math-and-meta

[vapid gorge]

you probably need even more considering distance and how much 1 car can hold

[thorny root]

@quartz pastureTrain inputs: 2 mk2 pipes. Max input possible into the platform itself: 1200 per minute. But it cannot continue putting fluid into the platform while the train is loading, so it STOPS putting fluid in, all your pumps shut down, and everything waits. Your PER MINUTE input during that DOWN TIME is going down to a number that is based on the number of trains you have coming and going per hour.

IT CANNOT POSSIBLY BE 1200 per minute.

Not unless you have 0 trains loading and unloading.

[quartz pasture]

Even with buffers?

[thorny root]

EVEN with buffers

[vapid gorge]

you'd need the round trip to be 2:40 or less, which is very very short

[thorny root]

The PIPE is limiting the flow rate

[quartz pasture]

Is that only true of gas?

[thorny root]

That is true of pipes and everything they carry

[vapid gorge]

of everything in the game

[quartz pasture]

I successfully port exactly 400m^3 to my turbofuel refinery

[storm imp]

Every item

[quartz pasture]

Ain't no way

[thorny root]

Don't argue with it. Accept it.

It's reality.

[vapid gorge]

this means it's impossible to get 2 full belts or pipes, because the STOP

if you have a belt moving 1200 pm, and stop it for 1 second, it no longer moves 1200 pm

[quartz pasture]

Wouldn't my turbofuel refinery have downtime if that was the case?

[vapid gorge]

there might be extra buffer it's slowly burning through, or you just haven't noticed power spikes

[thorny root]

No idea what it looks like.

[vapid gorge]

or youre only moving 400pm and it's doing fine

but you cannot move 1200 fluid pm on one platform. Ever

[quartz pasture]

Why does the wiki have fluid throughput at 1793.04?

[vapid gorge]

it's unlikely to even do 600 pm

[thorny root]

You also cannot move 2400 items per minute on freight platforms for the very same reason.

[vapid gorge]

are you looking at fandom?

[quartz pasture]

nah, wiki.gg

[vapid gorge]

link?

you might also be looking at the mk6 belt throughput, not fluid

[fallow siren]

always do single belt/pipe to one freight
treat it like how you normally do belts and pipe, how freight platform works makes it physically impossible to move 2 full belt and full througput

[quartz pasture]

Oh, no I understand now

[vapid gorge]

you can move more solid items pm than fluid

[quartz pasture]

Max throughput for one car is 896.52

The number I said was 2

[vapid gorge]

and that number would be very hard to achieve. That's just the theoretical max

[quartz pasture]

right

Regardless, if I plan to use all 1200 somewhere else, it requires two freight cars

[thorny root]

The double inputs on train platforms... are just loading docks. For catchup. Due to loading / unloading down time. Your real throughput should be treated as 1 input.

[vapid gorge]

you'll probably need more depending on how long the trip is

if it's more than 2:40? you'll need more

[storm imp]

Ill try explain by text, why u need more through put..

A Train Cycle is, example 5 minutes.. Meaning from it starts loading, till you start loading next time..

You're loading 1200 Gas/Min into the Cargo of the train - meaning over a 5 minute periode you'd expect 6000 units of gas.

But because your train blocks input for 30 seconds, you'll be missing 600 units of gas.. Meaning you're total will be 5400 gas.

Meaning in the example above, you'll only hit a 90% efficiency :)

[thorny root]

Trains block input for 27 seconds during docking, and the time to fill a train car @ 1200 items per minute is 2.5 minutes.

... might need to check my math on the last one.

[storm imp]

Yeah just took 30 seconds cuz it was easier to calculate with - and still emphazised the point i think :)

[thorny root]

Obviously that's gonna be dependant on stacks size... I forgot about that.

Belts move items based on single items per minute, but trains move entire stacks.

[quartz pasture]

it's really a shame the loading animation causes such madness

[thorny root]

So like... Train transport of things like wire and concrete: Oh god yes. Freight transport of things that only stack to like 50? Oof territory. Avoid.

[quartz pasture]

Do ya'll like to time the station round-trip distance/time?

Drone makes that prospect even worse with its insane animation length, right?

[wind spade]

yeah, but you'd need a buffer for the platform

oh nvm it was already mentioned

[storm imp]

Well its necesarry as you can only handle as much resources per trip :)

[thorny root]

Try to minimize the number of trains doing a single route by maximizing the train length, and minimizing the distance the route is. Then you basically can just run 2 trains and not have to worry about their schedules so much.

[vapid gorge]

drones are simpler, very little infrastructure and if you don't have quite enough you can just slap more on

[quartz pasture]

does their input/output suffer from animation lock timers too?

[wind spade]

no

[thorny root]

Drones are logistically simple but inefficient on throughput and operating cost.

[quartz pasture]

I like them for simple movement within the same station area

[thorny root]

Best used sparingly for things that are low volume over great distance.

[quartz pasture]

full load or single load?

[thorny root]

Ideally... while one is loading. The other is unloading.

in different locations.

Ideally... they swap locations every 2 minutes or so.

really depends on your buffering, throughput, round trip time, the difference in trip time between full and empty

Every train route is a math game, followed by obserservation and tuning.

What I'm doing is 1200 items or 600 fluid per minute into 2x industrial storage / buffer, with both of these manifolded to allow up to 2400 items or 1200 fluid per minute into the platform itself. This is a generous buffer, and ample catchup bandwidth.

I am not mixing items in containers. I am not mixing items in trains. I am only funneling exactly 1200 / 600 into exactly 1 platform into exactly 1 car.

Super easy to keep straight, this way.

[storm imp]

Kinda wanna quote Mandalorian here..

This is the way

[thorny root]

It's a way but it's solid and simple.

People are still free to complicate things further, or even go simpler. XD

Some things do just work better than others tho.

[quartz pasture]

I believe it. 1 platform = 1 item is the way

[vapid gorge]

if you have the throughput you can easily do multiple items, but you do you

[sick bear]

if I have mk3 belts and 2 sources of iron ore at 120 each and 1 coal source at 240, how do I best divide them up into how many foundries?

[smoky aurora]

Manifold

[wind spade]

open the foundry (or codex), check how much each foundry needs, do math 🙂

[sick bear]

https://tenor.com/view/animation-cartoons-toons-reaction-gifs-family-guy-gif-14827637

ok now whats the formula for calculating how much to underclock, if you want to spread out the underclock across the manifold and not just have 1 machine super underclocked

I have done it in the past but I cant remember what it is

[wind spade]

[total clock speed you need] / [number of machines you want to build]

[sick bear]

hmm

I mean by item input count

[wind spade]

?

[sick bear]

if item input is 120, but total machine input has to be either 90 (2 machines) or 135 (3), how much do you underclock each machine

for a 1:1 input to useage ratio

[wind spade]

if total input is 120, you have a total clock speed you need to process that

[frosty owl]

I think you may find this proportion useful
90:120 = 100:X

Where "100" is "100% of clock" and X is the clock you're looking for in case of input of 90

[wind spade]

(which is [amount of items you want to process] / [processing rate of one machine] * 100 )

[sick bear]

ah ok thanks

[civic bronze]

Will 1500m long pipeline (nitrogen gas) 2700/min split into 9 mk2 pipes be fine?

[smoky aurora]

just pack it ,..

[civic bronze]

Im assuming mk1 at full capacity at this distance wont be fine

[civic bronze]

Pack 2700...?

[smoky aurora]

why not ,.. 😄 ,... loopd the Bottles ,... i did like pre produced 3 Big storages ,.. so there are all the time enough bottles available in the Logistic loop ,.. but still easier than 1500 piping of 9 lanes or more ^^

[wind spade]

process it at source

[smoky aurora]

did this at my phase 5 build ,... workesd pretty well

aslo distibution at destination is way easier imo than to handle all the piping to the machine

[civic bronze]

Well greeny if it would be so simple...

[wind spade]

I mean it is 😛

[civic bronze]

I already picked a spot where i have 3 other resources, its either here or there

[civic bronze]

i guess, i can pack it and it will all fit on 1 mk5 belt

thanks

[zealous coral]

I plan to transport nitro by train in fluid freight

[oblique flame]

part of me feels like it might better to just make concrete at extraction with the vanilla recipie and have trains transport it, rather than try one of the alternate recipies and have to haul in 2 ingrediants to a factory. I'm saying this based on a need of about 16,000 limestone per minute.. which would reduce down to 5300 ish concrete per minute, the volume of crap would be much less to concrete in situe, does this sound reasonable?

(it would also require about 350 constructors but there are blue prints enough for placing hoards of those)

[dusky dust]

Heh, logistics on that scale are always sort of Up To You, in the end. Though since Limestone -> Concrete is a compression no matter what recipe you use, if it were me I'd make the concrete at the mining site and then transfer that, instead of transferring limestone

Actually quite a large compression, too, since concrete stacks so high

[oblique flame]

roger that!

[oblique flame]

if its 3:1 limestone to concrete... and 100:500 per stack.. wouldnt that mean 1 car of concrete = 15 cars of limestone?

[dusky dust]

Yep, for vanilla-recipe concrete, anyway. :)

(keep in mind that you're likely to need more than one car for concrete, since 5.3k/min will fill up a freight car in just over 3 minutes)

That was actually a silly thing to say since the most throughput you can get into a platform is <2.4k in the first place. :P

[obtuse leaf]

But you can parallelize with multiple platforms

[oblique hollow]

Yes, easily doable with gas, as it is compressed when packaged as opposed to normal liquids
At a rate of 4 to 1

[civic bronze]

yeah already did it, turned out really simple and im glad i didn't do long pipeline 😅

[frosty pulsar]

guys a got a stupid problem.. maybe I'm just overthinking;

I have two extractors with lets say 600/min and I want both to merge and split so the supply spreads on two belts if one belt gets over 600 consumption, is that possible? Because my current bottleneck is the belt (780) between merger and splitter (right?), I have T5 belts

[wind spade]

why merge them? just run each belt to machines that need 600

[frosty pulsar]

But I have several machines using different values, so just dont merge/split and run them seperately to two belts?

[wind spade]

you can clock machines so that they fit perfectly into the 600

[frosty pulsar]

ye I know I was just curious if there is a way to do what I though of, guess not

[wind spade]

I mean there's a way, but what's the end goal? you want two belts of 600 anyway, right?

[frosty pulsar]

Idk... reading my question now makes no sense to me anymore xd, I better go to sleep and look it up tomorrow

[pastel obsidian]

You can get 600 if you use a smart splitter, you have the overflow go straight and a mark 1 belt to the right answer mark 2 belt belt to the right,
@frosty pulsar

[fallow siren]

there is a way, but simpler to just do row of machine consuming 600 for each belt

[crimson moat]

you can overflow the belts into each other but it can only compensate up to the fastest belt speed

[pastel obsidian]

You can always use mods to make the game suit your player

[sand epoch]

2 splitters and 2 mergers. Not complicated (don't even need smarts)

[restive sparrow]

When you finally get around to melting down alien parts cause you hadn't for a lil while

[buoyant harness]

omg is there a way to make it less complicated

[vapid gorge]

couple ways

1. choose recipes that will reduce the base resources you need in the first place

2. choose alt recipes that will reduce the number of item types needed. For example if you use steal beams for encased beams and steel pipes for something else, use Encased Pipes.

3. break down the plan into multiple tabs, then add the product of that tab as an input of this one

[pastel obsidian]

Try out modeler

Also why are they all blank

[buoyant harness]

zoomed out

[buoyant harness]

ok i got it thanks

[vapid gorge]

nice 🙂 whenever a problem is too much? make it in smaller steps.

this, btw, can also help you design your overall layout. Those extra tabs you break off can be little factories that surround the factories they feed

[humble berry]

That's some SAM, Dayum!

[humble berry]

I do a similar dump once in a while

[frank vale]

I personally prefer to work with a listed display of machines instead of the net/flowcharts, which makes it easier to focus on individual parts.
You can also look up where things will end up, and plan your machines locations accordingly.

[restive sparrow]

When the DD finally couldn't hold any more hog meat 🤣

[thorn bane]

prio mergers
woooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
its happening

[wind spade]

I'm kinda sad that they are adding them 😄

[humble berry]

FICSIT Alien Protien Shakes when? 😂

[civic bronze]

In new video I noticed a clip where blueprint was autoconnecting belts with existing ones 🫠

[storm imp]

Not saying Packaging overflow water, into a prio merger before unpacking could solve fluid issue..

But its deffo an area I gonna explore if its working better than current piping systems

[wind spade]

can just process byproduct water normally

[storm imp]

Computer says no :P

[wind spade]

skill issue

[storm imp]

And that :D

[thorn bane]

can already do that with a 27/1 merger doesnt need to be perfect since both lines are moving
#math-and-meta message

[restive sparrow]

Is there a strategy I can use with STools to avoid this situation?

Scenario: I am making a factory plan and I put some random smallish number of a component/m to ensure I have extra I can throw to storage/dd and I end up with some weird stuff going on here with 2 separate refineries being used, one of which is at a WILDLY low rate.

So that 20 plastic/m I'd like to be arbitrary. Ideally I'm just storing cast-off from the system but I guess I was initially afraid I'd get into a situation where I had no cast-off.

[torn plaza]

you can disable recipes within tools to force it to use what you want

[restive sparrow]

Well it's not that I dislike the recipe. I dislike using 0.098 of a refinery.

[wind spade]

it's not weird, Tools reuse byproducts

[torn plaza]

though as far as the recycled rubber and plastic, what it's giving you is the most efficient way to get that much plastic out

[restive sparrow]

To be clear this wasn't meant as a slam of the functionality

[wind spade]

I'm aware, just explaining why it happens

[topaz jetty]

Tool just looks at efficiency it cares not for ease of construction

[restive sparrow]

I suppose it really is just optimizing input vs output so it's not bad. I totally got you.

[wind spade]

"ease of construction" is impossible to quantify anyway

[topaz jetty]

Agreed

[thorn bane]

still waiting for minimize building count

[wind spade]

also kinda hard, as people usually want "minimise integer building count" a.k.a. 1 machine at 100% is less buildings than two buildings at 20%

[restive sparrow]

I'm new to this phase of the game where output overflow really matters. What's the best strategy to use in a situation like this? I obviously can't just underclock to those specific values. I'm afraid if I underclock to "close enough" and sink the outputs that it might unbalance the system (IE: my recycled plastic is chewing up too much of my fuel and/or rubber).

I guess I just make sure that the recycled plastic is at the very end of the manifold for both fuel and rubber?

[wind spade]

you can underclock to those specific values

well, up to 4 decimals of precision

and the solution could be to have one machine making fuel just for the recycled plastic refinery

(or make a plan with better numbers 😛 )

[restive sparrow]

Ho-lee-shit I thought you had to use the slider. You can type!

[wind spade]

yeah, you can type and put up to 4 decimals of precision
which doesn't give you all numbers, but can get you "close enough" to the point that most people don't bother with the 0.00001% difference

[restive sparrow]

Would there be a way to approach this the other way, take out my 20 plastic/m and somehow otherwise optimize for cast-off.

I'm having trouble putting into words my desire here lol. I guess the idea would eb using the "maximize" option and go into my inputs and limit it down to what nodes I am pulling from.

Basically: My preference whenever possible would be to store and sink things instead of underclocking, since I could have a future use for that product

[thorn bane]

use something like this to only use the overflow fuel

[restive sparrow]

Maybe I shouldn't worry about future use and just build factory for now and get the fug over it

Ya might need to annotate that cause I don't have those buildings yet so it's not clear to me what's going on. Are you talking about the far one that's at the end of the manifold?

[thorn bane]

ye the elevated pipe
that makes it so the ones on the right have priority and only once its full it will fill and flow over to the left one
with this your rubber will always have fuel and only extra will be used for plastic

[restive sparrow]

More rando questions time...
Can anyone think of a reason I'd have a random ghost of 60 power capacity more than production? It has been there since my coal days. Absolutely rock solid constant gap of 60. (Don't mind the wiggles, that's just cause I'm feeding fuel into waaaay too many generators cause it's designed for turbofuel later).

[thorn bane]

thats 2 biofuel gens

[restive sparrow]

Oooooh I got you. Not just at the end of the manifold, but also elevated. That's totally untuitive now that I pay attention to the elevation. Thanks mate!

I have a TON of biofuel gens on my grid sitting there doing nothing (15 of them). You think that's the ghost of power generation's past?

[thorn bane]

maybe they all ran out but for 2

[restive sparrow]

You mad genius you. Out of all 15 these 2 jokers are green

[summer fox]

can someone double check my maths here:
pure copper ingot ( copper ore 15 pm and 10 water pm= 37.5 pm) is a better ratio and thus more efficient then copper alloy ingot ( copper+ iron ore 50 pm= 100 copper ingots pm)
i`m using scim with all the alt recipes turned on and it is directing me to use copper alloy over pure for a unknown reason

[thorn bane]

scim is weird
tools favors pure which is indeed the lowest copper usage
(why do i STILL have to go to fandom to see this????)

[summer fox]

thank you looks like i have 96 refineries to build and sort of- got 3600 ore to smelt 😄

[thorn bane]

i mean just because its most copper efficient doesnt mean its the best
there is lots of copper on the map (and iron) so most of the time its faster to just use the extra copper with alloy and save all the time of fiddling with refineries and water

[summer fox]

i've built a aluminium factory on the east side of the red forrest (near the sam node) near the big lake so i have plenty of water near by. as the copper/iron receiep i can do far away from it but then i be transporting ( likely via train) much more

[opaque quartz]

Copper and iron are commonly next to each other, but a lot of copper is not necessarily near water

So just like everything else w alt recipes, it’s situational. They exist to give you options

[dusky dust]

Because those comparisons, even though they looked like they were nice and clinical and objective, almost invariably contain all sorts of assumptions behind the scenes.

Recipe efficiency often depends pretty strongly on what other recipes are in use in the chain, and constructing a table to encapsulate all that would end up taking many screenfuls quite quickly

I admit that for "basic" stuff like ingots there'd probably be little harm in having something like that, but it's not something generally applicable

Just a case of something that looked very authoritative but was ultimately misleading. Folks look at it and go "well this recipe's obviously superior" when in reality there's no way to actually compare without knowing the whole chain

[wind spade]

SCIM uses whatever you select, if you select multiple it uses first one in its internal list

[thorn bane]

there is literally no downside to have it for copper ore
and yet im forced to go to fandom wiki which is outdated
and its been like this for half a year

[wind spade]

(why do i STILL have to go to fandom to see this????)
because it makes no sense and in most cases is misleading or even wrong

[thorn bane]

how does it not make sense for copper ingots or how is it misleading

[wind spade]

as said above, for ore it's technically possible (for some ores)

[thorn bane]

what do you mean technically
it SHOULD be there
removing it was a huge mistake
i get that the others might be misleading but id rather have that collateral than not have the information and have to go to fandom

[wind spade]

technically = if you put it there for ores, people start asking "why not put it for other items as well"

and the same information can be gathered by going to tools

and also technically = you can convert ores with SAM

[thorn bane]

the whole wiki can be gathered from tools what is that argument
the point is to have it available in 1 page

[wind spade]

for copper ingot it's already impossible due to tempred ingots needing coke which has multiple steps/possibilities

[thorn bane]

the copper column isnt
which was literally just asked
because people go to the wiki expecting to see which is most copper efficient
and it just doesnt tell you

[wind spade]

removing it was a huge mistake
that's your opinion, but that's not shared with most of wiki people's opinions 🤷

because people go to the wiki expecting to see which is most copper efficient
you can just check codex ingame, don't need wiki to do so

[thorn bane]

???
this whole thing is in the codex
thats the point of a wiki

[wind spade]

yeah and this table already tells you the ratios

[thorn bane]

not in decimal form which makes it realry hard to order
the question was literally "can someone double check my maths here:"
because you cant easly see that from this table
which is why a table that lists the copper per input makes sense

[wind spade]

again, if you do it in some limited cases, you get into "why is it not for other cases"

better to have same amount of information on all pages and leave dynamic calculations for dynamic sites

not to mention that if this bothers you so much, you can just add a line in trivia or somewhere and list the ratios on wiki yourself

[mellow lake]

Should I place the pump on the headlift or before the headlift?

[dusky dust]

You'd want to place the pump at a place where the fluid is still getting its full flow, so you generally want it a little below where the existing headlift stops

Back re: those old resource-efficiency numbers, also keep in mind that if you're not literally attempting to max out resources on the map, those numbers become even more arbitrary

'cause they're all based on calculations on the relative rarity of resources and such, and that's extremely unlikely to actually matter to like 99% of the players out there

And is almost guaranteed not to matter given the player's actual factory locations (ie: what resources are nearby? What other resources are being used?)

[mellow lake]

None of my pipes are at full flow

[oblique hollow]

you want to place the pump on sections of pipe that still manage to fill*

aka sections that still have enough head lift

but not so far down that you waste the head lift of the previous pump

[mellow lake]

Im gonna send u a screen shot of my issue

[oblique hollow]

you can send images in here too

no need to DM me

[mellow lake]

Oh I can

[oblique hollow]

in this channel, yes

just not in #satisfactory

anyway, the problem you have is flow related

[mellow lake]

The last 2

Don’t have enough fuel

[oblique hollow]

mk 2 pipes at max flow (600/min) cannot do that amount without some extra work

[mellow lake]

Or oil I meant

[oblique hollow]

you need to add a bypass at the first refinery's junction

[mellow lake]

A bypass?

[oblique hollow]

connect another pipe to the unused connection of the junction near the first refinery

and the connect the second end of that pipe to any other junction near the other refineries

preferably at least at the 10th refinery or later

[mellow lake]

When you say bypass does that mean junction?

[oblique hollow]

no, read what i said
its a pipe

[mellow lake]

Right there?

[oblique hollow]

yep, the left side.
it might be better to rearrange your pipes a bit

connect the oil input to the left side of the junction

and then add the bypass here, where it was connected

[mellow lake]

Like that?

[oblique hollow]

yes

now just add a bypass pipe

[mellow lake]

To the 10th junction

[oblique hollow]

any junction as long as its far down the line

you can even connect it to the very last junction

[oblique hollow]

yep, that looks good

now you wait

[mellow lake]

The game shouldve just let it work the first way I did it

Thank you mcgalleon

[oblique hollow]

you wish but we got confirmation that right now its not supposed to work that way

[mellow lake]

U want another puzzle?

[civic bronze]

What do you mean?

[oblique hollow]

its not supposed to give full flow when you build it as a simple manifold.

or at least: there is no bug at work. the system is working as intended.

Just means that this design doesnt work with the system at these flow rates

[civic bronze]

Oh okay i thought you meant the opposite

[oblique hollow]

call it a nerf to manifolds if you want

[civic bronze]

Tbh irl it should work, but im fine with how it works in game

[oblique hollow]

irl it might work, but irl you would have a lot of friction and turbulence due to the 90° bends

so irl, it probably wouldnt work either

[crimson moat]

I'm not really sure if it's intentional or not that you can only use valves to directionalise flow in certain situations

and through a junction isn't one of them

[opaque quartz]

Just remember, pipes are not belts

[crimson moat]

kinda ruins a lot of the point of valves existing at all, which is why i think it might not be fully intentional

[oblique hollow]

well we got confirmation that the backflow in these systems at max flow is intentional

and that kinda means everything that entails is also intended

this "water hammer" issue

[crimson moat]

it makes manifolded liquids quite bad

compared to Y splitters which just work

[oblique hollow]

maybe they want us to manifold less

(actually, you can still manifold, just means you need to do some different inputs)

the only thing this really nerfs is "manifolds with very asymmetric loads on junctions"

[crimson moat]

yeah, asymmetric junctions (exit pipes dont need the same amount of liquid/min) work horribly

[oblique hollow]

while every pipe system naturally has backflow and such, it is more prominent when the way the junctions have to split fluid is very asymmetric

[crimson moat]

symmetric ones work flawlessly

the backflow through the junction happens when it tries to split 50/50, fills one side, then it comes back because the first exit pipe is full while the main feed line isn't. You can't valve this because a valve cannot connect directly to a junction, and the junction itself doesn't have valving functionality.

but if you use Y splitters then it will not only not be unstable, but actually have a stable restorative force (the pipes which are underfilled will recieve more flow on future ticks until all pipes are equal)

manifolded liquids are always fighting and trying to minimise or compensate for active instability (a deviation will amplify rather than dampen itself)

[oblique hollow]

i did actually run a test recently and it seems junction splitting code is a bit more advanced that thought

if the flow rate (and thus fluid level) of 2 input pipes is within a few percent of each other, junctions will not move any fluid from either one of these inputs to the other input

which results in the very funny case of "2 half full pipes merge to become a full pipe output"

like they straight up dont have to fill completely

30% full pipe A + 30% full pipe B = 60% full pipe C.
without affecting A and B

only if the flow rate / content difference between the input pipes is above some threshold does the less full pipe start slightly oscillating

this also means that equal splits are very much doable and enforcable with pipes, as long as there isnt any major disturbance

aka verticality or a flow rate split asymmetry later down the line

either one could cause a backflow wave to travel back to input

[crimson moat]

Yeah and they are actively stable (will restore to equal if disturbed)

[oblique hollow]

Truly, the pipe system works very well as long as you dont try to do the one thing everyone's brain is laser focused on because big number = gooder and maximizing belt and pipe = dopamine

[pulsar lagoon]

is there a way to build outside of a blueprint designer or build bigger than 6x6?

[wind spade]

not really (in base game)

[pulsar lagoon]

i want to put 4 drone ports but i cant

[civic bronze]

I know there is some bug that would let you make bigger blueprints

But that caused me some crashes (i did it by accident) so dont recommend it

[pastel obsidian]

Mods

[daring surge]

so I built this setup for my turbofuel but well the last 2 refineries on the left dont get their 56.25/m
is this a pipe shenanigans thing where I should just do it differently?

(blue is the input pipes)

I am feeding 48 fully overclocked refineries

[civic bronze]

yeah im guessing 600 input pipes are struggling with backflow 😓 (common issue)

[daring surge]

hmmm yea I just put on a bunch of valves so once it reaches a machine it cant go back

it seems to work ok?

[civic bronze]

you could split the input into 2x 300 (mk2 too) pipes each connecting to 24 of your rafineries

[daring surge]

oh lol I just realised my drawing is for 24 not 48 I just have 2 of those setup but yeah im considering swapping to 300 instead of 600

[civic bronze]

you could also try to loop the pipe like so in this picture #math-and-meta message

[daring surge]

Somehow putting on valves works

[civic bronze]

now you are working with magic

[daring surge]

this is what I did

somehow it just works

I love pipe magic

[civic bronze]

i like the note on the right

[civic bronze]

im intrested to see if it will hold up at 100% efficiency for a longer time, let us know

[daring surge]

will do!

[oblique hollow]

a few ones near the start would probably have done it

once flow rate drops enough, you no longer need to try to mess with anything

[daring surge]

yeah probably, its mostly the first ones overflowing and bouncing flow around a lot I guess

bu ima just leave it as is and see if it keeps working

if its not broke dont fix it!

[crimson moat]

It doesn't really work that well because you can't valve a junction

junctions cannot connect directly to valves or function as a valve, so on every junction you have a pipe1>junction>pipe2 connection

and it will flow from 1 into 2, and then back into 1

semi related: This is basically a one-way gravity junction, but it will fail if overloaded

the pipe on the right isn't full enough to flow upwards, but the input from the left will flow right.

[daring surge]

yeah I probably shouldve went with 'always feed fluids from above' but wellll

[crimson moat]

That, and an easy dumb rule is to split evenly with junctions

like for example if you have 300 fluid, split it into 100+100+100, not 15+275.

The less of a pipe utilisation you have (like 100/600), the less chance backflow has to ruin flow in a pipe where you are splitting unevenly with junctions.

So for example splitting a 600/600 into six 100/600's will basically make a manifold work fine because it will flow back and forth but still be below 600

[daring surge]

well at least 296/300 gens are consistently running now

yeah overflow and fluids dont like each other I have noticed

aaand yup

it broke for second

it keeps filling up and then eventually dries up

[crimson moat]

continuation of previous pic (gravity one-way junction)

[crimson moat]

easiest hotfix is prolly just do this > "So for example splitting a 600/600 into six 100/600's will basically make a manifold work fine because it will flow back and forth but still be below 600"

[daring surge]

yeah its just annoying since it takes 56.25 as input

[crimson moat]

clock them down to 50

[civic bronze]

Can do x2 300/600 no problem too

[crimson moat]

most likely yes, i do love big safety margins though (especially on finnicky systems)

[civic bronze]

His setup doesnt look "finnicky" its leveled, just long manifold

[crimson moat]

Fluid junctions where you have unequal outputs and overload one or more outputs will backflow and be incapable of flowing at the full and otherwise expected rate, this includes setups like this and causes the issue that Wolv has

[arctic saffron]

i just made this sueprcomputer design

any tips on what to change or keep this how it is

[daring surge]

so this would be much better?

[wind spade]

if you have two 600 pipes, don't mix them together

[daring surge]

I have to though, I need the leftovers from both to feed another refinery? Or do I put that one at the end?

[arctic saffron]

if both 600 pipes r fully filled make them 2 seperates

[wind spade]

you can clock refineries so that they need 600/min together

[daring surge]

the numbers suck ass cus its 22.5 input for turbofuel and my brain is cooked atm

[crimson moat]

you can make the number whatever you want, make it 20

[daring surge]

it'd be 20.06

wait

[crimson moat]

what exactly is 22.5?

[daring surge]

the fuel input at 100%

[crimson moat]

what recipe

[daring surge]

base turbo

I overclocked cus I didnt want to have a million refineries as im using 2700 fuel

[crimson moat]

45 and 37.5 are indeed bad numbers to work with

[dusky dust]

If you didn't know, btw, you're not restricted to just the percentage slider. Most of the numbers on the machine's control panel are text inputs -- you can put in the exact numbers you want

[crimson moat]

30 in and 25 out (166.6667% clock) is a candidate

if you want to push the clock higher, the numbers will get more awkward.

[dusky dust]

The only one that actually matters is the percent clock speed, but it'll calculate it for you if you input it on any of the others

[crimson moat]

You can put 540 fuel into a pipe with 12 refineries at 250% and get 450 turbofuel out of it. That's probably the best compact option. If you put more than 540 fuel you are gonna end up with crazy numbers

[daring surge]

yeah I know I can just pick numbers but the numbers are just annoying

but yea 30 in 25 out seems the best

[crimson moat]

i think that last one i wrote would be best ^ but you have to tune the previous stage for it, to input 540 rather than 600.

[daring surge]

at least I can match pipes to refineries exactly

[crimson moat]

that is a great logistical advantage too where it can be used

[daring surge]

problem is my blenders produce 100 fuel each soo....

so 540 isnt working out

[crimson moat]

diluted fuel?

[daring surge]

yup

welp time to turn 48 refineries into 90

[crimson moat]

you can feed a 540 with three diluted fuel blenders at 180%, four at 135%, five at 108% or six at 90%

then you have a 540 pipe > split it to 12 refineries. Collect 450 turbofuel from output.

If you do that 4x then you can make 4x450 and you can recombine to 3x600 if you want

[daring surge]

oh right of course blenders also overclock lol, I almost never overclock, this is by far my biggest project so I keep forgetting it

[crimson moat]

https://www.youtube.com/watch?v=Sdea1MesW4Q I'm fairly sure that Y-splitting liquids (especially with a gravity gradient, but even without) is stable to practically infinite connection amounts and time durations with an input of 100% of the pipe capacity

a manifold on the other hand (which relies on breaking the rule of always splitting evenly at your junctions) will backflow because of that, and the longer the manifold and/or the closer the main pipe is to capacity, the worse the backflow issues as each uneven junction adds up cumulatively.

[daring surge]

that math isnt entirely right (its 10 refineries * 240% = 540) but that works out much better yeah. I do have to match my heavy oil residue but thats easy. I reaaaaally shoulda thought this out a little more before building it

[daring surge]

I only really thought about oil goes in and turbofuel comes out

thanks a lot for thinking with me and helping me out!

[crimson moat]

nps

600 > 6x100 splitter example to feed into 6 seperate manifolds

first splitter turns 600 into 300+300

next ones turn 300 into 100+100+100

if you split and consume equally they are basically belts

[oblique hollow]

if you split it in half, you can manifold the rest normally as anything below 600/min can usually be handled by the system

[crimson moat]

usually, but you can break it too

[oblique hollow]

eh, rarely

[crimson moat]

there's no good reason not to build with more margin for backflow

that's the main argument imo

[oblique hollow]

if you use 2 x 300/min in mk2 pipes (with pumps / valves installed to aid in backflow interception) it usually is fine

[crimson moat]

usually but not in all circumstances, and the pumps/valves aren't neccesary if you build in a smarter way

[oblique hollow]

thats the point where is becomes taste in buildstyle

[crimson moat]

eh i would say some stuff is objectively better/good/bad

like you shouldn't manifold 32 gens on one pipe in a row

because it's just objectively bad

[oblique hollow]

if you can get away with a manifold and not hit max flow, nothing really stops you

[crimson moat]

you "theoretically can", but in reality when you try to feed 32 gens 18.75 fuel each through a manifolded 600 pipe, backflow will break everything. It's objectively bad

[oblique hollow]

at very high flow rates, equal splitting is definitely good, but once you are back down to mk 1 levels of flow, you can do whatever you want

[crimson moat]

16 gens 18.75 fuel ea, it's probably fine, but 8 is definitely always fine

[oblique hollow]

you read my sentences like i said "manifold a full mk 2 pipe" but thats not what i wrote at all

[crimson moat]

Nah IK what you mean

my point was just that you can break a 300/600 pipe manifold

it's basically impossible to break a 100/600

so why not just use them if you're manifolding

[pulsar lagoon]

why this still happening?

[crimson moat]

it has no fuel

[pulsar lagoon]

i have 2 ports in this one it has not fuel, and in the other one has fuel

[pulsar lagoon]

🙂

[oblique hollow]

the drone ran out of fuel is his point

[pulsar lagoon]

this prot has no fuel, the other conected por has fuel

[oblique hollow]

it is supposed to pick up enough fuel on the other port to not run out

[dusky dust]

Yeah, if you're fuelling from the remote port, you still need to provide it enough fuel to get over there one one time, first

[crimson moat]

Ah i see

[pulsar lagoon]

thats what i thought and it works with my other ports, but with sam i dont know why dosent works

[dusky dust]

It won't take off unless it's got the fuel required to make it to the other side

[pulsar lagoon]

i have fuel in the remote port should be enough is not a long trip

[oblique hollow]

no clue, when i first played 1.0 my drone ran out of fuel, after i resupplied it it didnt anymore

[dusky dust]

Right, but it has to get there first.

[pulsar lagoon]

here is the remote port

[oblique hollow]

when in doubt, make the remote port the home port instead

so rebuild the drone there.
unless you use this as some kind of central docking point for multiple drones?

[pulsar lagoon]

what sorry dont understand

[crimson moat]

I am confused

[pulsar lagoon]

am too 🙂

[crimson moat]

so nvm

[oblique hollow]

its empty because the drone is supposed to deliver SAM here

and it didnt because no fuel

[pulsar lagoon]

yeah but dosent work, why?

[crimson moat]

which port is the drone built on

[pulsar lagoon]

i know, but in this port it has

[oblique hollow]

dont know.
Rebuild the drone

[pulsar lagoon]

i have 2

[oblique hollow]

make THIS port the home port

[pulsar lagoon]

what isnt the same?

[crimson moat]

but which port did you build the drone on

[oblique hollow]

home port is where drones start out. That may affect things

[pulsar lagoon]

i dont get it sorry

[oblique hollow]

you must always build a drone on a port. that makes that port the "home"

[crimson moat]

you build the drone either at the one with fuel, or at the port without fuel. It changes behavior. Which one did you do?

[pulsar lagoon]

isnt this a point of view? like base sam 2 is remote for sam 2 but for sam 2 is remote for base

[crimson moat]

No, they're different ingame.

[oblique hollow]

for you, maybe.
not for the drone.

[pulsar lagoon]

i build 2 drones, one in this port and 1 in the remote

[crimson moat]

you need to have only 1 drone

[oblique hollow]

oh they made a dual setup.
No clue about those.

[pulsar lagoon]

so how i make this port fuel not the remote one

[crimson moat]

If you're not sure, then:

• delete all drones on that route so that you have 2 empty ports and 0 drone

• make one drone at SAM place

• add a bit of fuel (like 30) for it to take off once, set route to the port which has SAM dropoff and fuel supply

• watch to see if it roundtrips properly

it should

[oblique hollow]

you can, usually, have a single port supply 2 drones with fuel.
assuming you built a drone on Port A and one on Port B, link them to each other and only supply port A with fuel

which is probably what they did here. But i dont know how stable that is anymore

ever since the fuel was changed to allow stuff other than batteries

[warm igloo]

It should work

Works well on my save i have the same setup with 3 drones refueling at the same drone port

So no reason it shouldn't work here

[crimson moat]

This needs more testing but it may be useful

[crimson moat]

600 > 8x8 gens via manifolds (9.375 fuel ea.)

[pastel obsidian]

Nice work

[crimson moat]

just testing stuff

I think i will actually split more to prove a ratio works with turbofuel @ 100% (then if it ever breaks, it'd be with nonsensical underclocking)

[civic bronze]

Does it need to be a junction? What if you removed it and left pipe - and connected right pipe to floor hole with horizontal to vertical build mode?

[crimson moat]

I don't think those behave the same

not 100% sure though

[glass harness]

Bunch of them are starving in the picture though?

[pastel obsidian]

It would be a good thing to test but I use what you did to recycle my aluminum water

[crimson moat]

It's not active, just showing the layout/connections

manifolds have a warmup time

for this many gens on a 600 pipe it's over 5 mins to warm up with all turned off, and hours with them on, so i just took the pic before doing that

[glass harness]

Yeah I was just wondering why they are starving

[pastel obsidian]

I find it is always best to give liquids two ways to reach their destination you still have a manifold system you just have 8 smaller ones

[crimson moat]

That's basically the same thing as using two half-sized manifolds

this second design is manifold based, yes

[pastel obsidian]

I more so meant as the end link all pipes together

[crimson moat]

that can be very problematic

[pastel obsidian]

If you have too much fuel go down one line it helps it flow to where it is needed

[crimson moat]

that is not a problem if you have evenly-distributed lines

like if you split 600 to 300+300, it will always take 300 down each line

if it starts to imbalance, it will actively stabilise itself by flowing less to the pipe that is more full

if you want to put 6 generators on one line and 10 on another, yeah, that becomes an issue

but generally best not to do that

[mellow lake]

How do I do the pipes

[crimson moat]

you're probably trying to move too much fluid through one pipe (the center one)

[mellow lake]

266.67 fuel

[crimson moat]

mk.1 or mk.2 pipe?

[mellow lake]

2

[crimson moat]

what's the issue exactly

[mellow lake]

One of my fuel things goes yellow

[crimson moat]

if you've checked that all of the clocks and numbers are correct, you can try using 2x as many pipes (like 2 instead of 1 to move HOR to fuel or whatever)

carrying less than before

[glass harness]

loop the last one back to the first and it will work, assuming the fuel generation side works properly

[mellow lake]

Like this?

So they on the same junction?

[glass harness]

Well cant make too much out of the screenshot but yes

[mellow lake]

It’s not a screen shot it’s a picture

[civic bronze]

discovered item teleportation (apparently these are connected lol)

[opaque quartz]

Put the generators on stand by, let their buffers and the pipes fully fill up

[mellow lake]

I hate pipes

[opaque quartz]

Also double check your math (fuel amount vs generator count)

[crimson moat]

stress testing distribution of 600 from 1 pipe into 96 generators (6.25/min each) - via splitting into 12 manifolds of 8

(IK it's not even since i expanded it, shouldn't matter)

[glass harness]

If you like the looks. Great but otherwise simpler solutions work too

[crimson moat]

That's the issue, they don't, not exactly

tl;dr

junctions split nicely, evenly, with no issues

if splitting unevenly like a manifold, limit the amount of junctions in series and limit the utilised % of pipe capacity to control backflow (ideally to <=8 connections and <=25% of the pipe's max flow rate, but a lot more may work)

Those manifold numbers are extremely conservative but i don't see a good reason for them not to be

[glass harness]

Fluid issues are always 1. Trying to fit more than it can handle. 2. not looping it back so backflow 3. Gravity

[crimson moat]

backflow isn't caused by not looping it back, it's caused in a manifold because of attempting to split fluids unevenly (e.g. 15+575, instead of 300+300) while being unable to valve pipe>junction>pipe connections. Fluid flows out to the side connection, fills it, and then flows back into the feed pipe because the feed pipe is emptier than the side one.

A manifold loop is essentially the same as making an extra row with another split in half here, that's why it helps.

It splits the 600 input into two smaller inputs, 300+300, and it makes it so that liquid has to go through half as many junctions in series to reach the furthest machines.

It helps a lot, but one split is not always enough. Having an approach with banks of machines that are of arbitrary length allows you to choose a size that works with margin to spare, lots of it if you want.

you can also split much more pretty than i did there, will get back to that later

it's stable (this machine is the furthest from the fuel source)

[glass harness]

What I was trying to say, the first junction is what does the trick here. otherwise its just matter of how you want to put them down.

for 2 lines of machines the end parts looping to each other is the other way to do it

aka not cramming 600 into one pipe

[civic bronze]

turning on new power plant like YES SAVE ME

i guess i can disconnect geothermal now

[civic bronze]

almost 2h later, it finally flatlined 🥵 ⚡ for this build to only have couple not connected belts and pure oil node mixed places with where normal node was supposed to go - i'd say it went smooth (gotta love math)

[thorn bane]

splitting 600 into 300 and then manifolding the 300 should work
the 300 would fluctuate between 200/400 not causing issues
dont think you need to balance it all the way down to 75

[crimson moat]

Half pipe i would say for the most part it does work, but it isn't a guarantee. Quarter pipe you can heavily abuse and it will be fine.

[thorn bane]

thats how the "just loop it" approach works
300 coming from the start 300 from the end

[crimson moat]

yeah, that breaks if you do non-overclocked turbofuel for example

you cant manifold 40 things on a pipe even if it's only at 300/600, the backflow starves consumers

[thorn bane]

oh really? the 300 hits a limit?
when i tested it i couldnt get it to break

[crimson moat]

ye

[thorn bane]

ye i guess 7.5 is very little hmmm

[crimson moat]

there is a.. flow inefficiency ratio which is dependant on how many imbalanced junctions you have in series. And it has gone completely to shit long before 40 of them.

[vapid gorge]

Number of input pipes and volume each machine consumes both affects stability

The more each machine consumes and more input points , the less stable

[thorn bane]

ye i never tested with as low as 7.5

[crimson moat]

i figure ballpark maybe 15 to 20 things is okay on a half pipe

[vapid gorge]

Pre 1.0 I got like 40+ TF gens on a non looped pipe because flow input was so low

Also at the same time, 25 input points to refineries on the same side couldn’t even be stabilized with a loop

[crimson moat]

It becomes quite sensitive to many minor aspects of the design i think

[thorn bane]

actually ye now that you mention it i always build it like this which does break it down into 150

[crimson moat]

Ye

Some people may try to do this

and that does not work

[thorn bane]

yeee true that would be an issue

i wonder if its better if you feed them from above instead of ground level
now that you can blueprint gens it might be worth doing it
but needs some testing

[crimson moat]

I think it might help a little for the pipe going out of the main feed pipe and into the gen to be downhill, but if it does, not by a lot

it will tend to fill up until it's full enough to flow back, before the fluid moves much further down the feed pipe

you can't fix too many junctions in parallel by flowing down. You might be able to get a few more before stuff breaks

splitting evenly in junctions to lower used pipe capacity is a fundamental fix though

so if you have issues it's probably best to just do another equal split

So i guess that a near if not absolutely optimal solution is to deliver liquid in 600/600 pipes, and then at the destination equally split that pipe enough times to be confident that the flow+backflow won't exceed the flow rate of the pipe - and then manifold the resulting pipe arms.

This kind of testing lets us say very specifically that high utilisation % pipes aren't the problem, junctions aren't the problem, but it's specifically an interaction with asymetrically split junctions (e.g. 15 going one way, 575 another -- as opposed to 300 going one way, 300 another) that causes fluid to flow back and forth through the junction (rather than uniformly in one direction) and thus effective flow rate to choke to only a fraction of the actual flow rate.

[thorn bane]

to my understanding it is because of the whole "water hammer" thing
the machine is open, eating all water causing it to flow into it at like 200
then the machine shuts off the flow completely which causes a huge backflow of the 200 that just tried to get into the machine since its running into a dead end
this 200 then flows backwards mixing with the 600/min causing the pipe to restrict flow since it cant flow at 800/min

[crimson moat]

we can prove that it's not even getting that far before it breaks by using valves

the problem happens in the junction and the 3 (or 4, if it's feeding to both sides) pipes connected directly to the junction

The junction does this

the excessive flow off to the side (more than is being consumed) causes that pipe to fill excessively while the forward pipe isn't as full, which then causes it to flow back through the junction.

The flow does adapt, mostly, but not entirely

and that counts multiple times against the flow rate limit of the pipe

so effectively you take e.g. 150m3/min of flow rate to actually move 100m3/min of fluid

the effect is also cumulative, so each additional junction in series causes additional backflow and effectively reduces the flow capacity of the pipe further

the problem doesn't exist when you split a pipe evenly into 2 or 3 via a junction

[thorn bane]

why pipes doing 600/min doesnt work in long manifolds

[crimson moat]

The consumption of machines not being constant probably plays a role in preventing the system from balancing properly yeah

it doesn't have time to stabilise before the next "gulp", which screws it up more until the one after, and so on ad infinitum.

[thorn bane]

ye the machines are on/off rather than constant consumption

[crimson moat]

overflow, while being the property that makes manifolds work in general, is actively really detrimental in pipes because the bidirectionality makes it go backwards instead of just sitting in place

and since junctions aren't valved themselves, pipes aren't valved themselves and valves can't be directly attached to a junction, you can't effectively mitigate that (which begs the question of why valves are ingame at all)

It's not so detrimental that manifolds aren't useful

but you must take backflow messing up your effective flow into account and correctly mitigate it or your manifold will not work properly

[wary tulip]

Would not having a small tank work to mitigate that hammering effect? So long as the tank is not fully filled… it should take that hammering back into itself and mitigate the flow, correct?

[crimson moat]

The ones at the start of the manifold would start to fill up, then flow backwards into the feed pipe. There's no effective way to prevent that if you're consuming 15 but the junction is pushing 300 into that exit, the extra 275/min will fill up whatever is on that junction exit.

Stuff like abusing headlift comes to mind (making the fluid prefer to flow forwards, rather than fill a tall buffer on a side path because it's higher up) but it doesn't fix the issue at all because that buffer will fill up enough to take priority over the main input, and then it will flow backwards.

[thorn bane]

would need a tank before every machine right?

[crimson moat]

the 300 sideways will fill any pipe or storage until it flows back UNLESS you are consuming 300 from that pipe, at which point it's not a manifold any more

being unable to flow sideways because that way is full is the only way that fluid actually moves down the manifold

but the same thing causes the stuff to flow back from the side to the middle pipe

so i don't think you can actually "fix" it

being unable to flow sideways because that way is full is the only way that fluid actually moves down the manifold

If you think about it

a 600 pipe connected to 30 things with a manifold

is trying to give 300/min to the first machine, 150/min to the second, 75/min to the third, 37.5/min to the fourth.. and so on

the when you open the pipe, the 30'th machine in the chain gets 0.000m3/min, essentially if not literally nothing. Fluid only gets there because all of the side paths are full or blocked already

[thorn bane]

junctions dont split even 50/50 though

[crimson moat]

they do

if you have 1 full input pipe and 2 or 3 empty ones, they will split into 300+300 or 200+200+200

and furthermore they will self-balance to maintain that ratio (if one pipe is fuller than the others, it will recieve less fluid and they will recieve more until they're equal again)

[thorn bane]

pretty sure ive seen junction split 400/200 in a T section
also the flow fluctuates alot and not always with both in the same way

[crimson moat]

You can force it to split unevenly via overflow but you have to utilise an additional mechanism, one that can degrade the pipe performance. When you first plug it in, it will send 300/300 each way until the side only consuming 200 is full.

[oblique hollow]

Funny thing: this is where a valve could potentially help if we go by some test results.

But only if you limit it. and if you limit the other output of the junction too

[oblique hollow]

If undisturbed, they actually try to. The problem is that the connected pipes need to be symmetric to some extent

[vapid gorge]

basically like this. Which isn't really reasonable expectation of a factory layout

[thorn bane]

yes that was the initial question...
#math-and-meta message

[thorn bane]

im like 90% sure that a T junction with water will favor the straight path over the side connection, hence the 400/200 split (as in feed from bottom, 2 outputs top and left)
think of VIP but flat
something like fluids preferring to flow the way they are coming
also watching turbofuel manifolds filling definitely feels like they are "viscous" or w.e. and fill the connection before continuing

[oblique hollow]

it doesnt

unless you put it vertically

[thorn bane]

isnt that the whole reason that the VIP works? that junction prefer straight lines of input/output

[oblique hollow]

VIP is vertical

and its a junction pair

[oblique hollow]

from tests i just now ran again, it must be the vertical arrangement of the junction.
flat junctions make perfectly even splits

and it seems pipes have some tolerance for imabalances. 1% - 0.5% difference in content is tolerated and wont lead to flow rate adjustments
(currently testing the precise tolerance)

[oblique hollow]

correction: it seems it allows a 40% deviation from the "expected pressure"

basically: if you have 2 exits on a junction, the junction expects to be able to do a 50/50 split
and thus it expects the appropriate pressure in each pipe (dependant on pipe mk)

for a mk 1 pipe split in half, this means it expects 50% pressure in each pipe (on flat pipes thats 50% fullness)

if the pipe now for some reason fills up to a higher fill level, the junction tolerates that difference

[remote flame]

If you wanna have fun with fluids looking at wild backflows/sloshing, this is an elegant 'sandbox' setup to play around with. Junctions do not have a flow limit going in or out of them, hence these water extractors at 250% each can feed the buffer at 1200m^3/min through a single junction. Then just tweak the flowrates from each line (clockrates), pipe mk's (1 or 2), maybe slap some valves in there to see how it changes the behaviour.. Basically this thing exaggerates the living hell out of sloshing and gives valves some real justice when close to the junction either side, I've seen 800m^3/min slosh effects while wasting time in satisfactory trying to get the worst possible effect haha

each and every time I've played with it, Once you create a pipe 'loop' on the extractors, no matter what I did, it always came good and supplied what I expected at the buffer, hence why its justified to just default to the 'loops just work'. But I have seen valve setups work good in here too! especially when set to the right numbers you expect

Sinking the buffer 1m lower in to the foundations from that screenshot shows even more cool behaviour (I say cool but this could definitely give people nightmares why haha) since the junction is then exactly at half the height of the buffer. On some different pipe/valve combos I've used, once the buffer gets above 50% capacity, it is above the height of the input pipes, which starts to create sloshing in the input pipes lol

(this isn't really related to any of the above topic, but its cool to play with!)

[oblique hollow]

i mean, if you give a junction an imbalanced supply, then of course the pressure differences will come into play

even a simple 120 from one side and 240 from the other will lead to the 120/min line sloshing around due to needing to fill up enough to "match" the pressure from the 240/min side

[wintry jewel]

quick question for you pioneers.
i was working on my nitrogen factory when i saw that the train for my SAM production was stuck on one of the ramps going up to it's next stop. it's 14 freight cars and 2 locomotives long. now i already know the recommended locomotive:freight car ratio is 1:5 so should i add a third locomotive?

[dusky dust]

It's not really true that there's a "recommended" ratio, btw. A lot depends on the specifics of the track and how full the cars are, etc (cars which are more full weigh more, and are thus harder to accelerate)

If your train doesn't have enough oomph to get up an incline, then yeah, adding another engine would be a way to fix that without having to reengineer much else. Other options would be to divide up your lines so that the individual trains don't have to be that long, of course

[harsh schooner]

is this geo gebra

[thorn bane]

nah its draw .io

[harsh schooner]

im doing so much 600 min pipes lol

i dont see a problem with it

other than it not working at 100% efficiency

[opaque quartz]

other than it not working at 100% efficiency

[wind spade]

that's user problem, they can work at 100% 🙂

@proud sigil pick any that you like. There's never a wrong choice, all recipes are good in ceratin scenarios and you can get all of them anyway. Whether you like their advantages and are fine with their disadvantages is fully up to you

if you don't know, just flip a coin or keep them until you decide later

[dreamy nimbus]

Nitro Rocket Fuel vs Rocket Fuel (for the same amount of Rocket Fuel / min)

-NRF
Pros : Less Crude Oil, Less Logistics, Smaller Factory, No Iron
Cons : Need Coal, More Sulfur, Nitrogen Gas and Water

-RF
Pros : No Coal, Less Sulfur, Nitrogen Gas and Water
Cons : More Logistics, Bigger Factory, Need Iron, More Crude Oil

NRF is so much better for "just" needing coal for me

[thorn trail]

not sure how Nitrogen Gas and Water can be a pro for one and a con for the other

or is the Less and More supposed to be including those

[opaque quartz]

the nitro recipe requires more resources relative to the vanilla, in exchange for not needing to make nitric acid. it can also be unlocked earlier (since nitric acid requires the particle enrichment milestone)

[thorn trail]

it was the wording that confused me

[civic bronze]

Yeah i made power plant yesterday, using NRF for the first time, and it was very straight forward, really really easy recipe, slapping 1296 fuel gens was the real challenge but thats not the point

The only downside i would say for me is just sheer amount of sulfur it takes, as i've used exactly 1/3 of the map's worth for it

[fierce ruin]

what do i pick

why 2

@thorn bane

[dreamy nimbus]

really simple one, just if you need less of x ressource compared to another recipe, it's a pro

[urban marsh]

Unless you need one of them right now i wouldn't take either one, if you keep them without selecting, the next hard drive you scan will not be able to get those 2, so you get more options

But in general terms i would say bolted plate is more commonly used since caterium is a rarer resource

[fierce ruin]

oh, i dont need them

when do i use them things?

[urban marsh]

what things

[opaque quartz]

Alt recipes just give you options

[fierce ruin]

bolt plates and fused wire

[opaque quartz]

Different ways to make reinforced iron plate and wire (respectively)

[fierce ruin]

yh but for what reason, do i need them?

[opaque quartz]

They are there to give you options

Alt recipes always have tradeoffs

[thorn trail]

In the first sentence, you said "Less Crude Oil, Less Logistics", therefore setting the pattern of prepending each thing with Less or More, then you follow it with a sentence where you don't use Less Or More in front of "Nitrogen Gas and Water". THAT is the part that was confusing, not the recipes themselves

[fierce ruin]

oof alr

[opaque quartz]

Fused wire, for example, makes more wire but requires you to add caterium to the mix. The default recipe only needs copper

So if you find yourself needing a ton of wire and have caterium available? Might be a good choice

[fierce ruin]

a good example is my HMF factory; iron alloy needs me making 288 iron ore and 72 copper ore, while without it id need 540 iron

drastic reductions can be done

Oh my this is beautiful

so, uh, what do i use for transport? a belt, train, or some car?

[dreamy nimbus]

cuz I changed the nitrogen water part later to make it cleaner and forgot to do the same with "less"

[dreamy nimbus]

I really like pure ingots because there's a ton of water

[civic bronze]

I know its either copper or caterium i've been playing too much lol

[thick plank]

I dont have the exact numbers avaivable rn, but i have a setup saved somewhere that takes just whole numbers of resource nodes for that recepie.
So yeah, if you have some caterium free in the area and are operating on a large enough scale its a good recepie

[fierce ruin]

caterium computer lol

i only need 78.75 of the ore

but itll be a massive pain making a long ass belt

[dreamy nimbus]

caterium computer ?
in what recipe

[civic bronze]

Just belt it

Not worth setting up train for this amount

Noone will see this belt

[thick plank]

In my save I have one from thrre to the waterfall in the grasslands, it takes shorter then youd think

[dreamy nimbus]

I'd rather use the normal computer than caterium computer just for the computer since computer only needs copper and oil (+water)

[thorn bane]

90/min is insane

[fierce ruin]

really? is there worse?

fun length

[dreamy nimbus]

But I guess you chose caterium computer since you're already making rubber xor quickwire xor control board somewhere

[fierce ruin]

yeah (the 3.75 rubber is being sent into a sink)

[static yoke]

I just made a Radio Control Units factory using Caterium Circuit Board and Caterium Computer recipes. It was actually rather straight forward.

Quick wires were needed for AI Limiters, so just making more of them was relatively trivial.

[civic bronze]

i've been enjoying putting my whole save in modeler - being able to interconnect different plans/factories there (about 90% done) just wanted to share

[harsh mason]

I've just started using the modeller too. It really helps me compared to the online calculators because I can tweak things midway through a production chain to model overflow

[dusky wren]

did anyone calculated how eficient fuels are ? i need some energy (100-300gw) for future and it is the best to make fuel/turbo/rocket/ion fuel?

[dusky dust]

In terms of using them for power, the wiki's got all the relevant numbers, if that's what you mean: https://satisfactory.wiki.gg/wiki/Fuel-Powered_Generator

For Fuel -> Turbo -> Rocket, there's a clear escalation of power production. If you've got access to Rocket Fuel, most folks will go for that (assuming you don't mind endless fields of fuel gens)

[robust raptor]

If you're wondering which consumes the least resources, it's up to what you want to consume

[dreamy nimbus]

indeed

[dusky dust]

Ionized Fuel is sort of the one exception; it takes so much energy to produce that it's not worth using in Fuel Gens

[robust raptor]

Diluted fuel only consumes oil, but consumes a fair bit of it

[jovial moon]

Like, i never understood the modeller hype, its more of a toy rather than something that could help you out, everything has its limited amount, so everything has a constant cost that can be mathematically optimised, narrowing things to using the best recipes, and you dont really have to plan a layout because everything can be manifold

[dusky dust]

You can make Ionized Fuel net-power-positive but you'd have to do a lot of judicious underclocking to get there (and it just wouldn't be worth the bother)

[robust raptor]

Rocket fuel trades a lot of oil for sulfur and nitrogen, which is pretty great, and depending on the alts you use, can require very little infrastructure for the fuel making part

[dreamy nimbus]

the only reason you'd want ionised fuel is for jetpack and drones

[dusky dust]

Yeah, Ionized is great for those. Just not as a power-producer

[dusky wren]

thanks guys. i was only considering is it worthy to do ion if its that complicated

[dreamy nimbus]

yeah same for fisconium, it takes so much that it barely gets power productive, its only uses is to get rid of uranium waste at all but you could also sink the plutonium fuel rod

[dusky dust]

Yeah, I suppose that's a fair point. The real tier is basically: Fuel (with vanillaish recipes)-> (HOR+Diluted Fuel or Turbofuel) -> Rocket Fuel

[robust raptor]

Diluted fuel is surprisingly efficient

[dreamy nimbus]

also it's phase 5 stuff, by the time you get there you realise how not that complicated it is

[dusky dust]

Well Ficsonium I'll stand up for; that's power-positive by a fair amount. The power efficiency is pretty abysmal compared to Uranium/Plutonium but you still get a reasonable amount out of it, and it lets you take advantage of Plutonium while still having a "clean" nuclear solution

Whether or not Ficsonium's worth it is a personal question, of course, but it's in a different class than ionized fuel

[dusky wren]

i barely finished phase 5 milestones and i see the real game is starting here XD

[dreamy nimbus]

I finished all of them except for mk3 blueprints, I'm starting to do my nuclear pasta facility

[dusky dust]

Heh, yeah, specifics depend on recipes, of course, but about 66% of the power you get from Ficsonium goes into making the Ficsonium in the first place

[opaque quartz]

Tier 5, not phase 5?

[dusky dust]

Still, on my own (IMO modest) nuclear buildout on my 1.0 save, that was still a net +41GW, which ain't bad

[dreamy nimbus]

I'd rather maximising my power my just sinking the plutonium and spending more time making the uranium/plutonium facilities

[dusky dust]

Technically maximizing nuclear power means doing Ficsonium (and not sinking Plutonium). :)

[dusky wren]

phase 5 (only quantum somthing to unlock) , tier 5 with oil things was easy for me

[civic bronze]

when maximizing with some sloops i assume the best thing to sloop are rods?

[dusky dust]

Probably, yeah. I've not worked out that math though

[dusky wren]

for now i have some small coal power plants and 70gw dilu fuel pp

~85gw of power with boost

[outer vale]

also if you're comparing clean power specifically, then Ficsonium also gives you the Plutonium you'd have had to skip, and that in itself gives you a ton over just uranium (at least double)

[dreamy nimbus]

what a great idea you've given me

[dreamy nimbus]

well if I spend 100 hours on doing uranium/plutonium and 100 hours on doing uranium/plutonium + fisconium

I'd have less power by doing fisconium I reckon

[crimson moat]

depends on the context, sometimes backflow (and thus effective flow rate reduction) is inevitable

[dusky dust]

On that example you're maximizing your time, not power. :). But sure, by that criteria I agree.

[wind spade]

whichever you want

[mellow lake]

Is there a cave under that?

Where my crosshair is

[dreamy nimbus]

More time = More power

[crimson moat]

I use 1:4 and round up so that they can climb the inclines that are easy to build. That demands 4 locomotives for 13 to 15 freight carts, which is twice as much power as you have

[dusky dust]

Yes, but unless you're including burning Plutonium + burning Ficsonium, you will not be maximizing power. That was my only point. :)

At a certain point you will run out of Uranium Power to produce, and Plutonium Power to produce, and adding Ficsonium will always give you more power than you had before

(Which is fine, I'm not trying to convince you or anyone else that they should be making Ficsonium. Just that maximizing nuclear power (for any given amount of Uranium input) means utilizing Ficsonium)

[dreamy nimbus]

truege

[dreamy nimbus]

but mid-maxxing my time would be sinking plutonium

[dusky dust]

(Could get closer to also maximizing power by burning the Plutonium and storing the waste, of course. Nuclear waste storage is ezpz and generally without practical downsides)

[opaque quartz]

probably? there's lots of caves all over

[obtuse leaf]

What is your preferred production chain for supercomputers?
I am using up almost all the bauxite and quartz on the map already, so it will be tough to make enough radio control systems for the OC supercomputer. Batteries will have a similar problem with the super state computers.
Regular supercomputers will also be some trouble I think
I'm aiming for at least 100/m

[crimson moat]

I'm doing OC because i can do most of the work in automating many RCU's and then sending some off to a few other factories

what is your baux/quartz used on?

[obtuse leaf]

fused modular frames and the rest is alclad sheeting in preparation for future ones, but i guess i should re-evaluate the distribution 😂

my goal is kinda to brutalize myself so space elevator demands are a bit higher

phase 3 i needed 100k versatile framework, but i made about 800/m

[crimson moat]

electrode and pure aluminum ingot saves on a lot of quartz for a little bauxite

you are gonna run out of resources if you try to make 10-100x more of stuff than usual, you can do it at p1-p3 but p4 is way heavier

[obtuse leaf]

i have to check
so i am doing sloppy alumina -> electrode alu scrap -> pure alu ingot

[crimson moat]

that doesn't consume quartz

so you must be using it somewhere else

[obtuse leaf]

oh - the quartz is being consumed by the oscillators for electronics manufacturing

at some point i'm going to have to either give in a modded "mk4" miner or stop trying to produce so many/m, but we'll see

[crimson moat]

supercomputers like this kind of is electronics manufacturing

just with a big scale up that you likely dont have resources for

since you already scaled up with mods

[obtuse leaf]

ah yeah all the computers/etc are being consumed by adaptive control units, so that has to come back somewhat probably

[obtuse leaf]

i made it to end of ph4 a couple times in the past but i didn't like just hand-delivering stuff to the elevator i wanted to feel like i had to have a logistics network in order to make it

[topaz sundial]

Im gonna need 1e+2 Paleberrys

[pulsar lagoon]

do you like the bus system?

[pulsar lagoon]

xd thats too much

[topaz sundial]

theyve achived mitosis

[obtuse leaf]

i like the idea of keeping it clean, we'll see how long until spaghetti starts 😅

[topaz sundial]

and red obital strikes

[pulsar lagoon]

first time playing it, since i discoverd it am more happy

[viral sparrow]

actually this seems kinda fun, heavily producing spelevator parts

not to that extent by any means

[pulsar lagoon]

at the begining i was reallly worried about shards, but after knowing you can craft it it not a need

[pulsar lagoon]

why too much? is a bug?

[viral sparrow]

mod

[topaz sundial]

real explanitory name for these things

[obtuse leaf]

it's a mod that multiplies them

[topaz sundial]

My mod?

[pulsar lagoon]

why play like that? it seems much harder

[topaz sundial]

with the lizzard doggos?

[pulsar lagoon]

am doing just 5 xd

[obtuse leaf]

i know that whenever i finish ph5 i'll probably never touch the game again, so dragging it out
this is also like the 40th time i've started a world probably, but the "final" one, so i have a fair bit of knowledge of the basics although not so much of late ph4 and none of ph5

[topaz sundial]

Do not ask why i have so many

[pulsar lagoon]

ok, this is my first time but i did finish the game, i dont know if am the only one but it taked me 700 hours xd

[wind spade]

no

[pulsar lagoon]

the first 500 ish was like just doing nothing

[pulsar lagoon]

why? is really fun

i love it, before was really hard

[wind spade]

you build 4 times as many belts and centralise production for no reason

[topaz sundial]

Can i have some help or is my base owned by them now?

[viral sparrow]

some people want a challenge

[viral sparrow]

go far away and crouch up to it so you can get rid of them

[topaz sundial]

Hold my homing rifle ammo

i just got about 125K in my envintorry all at once

these things are dying or my soul is

Megalovania just starts playing

wait how did they spawn in there?

my base is in the sky

[viral sparrow]

my concern is how is there so many

[empty quartz]

Replaying the game for 1.1 bc why not… and just remembered how much this game kicks you off a cliff after phase one

[topaz sundial]

dude we are in the air

[empty quartz]

Why the hell do y’all play w booty graphics

[viral sparrow]

pc issue ig

[topaz sundial]

i threw three clusters at it and it didnt die

[viral sparrow]

throw nukes

or use explosive rebar

idk if that actually deals splash damage if theres multiple in one which there seems to be there

[topaz sundial]

two at once didnt work.

it took 15 at once

in the air.

Now the next corner.

it has the OP stingers

[outer vale]

which part of this is math and which part is meta?

[topaz sundial]

yes

[pulsar lagoon]

this way the game is more fun and you dont have problems

[pulsar lagoon]

seems like a job i prefer have fun

[thorn trail]

play the game in a way that you enjoy it

[pulsar lagoon]

agree

[dreamy nimbus]

and they told me supercomputers were annoying
#screenshots message

[wind spade]

you do you, but most of the people I've seen here have had problems

and "more fun" is ofc valid reason, though I do not find having to build 5 times more belts "fun"

[pulsar lagoon]

what would happend with this set up?

can drone fuel 1 go to 2 drone ports?

[pulsar lagoon]

i dont feel like building a lot of belts, the thing you get are better so i prefer build more

[wind spade]

but with bus system you build way more belts than with f.e. independency

[dusky dust]

Priority Merger's gonna encourage more bus users, I'm guessing. :P

[pulsar lagoon]

and you get crazy xd, also is easier to craft more items without thinking too much

[dreamy nimbus]

that's coming ?

isn't that insane wait a min ?

I've just plugged a factory of 5 Alien Power Matrices in Modeler and get a result a 33'982 MW used max

thing is, it doesn't require a ton of different factories, but it requires a lot of the same factories

[opaque quartz]

this should work. set up one drone port as the outbound fuelling port, and then build however many remote drone ports to "fetch" from the fuelling port. build the drone on the remote port and manually fuel it one time (enough to make the trip to the fueling station). it looks like this is how you have it set up

[opaque quartz]

APMs are silly stupid to make tho. so you have to decide if that 30% boost is worth all that complexity and power cost to get that production line stood up

[crimson moat]

building power on a large scale is also generally easier than the other stuff that sloops are useful for

[dreamy nimbus]

I guess you could be building like 10ppm slooping to 20ppm for your 4 APAs, then you can use sloops for other things

[crimson moat]

That would be using 40 sloops to create APA's, which is the part that is not really worth it.

[dreamy nimbus]

true and there are like 100

that leaves you with 60 sloops for your world, that's a lot no ?

[crimson moat]

a lot less than 100

[dreamy nimbus]

because 1.3^4 = 285.61% boost

that's huge for just 40 sloops nah ?

I mean I guess power is never really a problem but still

[crimson moat]

40 sloops and all of the power matrices

it's also not 1.3^4

it's 100% +(4*30%)

[dreamy nimbus]

it's 30% per machine though

[crimson moat]

yes and they stack additively, not multiplicatively

[dreamy nimbus]

who tf said that they stacked multiplicatively then

[opaque quartz]

wiki has the formula if you want to math it out: https://satisfactory.wiki.gg/wiki/Alien_Power_Augmenter#Usage

[crimson moat]

You did, when you said "1.3^4 = 285.6%"

[opaque quartz]

For example, existing generated power produced is 5,000 MW. Building an APA generates an additional 500 MW, bringing the base production to 5,500 MW. Next, the base production (5,500 MW) is boosted by 10% multiplier, resulting in a total power capacity of 6,050 MW (5,500 MW + 550 MW = 6,050 MW). If for example two augmenters are built, the base capacity becomes 6,000 MW (5,000 MW + 1,000 MW ), and is then boosted by 20% becoming 7,200 MW (6,000 MW + 1,200 MW = 7,200 MW).

[dreamy nimbus]

someone else did it before me, that's why I thought it was this way

[crimson moat]

well, it's not

you can choose to spend 40 sloops and make all of the power matrices for 2.2x power gen, or you can just build 2.2x as much power

augmentors are more useful when gross and net power are closer together, like for ficsonium, as the gross power is multiplied rather than the net.

for example if you make 100gw and consume 80gw, then your net is +20gw

but 4 augmentors turns that into 220-80 = +140gw net.

which is 7x as much net, useful power

[dreamy nimbus]

what

it doesn't take the whole power you produce but only the one that's being used

[crimson moat]

Augmentors multiply your gross power

[dreamy nimbus]

what do you mean by gross power

[crimson moat]

Total power gen

for example if you spend 80GW to turn waste into 100GW of ficsonium, you only get +20GW out of it because you spend 80GW to produce 100GW.

[dreamy nimbus]

isn't your capacity supposed to be your max production though ?

but nuclear plants only work IF you need power

[crimson moat]

However if you use 4 augmentors, your 80GW of processing costs stays the same but now produces 220GW of power.

The sum becomes 220-80 = 140, instead of 100-80 = 20, so you get only 2.2x the gross power but 7x the net.

[dreamy nimbus]

I'm analysing

[dusky dust]

Yes, in 1.1

[dreamy nimbus]

when priority fluid merger

[crimson moat]

same time as the valved pipe

[dreamy nimbus]

valves suck

theyre weird

[crimson moat]

Valve and valved pipes are two entirely different things

[dreamy nimbus]

you really did meant valved pipes ?

what will be the point of valves then ?

[crimson moat]

what is the point of valves now

[dreamy nimbus]

fair

[crimson moat]

they can't force directional flow through a junction or connector (like a pipeline support)

but if valve was part of the pipe itself, or part of the junction, it could

think smart splitter that can toggle flow directions on different connections, or pipe that only goes from A to B and not back.

They probably don't wanna do that though because it basically makes pipes into belts with headlift

[dreamy nimbus]

you took it to the nuclear point, I haven't even done nuclear since I didn't needed it, I'll do it tmrw because I'm hella close my capacity line because of my new nuclear pasta factory, so I had to remember that nuclear plants only work if you need said power

Then I had to realise what you said, comparing what my brain was mathing along you and finally understanding

[crimson moat]

It's even more specific towards Ficsonium even, but it's true for coal onwards. You need to spend power to make power.

If you spend 10mw to make 100mw, you only get +90mw of power

but the 100mw gets multiplied by the power aug, not the 90.

Ficsonium takes it to an extreme where it's like, you must spend 80gw processing it to make 100gw of power.

so it's much more appealing to get (100x2.2) from your 80gw, rather than 100.

[dreamy nimbus]

makes sense it multiplies your whole network, not only what is being produced by your new facility which means what you now produce and what you were producing, in those thing you were producing there was this 10MW (i'm making it more complex for no reason don't mind my wording I agree with you)

[dreamy nimbus]

well, it's what the pioneer think is worth it, is x amount of somersloop invested in APAs worth it or no

and then when you invested them, even though the factory is costly, the Return On Investement of the Alien Power Matrix is insane

100*1.4 = 140 ; 140-80 = 60GW of profit
100*2.2 = 220 ; 220-80 = 140GW of profit

[crimson moat]

I still don't think that you are understanding gross vs net, and why having power augmentors or not makes it better to use certain types of power

[dreamy nimbus]

10ppm alien power matrix factory might be even bigger than BWP 1ppm

[crimson moat]

but yes, each one has less value than the one before it

[dreamy nimbus]

gross power being the power you used to make net power which is the profit
APA, multiplying both of them makes Power that takes a lot of gross power for way less net power way more profitable

which can make it more appealing to the player to make harder power plants

It took me a while to understand but at last I did

[crimson moat]

"gross" refers to a total amount before any deductions, while "net" refers to the amount remaining after those deductions are made
Yeah, think you got it other than this backwards 😄

[dreamy nimbus]

fuck

[crimson moat]

spending 10gw to make 100gw
= gross 100gw, net 90gw
if we add an APA then the net increases from 90 to 120gw (+33.33%)
if we do it again it increases from 120gw to 150gw (+25%).
A tenth APA is only +8.33% net power

spending 80gw to make 100gw
= gross 100gw, net 20gw
if we add an APA then the net increases from 20 to 50gw (+150%)

So you can see that the APA is much more valuable when you have fewer of them, and when your power plant is consuming a large percentage of the power that it generates. Ficsonium-based nuclear is the best case scenario for using APA's.

[dreamy nimbus]

oh gross power being the power made and net being the actual profit when you remove the power it took to build the plant

[crimson moat]

yep

[dreamy nimbus]

thanks for your insights

btw I like your pfp, it reflets the calm and composed person you are

like you deserve it or something

[empty maple]

haha wow, I was going to share some knowledge about nuclear power waste chains, but ya'll teachin me. TY math nerds. you're the best

like i'm rethinking my entire nuclear setup with sloops now...

[tawdry blade]

Red- any undefined
Green- overflow, into awesome sink

Conveyor lift feeds back into the storage- to keep it cycling.
Overflow should only occur once the storage is actually full,
correct?

[thorn trail]

if you put the smart splitter at the entrance to the storage you won't have to loop it

[tawdry blade]

Yeah no space there left. But I will remember for next time.

[thorn trail]

that splitter configuration will also limit the speed of the DD filling to half of the belt speed coming out of the container

[pastel obsidian]

The spitter has no clue about how many items are in the storage

You would want to put the overflow splitter before the storage for overflow into a sink

[dreamy nimbus]

what am I looking at

[crimson moat]

Cleaned up example of distributing a 600/600 pipe of fluid to many consumers via split manifolds

This particular example splits 1 pipe into 12 equal arms, but as few as 3-4 arms is probably sufficient for any reasonable setup

[wary tulip]

And because it is balanced... there is very little sloshing, correct?

[crimson moat]

Yeah, there isn't problematic sloshing when pipes are split equally with a junction

or when the usage of pipe capacity % and the amount of serial unequally-split-junctions are limited

If you put 96 consumers in one serial line, even if you feed it from both ends, you can get problems. If you split like this then it's extremely robust while adding little complexity or build work (though again, 4 arms is probably fine)

[pastel obsidian]

How are you pumping enough fuel to power so many fuel gens

[crimson moat]

They're underclocked to 6.25 ea to exaggerate the sloshing problems

that is very close to turbofuel @ 100% though (7.5) which can run 80 generators on one pipe (but not if you put 80 junctions in series)

[tawdry blade]

I have heard sloshing is caused by pipes that arent elevated above the end point, resulting in backflow into the main pipe, which then hempers throughput

[civic bronze]

Maybe it works like that, but i wouldn't call it a "cause", sloshing is just what pipes do/ programmed to do

[pastel obsidian]

Sloshing is just pipes trying to balance themselves, it only becomes an issue when the pipe doesn't have the capacity to handle the extra volume

[tranquil siren]

Making 3600 steel ingots/min at new iron project. Is an 80:20 ratio of Pipes to Beams reasonable?

[dreamy nimbus]

why do you need so much of them

[frosty owl]

You might find this convo interesting ^^
#math-and-meta message (topic: pipes' "load balancing")

[tranquil siren]

got 9360 iron ingots being made, why not?

[topaz sundial]

Um what happened to my reactor?

[civic bronze]

Math happened

[frosty owl]

A cosmic ray flipped the "visibility" bit of that building

[wind spade]

I'd suggest making them when you need them rather than guessing in advance

[pastel obsidian]

Melted

[topaz sundial]

well thew radiation hasnt stopped so idk what the hell happened

[wind spade]

@charred delta all alt recipes are good in some scenarios, just pick whichever you like or flip a coin if you don't know. You can get them all anyway

[viral sparrow]

^ go for what you need currently, e.g if you need supercomputer production and you want to use that recipe, go for that, but if your setting up nuclear, go for infused uranium cell

[hoary wigeon]

ouchy me brain

[wind spade]

completely unreadable lol

[hoary wigeon]

for you*

[wind spade]

definitely not just for me

[hoary wigeon]

once you embrace the spageth you cant really go back

oooo wait i had forgotten that sommersloop existed

how many are there in the world ? like 108 or smth ?

[fallow siren]

106, 3 for mam

[hoary wigeon]

oh alrighty

[vapid gorge]

yeah this is a pile of poo, sorry

[oblique hollow]

i finally did it and went down full insanity path:
trying to model SF pipes in Excel

[vapid gorge]

I'm pretty glad I can't read any of that either xD

[oblique hollow]

Biggest pain is getting it to work like ingame

[oblique hollow]

the funny blue bars are pipe content
and the big bold numbers down below are flow

thats all you need to know

[vapid gorge]

are they individual , non connected pipes?

[oblique hollow]

this is a pipeline made up of 5 segments with a source (left) and sink (right)

they are all connected

[vapid gorge]

riiiight. How do you decide how much 'pressure' there is/

[oblique hollow]

pipe content

as they do ingame

problem is that this isnt close enough to ingame

as pipes can have the same content and still have flow

in my model here, only a content difference leads to flow

and getting it to act stable with equal content is a p a i n

In SF, pipes determine how much outflow they can have based on their content.
With vertical pipes, it is the same, minus vertical differences

[vapid gorge]

I'm very happy you managed it. For you. I'm.... not even going to try to break it down after 3 months of insomnia

[arctic saffron]

@crimson moat does this work?

its 10 ontop and 10 below and 200 in from left

[oblique hollow]

i didnt, its still terrible and not like ingame

but its a start i suppose

[oblique hollow]

10 generators on the top and 10 on the bottom?
what fuel type

[arctic saffron]

rocket fuel

[storm imp]

How I felt after setting up a factory to make 20 Iron Plate / Min 🤣

[crimson moat]

probably

[oblique hollow]

thats only 41.666/min for each floor

[crimson moat]

i prefer not to vary height with the same pipe but it's probably not an issue

[oblique hollow]

unless you overclocked em

[arctic saffron]

240%

[oblique hollow]

aight, works out

should be fine

[arctic saffron]

i am so stuck with this layout then

[oblique hollow]

though the input should be a mk 1 pipe if its only 200/min

[arctic saffron]

i have a 600 pipe going into 3 mk1

[crimson moat]

mk2 will always be better than mk1

[oblique hollow]

i prefer not to use a mk 2 if its not needed

200/min input => mk 1 suffices, has 100/min spare capacity

PLUS, it goes into a junction that goes only into a mk 1 anyway

so why have a mk 2 on top

[crimson moat]

The way that i look at it is that mk.1 can break a setup via overloaded flow rate, but mk.2 won't harm it, right?

[arctic saffron]

that mk2 is the 600 splitting

[crimson moat]

you can even limit a mk2 to 300 with a valve

[oblique hollow]

its better to use the lower mk IF there is a junction on top where you split this from

[arctic saffron]

thats the rf being made

[oblique hollow]

lower pipe mk is better than valve when connected to a junction
because you cannot put a valve directly onto a junction

the junction can deal with it more easily

because it doesnt expect to be doing an equal split into mk 2 pipes when the connected pipe is mk 1

[obtuse leaf]

also if you set a valve to 300m3 it lets in more than 300m3, but a mk1 pipe won't

[oblique hollow]

not anymore from what i (and others first) found out.
Its a display issue now again

pipes, pumps, valves - they all terribly round the flow rate

[vapid gorge]

In general, even with gasses, I wouldn't split a manifold over multiple floors and branching it

[oblique hollow]

the actual flow is pretty accurate

[crimson moat]

Yeah, my 600 splitting into two reads as 2x298.

but it doesn't starve, and it would if it was any less than 300.0

[oblique hollow]

yeah thats just the flow rate display at work

same way you sometimes get pipes that have a flow of 0 in a looped network

but you can bet that there is flow

[obtuse leaf]

that's good to know

[arctic saffron]

should i just delete my current 20 fuel gens and move them to over the edge of map so i can just lay them all flat like that

would make it simpler

[vapid gorge]

simpler to just have 2 pipes from the fuel producers

[obtuse leaf]

simpler is boring!

[vapid gorge]

don't merge all of them in the first place

It's less boring than just making a sky factory over a cliff

[obtuse leaf]

world grid is overrated, embrace lack of grid

[crimson moat]

Yeah don't merge them, or split the pipe before manifold

[vapid gorge]

better to not have them merged in the first place. As simple as possible. As few splits and merges as you can manage

[topaz sundial]

1 bc it only requires a assembler,

[arctic saffron]

so that or?

and then 20 on eat side in a manifold

[wind spade]

I'm not asking though?

I'm saying to them to pick what they want

[arctic saffron]

am i able to connect 2 uel gens to the same intersection or should i avoid that

[grave grove]

Haven't played in awhile but it is still a strat to increase pipe valves by 10% to allow for proper recycling flow? Most specifically the encased uranium cell recipe.

[junior escarp]

is there a way to make factory layouts not in game?
Trying to make a factory for both motor and stators :P
(sort of like plan it out first hand before i start building it)

[dreamy nimbus]

like designing your factory chain manually ?

[junior escarp]

sort of yeah

[dreamy nimbus]

Satisfactory Modeler, on steam

[junior escarp]

ty

[dreamy nimbus]

np

[junior escarp]

not really the thing I was looking for but it's still useful
Like how to actually make the layout, where to put the smelters/foundries/constructors etc.

[dreamy nimbus]

I'd like a tool like that sadly I don't think it exists

What I usually do is do the thing in reverse, placing the last bit of factory then working my way around it