#math-and-meta
1 messages Β· Page 233 of 1
wtf is that
a 1/27 splitter to deliver fuel rods to drones
one drone nees like 1 fuel rod every 5 hours of operation anyways
so even with significant lag its not happening for me q_q
putting a full 60 item/min belt is too much radiation lol
even after i restarted the game nothing happened
Youβre a chad plutonium drone gamer
yep
Love it
red dots all across the map, beautiful
Spontaneous death in case you forget your filters
Oooh i thought it was put into the drones themselves
nope, what's taken out of your power grid
use a lot of drones, make sure you have plenty of power π
i'm not complaining, but they are pretty expensive considering you need at least 2, lol
legit some european railways here assΓ₯
wdym?
Swede spotted
it's always funny to me how long cables on power towers look from afar
default station names?
no it's not stations, just power towers and cables
meh, ok
though default station names are funny too
when i saw Krasnoyarsk as a name, i wanted to leave it and use it
not because it's the funny name, but because it's funny to see russian name in a game with english as a main language
Is there a tricky way to jury-rig a priority merger?
Like "take stuff from this input if you can, and only then take it from the other input if you can't"?
for belt? I think you can do with belt speed.
Yeah for belt
a lower speed belt takes priority normally, so you could split one side up into a bunch of slow belts and merge the pieces, idk?
i'm trying to think of a way
you can also split the less priority one into the more priority belt.
Yeah it's tricky. This is the recycled rubber/plastic loop thing again, and I want to prioritize the residual rubber coming in over the rubber coming out of the recycled rubber refineries.
Maybe it'd be easier to just let them merge naturally and have a smart splitter after the merger to dump the overflow as the main rubber output?
ah, I just finish that one. you can just put a smart splitter at the end to to send out the excces out of the loop.
Or dedicate certain machines to recycling and the rest to output
Thats what I did, merged all the lines then split that into my feeding belts, overflow off to where it needed to go.
serendipity - I tried using trucks to ship packaged oil to my main bus, and then was able to take the empty canisters and upgrade my solid biomass burners to biofuel burners and spin down coal generators. Probably get to spin them up again soon to burn coke
you're trying to manage recycled plastic and rubber?
Yes but mainly just merging the recycled rubber with the primary rubber injection from the residual rubber earlier in the process.
the answer to that one is 'don't'
make residual rubber from the resin first
then use that to make recycled plastic.
At that point you design the layout so that you wind up with your final desired output
much simpler
Yeah that's what I'm doing, the residual rubber is already set up and being sunk atm. Mainly just trying to figure out how to merge the recycled rubber and residual rubber into one belt to send back into the recycled plastic refineries, while still being able to extract the factory output from that same system.
why are you merging them on the same belt?
Because they both need to go into the recycled plastic refineries
make 2 different groups
I thiiiink I have something that'll work? I have to pipe up the fuel before I can test it.
The recycled plastic needs more rubber than I'm making of the residual rubber though, so the recycled rubber needs to loop back around to meet with the residual rubber.
The ratios all work out, just need to figure out the belting
but you ... don't need to merge them even then
I'm only making 80 residual rubber/min but I need to feed the recycled plastic with 213.333 rubber/min
resin to Res rub, to rec plastic, to rec rub again until you have your 'starter batch'
and you don't even need to do that really, except you seem set on a specific starter number
Because I've already built the starter number, which is being fed by earlier stuff in this process.
if you don't mind not making it work the most efficient way possible, you can try using just recycled recipes, they make nice even loops and simplify the math a lot
like, half refineries of output go into all refineries of input, the other half is your surplus
I'm doing the recycled loop, yeah. It looks like you're not feeding yours with starter rubber from residual rubber though?
That's interesting, I guess you'd have to toss in a stack of something to prime it? Making cloth instead... hmm
you just need a tiny bit to kickstart it and let it spin up
im to tired to end this thing
That's a lot of resin to sink haha
yea, you prime it up and it work indefinitely. in theory, at least
you can refine it as well if you want, i just don't wanna bother
this is what I just finished.
Nice, that's approximately what I'm doing, just with equal ratios of fuel/plastic/rubber, and I probably should've done petrocoke but didn't, and I'm gonna need it for aluminum...
I suppose if I can find a spot to squeeze it in, I could just overclock one of my heavy oil residue refineries and one of the oil pumps.
going to setup last half of the factory, probably won't finish untill weekend. goal is to make stuff need for recycle uranium waste.
more is better, maybe π
how much power do you guys think one might be looking at to finish the game?
yea i suppose your right its pretty variable
if you need more than 40gw or so I highly recommend going nuclear though
my turbo plant is looking to be 60gw
then im thinking of going rocket fuel slooped benders
blenders*
eh.. pls go nuclear fields of fuel gens are super awkward and honestly ugo imo
they made gens use more fuel per gen specifically to help people avoid fields of gens. Now they're just going rocket fuel
Convenient they made it so I can just pack my fuel gens with power shards, too
So many fuel gens...
i might skip rocket fuel and go nuclear just for the experience of building a nuke plant
you know how many nuclear plants you need for 60gw? like 9
damn thats crazy
I'm leaning towards specifically not making rocket fuel, just to have the experience of doing nuclear
^^
and remember, if you do go nuclear you don't have to burn the rods where you make them, easy to drone
I dislike the spicy rocks, I'll stick to rocket fuel
it's their problem for making rocket fuel so simple to set up π¬
Yeah it actually needs a nerf imo
i saw the numbers after you overclock and sloop rocket fuel blenders and im like omg
thats crazy
agreed, but I'm not sure if they're ok with potential wacking of the existing setups. they could've afforded that if the game was still in EA, but maybe not now
Yeah, kind of an oversight imo
Fuel gens are fine up to 250GW IMO. Now that you can easily fit 4 in an MK2 blueprint, overclocked.
You can actually fit 5
I think nuclear mainly needs a way to stop being dangerous. Doesn't really matter the power output.
Dangerous?
Mind you, that's a mk3 BP, but I made it into quarter segments and just stick the fifth gen in the middle once built
Rocket fuel is like the simplest thing ever if you have the nitro blender recipe. You can even OC and sloop it for fun and profit
It kinda stands out as the meta power production atm
Yeah, being around uranium kills you or burns up suit filters. I'd really rather not deal with that
Especially if my plant crashes and forces me to spend a long time troubleshooting it
was rocket fuel new to 1.0?
yea, as well as ion fuel
though some say ion is kinda expensive and should be produced in small amounts for a jetpack
thats what ive read
can't confirm myself though, didnt get to it yet
Avoiding tech spoilers, but it's very expensive
The tech spoiler -> ||Ionized fuel requires synthetic power shards to make||
probably because of how simple RF is. you just take some existing TF and slap a bit of N2 gas, water and iron plates on it
and voila, 7.5 consumption is now 4.16667
yea and that spot in the blue crator lakes, has all of that right in a central location
there's one in swamp as well, but it's on an elevation though
it was a no brainer, but i think like someone said earlier i dont know if id want to build such a large plant if i needed more than acouple hundred gw's
yeah but its too easy, nuclear is fun cuz of the challenge, also rocket fuel eats away at your oil which you need for plastic and rubber, while nuclear eats away at uranium, which only has one purpose, nuclear power
have a dimentional depot have filters fed into it. you're invincible
Do filters get automatically withdrawn from the DD? I know ammo doesn't.
idk but if you get low its pretty easy to just restock
Oh yeah, for sure. I was just hoping it'd auto-DD-pull them
Nope, doesn't autopull
oil is suprisingly plentiful though
tbh there's so much oil on the map that you can basically ignore that in a normal playthrough. unless you're one of those minmaxing all the nodes (cough Kibitz cough)
no, but a whole stack goes a long ways, just gotta click
sometime, I feel like this 100% is a lie
(I'm not at nuclear yet)
right now i feel like there is more oil on the map than ill ever know what to do with, with the diluted fuel and recycled rubber/plastic recipes
I've actually realized the way satisfactory handles radiation makes more sense in the context of the small number of employees sent to the planet. It's a smaller expenditure of resources to do radiation shielding for the small number of employees, versus radiation shielding on every structure you build.
there's more of everything on the map than you can reasonably use
fair enough
especially if you use the more efficient resource recipes
resource efficient alts are pretty much just there to reduce needing to import stuff
?? 12000 ingots is more than enough
Irl though, a suit isn't going to protect you against gamma radiation
I guess so, but if we have space elevators and other just about magic tech, can we please have a tech that lets me stop worrying about stay times?
has anyone speed run the game yet?
like i wonder whats the fastest time to finish the elevator
isn't it ~2 hrs with the duplication glitch?
FutureTech π
irl holographic gun isn't going to build gigantic factories with just one button either :P
When you sushibelt your plastic and rubber and then realize you need an 800/min belt and only have a 780/min belt... π³
(Probably going to just have the sushi go out both ends to divide it in half)
ohh thats right i saw that video
It would be really great if the programmable splitters could let you configure all four belt ports. Just to make them more distinct from the smart splitter. Two inputs, two outputs would be amazing.
@vapid gorge
inputs and outputs are looped, headlift is also fine (lower pumps sit at roughly 40m, upper pumps are nowhere near max)
but somehow the left side in the first screenshot is still getting less fuel than the right side, even though both come from the exact same production chain
it's also just one generator that is sometimes not getting fuel and then only for like 4 seconds...
are you cannisters flooded? usually it's simpler to do 1:1 in front of the refinery and do a small loop
cannisters were pre-filled, belts are all moving and no packagers are backed-up
are your generators actually starving? I don't seen yellow lights
will probably do the setup in front of the refinery next time though, never thought of that lol
one gen is starving every full power graph cycle for like 2 graph updates
currently pre-filling the setup again and checking if that helps
starving gen is the one in the lower left corner, second from the left third from the bottom
ok so you're kinda bottom feeding the input of the pipe loop and it's not the most stable loop option
I'd try it like that with a pump right before the entrance of hte manifold
okay, re-filling looks like it has helped a bit, will let it run for a while to make sure. If that doesn't help i'll try out the pump idea. Might also have to throw in some flow rate compensators to combat the sloshing a bit, it's quite bad
looking promising so far though
ah nope, pipe is already half empty again...
okay, so you want the pump right at the manifold entrance. So at the point where the pipe comes up through the floor right (image 4)?
yeah okay, put one there (kinda akward as it creates a super small pipe segment, but thats the only way it fits in at that point)
I really suggest reworking the loop like in the image though.
if you're going to keep this kind of loop at least re arange the pipe from below so you have more space for hte pump
that's a lot of jank there
and while yes what you do have there is a loop it may not be a very stable sort as what we're seeing is starvation
did you double check the fuel producers? they should be clogged right?
kinda considering reworking the whole fuel part of the factory lol. Moving the packagers in front of the refineries, re-doing the pipe loops for the generators and so on
not clogged, but one or two will sometimes sit on a cycle of fuel before it gets drained. Not enough to stop production, but enough to not fuel all gens completely
yeah that's a clog. any stutter in their production
I'm of the opinion that it's fine to have a less than 100% efficient fuel set up early on. Most people have a bit of a broken first big fuel set up
Learn from it and use what you learned on the next fluid set up
right now it's stable though, will do some more testing tomorrow (past 3am here)
ohh, this is the 3rd try xD
the first two had so much bigger issues, in the first one i belted the packagers wrong and all canisters ended up at the same group of packagers. Second was better but not aligned to world grid because i forgot
fair enough! pipes aren't that bad, there's just more and less stable ways of doing things and if you pay attention and ask around a bit it's easy enough
imo avoid the world grid, all it does is restrict your design choices and forces your buildings to be all at 90 degree angles
yeah, normally pipes aren't that hard for me either. All production chains so far run on max 600/min pipes and its fine, just the gen fueling is getting me for whatever reason π
managing height changes is a thing, plus the more machines on a single manifold seems to make it less stable
yeah, i was also considering just splitting the fuel loops into 2 300/m loops in MK2 pipes
but not sure if that would help at all apart from reducing manifold size and giving me more room to compensate for sloshing in the flow rate
gens are still stable though, even if some pipes are sloshing from basically empty to full constantly
also kinda curious why one loop is working perfectly fine and the other one is not
probably just luck
There is a bug with junctions at the moment it might be that but who knows
what bug?
I think it's something to do with junctions that you snap onto pipes
then it shouldn't be an issue with the gen manifold (maybe the production pipes below). Gen junctions are all placed on the ground, all production junctions are snapped on though
anyways, imma do some more testing tomorrow, for now its fixed though. So thanks a lot!
From what I have seen junctions that you place onto a pipe have slightly different properties to junctions you place and connect pipes. Most of the time it doesn't cause any issues but sometimes it can, it impacts pumps and the hated valve.
Take care and ping me with the update
will do, cya!
snapping junctions to pipes can jsut allow for really tiny pipe segments to exist which can do weird flow, it's best to always rebuild pipes if you snap a junction on them
What's ya'll's go-to method of dealing with water byproduct?
feed it into refineries that only use it
just spent an hour figuring out how to make 420 canisters/min
i bet i can make that work
all you have to do is figure out the ratio and clock things right. It's the more reliable and robust way to manage waste produce
and you can do something similar for all the waste cycles in the game
dark matter residue is a bit of a outlier as I think depending on the recipes you use and production goals you can wind up with more residue than you use, but then you can just make crystals with it and sink it
Do pipe junctions have to have mk2 at all connections to work at 600/min? I seem to have having flowrate issues well below 600. Main line is mk2, but refinery is being fed with a mk1 because it only needs 45/min.
Is the mk2 the pipe having flow issues?
No, it's not peaking at 600
I'm getting spikes into the 400-500, then it drops back then to much lower, then back up, etc
Almost flat, so it's not a headlift thing
Maybe 5-6m at most
And yet, the ones on the right aren't getting fuel
And on the other end, I've got blenders making diluted fuel that aren't hitting 100% because they keep getting blocked up
Yeah suddenly it's just close to solid 500-600 min adding a pump lol, whyy lol, it reports a headlift of only 3.2m
Thanks for the tip. I'm really confused why it was necessary though.
np, fluid dynamics are weird in this game lol
I set up a VIP for the process, so the recycled gets used first. Then set the extractors to only provide what's left.
12 refineries, each 30-45 fuel/min (some are overclocked). Should work out to exactly 400 fuel/min required. Diluted fuel blenders are making 640/min.
VIP is amazing
Yeah, now all the refineries have maxed out fuel. Really confusing why a pump was needed when headlift was negligible.
Some people have PTSD when you mention recycling
My best guess would be pumps provide pressure that push liquids to a low pressure area but who knows
Anything anywhere close to 10m, becomes iffy for export out of a machine.
Maybe the drop down into the refinery causes a dip in the pressure
We need a fix it slide deck on how fluids work
Could be it's preventing sloshing back down towards the blenders, even though they can provide 10m of headlift?
Like I'm right above the blenders in this pic
I have all of my pipes set at 3 or 4 stackers high, then drop the fluid in.
The pipe that directly connects to the refinery, has an easier time if the pipe can hold at least one use worth.
It'd be handy if pipes had an efficiency % when you open their UI. See if you're shoving too much through them
I imagine it also minimally helps with performance, because the math on the pipe only has to be done once.
What do you guys mean by VIP?
I forget the acronym, but if you set up a pipe in a certain way, it uses the fluid from the lower pipe first.
There a reference for that design?
how would a pipe know how efficient it is?
it's a less reliable option to keeping fresh and waste seperate
Not sure on the accuracy of everything in there, but the VIP seems to work.
I guess like it's max capacity vs how much it's getting through it on average.
the max capacity is always either 300 or 600, and it shows the flow averaging it every few seconds. The real way to test the flow of a system is 'are my machines working properly' . Just look at the lights
Yeah I guess what I want is for the flow to be averaged over a longer period of time.
well since headlift is after a pipe is filled, maybe its volume filled and flow rate in some sort of combo %
Because it spiking up and down constantly depending on whether a machine just instantly slurped a chunk of pipe for its next cycle makes it hard to judge
I guess what would really help is if machines consumed fluid at a constant rate over the course of a cycle
I don't think that would help either unless it's moving at absolutely max capacity as your snap shots will keep changing
like if you need it to be exactly 568 pm you'll always be plus minus
just faster and more accurate to look at your machines after you flood the system
are they starting to starve?
no? flow is good
yes? flow is bad
Yeah, I mean none of the recycled plastic/rubber refineries are shutting off now, so I guess it's good. Still really miffed that the pump is needed, like I'm glad it works, but y tho lol
depends on the whole layout.
sometimes a pump is helpful to keep directionality
also the longer the manifold with more machines on it the more points of instability
Wonder if that means a valve would fix it the same way?
sometimes it needs a little push
unlikely, it doesn't stop back flow
Cobalt hates valves
Valves are supposed to stop backflow, aren't they?
valves have the issue that if the system isn't fully flooded it only lets X% of the flow through
They do but not in they way you think they do
they stop specific units of fluid from crossing back, they don't stop those units of fluid from knocking other fluid backwards
Oh weird. That sounds unintentional. So a pump does completely prevent backflow?
pumps, afaik through all the testing, do not allow knock back
Huh neat
Also sidenote, mk2 pumps are >>>> mk1 pumps just on appearance. I use them even when I don't need 50m of headlift lol
that's a hard question to solidly answer.
Yes. Unless your piping is truly a mess? but then is the pump allowing itself to be completely clogged up or is the layout the culprit? I'd say the layout
I imagine you'd still get back flow behind a pump if everythign in front of the pump is at a standstill
but that'd be less the fluid in front knocking it backwards and more that the fluid behind the pump hits a wall
see, I like the industrial look of mk1s xD
I wish mk2s were mk1s but with pumps on both sides. Like an old style roadster car engine
I think I'd like the mk1s better if the pumpy things stuck out at the same angle the power connector is at
Like this is mostly symmetrical-looking if you put two side-by-side
boo the image on the wiki doesn't show the power node on the image
and I don't hate them, there's just no real benefit to using them and a lot of issues that can come up with them. The few times where they possibly have a use, they're always far more finicky to set up and there's always ways that are far far more reliable.
What I hate is people just spreading around false info to people. Cause I then see those people and fix their pipes.
Yeah after finding out about the weird precision they have with values I only ever use them to prevent back flow at 600/min flow rate now
They however donβt prevent backflow
powered pumps do, valves do no
with valves water can hit the front of the valve and knock teh water behind the valve backwards.
and since all we care about is flow in systems, not specific units of fluid, that's the main issue
if you ever want to keep directionality? powered pumps are a much safer bet. If you're using a valve that way chances are you didn't need it.
There's a few niche systmes people have made that use a valve and a buffer for waste water but they are very fiddly and not super reliable
π«‘
i think there's a few other uses in shaping the flow of uneven manifolds, but such things are special case at best
i still feel that the buildable should just be removed from the game to prevent confusion
i.e. we've all had the experience of unlocking pipes for the first time and think they work as described only to find that they don't work as we thought
I thought valves were begrudgingly added because otherwise people were using unpowered pumps to achieve part of the effect?
i think they may have been added at the same time with update 3.5
i gotta say, they're really sort of a redheaded stepchild in the game
Didnβt know you could have expressions in the valve textboxes lol
Wonder if thereβs any reason to even use them in that case
probably to enter an equation like you would the search bar
even then - they only allow PART of what they are set to based on how full the pipe is, which is an issue, and should they be in perfect situations like that? the flow is controled by the machines consumption that you clock however you like.
Where they can be nice is limited overflow,
e.g. on a 600/m pipe, you can loose 50/min without issues to other chains, so you valve 50/min
so if you say had a buffer before your fuel packager, it would not steal everything
once it's full it wouldn't either? and you can just have a hump in the pipe as an over flow
which would in fact prioritize the machines
zero needs for a valve
And having a buffer in line with a manifold, even with a valve, can create flow issues .
One of the reasons to have buffers nowhere near a manifold
(personally I have 0 valves in my 1.0 save, :D)
What's the most efficient way to sloop SAM? Doubling the Reanimated SAM?
your after same for trigons right?
Ya
Huh, Converters have 2 sloop slots, not 4
so it doesn't really matter if you put the sloops in the converters or the constructors
yeah, pretty sure most sloop efficient by doubling the constructors
ficite ingot is next best
I checked my math by doubling the SAM to account for the sloops
trigon is pain π
which threw it off
anyone down for double checking some ratios?
5 RF @ 100%
10 TBF @ (2/3)%
5 HOR @ 100% or (10@50%)
1 DF @100%
2 HOR @ 62.5%
why not just toss it all into a planner like Tools to see what you get
it spits out how many machines you need for X end product but that doesnt translate well into ratioed cells for mass production
at least not as far as i can see
like I know I need 40 total blenders for rocket fuel, and 53.333 TBF blenders
8 DF Blenders
60 HOR refineries
10 Petro Coke Refineries etc
but for ease of building sub cells to make fluids not a beast,
5:10:10:1:2 under clocked seems to make more sense
ratioed cells?
if it says you need X parts per min down a path and your belts can do Y parts per min you know you need X/Y belts at least tand you go from there if you want to adjust it
of course it doesn't break it down into groups as there's an infinite number of combinations
use a planner to 'check your numbers' and then break it down into groups as needed
your cells are like a subsystem are they not? just take the expected production of one such cell, and plop that in the manager, then build it 6 times, or whatever
so the production for Rocket fuel,
1 RF blender to 2 TBF Blenders being supplied by 1 (or 2@ 50%) refineries, make nice little subsets and are a nice clean ratio, no need to mix long fluid chains.
5 of those cells, supplied by one diluted fuel blender, being supplied 2 HOR at 62.5%
fluids sub sets, belts systems i could care less about, those dont have weird issues with back flow and too many splits or merges being weird.
Hi, need help with those damn trains again. Made a single line train for 3 stations and a 4th one for the factory. I did get it running but for some reason it dosent run anymore by itself after adding a few more freight platforms. 1 or 2 stops then says station unreachable. As for stations directions I did <A <B <C D> which was running previously
Set it again to be <A <B <C <D, still dosent work.
Can anyone help?
Open a #1038092680493801533 thread with screenshots of each station
is the timetable A-B-C-D?
yeah timetable A-B-C-D
Its stuck at A
super annoyed, it was running just fine with 3 fewer platforms. and it can be manually driven so no track issues
if it is a single line, try <A B> C> D>
no idea why <A <B <C D> worked before, but it works when it is DCBA
That would work if the timetable was DCBA
Okay I have it at <A <B <C D>
I think I got it to work with A-D-C-B timetable
that'd work yea
Do the first two need to be opposite and rest as is?
the rule is that a train can only enter the station from the back
is this a single rail line?
yes
but it can leave from the back (*if the train is double headed) or the front
ah, interesting. Thank for the tip mate.
sure you dont want a double line for more trains? its quite the distance
first train so kept it simple, might redo it later down the line. It picks up 7 resources. For the most part its fine but had to add 4 extra stations for similar resources to compensate for the single rail
this seems pretty comfy
it can eat one 600 HOR line completely when at 2.5x
the rest of the oil I plan to send straight to coker
also when @ 2.5x the turbofuel is exactly 42/min
which sets me up for Turbo Diamonds
nope, stuck at B. Why tf dosent it go to A
the train is double headed I assume
yes, in opposite directions
why tf indeed, I admit I have no idea
the only thing I can think of is some sort of rail connection bug but it went from A to D, it ought to be connected
and I assume you did not use blueprints anywhere while building the rails
no, only platform base and support bases were blueprinted. Rails and platforms were manually made
it super weird I had it running last night with a table somewhat like this
added 4 stations to all 4 stops in the same directions. Gets stuck randomly now in either A or B
you are not powering the stations manually, right
if the rails are connected the station will power via rail even if you are not connecting cables to it
yeah checked all the stations, they do have power. This is station B where its stuck at
(Station A is toward the left and C is toward the right)
and this is the rear of station B
retrying again with this, rebuilt the train, docked it at A and ready to go
Works finally smh
idk why. I feel like a software engineer.
facing a small issue here. From machine A and Surplus storage B, have 2 conveyor outputs of the same resource going to machine C. I want it to pull from machine A first and then if it needs more it would pull from surplus B.
How would I set it up?
Cuz machine A is a refinery so dont wanna get it clogged up
set two machines C, one that exactly eats what A makes, one that handles excess from B
here, i've been trying to figure out what the train pathfinder's tie-breaking criterion is, assuming that floating point imprecision isn't causing identical paths to have different lengths. it doesn't seem to be random, trains always take the same path between two paths of apparent identical distance; my best guess so far is that it takes the first path that was built
must be FP then
Update: The pump has created a pipe segment so small dismantling the pump crashes the game, bruh
scim to the rescue!
might be able to dismantle it if i remove the pipe piece first
and its the ags test save anyways
First come first served. The game can always find a train that came "first"
Or do you mean the A* criterion for which path they set?
I assumed you meant "which one goes first at an intersection"
dont forget to report it
looks like i fixed it now though, added a second loop above the first that injects fuel at the "ends" of the manifold loop. Input pipe is now always at 600mΒ³/min. Will monitor the packagers a bit to see if they clog up again, thanks again for the help
will do, already send the crashreport as well. Will keep a backup of the save as well
this: i have several paths with the exact same distance (not accounting for floating point imprecision) because i'm building on a square grid, and i'm trying to figure out what the tie-breaker is, which is possibly an implementation detail.
How can they be identical
of course, that's assuming there's an explicit one: if the dijkstra / a* underneath just returns the first path it finds, then it's down to implementation details / memory layout
yeah, i've been trying to figure it out with test cases like this, but i am not convinced i have the answer.
probably a quirk of the path-finding algorithm and world coordinate system
I assume rail build order doesnt affect it?
i'm sure the game has an internal stack that decided what gets executed first, it might be based on location or an ID number that was assigned when placed.
the reason i'm asking is: since i'm building a square grid, there'll be times when there will be several "identical" paths: knowing ahead of time which path will be selected means i can predict where the congestion will be and plan my stations accordingly ^^'
like a game of GO it is impossible to know the state of trains at any time in the network
as trains load and unload you will never have the same state twice as the network grows
What's your opinion on OC supercomputer recipe? I kinda like it, because you can sloop it for twice less sloops than the other recipes.
Also the yield per machine is larger
Although, you don't really need that many supercomputers. They're mostly used for building.
Superstate looks the least attractive to me. It needs batteries and control rods
I don't have any incentive to make batteries for anything else. And control rods are mainly for nuclear, which would be a dedicated chain.
it maximizes the production iirc, which is all that is important to me
I know it varies by travel time, but is there any sort of average rule-of-thumb for how much throughput I could expect from each train car? LIke 100/min per car or something like that?
Let's say over a quarter-map trip
Also is locomotive+4 cars the meta? 3 cars?
Check the wiki freight car page
!wikisearch freight+car
Freight Cars are non-motorized vehicles that can be attached to Electric Locomotives or another Freight Car to form a train. They have a capacity of 32 item slots or 1,600 m3 for fluids and allow for the transportation of resources over Railways. Freight Cars can be loaded and unloaded via Freight Platforms...
If your train arrives before your freight platform is 50% full, its usually full throughput.
You cannot run 2 full belts per platform, but 1.5 should work
3 stacks/min per car is the "you don't have to think about this, it'll just work" bar IME
Yeah thats about what i go for
if you need more, then you want to actually start doing math
On the higher throughput lines, ive put large storage containers in/out of the station so that the flow of items doesnt stop while the train is loading/unloading
Which level of full belt are we talking about here?
Any belt you hook up to it
belt level is usually a worse bottleneck than trip time in trains
I would ISC every station, it's basically free and ensures a steady flow of resources
Whatever you do, the freight platform cannot support 2 full belts (of any mk) of throughput.
But if you use mk 6, then its like 1800/min best case which is a lot of lower tier belts
The reason for that is 27 seconds of docking time
Oh is there something about the platform that means it can't accept two full belts at a time? Like even a mk1?
during which you get no flow of items
Ahh ok
yea, the animation is basically picking up the entire contents of the freight car and placing them into the station all at once, rather than at a steady rate. While this process is ongoing, no belts can extract, since the items "aren't there yet".
of course vice versa for loading
So would two full belts be possible if there's an ISC before the platform with both belts running through it?
no, no matter what, you lose 27 seconds of transfer time every trip.
The buffer just reduces your losses
the ISC will slowly empty
Ah ok
the buffer is there so that you still get your < 2 full belts demand met during the 27 seconds without relying on belts alone as buffers
The wiki has math for this of course
But from what i remember, 1.5 belt is about as much of a rule of thumb as i could hand out for this
So instead of like 2400/min for 2 x mk 6, its 1800/min
Ah nice. I'm only going to be dealing with 480/min of plastic and rubber each, so it sounds like I can probably get away with a single freight car for each.
yea, that should be fine unless the route is egregiously long
It's just going for the northern coast to the titan forest. Might be a bit of a winding route, but it's not going across the map or anything.
Just hook the 1 mk 4 belt up to an ISC and the connect both belts between station and ISC
Oh that's a good idea - to make sure the ISC empties into the station quickly.
and vice versa
I'd recommend doing the math, it's a lot more reliable and rather easy
Its a rule of thumb.
If you wanna do the math then like i said theres the wiki page
transport the oil and produce there preferably, oil to end product is a 1/3 decompression ratio
edit: actually never mind 480 is not that much, run 2 cars on a loco and you should be fine
Haha yeah too late, I already built a huge oil rig on the northern coast
Would you like water like this in game ? https://matthias-research.github.io/pages/tenMinutePhysics/18-flip.html
Vote now ! https://questions.satisfactorygame.com/post/670c8ee7ddb9d97e071ff742 
the existing semi-realistic fluids system is not fun, making it more realistic won't make it fun
I think they should simplify the system in favor of fun akin to Factorio
Iirc Factorio used to even simulate static pressure, so you needed pumps every so often just to keep things flowing on flat ground.
what would you like to see in a simpler fluid system
i havent played factorio so idk how it works there
But yeah, removing the slosh simulation and making machines produce/consume at a regular rate rather than in batch chunks would be handy.
Yeah I had to add a pump that only reports three meters of headlift just to make my diluted fuel->recycled plastic/rubber system work, otherwise refineries at the end of the manifold were starving. Makes no sense at all.
do pumps work on gasses? it feels like they do, but i dont have any actual evidence
Gases don't have headlift, so I'm not sure
would be weird
Like, diluted fuel still makes 100 fuel/min, but it'd be added to the pipe system at 1.667/second rather than 10 every 6 seconds.
I'm mentioning also hotfix rename stuff (at least dont call pump a pump) and add a buffer with bottom In/out solves a lot in current system
If my example give them some help or insights then maybe they look to be great at optimizing (looking at how much they are able to represent on a screen)
maybe they can simulate 3D volume and create transparent pipes - for like 20 tickets in MAM as punishment for skill issues π
GPU vector units can simulate this stuff crazy good (maybe complicates Dedicated server stuff) if you have good base. They can make molecules bigger and other programming Tricks (SIMD) , DOD
https://www.youtube.com/watch?v=IroPQ150F6c
other examples of optimizing
https://www.youtube.com/watch?v=U16RnpV48KQ
https://www.youtube.com/watch?v=pSvSXBorw4A
Copyright: Belongs to Handmade Seattle (https ://vimeo.com/649009599). I'm not the owner of the video and hold no copyright. And the video is not monetized.
In this video Andrew Kelley (creator of Zig programming language) explains various strategies one can use to reduce memory footprint of programs while also making the program cache friendl...
7 Steps it took to make an algorithm 1,606,242% faster!!!!
Become a backend engineer. Its my favorite site
https://boot.dev/?promo=PRIMEAGENYT
This is also the best way to support me is to support yourself becoming a better backend engineer.
David Perez
https://github.com/david-a-perez
B3NNY
https://twitter.com/_B_3_N_N_Y_
Bi...
Dave tests almost 100 different languages to find the ultimate champion in generating the fastest code. Feeling a little bit autistic? Check out the free sample of my book: https://amzn.to/40HUwgu
Primes Code on Github: https://github.com/PlummersSoftwareLLC/Primes
PrimeView: https://plummerssoftwarellc.github.io/PrimeView/?rc=30&rs=0&fr=&sc=d...
Would make the pipes less surge-y.
would need like 95% of production-related code to be redone
does anyone know what the delta speed is for trains to derail vs halt?
Anyone have a preferred alt for canisters? 30 plastic/min seems like a bit much. Maybe Coated?
iron and copper one is pretty good
its faster than the steel one, although steel one is simpler to make
do you need so many canisters? if you pack it for transport an personal use, 30 plastic is not a lot. if you need to transport packaged fluid - just make a big batch and loop it to reuse it
It's mostly just for jetpack, maybe drones at first once I unlock them, and also thinking maybe 30/min (or 60?) eventually for the fused modular frame alt recipe.
id make this temporary because ionized fuel uses aluminum cans, and thats kinda the end goal wiith fuel
the main thing I want to package are oil products, at the site where I am producing them, so I find the default plastic simply easier because it doesn't involve shipping in any metals
Would the batch loop involve a train with a car dedicated to empty canisters?
i like to have half the train for full and half for empty so i essentially cant run out
Yeah that's what I'd be doing, siphoning off some of my plastic to package the fuel made at the same place. Maybe I'd just underclock the canister constructor.
if you pack it on the spot - yes, you'll need 2 cars. if you decide to transport liquid (why would you do that?) - one is enough
I might transport the fuel as liquid just because I can, and trains with tank cars mixed in look cool π
In which case I'd only need the canisters for jetpack and maybe vehicles until I get a better fuel.
like package and load to car 1 - deliver - unpackage and load to car 2 - return
although Residual Plastic is a worse yield than Residual Rubber, it's handy in a pinch for getting fuel canisters in clean numbers with minimal steps
though if you're working with simply fuel (the orange one) - if you need to burn it, do it on the spot, if you need to deliver it - make it so you don't have to reuse the canisters, like burning it in drones and tractors
I'm having train troubles
this also eliminates the need to reuse canisters
Im trying to accomplish this without collisions
I thought it'd be as simple as blocks, but its notwokring
signals should work perfectly for that
whats your signal layout? maybe youre missing a bidirectional block
This is the first time I've had more than one train, so im brand new to signals
i just put a block at the front and back of station a
I can see the train going from bottom right to D and then trying to go back to the bottom right
at minumum you need one more block than you have trains
i forgot to illustrate, this all moves clockwise
ok
so my first thought is the merging of the rail to D.
a train may come from the bottom left corner into d, and then try to exit back to the bottom left corner
so far that hasnt happened
if that rail is one direction, then the train will say signaling error
are the trains not moving at all?
waiting at signals that are occupied?
without any signals, it moved like it should
I think this would be the bare minimum needed?
but the crash is inevitable
whats the direction of the signals?
lemme draw
np
the section with Station A is the only place collisions would happen
so a basic solution is to make that into one block; Train 1 will not enter the shared track if Train 2 is using Station A, and vice versa
(to be more efficient, you can then chop it into sub-blocks so the trains can proceed further in before they have to stop and wait)
the two bottom corners switch direction according to the timetable of the passing train
Ok heres my idea: More blocks is (almost) always better, it keeps the trains moving
er i guess the bottom left corner
its especially important to put signals at the beginning and end of train stations
Make sure that any signals on that bidirectional rail have block signals on both sides of the rail
oh, do the trains enter the station from opposite directions?
that does add complexity and does basically mean you have to make that entire bottom rail its own block
They dont
oh, I'm not sure what you mean by switching direction
just as in, an ordinary junction?
yeah
id at least add signals there
red lines are blocks?
signals
I added arrows to clarify
I'll try this real quick
theres not much reason to be conservative with block signals
the bigger a block, the less trains that can be on it
well, laziness is a reason
so true
just not a good one
for that one bidirectional rail, it needs to be its own block. Having a train staring contest is not a fun block to clear
It's only a concern if more than one train uses station D
Hi
hello
i have a problem with train a rails, can someone help me
sure
is there a voice chat here?
i cant vc (definitely not in class rn or anything what are you talking about)
ok
but is there any vc? so i can join
no
IIrc they used to have a voice chat here, but it was really hard to moderate and people were being rude, so it's gone.
can you help me with my problem?
Is that ok ?
Sorry, I haven't done trains yet in my game
i am tired of this
if it fits what you need, then its good
ok thanks
whats your layout look like?
thats why i want vc
ah
will be imposible without sharing screen
Can you draw it like was done here?
the sad part is really simple and i am trying to learn how this works but you cant do anything
if you can draw it, that works otherwise there isnt much i can do
theres no signal that a train entering that block can use to leave
what xd
id put a signal on the other side of the track too
add a signal on the other side of that rail
wich side?
hold on lemme open paint
so your rail is bidirectional meaning you want trains to go both directions on the rail
so you need your signals to support that
the red lines should have a signal on both sides of the rail
nvm actually all signals should do that
so....
signals point in one direction, so put one on both sides of the rail for both directions
what, i dont get it
Just like in real life, train signals only shine a light in one direction. You need to add a second signal in the same place on the rail, but facing the other way to the existing signal.
yeah
i have no idea how trains works neither in real life
sorry im busy i cant call if i wanted
i will try this
look up a video about this
no problem, but vc is easy just share screen
there are plenty and theres probably one that makes sense
You have something like this. A train coming from the opposite direction can't see the lights we see in this photo, unless there's a signal facing away from us too.
now this happen
you have different entry signals
train move but stop again
you have some path singals and some block signals pointing into the smae block
open the signal and read what error it gives
Did you add signals on both sides of every signal you have? Even beyond this intersection you've shown us? If you're doing bidirectional rails, you're going to need bidirection signals in every place.
this may or may not help you... now & then when i'm building rails in configurations that should work, the signals just get stuck in a weird state and i need to delete a bunch of rail & rebuild/resignal it to get the signal states to reset. I suspect that you probably have some of that going on, but probably more what your problem is that you are doing things like crossing bidirectional tracks over themselves
i did
Not train stations, signals. Every block signal needs to face in both directions.
Again, I haven't done this in my game yet, but this is my understanding of how they work. I'm not planning on doing bidirectional trains though, just going to make two rails, one for each direction.
which is much simpler to reason about
bidirectional tracks are very difficult to wrap your head around even in very simple topologies
the train stuff is so anoying
Bidirectional is much harder than single-directional
there's truly a reason why most real-life railways go with unidirectional tracks
i have no more warning blocks signals but the trains still dont move
sometimes you have to hop in them and drive them manually a bit before re-enabling autopilot to kick them into moving
idk why that is, but it happens
PLEASE stop using bi-di. esp on your first playthrough, it needs to be absolutely perfect for them to not shit themselves
is anoying as hell
what part of it is annoying
everything xd
its pretty simple, take your main line, branch off 2 rails into a station manifold and loop em back around
path signals wherever two trains may occupy a junction simultaneously
block signals otherwise
can you do vc?
why people is afraid of doing vc you dont even need to talk
really hard with just pictures
im not home rn, can't vc very easily
why are people afraid of sharing pictures
Sometimes people aren't on their PCs, so they can't voicechat
its weird and dumb (though looks cool in action), but a "shuffle balancer" like this would technically work, right?
in a situation where either the output demand is unknown or variable, any input can reach any output, and if any output backs up the corresponding input will just overflow into another line?
there's no situation in Satisfactory where input or output is variable
there is when I put temporary/unfinished factories up that may or may not be using everything on an input line
so what is actually a sane and reasonable amount of Phase 4 elevator parts to make per minute?
or rather, say in that example i removed the middle output - the other 4 would still get an even balance
Assembly Director System / Magnetic Field Generator / Thermal Propulsion Rocket / Nuclear Pasta?
need 500/500/250/100 of each... for now
in that case just build for the permanent state and ignore unfinished state
i thought maybe 5/5/2.5/1 might be reasonable but... 4600 copper per minute is daunting xd
Make 2,5 and upgrade your power generation during those 200min
2.5/2.5/2.5/2.5?
Half your number
Iβm only planning on making 1/min of each whenever i get around to it, you donβt need a lot and having it trickle in will get you there eventually
is this valve done correctly? the machines on the right produce 400 wastewater, the machines on the left need water and are also connected to my main water supply.
I just want the valve to not allow back fill left to right.
this intrigues me tbh
I would have never thought to build something like this but it's creative problem solving imo
though I don't think the throughput is great in the worst case possible
i.e. all but one inputs off, and all outputs off except the farthest output from the input
Which will I likely need more of in the long run, plastic or rubber? I am planning a large oil refinery, and am wondering if I should prioritize one or the other?
have you decided on what alt recipes you want to prioritize? that's your answer. if you didn't - maybe consider making equal amounts of both?
Really depends on the alts you use.
That's been my main point of contention. I'm not nearly as familiar with the alts for later game items than I am for the midgame, but it seems that on a whole the base recipes at least use more plastic
caterium alts often use rubber afair
i usually go with equal amounts
the overflow will go into sink anyway
the underflow will just receive one more factory
Sounds good
Plastic is used pretty regularly throughout the tiers, rubber isnβt used much at first (except in alts), but comes more into play later on
Equal amounts of both is honestly the best way to go
yeah start with a mix then oversupply what you feel is best for your alt choices.
I knew I wanted rubber earlier on so my first loop was 700 rubber/200 plastic.
I think plastic and rubber usage equals out as you get further down the line
dont use valves
so 2.5/2.5/1.25/0.5?
valves for wastewater are fine, (seeing as you want it to be able to drain without being blocked) but if you want to achieve 600/min flow without spending ages and having many headaches you also need to create a height difference after the valve (and add a pump if necessary too). The height difference, made by simply putting one pipe that goes a few foundations/walls upwards and another one connecting it back down to the ground, allows the wastewater to more easily flow elsewhere.
but can't you just reuse the waste water?
A wastewater return line is how I use it as well. To stop the main line from line from filling up the return line.
yeah, you can reuse the waste water by plugging it back into the main supply.
can result in issues, but most of them crop up from having too much water flowing in from pumps or otherwise
i mean it seems to work for fuel
there is a chance it could work. Direct feeding waste water with a valve is the least reliable option
i have to start doing aluminium production myself soon so yea..
whats a good spot for that?
i was thinking of the ones near the oil on the west coast
this is not accurate - valves for waste water is like juggling plates instead of carrying them
generally you want to be near bauxite, which tends to be in the middle segment of the map (in between north and south)
i can then have it piggyback on my trainline that im already using for plastic and rubber
and i already built a bunch of stuff on the water there anyways
I use valves for wastewater simply so that there is a spot in my waste lines that remains empty. The height difference and pump sorts out any throughput issues
i was thinking the 2 bauxite nodes here
yes but these are highly unreliable set ups that are prone to failing unless you follow a very particular shape and layout.
things people, when people suggest valves, don't show examples that dcan be copied exactly
i can get the geotherms while im there too
which is why 'fuck valves' and the people who suggest them because then I see a bunch of people with broken pipes since all was suggested was 'use valves'
you mean returning it to the same refineries? this will work but only if your refineries never ever stop. the moment one stops - pipes may get flooded with water and you'll have to manually drain it
how is that a bad thing?
Valves are not the be-all end-all solution. I've found they mess up stuff almost anywhere I put them except wastewater. Which is why I use them there
much safer is to make a bunch of refineries working solely on waste water and a bunch of them working on extracted water
i mean with crude and fuel you flood your pipes anyways =|
They can be a solution, but I'll give you that they are 'more' of a problem than a solution.
you can get by perfectly fine without valves, I just feel more satisfied if my wastewater removal has a spot that remains empty for more to flow in
That's the only thing I use it for as well.
except by habit you probably have an extremely specific layout where the direct feed works.
a pump is a more reliable solution at that poitn but not much better
my main issue is people say 'use a valve' w/o also going 'this exact set up works' with a diagram and images
because if you don't do that you need to also add
'also this is like 50/50 it'll break and you could spend half an hour pointlesslly trying to make it work'
Yeah
say you have an alumina to scrap refinery, which makes waste water. it makes the maximum amount of scrap and waits to be emptied, thus stopping to produce water. meanwhile the extractor keeps working, filling your pipes, and by the moment the refinery starts working again - it can't output wwater because the pipes are full
you have no idea how many people show up with broken systems cause someone said 'just use a valve/buffer, it'll be fiiiiine'
thanks, guess ill have to start working towards that stuff q_q
ah yea i guess, how do people get rid of the waste water then? just using it in a different process like diluted fuel or something?
as I just said - one section of refineries works only on extracted water, the other part - only on waste water
no, just use it in a dedicated part of the refinery system
blue line is fresh
red is waste
unbreakable and reliabe
^
this is why specifically wastewater is what I use valves for. Anywhere else it creates far more problems than it can solve, but wastewater is simple enough with simple enough requirements (just don't let a machine fill up) that a valve can be used. Not completely necessary, but it can improve some setups
we just showed you an image with the example
if the problem is the pipe filling up so it can't continue producing
you recycle it back
depends on the process, but basically make sure that the amount flowing in from pumps + the amount being made by machines = the water you're using
It let me make a base wide return line for waste water, instead of each set up having a different one.
in other words, plug it back into the water supply
except its simple for you because you build in a specific way. I help people every day with this sort of thing, including broken waste valve set ups they've done
there's this jack ass who made a video recently espousing this damned valve buffer system for that that keep failing for people
the pipes here will never back up
becaues it never mixes with the fresh water
ah so you guys just use a mk2 pipe with 150m3 water (or whatever the number is) and under/overclock it?
I honestly do not know what other way people build wastewater systems that is similar enough to mine to have my thoughts applied but different enough for it to break.
oh no i see it
it depends on the ratio? just need to clock the groiups right π in that example I ahve each solution refinery feeding 1 scrap refinery
I don't recommend messing about too hard with valves but if you add one to a system and it works, it works
if you're having trouble, remove it
People use it for a lot of things that don't work, other than waste water.
I mean sure but if you're going to suggest direct valve feeding, which is possible but needs to be set up specifically, pls give them an image and layout of how you do it cause it's very easy for them to fail. And tell them to follow it.
it's like setting up a VIP junction, there are other ways than the image in pipe manual but it's super easy to fuck up if you don't copy an existing and tested sample
whatever i don't get it
I'm at the end of phase 4, and have used all of 1 valve.
again, 2 choices, taking alumina as an example:
- you place a bunch of alumina refineries - make them feed on extracted water - you route the alumina solution to scrap refineries - these refineries produce scrap and waste water - the waste water is being fed into another, separate bunch of refineries.
- you recycle the water from scrap refineries back into alumina refineries as usual, but all your refineries have to work 100% of the time, otherwise the extractor would fill the pipe and the waste water would not have space to come out
is direct valve feeding where you have a valve right before every fluid input in a manifold? Never seen the term before
ok this is both an image of the set up and the top diagram with the numbers are what they are clocked to. What is the issue?
this processes 780 baux pm
If you have a vip line set up that doesn't backflow, that is not the case. Because as soon as it starts up again, there's room.
direct feeding is what you desribe just slapping the waste pipe into the fresh, usually you want to use a pump and make sure there are NO repeating decimals anywhere near it
Also, aluminum would be a bad example for me, because I wouldn't ever stop it.
someone said something stupid and its confusing me
prolly me lol
so im just gonna go away.
If you did stop it, you are inches from the pipes, and can flush them.
imma defend the viewpoint that valves have rare niche uses but the very first step to solving fluid problems is removing your valves
Agreed
bilbo is right
I was gunna mention, the one guy showing off that weird extractor thing with the backflow. (but I was eating)
(the sloshing thing)
does the game still run into issues if you carpet bomb the biome?
iirc back in Update 6 or 7 or whatever people generally didn't recommend removing toomany trees etc. because it caused some issues with lag or save size or something like that
finally emptying out the storage after building a plutonium rod sink plant
brought a tear to my eye
idk if this is the right place to ask but why is a hypertube cannon with 12 entrances stronger than one with 13? they are facing opposite directions
initial velocity affects it
ive used both of them many times in the same way
possibly slightly different layout, spacing ect.
they're exploits so really can't expect strict rules
thats fair ig
with finicky things like that the order of operations also can matter sometimes. Like which objects you build first in the system. Plus floating point errors in positioning
in theory a blueprinted cannon might be more similar? but there could be effects from being pointed in different directions
best way to get rid of wastewater when limestone isnt nearby?
coal generators
Are you sure those are tears and not blood
Is transporting liquids via the train a good idea?
I am trying to keep everything as centralised as possible
it's fine, you just need to buffer it properly
it works, yeah, is it a good idea, no
Liquids yes, gases no, but you have to do it right or it'll be weird
#design-and-architecture message and #design-and-architecture message are reliable fluid buffers, save the images and replicate them exactly
Oh those are the ones on the wiki lol
not surprised, Sevrahn did a fair bit of wiki stuff
Why do it exactly like that?
Is it so the liquid does not back travel into the trainline liquid storage
to allow pipes to burst faster than 600/min into/out of station and tolerate the train lockout period
unfortunate how some people are great in some ways not so much in others 
Uh oh lol
Was never a fan of the "vibe checks" and public shaming with the block lists, also the mods thing EVERY convo
yeah essentially there's a lot of little ways that fluid have tricks to them. These iirc use VIP junctions to function?
eh, he was just abrassive, never actually attacked anyone
I know about the water tower one but I do find that a bit cheap to use
it isn't that useful, honestly
yeah the pipe manual has a bunch of technical ways to use pipes - imo you don't really need to use almost all of it. It's more about documenting how pipes behave so you can have an idea of what a set up is doing
good to know what to avoid too
but what they were saying about gasses is true, package them. they behave weirdly in buffers and packaged gas is more compact anyway even shipping the containers back
what exactly are the rules for gas in a buffer?
btw @shrewd oar , you were testing gas buffers right? I can't remember did you end up getting it to work smoothly or was there wonk?
possibly just 1 rule of 'don't'
but I can't remember of Ladyhawk got them to work or not
if she did I'd suggest putting it on the wiki
no wonk at all, 600/m steady
i'd imagine they just fill up completely and empty like long pipes
nice! definitely make some clean example set ups for the wiki π
that's been an issue that plagued the game for ages
What causes the weirdness? Iβm hearing this for the first time, meanwhile my nitrogen train has been supplying my rocket fuel plant with no issues for dozens of hours now.
#1298454344986923039 message that's a working setup at train station
'someone on the net once said...'
and #1298454344986923039 message that's the clean setup blueprint version
I think the theory is that gas tries to equalise all spaces unlike liquids? moving gas by train tends to not get full throughput. Are you oversupplying?
powered pumps don't work on gas lines but powered pumps on a buffer forces buffer to output steady at pipe cap
gas or liquid makes no difference
do you have to have a double buffer?
for a train station 1 small buffer isn't enough to cover the train docking time at 600/m so yeah, you either need 2 small ones or 1 large one
2 small ones just takes up less space than one of those giant blobs so i prefer them
hmm although this one might be a bit confusing, it looks like it serves only 1 train in/output?
you can cut those off
depending on whether you're loading or unloading a train you'll need the 2 pipes on the other side
that serves 1 pipeline at 600/m
but obvs you need 2 pipes from the station side or supply side to refill the buffer during the time the train's not there <<< bad description.. need double station output to refill buffers when train's not there on unload side, need double station input to empty buffers while train's not there on loading side
pumps always face the output side, so station loading pumps face station... station unloading, pumps face away from station
I don't know if its still a problem, but the game does track what you destroy.
rip
I mean it has to - it needs to keep track of the changes made to the default world state
Otherwise all the trees would respawn every time you loaded a save π΄
but whats the issue, just the save file size or lag?
lag while saving π
honestly, a lot of the world state modification is marshalled into the save file asynchronously from the autosave events in 1.0 so it isn't a huge problem, but after a few 100 hrs in the game, the autosave isn't an unnoticeable blip anymore π
ah okay
alright for Nuke waste I got 2 sources which go into my drone.
I wanna fly the active waste first then focus on stored waste.
For 2 inputs to 1 output how do I make it so that the active conveyor is first used by the drone and then the storage conveyor
welp got it somewhat solved by using a mk1 belt
imagine 4 days of work, just to sink them π
How do you usually do Battery Drone Ports? A battery factory with several Ports that sends drones to every other stations, or just 2-3 ports that recieve battery movers from multiple stations each?
@flint rapids Afaik the standard way to deliver batteries (fuel of any kind) is to have a battery production line feeding a drone port. That drone port is empty. You then build drone ports with drones at the destination locations that go to the battery production line port. Ala a many-to-one system.
If you need to send batteries to a large amount of locations, you can duplicate that setup as there is an upper limit on amount of drones that can get filled at one port due to docking time and input belt throughput limitations. Reddit has some good posts on the topic.
Shrink it and use a normal T intersection?
-_-
I don't think there's anything that could be done to improve it
and you probably wouldn't even get bottlenecked by a normal t intersection
this is a pretty high-traffic line that's why
after everything is complete there's probably gonna be 10-15 trains passing through it a min
making multiple directer paths would probably be the only way to improve it, or reducing the amount of trains you need
Make it stacked #screenshots message
Anyone else kinda bummered we cant package dissolved silica?
I was thinking about logistics and it came to mind that it wouldnt be "that bad"
isn't theirs already more efficient?
also I doubt there'll be a train going through that junction every 4 seconds
Not in space and looks! π and you could max out the efficiency on that stacked one easily, just wouldnβt look as good
Need basically the same stuff to make the silica , not sure I could come up with a situation where Iβd want to move the fluid
i wonder how a smallabout compares to no crossing junctions in terms of efficiency
Small about?
a small roundabout
If itβs a roundabout afaik itβs the worst efficient
stop trying to make fetch smallabout happen, it's not going to happen
I mean. Its defo more compact, like 900/min fluid = 900/min packaged vs 2025/min silica.
I mean a small one tho specifically, to minimize time in roundabout
Thatβs not the issue, itβs that all train paths interact
Thatβs why turbines are the most efficient, they only interact with their specific path
the fact that they interact isn't the only thing that matters. how long they interact also matters
that's why I think a smallabout might be more efficient than a regular t intersection, and it's definitely more efficient than a regular roundabout
what about two-lane rounabouts
tgeres nothing better than your setup in terms of possible trains per minute afaik
unless you're trying to squeeze out a few percent more by having the trains slow down less in the corners, but at that point it's insanity
assuming fully loaded trains carrying plastic, 15 trains per minute would equal 96000 items per minute
I personally love stacked rail
which is why I doubt you're going to hit 15 trains per minute
80 belts of throughput π
yeah i probably wont, just a lil over future-proofing here hah
yea at that point the only option is alternate routes. shorter alternate routes.
apart from horizontal footprint whats so good about it
and doing the math to make sure you use as few trains as possible
I think it looks cool, the shorter footprint allows it to snake through the world more nicely
I'm sure @vapid gorge has some nice example pics
Itβs the biggest part though
Two lame roundabouts are basically just 4 T junctions in a row
but look how small it is!
a train barely spends any time inside
also, not all paths block each other
3 trains can use it at the same time
Imo itβs easier for stacked rail to set up under and over passes for better throughput, easier to snake through tight spots. Those are the main points. It mostly functions like parallel tracks mind you
But they still need to reserve xD youβd have to do throughput tests though
I'm using blocks only
- a main line bypass route would make a lot better imo in terms of throughput
but thats a lot of blocks signals
they won't jam? i get uncomfrutable even looking at that
Thatβs what turbine junctions are, just lots of slip lanes
eH, im sure it'll be fine
but now that I think about it, maybe I should actually put path signals on there
I'm not good with trains, cobalt is there a way for that smallabout to jam?
except the 4 bypass ducts are merged into one big circle
I guess if a train wants to turn around, but that shouldn't really happen
The one circle bit though suggests thereβs a lot of path interaction
Some stacked rail #design-and-architecture message
I did make a prototype Stacked Interchange. Looked pretty cool
Could probably make it more tidy , just was curious what itβd look like
You could sacrifice some throughput and make it much more compact though. A half turbine
also, need some input: 4 platforms can do 6000/min with this setup? running 2 locos per node group
Prooooobably not
Especially if the same platforms are used car the shared station
The more trains stop at a particular station the less throughput the platforms have
i did some great planning and have nowhere to expand my station manifold but up ughhhhhhhhhh
Years ago I did this ?? xD
Interesting, single car trains?
Not all, the plan was to move all the raw goods from that spot
Iβve since gained logistics and planning wisdom xD
But still some functional station floors
hmm, we go a) up, b) right, or c) stack them doubly π
D) Nuke it from orbit, it's the only way to be sure
Amy way works really. You can always build the factory above the stations
what i was planning to do
probably going to be 6-7 floors of hell
15 BWD's/min
also planning to bring 1300 plastic and 1650 rubber in via drone, shouldnttt be thaaaaaaaaaat many drones, right...
Drones can carry 9 stacks iirc
yep, 9 stacks.
you can probably pull 4-500/min per drone
Is Nitro Blend Fuel worth it if I A) already have nitric cid production and B) don't currently have a need for compacted coal?
its easy to build. i prefer turboblend+default rf for the better yield, but either one is pretty goodβ’οΈ
as for the compacted coal... there's a number of good uses for that
The real trade off for me is, Oil or Sulphur.
what do you have more of and feel like spending
tbh, i think where it should be utilized is to make turbofuel for diamonds
nitrorocket uses a lot of sulfur and isn't great on coal
I feel like I have a billion leftover oil
I was tentatively planning to just do oil diamonds
yeah, it can be sunk so punting on its use for a while is ok
i started looking at the diamond recipes last night and realized turbodiamond is a heckuva lot better than the other alternatives
for reference I did the alt heavy recipe for Turbo Blend -> Rocket Fuel
1200/min oil got me 2600/min total RF + a tiny bit of extra base fuel.
i realized i was going to be using a bunch of oil and coal for diamonds one way or another and saw that the turbodiamond recipe is pretty good, for less oil and coal than i'd be using otherwise for a lot of diamonds, it ends up being much smaller of a footprint
Better by what metric?
I'm pretty sure it's substantially less efficient than just doing oil diamonds
well, ok, how many oil diamonds can you make with 600 oil?
120
And like, yeah, you can get 2k diamonds out of 600 oil if you use turbo
... but that's also costing you 1k sulfur and 2k coal
i'll let you draw your own conclusion
24000 cool per minute
yeah, big numbers, that's ballpark what i'm projecting to be needing
Yeah the diamond thing is only really a great idea if its 100% based on your byproduct than straight input imho.
I'm just planning out my first train station are three-car trains the usual? Or is it four cars?
Im under the assumption you want to base that on your throughput?
Im not sure myself 
I'm loading multiple things, but not sure if four is too much
I'd like to do four, for plastic, rubber, fuel, and petrocoke, but maybe four is too many?
I mean, if the track is not pushing the limits for an upwards slope, should be fine.
four cars work
i go with 4
you can go up to 5 without needing another engine on inclines. i find 4 to be a better way though - that's more of a 4 being a nice number for splitting & merging more than anything
whatever size you settle on, you're going to be kind of married to it for the rest of the playthrough
any train not that size is going to be problematic in some way, lol
(there's a joke in there somewhere about having an affair & a mistress, but my funny bone is broken)
is there a convenient way to illuminate a large indoor area?
all the light sources ive tried (ceiling lamps, wall lights etc) have really intense point lighting and dont really illuminate big spaces
i dont wanna have to plaster them everywhere
really just the light buildables or billboards mounted on the ceiling
how do you do billboards on ceilings? the build prompt for me is like
hold ctrl
i rebound ctrl to crouch
and C to the functions i saw that used ctrl
but for some reason after doing that i cant find a keybind that works at all for any of the ctrl functions that arent mass dismantle
like the 0.5m nudge doesnt work either
i worry ive permanently unbound something internally π
have a feeling the game just has some very messy keybinding code and a lot of edge case bugs esp when you do rebinds like that
resetting to defaults fixed it, i can deal, it just annoys me that crouching is C
well, it is also the down button for the hoverpack
you probably want to keep that near wasd
rebinding crouch also rebinds hoverpack down, with the *interesting* side effect that using copy+paste midair makes me descend
Yeah this is why I'm being super careful in my initial design, don't want to regret it later π
Speaking of trains, im very glad I swapped my planning over to drones for packaged gas, would save me some logistical nightmares.
here's an interesting little comparison
for what matters, default ECR is cheaper with plastic ai limiter
Do I need to know anything special for a tower of generators using rocket fuel? Ik that gasses behave funny
They should be straight forward
in theory, you should just be able to pipe it however
no pumps needed, loops can be horizontal, and remember to keep the pipes SIMPLE
i haven't really tried pushing gases around through height changes
Ill see if I can get a decent snapshot of the pipes in SCIM
i should probably start working on the conversion for that all soon
Each floor is just a copy paste, and it works. One floor per pipe.
makes sense. i find you get better results in general if you have one level of destination for any pipe
I also forgot how comfy this clockspeed is
my setup for it all is just really basic
Oh I was going to explain this better. 
I have to send the nitrogen not just 2 directions but 3, and if I sent them via train, I would have needed pipes to go up that mega cliff arch, defeating the purpose of packing/unpacking.
I need help.
I have 2 conveyer belts.
One should always be at max capacity, the other one is supplying the main belt.
I tried a splitter going from 2 to 1. But that resulted in line 1 backing up til the end and into the machines and in general lower throughput and low efficiency.
Priority splitter would be nice. Always taking items from belt 1 and then the remaining items from belt 2.
How can I solve this problem?
there's 2 hard drugs i've tried in my life and which I'll consume for life - one is hoverpack, second is dimensional depot.
i've gathered those in a single run w/o returning to base
Are the two belts carrying the same item?