water up using only 10~11 pumps total, so about ~88MW to move 19,200 kL water/min
just don't be surprised if it stops working after an update... I don't think this is a feature 😉
Thats fine, I should have nuclear up and running before then hopefully and then I wont have an issue needing a couple extra GW for all the pumps proper
15 uranium fuel rods is 3.75 plutonium fuel rods not 1.33
also i feel like with that many space elevator parts youll be done before you even get all of it running
so maybe go for something more awesome points oriented
i did 20 ADS 2 PTR 4 magnetic field generators 1 pasta for example because by the point i got the 20 ADS ready i was already almost done with pasta/ magnetic field generators.
I dunno if it matters a tonne which you sink for points. This is also just phase one.
In other news... I think I have figured out a way to achieve 100% throughput on trains.
If you balancer split resources into 2 separate train stations and alternate which is picked up, and 1 station can handle 100% throughput, your resource will just go into the not-docked station, so items always flow
I typically attach an ISC to each freight station to buffer the 15 second loading delay
every machine in the game except for miners has enough buffer for 30 seconds of overproduction
I wonder is it worth it to package crude oil for train transportation, since packager does this at measly 30/min rate. That would mean 40 packagers for MK2 pipeline (20 to pack and 20 to unpack, now thats 400MW, on the other hand I can just add more carts for way less power (assume 1 locomotive-4 carts train, that's 30ish MW per cart)
So 40 packagers vs twice the carts, or is there something I'm not seeing?
Yea, that’s not a water tower, that’s a water pillar all right.
@wind spade Um, if you switch tabs, are things you moved around supposed to reset? Not sure if that's a bug or if it's always been like that. Just wondering as I wanted to isolate things I needed to change in order to add in something I forgot, but it's not that big a deal.
Yeah, they do reset... it's planned to change in the future
'k
one thing is that packaging requires twice the time of unpackaging.
but I've always thought of the same idea but in the end I've never done it, because of all the space the packagers would require
So, um, a setup like this to act as the hub for the whole nodes extractors (max 2,100/m) should work fine, right? I made it so that I wouldn't need to calculate each individual extractor and instead just plug into it as needed and mark the consumption for them as a whole.
Can you make it more wavy and noodly?
Why? lol
Chaos. 😈
Also, the game is lacking nitrogen gas icon for the signs, pfft.
I was playing on Update3. that time, the all-resource meta was turbo-motor production, and oil was the bottleneck. What the equivalent now?
Plutonium rods? I don't know.
Sulfur and ADS/TPRs.
2 layers would be enough already
But 600*4=2,400?
My logic is that it needs room in order to flow at max 2100 if needed.
What about more than four at random spots?
you said 2400 output, right?
if you have more pipes at random spots then its just random shifting of this connection pattern
point is: 2 layers is enough
give me any combination of output pipes and ill draw a flow connection diagram
My thinking though was that I'll need to allow the flow to be able to reach 2100.
nah
junctions have no limit
so thats no issue
you can stuff 1200 into one junction
Oh, I was thinking in terms of like how coal plants work vs MK1s.
its a similar case too, partly
if you had 2 layers of pipes, you could easily supply 13 gens
you dont need 3 or 4 layers
just try it, play a game of "connect the dots"
its really that simple
wouldn't it be better to not mix the pipes and just use what's in each pipe? 🤔
I didn't want to go and calculate how much I'm using out of each extrator as I use it.
I'd just use one node at a time 🤷♂️
Anyways, did a bit of math (based on 480/m, which also happens to be close to current usage) and I can add four more outputs of that, while the fifth output is what's left over. So, 6 total (including the existing output) @oblique hollow ?
Actualy, no wait, that's five total. @oblique hollow
What are you pulling off these pipes? water?
N2
Are you pulling the same amount each time you pull off it?
Or are you doing multiple recipes that use different amounts?
I'm not even using it halfway yet.
Since 5 uses less than the full 600 capacity, it balances in the 2 pipes
2 pipes can pretty much transport 1200 anywhere
If you mean my math, the 480 is from 2x N2 canisters production.
Well you do have one option if you wanna make life easier
Ah yeah so I was gonna say
if you just package the entire thing, you can fit up to 3120 nitrogen/min in a single Mk5 belt via packaged nitrogen
and then you can just have a single manifold
Only takes 9 packagers, and they are pretty small buildings
The current one is from my production of turbmomotors (pressure alt), heat fused frame, and the cooling device (give me a min to check the alt used)
Could just slap 9 packagers down or whatever at your nitrogen source and have 1 belt running to this spot
Or sorry I guess you want 2 belts, you want a second belt coming back you ship the empty packages back on when consumed
I haven't firmly decided whether I want to make the packaged nitrogen facility in this area yet though.
My point is you can replace that entire pipe mess with just 2 Mk5 belts in a loop
I still need to bring it to the site.
I meant the packaging location.
You package it right at your nitrogen source
And was gonna transport those by train, or maybe drone.
And then you bring your packages all the way to that spot in your pic
and then manifold the nitrogen packages out as you need em
I dunno yet,
Nitrogen has insane density on mk5 belts, way more efficient than pipes, you can put just over 5 Mk2 pipes of nitrogen on a single belt
You just need another belt in the opposite direction to get the containers back
I wasn't planning on piping it all the way over to where I needed it.
yup, 2 belts stacked in opposite directions
Or rail or drone route.
@wintry aurora Im talking about your pipes you have in your pic
The whole thing
I would tear it all down and replace it with just 2 belts
There's not many oil nodes decently close to oil though. Gonna take a look at the other scrap recipes.
?
You mean for making packages?
You could have your package making anywhere on the return line, since you will have the return line touching everywhere the input line does
Yea, I was gonna make them on site, I dunno.
You said you need 2100 nitrogen/min right? Whats your total math on that?
like what all are you using nitrogen for
Haven't gotten used to the outposting mentality maybe?
Or is that 2100 just how much the extractors are supplying
Yes
gotcha
I don't need it now, it's the max amount of the node, I just wanted to create a manifold to balance it all or something.
Right, so yeah I would say a belt is the most head empty way to do it, very simply and then it just becomes a plain ole belt manifold
did you write that number in the sign manually?
Well duh theres no automatic signs
BALEETED, lol
what an atrocious looking pipe network, i love it!
lol
you know what they say, the worse it looks, the more efficient it is
K E E L L
Well, anyways, I'm just trying to figure out how to outpost things here I guess.
Current map
Uh oh, I haven't planned for nitrogen yet, I have that sinking feeling that there isn't any even remotely close to where I built my entire nuclear setup.
how much do you need to get the nuclear waste dealt with?
1260 according to the plan.
Hey, someone here who exactly know how to work with the production planner from statisfactorytool.com?
pack it on site and ship it over imo, especially if you don't know how far it'll be
Looking at it now I might need to make the nitric acid on site and move* that*
bruh... why the hell did i never think of that lol, i'm already committed to my packed nitrogen being moved, but ugh, packing the nitric would be so much simpler
actually, no, nvm, there's no water near my nitrogen well i think
How would you balance these two (the right one doesn't have any current consumption) between....
These (or more) drone ports?
So that they both fill as fast as possible?
chuck them both in a isc and make the output lines both 780?
That makes more sense I suppose.
Can someone please help. This train isnt moving
What error does the train give when you look at timetable?
Also, the error could be somewhere out of sight.
I realised that mistake when I saw the water for the nitric acid is exactly what the sulphuric acid throws out as waste when I looked at the non fissile uranium part of the plan 😦
Says "Signals are making next stop unreachable" however all the signals for this train to reach its next stop are green
nevermind.. found the problem... lol
Looks like you put two left hand signals down there on a right hand track.
Indeed. Now I must step in front of a train for my incompetence
thanks for your help
It's easy to make that mistake on a complex junction though.
If no one has done the math yet (which is unlikely)
A wonder star is worth 150 gifts
Looks like my 120 gifts generator is obsolete
Beware of exact fluid loops due to the load loss bug 😉
I suggest topping off at least the water (easiest solution)
That's something I'll need to be cautious with, I don't need a repeat of my alumina production where the new in feed clogged the system
Best would be to refill both the water AMD the acid, but that's the most annoying solution
One approach since it isn't a large system would be to add a spur for a small fluid buffer, that should handle the magical fluid loss.
Fluid loss is only on load up though right?
Yes, but if you buffer you're just adding loads before your machines will run dry
Even without 1 small buffer, the lost fluid should be self correcting since the planned loops will be tiny.
There's no correction if you're providing as much as you're consuming. Just loss everytime you load
Regardless of the size of the loop 🤷♂️
That's why I think the easiest solution to have a loop never run dry (without manually refilling it somehow) is to top off the water part via a VIP
Looking at the water extractors that should be easily achievable by just having all three of them going at 100%, 315 water is called for, 360 should keep the nitric topped off without choking the sulphuric thanks to said vip
just use the water for wet concrete for encased plutonium cells
Does anyone have a factory design for completing phase 3
Prob best to just be making a bit of everything and then use containers to buffer the phase 3 parts with limited production lines with saves materials
So though the pipe tower headlift trick is sort of silly, it does encourage you to make very intricate and cool looking pipe setups.
Crystal Oscillators are annoying as always... I am lucky they are "done" now...
I'm kinda thinking, is the creation of elevator parts directly corresponding to the lead up to nuclear power?
@fluid mist
you can't finish the elevator without nuclear right?
Turing complete pipeline logic when
If you can do a NAND or NOR gate, you are already turing complete
nice
@cinder silo I just had an epic idea for a factory
a
essentially, you would have a production floor for a given part, that floor would consume all your available power, but then you'd shut down that floor and enable the next floor which would construct the next part from the parts you made from the previous floor using the same available power
if things are overclocked, I would imagine it would finish faster than operating half of both machines in the same available power
Consuming all my available power would be difficult, especially when my nukes online.
let's say, 20 constructors and 20 assemblers working at 100% would finish slower than 10 constructors and 10 assemblers working at 200%?
or is it identical?
Should be identical, the only difference if I'm recalling correctly is the power consumption shoots up several times for that doubled production.
so then my proposed plan is a trade-off of space to power consumption
err rather the other way around
I am proposing isolating production floors which require more power but less space
to achieve the same result as a larger facility which is running everything at once
sound like a plan? 😅
so, my mate started a new dedicated server, and cause he's new i'm not allowed to rush up to tier 8
Space isn't exactly a problem in this game, you'll crash your game long before you run out.
which is ok, but in the interim, i want to maximise the shit out of all nodes, so, question, 60 limestone p/m out of a normal node, am i better off leaving that in one 100% constructor, or splitting into 2
and let them run at like 60-70% each, keeping in mind i'm running on bio power atm
somehow yeti's question is related to mine
Yeah, right now your target really is going to be coal to get the generators and steel for mark 3 belts, I wouldn't dare expand machines much on bio, too much babysitting and blown fuses.
i have no choice, i can't unlock tier 3 stuff, he has to
and i have like 10x more time to play than he does right now (new g/f for him)
Oww, best to have the factory churning construction materials in to cans that you can use at a later time, too many machines will make it a game of weed whacking just to feed the generators above all else.
right now i have like 5 generators on their own networks, which can be chaotic, so i'm thinking of moving them all to a central hub
it'll be easier to reload, plus i've already built a simple leaves/wood to biofuel line
with smart-splitter sorter?
hah, no
Too early in the game for smarts, he doesn't have coal even 😦
it's been a while since i massed used bio burners, but from memory, they don't really burn if they aren't needed right?
so i could build like, 20 of them and if i only need 5, they'll just burn 5 at a time?
Smart-Splitters only need a bit of Caterium I think 😉
that requires unlocking
i know where the caterium is, i know the tree in the MAM intimately to get it
but to be fair, it doesn't take that long to get to coal
tier 3
afterwards all the biomass thing only becomes necessary for the chainsaw
and even this is not necessary anymore once you have Nobelisk!
EXPLODE EVERYTHING!!!
I’ll be honest though why is Nobelisk more precise than the chainsaw
I go to cut down a small tree and it cuts down the 5 massive trees next to it. Nobelisk just gets rid of the single tree
and even better, currently you can produce Nobelisk from frozen Water... no need to look for Sulfur! 😉
Wait what
Don't ask me how ficsit make frozen water explode
How?
since when did we have frozen water in the game?
Since end of November?
Most useful part of the ficsmas event .. exploding snowballs
And here I thought the most useful part was the menu checkbox that let you disable it.
😉
As much nobelisk as you want without any resource node?
🤣 💯
To be honest, I disabled ficsmas last year
Is there a way to take two outputs from a truck station where two items come in.
Making the two items stay on separate conveyor belts, and then overflow would go into a sink?
Where it won't sink items before the conveyor belts have truly stopped?
Use a smart splitter.
I am already doing that it's just very slow, and might even be too slow
Actually, make two smart splitters and merge the overflows.
Make two separate sinks?
May also just need to upgrade belts.
Can't upgrade yet 😭
‘K then.
Smart split each of the 2 outputs in 2 non-mixed belts and send each item-type belt into an ISC (2 belts each)
Max unload speed of 2x max belt speed for each item, can add overflow rules either at the smart splitters dividing the items or after the ISCs
It’s the only two input and output merger/splitter we’ve got.
For some time the whole system is totally clogged, but apparently it should be able to handle it once the station is cleared.
I only produce about 120ppm that goes into it, thankfully
There should be no clogging if you unload properly
Well yes, but then again I have a server running 24/7 most of the time.
So it filled up in no time and got clogged so I had to stop the tractor.
And I didn't consume much other than beams so it was mostly full of that.
Now I need beams so I had to restart the whole thing, but this time making sure overflow is sunk cuz steel gives tons of points 😄
May I ask what is being made that requires the items to be separated?
(Ping me since I'm on my phone atm)
Steel, so the station is bringing the iron and the coal?
No it's bringing the steel beams and pipes
Which are then being used for what?
Storage
So your storage system is clogging?
It's the old storage so yes
Ok. Was a bit confused. Sounds like a design flaw in storage more than a truck station issue.
But it looks like you guys figured it out.
Making sure excess is sinked or dealt with (if needed) is part of "unloading properly" imo 😉
xD
I had the same problem a few days ago... my "storage input" train station and its output belt couldn't handle the incoming trains anymore
so I added a second train stations... put a programmable splitter to each of each train stations output to sort the traffic in two groups, merged the corresponding 4 outputs and let them run on separate belts towards my storage... with a dedicated sink at each end
Yeah I plan on adding mergers every so often inside my new storage system so I can have multiple belts coming into it
just instead of doing mergers I would just keep parts of the storage systems "working belt" separated... because its difficult to merge belts AND control what gets thrown away
Not really
The system most people use is where you have the smart splitters setup like this:
C-+---
|
C-+---
|
C-+---
Where left would be the item for the container, forward would be any undefined, and right would be overflow.
That's just stupid and over complicated.
Instead I did:
C-+
|
C-+
|
C-+
Where left is the item just the same, and forward is overflow, any item that can't go left would then go forward 😉
If you add mergers but not sink connections, your system will become more prone to clogging again
... I thought most people used the bottom one since it's the quickest to set up 😆
That's where I'd do something like
[...]
|
C-+--&----
|
C-+--+---S
|
C-+
|
C-+
|
C-+
In this way the last smart splitter would go left with any, and if it overflows it would just go straight into the sink.
Or alternatively you could just have setups where only n items per belt. (Which might be easier).
I think the solution with the least complicated beltwork is a factory sending a few mixed lines directly to storage, so no balancing shenanigans, just merge a few machines with the right clock
I saw a thing on the Wiki about setting up a battery type solution for hypertube cannons, so they're not constantly draining power, has anyone set one of them up successfully?
@vast jungle
yes, I do it regularly...
you just need one switch and one battery for each "hypertube hub"
connect the battery together with all hypertube entrances on one side of the switch and your "power" to the other side
when travelling, activate the switch for a few seconds to charge the battery, then deactivate and quickly hop into a entrance
just an advice... it works best without the Hoverpack
because if you wear the Hoverpack you might draw an additional 100 MW from the battery, which can decrease the time to hop into the next entrance 😉
Now it would've been cool to have proximity sensors that could active and deactivate power
The hypertubes have proximity sensors themselves, so, in theory.....
Although you wouldn't want it shutting off when you're in the tube.
the other option is to allocate a few nuclear powerplants to the hypertube entrances... and skip the whole "shut them down" thing
How the heck many hypertube entrances do you have that you need a few nuke plants to power them???
if you consider map-wide a mesh of them instead of all going to a central area, you need quite a lot 😉
I'd get that, but I don't think they take THAT much power.
each intersection with 3-4 tubes will cost 75-100 MW...
I think I have nearly 30 HT-entrances... and I still occupy only a small part of the map
But those are baby numbers when considering each of my tubes will probably have 64 entrances on each side (two way)
Also just want to be sure, those snowballs act like nobelisks right?
I mean that's what the prompt says
Guys I found a way to have two way split into two separate lines at the ends 😄 (Hypertubes that is)
an exit and immediately two entrances, one a little left and one a little right?
| |
E E
E
/
Something like this
The exit would be in general be offset by one in the opposite direction you realistically wanna exit at.
yeah... it works, as long you angle both (upper) entrances right... (so going from top to bottom works)
No I don't think you have to angle them the same
my problem is that I normally have more like a crossroad, so I don't necessarily go to a central place
but if you built a really large HT hub, this could be a fun way to organize the whole network... and even have little "jump off" stations
I am going to build a large ht hub where all tubes make you travel at the max speed of roughly 1 km/s 😄
you could add an HT accelerator/cyclotron at the beginning to get up to speed ^^
unfortunately as soon as you have trains, HT's are kinda slow
Well I'll have 64 entrances getting me up to roughly 300'000 units/second which is equivalent to about 3 km/s but hypertubes are limited in speed to roughly 1 km/s
but this might mean you need an accelerator at every Y-crossing (you described above), which will increase power consumption even further
O.O
"Warp 5 Mr. Worf..."
Nope it retains speed through it 😄
Also you can slow yourself by yeeting yourself into u-jelly landing pads.
Though at 300'000 units/second you absolutely need tons of them xD
how many entrances would you use for a "flying" HT cannon?
