math-and-meta

[ashen girder]

Getting that far and being like "wtf am I even doing?"

[feral breach]

Moving just 60/m copper with train will feel weird but... its a start 😄

[ashen girder]

That's cause it is a lil weird. I hate moving ingots and ore by train.

[glossy schooner]

feels like you'd want to scout more nodes than that

[feral breach]

Ill move cable and sheets

but its only production of 60

for one computer machine 😄

[ashen girder]

Considering the sheer number of ways you can make cable, I wouldn't.

[feral breach]

I either need to move 80/m plastic to a copper vein or move 60/m copper production here 😄

[pallid spoke]

Whats wrong here?

[ashen girder]

4 different recipes for cable itself, 4 different recipes for wire, so many recipes for ingots.

[ashen girder]

One of those Ps is a B.

[cursive heron]

only if you use basic oscillator recipe, otherwise insulated crystal osci needs AIL which needs quickwire

[pallid spoke]

Its taking the whole station what?!

Placed the block further into the station fixed it

[gaunt umbra]

that path signal after the intersection seems weird

[pallid spoke]

something is wrong with the junction

[gaunt umbra]

the train would stop at the station so it can't leave the path signal block

[cursive heron]

yeah was about to say

[ashen girder]

Your P is still a B here.

[cursive heron]

the entire middle section is one big block for some reason

[ashen girder]

Probably because it isn't a valid junction with that species-swapped signal.

[cursive heron]

something janky happened with the signal being on the merge itself

[pallid spoke]

Which one?

[ashen girder]

Oh, is it a P now?

It's hard to tell. 😂

Anyway. I would pull all of your signals back slightly from the splits. Signals right on splits sometimes get Weird ™️.

[pallid spoke]

that works but that path its it

[gaunt umbra]

just had that issue actually

block signals on splits seem to allow both rails still

[pallid spoke]

The block on the station entrance doesnt like being close

[ashen girder]

It's because it's on the split, I think.

[cursive heron]

the issue was because this entire section is one big block

[ashen girder]

I mean, everything you've drawn red on here but the station should be.

[cursive heron]

when it should be like this

[teal urchin]

My brai nis failing me atm. If I produce 300 oil/m, using the heavy oil residue alt + the diluted packaged fuel alt, how many fuel gens should I be using?

[ashen girder]

Yes. It would be like that if the block in the station weren't on a split.

[pallid spoke]

Adding the block to the station entrance doesnt shorten the block weirdly enough

it dont care

[ashen girder]

Because it's on the split.

[pallid spoke]

Thats never been an issue till now

[ashen girder]

Then you've been lucky.

[pallid spoke]

ig I have a block inside the station ig

im probably gonna move the intersection back more

[ashen girder]

That's probably gonna block the junction any time a train's stopped.

That's the best solution IMO. Add some space between the station and the junction.

[pallid spoke]

and or a train is gonna get rammed in the back

[ashen girder]

Path signals will prevent that.

[pallid spoke]

true but still, trains are gonna stop for no reason. So ill move it back a bit ty guys

[cursive heron]

try deleting and replacing the track at the entrance of the station see if that lets you put the signal at the split

[ashen girder]

Most likely, yeah.

And yeah, that might work. Most of the time, signals do work fine on splits. But sometimes they just decide not to.

[pallid spoke]

First time for me :)

[ashen girder]

You might try rebuilding right there first, like Rai suggested. Might fix it for you.

I definitely wouldn't want to leave a block inside the station though. That seems like it's asking for trouble.

[pallid spoke]

💯

[remote ice]

if a splitter has two belts coming out of it with different speeds, it generally distributes more items to the faster belt, right?

[ashen girder]

It's not consistent.

[edgy leaf]

i made some new funny discoveries with the wonky pipes!

[ashen girder]

oh no

[edgy leaf]

i multiplied the amount of refineries by 4

[versed violet]

from observed behaviors - splitters and mergers work in round-robin fashion, unless one of the belts is too slow to accept/give items.

[remote ice]

ah okay

[edgy leaf]

now it gets wonky even with 50 meters of headlift.

[ashen girder]

See! I told you! It's not just a headlift thing!

[edgy leaf]

you might think, its the sloshing! its not related to headlift. but you would be WRONG. if i use 320 meters of headlift it works

[ashen girder]

No.

[edgy leaf]

yes

[ashen girder]

Just no.

[edgy leaf]

yes.

[ashen girder]

I fuckin' quit.

[edgy leaf]

my hypothesis was that each junction somehow "removes" headlift to the next junction

so more junctions would mean more headlift required

and it seems to be true

[ashen girder]

Quick question: how many pumps is that compared to one between each junction?

Alternately: can you try using the same number of pumps spaced out evenly between junctions?

[edgy leaf]

uhh its the equivalent of 6.4 pumps, however its not that it started working at that amount of headlift

i built the refineries, then i built the tower, then i tested the mk2 pump and then the tower

[ashen girder]

Oh, you just decided to go big.

[edgy leaf]

so 100 might also be enough

or even 60. i just went straight to 80 4 meter foundations

i can send u the save if u want

[ashen girder]

Nah, I'll just 🍿 from here.

The saga of perfect bottomfed machines continues. 😂

[wind spade]

Depends on belt speed and input rate

[edgy leaf]

just add a kilometer of headlift everywhere, DUH

[glossy schooner]

see that's what i thought

[bronze coral]

Well the pipeline manual says this isn't the case

[ashen girder]

But really, if what you're saying is correct, then there should be a way to calculate how much headlift you need to account for through each junction right?

[glossy schooner]

if you have input of 90 and 1 output is mk 1 and another is mk2, wouldn't it still split 45/45?

[ashen girder]

We're very, very far beyond what the manual says.

[bronze coral]

Fair enough xD

[ashen girder]

Nothing so simple or straightforward, no.

[edgy leaf]

yes, thats what id love to do. it sounds like itd be relatively easy to do if i had access to the debug commands :/

[ashen girder]

IIRC it tends to over favor the slower belt?

[edgy leaf]

even mcgalleon isnt sure whats going on

[glossy schooner]

once again proving that uniform belt speeds are superior factory design

[ashen girder]

I tried messing around with them, and while I got some of the debug commands to work, I couldn't get any useful info out of them. 😦

[edgy leaf]

yea, the good ones are locked away. i hurt running satis on a dedicated server might give u access, im sure it can definitely work with mods

maybe mcgalleon has an snwer

[ashen girder]

Yeah. I think it winds up doing something stupid like 48/42 with 48 on the Mk1. Ultimately though, I'm pretty sure it just isn't consistent no matter what.

[ashen girder]

And yeah, like 0laura said, we've been involving McGalleon in these conversations too. He's just about as mystified as we are, I think. 😂

Mind, we're talking about a super niche, specific scenario.

(That people happen to run into a lot because it looks way better than the alternatives.)

[bronze coral]

Well in that sc I don't see any loops 🧐 I think that's something you should try

Unless the loop is behind the pipe

[ashen girder]

Nope, no looping.

[edgy leaf]

theres no loop, but it works. im not confused by it not working, im confused by it working

[ashen girder]

It's evil and defies all logic.

[bronze coral]

You mean it suddenly working after a threshold of headlift? Yeah that's pretty wonky xD

[ashen girder]

It tracks with the other solution we've found: throw a Mk2 pump between each junction.

[edgy leaf]

yea, the theory that makes the most sense to me right now is that each junctiion somehow swallows some of the headlift

[ashen girder]

Honestly, I think that makes sense.

If you watch the input pipes, they drain every time the machine cycles.

So the pump has to literally push the water back up.

[edgy leaf]

that should cause sloshing and maybe stalling but it shouldnt be fixed by more headlift

[ashen girder]

Unless headlift actually does apply vertical pressure.

[edgy leaf]

like, in the past the wisdom was that because the fluid level goes down in one pipe, it goes down in them all because they try to equalize. that causes sloshing and makes the fluid go below the point where the machine can consume

[ashen girder]

Then more of it means more pressure.

[edgy leaf]

but that should make there be no fluid due to there being no fluid, nothing with subtracting headlift

[wind spade]

afaik snutt confirmed there's no pressure

[ashen girder]

Even without sloshing, though, the difference between bottomfed and topfed is bottomfed the pipe does drain.

[wind spade]

just headlift + amount of water in pipe
the only two parameters that dictate flow

[edgy leaf]

yup. and headlift is either enough or it isnt (besides some fuzziness at the boundary maybe)

[gaunt umbra]

alright cool, intersection signals is complete

[ashen girder]

Your own evidence says that's not true, doesn't it?

[edgy leaf]

but this thing doesnt work with 50 meters of headlift, it does work with 370 meters of headlift...

[edgy leaf]

yes, what i said is how it should work, how we thought it works. that seems to be wrong

ooh yes, i have a good idea. ill place a mk1 pump in line with it and see what happens...

i think this would be the definitive proof

[ashen girder]

Because it would reset the headlift right?

[edgy leaf]

yes. one could argue that the big tower makes it work because it acts as a buffer

well one cant argue that now

the red thing in the distance is hte mk1 pump, i loaded the savve i made when the system was flooded, deleted a pipe segment, placed the pump on the ground, reconnected it and turned the machines on...

its stuttering HORRIBLY

this is hilarious

[ashen girder]

b-b-b-b-b-b-b-b-

[bronze coral]

Yk your theory that junctions suck up headlift might actually be true

Testing it right now and it has some really strange behaviour

[edgy leaf]

i replaced it with a mk2 pump and it still doesnt work, removing it now

i removed it and it instantly has nearly totally consistent flow...

like, i didnt even flood it again... i just removed the pump and re placed the pipe... instant perfect flow

thats the result of 370 meters of headlift babyyyy 🔥

eating dinner now cya

[bronze coral]

W/o the junction there it fills the tank fully, once I added the junction its only able to fill up 15 m^3

[proud totem]

So the solution to all of our problems is an obscene amount of headlift

[bronze coral]

No, the solution is to keep your junctions above/ on the level of the machine

[proud totem]

Unless you want to bottom feed

[bronze coral]

If you really want to bottom feed you could add a pump after every few junctions I suppose

I think it'd be ugly to have a massive tower in the middle of your build, but that is a solution as well

[proud totem]

I'd be curious to see how much headlift each pump has to add/how much head lift each junction removes in a bottom feeding case

[oblique hollow]

Hard to say as its not as much "removing it" and more "interrupting it"

[fringe crow]

What Alt Recipes should I focus on early on?

[ashen girder]

Solid steel! Cast screws!

[proud totem]

I guess by removing I mean how much additional headlift is required for normal operation to resume

[oblique hollow]

Any that can help with deficits jn your factory

[kind silo]

does anyone know the avergae time a lizard doggo generates an item?

[bronze coral]

But the fact that you can overpower it with sheer headlift makes me think otherwise

[proud totem]

So perhaps "offset" may also have been a better word

[glossy schooner]

These are the lengths bottom feeders will go

[proud totem]

I wonder in the case with 370 headlift if at the end of the line how much headlift remains. Like can the water get all the way back up 370?

[bronze coral]

How do buffers work by the way? I put one at a certain height and it fills fully, I try it again 1 meter higher and no fluid flows into it, I figured it'd fill to 7 meters

[proud totem]

And if not 370, then how far

[oblique hollow]

Seems weird.
It should fill 1 m less, not not at all

[livid meteor]

Is 4 freight wagons enough to transport that?

[orchid pond]

maybe. you can always just add another train

[pallid spoke]

depends on how many trains and distance

[livid meteor]

If I remember correctly 1 could carry 900~ per minute

[orchid pond]

entirely depends on the route distance

[livid meteor]

I'm building the trainstation right now that's why I ask. Need to know how many platforms

[oblique hollow]

Gas is usually packaged

[bronze coral]

Huh, I'll try restarting my game I suppose

[oblique hollow]

As thats a lot more efficient

[livid meteor]

Wait, I thought that stuff is liquid

[oblique hollow]

Nitrogen is a gas

[orchid pond]

the clue is in the name. nitrogen GAS. lol

[livid meteor]

meh. I think feeding trains with pipes is pretty cool

I just wanna know how many train wagons I need for that nitrogen gas

[orchid pond]

valid

[oblique hollow]

Just know you cannot buffer the gas

[livid meteor]

wait really?

[oblique hollow]

Fluid mechanics doesnt work out the same as liquids

So buffers cant do their job

[livid meteor]

that sucks

I guess I lose some then

[oblique hollow]

Which, again, is why packaging is preferred
4 m3 per bottle instead of the 1 m3 as with liquids

[orchid pond]

ah! that explains why packaged gases require aluminum. thanks for that info

[livid meteor]

so basically the train station can't store the gas?

Oh wait fluid buffers can't

[oblique hollow]

The station can, stations are not normal buffers

[livid meteor]

But what if I build 2 train stations?

if 1 gets loaded the other could work as buffer, no?

[oblique hollow]

Well the answer is always " what if i build more of x"

[livid meteor]

Yea. I know. I just wanna use fluid transporters ngl

It just feels bad know that the station won't get filled while a train is docking

[pallid spoke]

When mods get added back im getting that nonstop train mod

[livid meteor]

huh? @oblique hollow

[pallid spoke]

The one that fixes input/output loading stops during loads

[livid meteor]

It seems to work

[oblique hollow]

Just cause it goes in doesnt mean it will do what the buffer should act like in order to buffer

Buffers are supposed to fill up and then discharge liquid at max speed

[wraith ruin]

I bet it steals flow from the station instead of outlfowing when it should

[oblique hollow]

Gas does NOT act that way in buffers

Buffers act like pipes when it comes to gas

all you gotta find out now is how pipes interact with gas

[livid meteor]

I see

damn that sucks

Man I really wanted to work with train stations badly 😦

well fluid trains*

cuz I never use them and I thought that would be a good use case

[proud totem]

This is only for gas though, right? Like liquid should still work (even though packaging is still better)

[edgy leaf]

yes 💪😎

[proud totem]

0laura you should test this if you could? I am curious to see the result
#math-and-meta message

[livid meteor]

why is gas getting discriminated 😢

[oblique hollow]

Because gas is gas

[bronze coral]

I'm trying to test that as we speak, I'm trying to figure out how to measure the current head lift in te pipe accurately

[oblique hollow]

And has no head lift

[ashen girder]

That means trains are gonna have trouble moving nongasses?

[bronze coral]

Ah it looks like buffers are working normally now, so I can use that

[livid meteor]

more like gass is @$$

[oblique hollow]

Yes non gasses are basically gasses

[proud totem]

Oh cool! I figured you could just build individual pipes up and see where the fluid stops going to

[next pewter]

can anyone explain me, what's going on here?
(a refinery, doing alu scrap)

[bronze coral]

Yep but I can only look at each pipe segment then, a tank would be more precise

[oblique hollow]

All this means is dont move your freakin dark matter over long distance

[oblique hollow]

rebuild

[next pewter]

well, its working

[proud totem]

Either way I suppose. More curious if it is drastically lower than 370 (or whatever you have) or close enough

[next pewter]

and going brrr

but its a bug, then?

[oblique hollow]

No clue what you did

If you cant replicate it its not worth reporting much

[urban kite]

480/min of rods, screws, and plates 🙂 right into the depots

[bronze coral]

I have a question about the buffer, if it is filled, say 300 m^3 (3/4th of capacity) will it output 6m of headlift? or is the volume not proportional to the headlift

[oblique hollow]

It is proportional

For further questions, please check out the manual

Found in the pins

[bronze coral]

Ok thanks! With that information it looks like junctions remove 0.376 m of headlift, though if I had to guess this is probably not consistant for all systems

[bronze coral]

Oh is that mentioned in the manual? I forgot, my bad

[bronze coral]

Yeah this is pretty erratic... A second junction removes 0.252 more meters

[ashen girder]

#thisisfine

[bronze coral]

And even more weird, If I actually attach a pipe to that junction then it goes down even further

[ashen girder]

I dunno, I just work here. 😂

(I don't actually.)

[proud totem]

I wonder if junctions are splitting/distributing head lift?

While also outright removing some

[oblique hollow]

Please check the manual for that

[bronze coral]

And it depends on the length of pipe attached

[proud totem]

What we are observing seems to be contrary to the manual

[oblique hollow]

The head lift sharing is normal

and stated in manual
You are now looking at something else in addition to the normal behaviour

[bronze coral]

But adding a junction should not reduce headlift, no?

[livid meteor]

Theoretically I could just package and unpackage the gas without consuming the empty canister right?

[oblique hollow]

I will not answer this for i cant

[oblique hollow]

Yes

[bronze coral]

This is my setup to test this fyi

[livid meteor]

Nitro rocket fuel doesn't seem fun to make

[oblique hollow]

Mind you that if i test junctions at sea level (swamp or west coast) sometimes a single junction just shoots fluid 80m up

Oh wait i just remembered head lift loss exists. I am silly

[proud totem]

That is what I mean, at a junction I wonder if it is actually splitting head lift between the pipes (i.e. the pipes are receiving significantly less head lift than "earlier" in the chain), instead of sharing. Unless that is what the manual implies?

[bronze coral]

Huh?? Like as a bug, it just generates 80m of headlift?

[oblique hollow]

I dont know how often it is applied, but the game does have code to deduct a fixed percentage of head lift
As to prevent self-inductions

[oblique hollow]

Yes

[oblique hollow]

Its possible this is where head lift loss is applied

Sorry but i really JUST remembered this mechanic exists

As it is obscure pipe code knowledge

[light fox]

anybody know how much petroleum coke a coal generator takes per minute?

[bronze coral]

Well that would explain it xD

[oblique hollow]

25

[oblique hollow]

I havent tested this in 1.0 yet at all

But it would explain partially why an old bug is now gone

[bronze coral]

Ah I see, I would look through the code but I have no experience with that sort of thing

[livid meteor]

Meh, I go sleep. After hearing that gas via train is a bad thing I'm demotivated :(. What about drones?

[oblique hollow]

An old bug that could make 2 mk 2 pipes on a ramp power each other to infinite height

[bronze coral]

I've heard drones are good for nitro

[proud totem]

So you say this is basically what happens at a junction, where epsilon is some smallish amount?

[oblique hollow]

That bug is definitely gone now

[next pewter]

i didnt, its in a multiplayer server. i just see this in a machine 🙂

[bronze coral]

Ah yk what I should test the headlift in the branching pipes as well, I'll get back to you on this

[next pewter]

yes, packaged can take 6x more than the m³ value.

[proud totem]

Though from the original example that started this discussion, it seems epsilon may not be so small, since it made a system eventually need 370m of HL. Unless something else is at work

[livid meteor]

Do drones actually consume fuel rods?

or do they output waste?

[edgy leaf]

I think the height which the pipes goes up after the junction matters

[urban kite]

yes, no ( i think )

[unique thicket]

am i stupid or does this make no sense at all?

[edgy leaf]

I went from 50 to 370

[livid meteor]

Cuz if they simply consume them I think about just feeding them with rods

[proud totem]

See, I wonder if something like this may be happening in 0laura's example

But with different numbers obv

[proud totem]

Wait, at one junction you lost 320 HL?

Or over all the junctions?

[edgy leaf]

nonono, it didn't work with 50, so I tried 370. it worked with 370

if might also work with 60

[proud totem]

Oooh, I see

[edgy leaf]

it also worked with 320

[teal urchin]

Quick question, when using pipelines, do i need to worry at all about swapping to mk.1's after I get lower than 300 water from my mk.2's?

[proud totem]

I wonder what the minimum amount of HL needed for your setup to work is

[livid meteor]

Maybe I should ditch trains and embrace drones

[oblique hollow]

epsilon as far as i know is around 2-5% ish?

[proud totem]

So its a percentage of the "incoming" headlift?

[spark cove]

Anyone know what is a good amount of reinforced plates per minute is

[edgy leaf]

I think the amount of headlift that gets deducted might depend on how high the pipes goes up from the junctions, not just a static percentage from the junction

[spark cove]

trying to design a farm for it and don't want to under-value it

[bronze coral]

Yep that is the case

[oblique hollow]

should be as its a damping factor "to prevent pipes from pumping themselves up"

[proud totem]

That is why I wonder if the 2nd diagram I posted may be more accurate. 10mHL in, 2m is "needed" in one direction, leaving 8-e remaining (-e from the junction itself)

[proud totem]

Interesting

[edgy leaf]

here you can see that the pipe is directly below the refineries, it felt like having it on the ground made it worse

it's from a different setup (less refineries, higher clock) but same blueprint

with 4 refineries 50 meters was enough

clocking them to 25% and using 4 times as money made 50 not work anymore

[proud totem]

Wait, so same # of junctions and refineries, just slightly higher clock?

[edgy leaf]

it feels like this, but I haven't been able to test it. someone should do that

[oblique hollow]

Mind you pumps actually apply 55m and 22m respectively

[frigid thorn]

wouldn't it be easier to just split the right side off 3 ways, one full belt speed, the other two outputting to mk1 belts. split one of those again 3 ways into 3x 20, then merge one of those into the other mk1 belt with 60 on it, and remerge the two remaining 20s back to the original full speed belt, then load balance the two lines?

[edgy leaf]

no, one had 4 refineries and the other 16, one junction for each refinery.

so one had 4 times as many junctions but the machines were at a fourth of the clock

[proud totem]

Oh, I misread the 4 times a many lol

[unique thicket]

what

[proud totem]

I understand what you mean now

So I still wonder if my 2nd diagram is somewhat accurate then

[edgy leaf]

yea like I said, it feels like it. if you want I can send you my testing save, just fyi sending saves in the server isn't allowed because apparently they could contain malware

[bronze coral]

Man I feel like I'm going insane, just flushing a system and letting it refill gets me a different number for the headlift

[frigid thorn]

you can restrict outputs by using slower belts - so you can make a 60 and a 20 easily, make your 80/min and send it off. Then just balance what's leftover

[oblique hollow]

overshoots / undershoots happen

[unique thicket]

@frigid thorn oh man i have no clue how to even go about something like that, i usually just do manifolds but with the amount of stuff i need per min it doesnt really allow me this time, so it just kinda hurts my brain xD

[proud totem]

I won't really be able to test it unfortunately, don't have access to a PC that I can play on for a bit

[oblique hollow]

This is just something you have to account for with pipes

[bronze coral]

Ok wow then I just have no idea how to accurately measure headlift, would just using 1 m pipe segments not have this problem?

[proud totem]

But my hypothesis for your situation would then be

[oblique hollow]

You never get precisely the same result. You must work within error margins

[proud totem]

By the end, after 16 machines, you've lost a lot of HL

BUT this is just a hypothesis

[oblique hollow]

Repeat one test a few times, Note the data, change the test up (ideally, drastically at first so changes are very noticable) and note the new data

[bronze coral]

Ah man, alright I'll keep that in mind

[proud totem]

That is why I wonder how drastic the drop in HL is between the start and end

[ashen girder]

I just ran out of RIPs while building my Turbo Motor/Supercomputer factory. 😂 Whoops.

[oblique hollow]

i wish i had my test data

[proud totem]

Which I guess is why the 370 and 16 splits is a good test, since it has a pretty drastic change

Since apparently, the drop must be at least 50 (or maybe even 55) since a MKII pump could not handle it

[bronze coral]

Yeah Ill probably do a 100m test as well

[proud totem]

Since if it were just a subtraction of about 5% of the incoming HL, after 16 splits 50m of HL there should still be ~22mHL, which should've been sufficient for the factory to work

[oblique hollow]

Presto! I found it

[edgy leaf]

what

[oblique hollow]

#1071542907091304539 message

[edgy leaf]

ah yes, the good old days, i rember

ive always been obsessed with this

[bronze coral]

Ah wow that seems really fucky

I guess the conclusion is don't bottom feed (🤯)

[proud totem]

And that data seems to coincide with the about 3-5% loss per junction

[edgy leaf]

imma head to sleep, cyall

good luck

[oblique hollow]

Summary of test data in one image:

[light fox]

what recipes could i stick petroleum coke in

[proud totem]

Since at more than 18 junctions with a MKI pump it cannot support the 10m tall pipes, which means that the HL had to be <10m at that point, giving a loss of about 3.8%

[oblique hollow]

Only alternate recipes really

[proud totem]

Lemme check if that is consistent with the MK2 pump and 32 connections

[oblique hollow]

Sure enough 55 * 0.95^32 is about 10

[proud totem]

Very close, now it is about 5% loss (slightly less)

[unique thicket]

thank you, i've done it a little bit differently but pretty much the same, i hate that i didnt think of this sooner lol

[proud totem]

Yep, and I'm sure part of the discrepency between 3% and 5% loss is due to integer amounts of splits and HL success is a binary state: either we make it to 10mHL or we don't

Though I still wonder why this didn't work with 50mHL #math-and-meta message

Let me do the math real quick I guess

With 5% loss at each junction, the HL at the end is 22m. I would assume that would be sufficient for the refineries, but maybe not?

[oblique hollow]

Mk 2 pipes are a lot more finnicky when it comes to backflow

[bronze coral]

Yeah I think fluids needs a revamp, at least in code xD

[oblique hollow]

They prefer to have more excess head lift, else theres more flow rate interruptions

[proud totem]

So solutions:
Keep manifolds short, and if bottom feeding, you are probably gonna need more pumps

[oblique hollow]

yes

One thing i didnt consider: loops

Loops should offer a path that doesnt have as many penalties for head lift

[proud totem]

Like connect the start of the manifold to the end with 1 non-split pipe?

[oblique hollow]

Yes

Or just start to middle

[proud totem]

Was just about to say that

Halves the # of consecutive splits

[bronze coral]

Haha so my original solution works after all lmao

[oblique hollow]

And thus reduces the number of head lift penalties

[ashen girder]

Does this mean you figured it out?

[oblique hollow]

Likely?
If the loop performs better then even more likely

[gleaming shuttle]

how much should i care about wasting sulfur? tempted to use the nitro rocket fuel alt but that thing burns through sulfur relative to more complex rocket fuel recipes

but also i feel like theres way more than enough sulfur on the map as of 1.0?

[bronze coral]

I mean even looking at later messages in that thread it looks like it had some really erratic behavior, but we now know how to mitgate it succesfully I believe

[deft lichen]

don't worry much about wasting resources, it's unlikely you'll run out of them

[gleaming shuttle]

yeah im just imagining long term i might want to transition into nuclear which i know also uses sulfur

[deft lichen]

unless you're knowingly maxing out someting, I wouldn't worry about it

[gleaming shuttle]

but i did some napkin math and it still seems like theres more than enough

[obsidian trench]

So I just had a thought...since Underclocking a machine decreases the power consumption exponentially shouldnt you theoretically build a bajillion machines if you have a lot of space and are optimizing for power consumption?
Not far enough into the game to really be able to tell how much power you are ultimately gonna have, but it feels weird that this is theoretically a strategy...
I guess it really isnt worth it in most cases.

[deft lichen]

it's silly beyond reason

power isn't such a problem

[obsidian trench]

Yeah figured. Is there some crazy power hungry late game machine though?

[remote ice]

its a known strategy and it is, indeed, fairly silly

[deft lichen]

yeah, but you also get better power sources too

it might be reasonably applicable to some tier 8/9 setups, now that you mention it, but I still wouldn't say it's a good strategy

I mean if power is something you have to work around, you're just not building enough power capacity

[obsidian trench]

Yeah was just a "mathy thought"

Since space isnt something easily quantifiable

[deft lichen]

totally valid question though

1.0 made power a bit easier with power augmenters and overslooping

you can oversloop fuel rod manufacturers to double your nuclear setup, if it's small enough

[obsidian trench]

Just checked for a Foundry with the Solid Steel Ingot recipe at 1% you get 0.1 MW which means you could run 160 Foundries for the same cost of a 100% Foundry, which would produce 96 steel/min instead of 60. However you also have to build 160 Foundries xD

[deft lichen]

if you manifold that, it would probably take days or weeks to start up

[ashen girder]

Sloops make it so you probably don't want to do the 1% thing.

Especially on the recipes that you'd normally want that kind of power efficiency.

[prime cipher]

does any one know why my rates are some how above 600

[obsidian trench]

Ohh btw is there a way to meassure the throughput of a belt in-game easily?

[deft lichen]

I keep forgetting that OCing and slooping can be combined when considering which setups are worth slooping

[deft lichen]

no not really

[obsidian trench]

Thats actually a huge shame for such a math-focused factory game

[deft lichen]

this might be another of my old fart opinions, but I don't think so

[orchid pond]

agreed. throughput is the main barrier to progression in Satisfactory and there's no way to cleanly measure it

[deft lichen]

you shouldn't be measuring it in the first place, you should instead be keeping track of what you're working with

[orchid pond]

idk they have a bunch of buildings that calculate it already, i dont think it would be that hard to implement

[ashen girder]

You can track efficiency on the producers and consumers.

[deft lichen]

there is no variable production, so the belt is exactly as full as you designed it to be

[obsidian trench]

Keeping track is kinda meh, when you need to take a few days break in a huge factory

[ashen girder]

And there used to be a mod that'd show you throughput if you really want that. 😂 Mods haven't been updated yet though.

Why are you keeping track?

[orchid pond]

that's true!

[ashen girder]

(Also, signs...)

[obsidian trench]

Signs can lie

Meters dont

[deft lichen]

they are as much of a liar as you are ... 😛

[orchid pond]

i guess if you're constructing your factories properly in the first place you shouldn't need to calculate

[obsidian trench]

Hence, why I want a meter 😛

[ashen girder]

Liar.

[orchid pond]

i think a througput meter WITH a display would be a fantastic addition regardless of necessity. i just think itd be neat 😄

[deft lichen]

yeah that's my point, it kinda pushes towards the "fully planned beforehand" philosophy

[obsidian trench]

Its also great to validate you actually getting what you expecting.

[orchid pond]

satisfactory is actually a serotonin management game

[obsidian trench]

I sometimes forgot/misclicked a power connection and then one of my hundreds of furnaces wasnt running. So the belt was slightly undersupplied and I noticed way way later

[deft lichen]

but I recognize that this is an old fart opinion (I'm still kinda against dimensional depots and augmenters... oh, and ||portals||)

[ashen girder]

https://ficsit.app/mod/CounterLimiter

[ashen girder]

I'm just as much of an old fart, and I love those. So.. 😛

[orchid pond]

dimensional depots are the biggest smoother of gameplay ever and i cannot complain about them, esp since they cant be used as factory inputs. ||converters and augmenters i can see the argument against||

[deft lichen]

how much?

[ashen girder]

How much what?

[deft lichen]

when did you start playing I mean

[orchid pond]

||portals|| are harder to make and less fun than hypercannons so i dont plan to use them anbyway lmao

[deft lichen]

older fart than me, I am defeated

[orchid pond]

chuck me across the map at an obscene yet measureable speed please and thank you

[slate silo]

So just looking at these recipes theres no difference between diluted packaged fuel and diluted fuel correct. Just one is avaliable earlier then the other cause of blender. If so I'm just going to go with diluted fuel for my nitro rocket fuel plant.

[orchid pond]

i've been playing since U6 so i think im allowed to complain about some of the new stuff, but honestly mk2 pipes still being buggy is my biggest issue with the game

[ashen girder]

Oh, they aren't buggy.

[amber jacinth]

Correct. Packed diluted takes more machines and an input of canisters, but is available much earlier.

[slate silo]

Alright thanks

[deft lichen]

I think the border between old and new is update 3, the game had a different feel prior (it was easier)

in ways, 1.0 returns the difficulty level to how it used to be, or at least tries to work against the new higher complexity

[river night]

We had oil in barrels on conveyors, and we liked it

[proud totem]

I remember running a conveyor all the way from the gold coast to the plains to have my oil

Cuz there weren't trains yet, and I guess I didn't want to use a tractor

[deft lichen]

I remember doing the same

[deft lichen]

pre-U5 trucks were a nightmare to use

[river night]

I first played in U1 I believe, we had lifts but no pipes or trains

[ashen girder]

DPF is 50MJ per fuel, DF is 45MJ per fuel, is the only other difference.

I remember when we didn't have lifts. 😁

[deft lichen]

huh, DPF is better power-wise?

[ashen girder]

Er, no, other way around.

Thanks. 😂

Packagers add just enough to make it worse.

[fierce ruin]

guys which one should i do for all of them

wait is there a better channel?

[obsidian trench]

nah its meta related

[ashen girder]

I'd reroll #2, stitched iron plate for #3.

[deft lichen]

huhh adhered plates are goated

[ashen girder]

#1 I would leave until I ran out of things to pick, and #4 I'd flip a coin.

I don't like adding oil to my basic iron/steel products. 😦

[deft lichen]

I recommend reading #math-and-meta message

[deft lichen]

the trick is to go all in

they removed steel coated plates in 1.0 so it's not as good now, but still applicable

[ashen girder]

Steel cast plates is solid tho.

[deft lichen]

coke steel, steel cast plates, adhered iron plate, steeled frame

if you're already bringing oil to the factory for a recipe that requires it, then it's worthwhile, because it doesn't cause any extra logistics

[ashen girder]

That's fair. I'll have to check that out.

The WP monsters have definitely left their mark on me. 😂

[deft lichen]

I think it actually works out fairly well WP wise

I mean SFTools will pick them if given the choice

[fierce ruin]

alr ty guys sry i didnt respond i was reading that thingamebober

[deft lichen]

good, no other response is required

[sacred pilot]

ok, so it all works! The only issue is that the rods don't load-balance into the plants. Am I correct in thinking I'm better off just waiting hours for them to fill than coming up with some convoluted load balancing solution?

[deft lichen]

in this very specific case, I think a balancer is worthwhile

how many NPPs?

[wind spade]

well, only if you need all the power right now (which you usually don't, so even a manifold would be fine, even if it stabilises in a day or so, you probably won't need all the power right now)

[deft lichen]

the stabilization can take a pretty long time though

[fierce ruin]

also ig i should have said context im like mid..? ish phase 3

[sacred pilot]

Oh, my power went from problem to FINE lol

It'll take a long time simply to build the new factories to use it

[wind spade]

as long as it takes less time than when you need the NPPs to run on full power, it doesn't matter

[sacred pilot]

word, that's what I thought

[deft lichen]

radioactivity

[wind spade]

well, that's a thing no matter what, also NPPs are mostly built above water on the edge of map 🤷

I'm not saying it's a bad thing to load balance NPPs, but it's imo a bit overrated to loadbalance them

[deft lichen]

greeny, NPPs are the only thing where I advocate for load balancing

wouldn't call it overrated 😛

[cloud comet]

What's a good starting place considering nearby resources? I'd prefer a place with a little of everything and draw conveyors from other places afterwards, but it's fine if it's more than a little. Also with lots of horizontal space if that's a choice.

[ashen girder]

They're all good.

[deft lichen]

probably best if you find that spot yourself: https://satisfactory-calculator.com/en/interactive-map

you can toggle nodes of all purities, not just pure as is default

[cloud comet]

I feel like that's too much spoilers though

Or whatever, I'll just pick a random place

[unique thicket]

anyone got any idea as to how i can get a 308/min belt line easily

[cloud comet]

Make the machines take only 308/min by changing the clock speeds

[unique thicket]

i need 4 308 lines, i dont know how to make the 8 easily tho

[cloud comet]

Do you have smart splitters?

[unique thicket]

i got one 450, 480 and 310 and i need it to be 4 308

[unique thicket]

yes

even programmables but i dont see how those could help me

[cloud comet]

Lead each of those 3 lines to 3 different 308s with smart splitters outputting on any for those 308 lines

Then make an overflow split on all of them and merge those 3 into a 4th 308

[unique thicket]

hang on i may be a lil stupid, i have some left over iron, it wouldt hurt the system to just round them up to 310, would it

[cloud comet]

It wouldn't

[unique thicket]

i already have a 310 so just the 450 and 480 need to be split

[cloud comet]

If you're bothered by it you could underclock the miners to produce 8 less in total

[amber jacinth]

Could consume thr amount per line, and have a smart splitter overflow onto a line that merges with the other three overflows

[cloud comet]

That's also what I said

[unique thicket]

nah i dont mind it really, i would usually do manifolds but belts aint speedy enough yet for me

[amber jacinth]

Manifold but wackier, basically 😛

[cloud comet]

If it works, it works

[unique thicket]

thanks for the help

[cloud comet]

4 pure iron and 2 pure limestone within like 150m of my hub but no copper, should I be happy or sad

[earnest tide]

how is the drones now on fuel? Is Battery gona be the best option, or how is the other fuel compare?

[amber jacinth]

Batteries are still a great fuel source- the best non-radioactive one.

[prime cipher]

i have never made it this far before. will this be enough power to make it to phase 5 ???

[earnest tide]

when i finnish the story and all unlocks i had about 46MW

[ashen girder]

Rocket fuel and Ionized fuel are better..

[earnest tide]

but it all depends on your build and what you want

[earnest tide]

so rocket fuel and ionized is better on the drone?... how is the consume?

[ashen girder]

Yeah. They last longer than batteries do, and go same speed or faster.

[amber jacinth]

Hmm, didn't realize that

Cool

[remote ice]

How do i make a priority fluid output agnostic of later height changes in the network? Take this one for example

i'm pretty sure the priority afforded by the initial height change is made irrelevant by the height of the pipes later on, so this junction would not prioritize

[ashen girder]

Add a pump after the priority fluid.

The only reason it's impacted is because of backflow, I think?

[remote ice]

the pump should only be on the priority line?

oh, hmm... does that mean a valve would work? The pipeline manual does show valves in it's VOP design

[ashen girder]

Might? I dunno, honestly. I don't usually mess with priority stuff.

[kind prairie]

Speaking of pipes, anyone have a link to that old manual with methods to make sure pipes are set up/working properly?

[remote ice]

https://satisfactory.wiki.gg/images/3/39/Pipeline_Manual.pdf

[kind prairie]

Damn, that was fast. Thank you.

[remote ice]

already had it open

[ashen girder]

(Also in the pins)

[tame harbor]

Could probably beat phase 5 with that

[tame harbor]

So if a route consumes 6 Batteries/trip then Packaged Rocket fuel would consume 5

[valid plume]

Question, do you guys know if rocket fuel in fuel gens is consumed at 4.16667 exactly or is it just 4.1666666.....

[deft lichen]

1 m3 takes exactly 14.4 sec to burn

The per minute value is derived

[valid plume]

cool thanks

[urban kite]

I think it's easier to say you need 144 gens for a full pipe of rocket fuel.... That is, if you can get 600 throughput to actually work

[earnest tide]

thanks a million. that was perfect info

[tame harbor]

what i go for is making rocket fuel in batches of 500. gives 30,000 GW of power and is the output of 2 Blenders at 250%

[deft lichen]

Better, it works in multiples of 6 fuel generators consuming 25 m3 rocket fuel/min

3 gens and 12.5 m3 is also clean

[violet turtle]

Am i doing the math wrong or is this gonna take 100 hrs to finish

[valid plume]

that would be 100 minutes no? unless im misunderstanding

[kind finch]

Thats what I thought lol

[violet turtle]

Ohhhhh

so 5/min * 250 / 60 would be a better estimate?

[valid plume]

why *250?

just divide the amount you need by how many per minute and you get the amount of minutes

[violet turtle]

o

so itd just be 500/5?

[valid plume]

if you need 500 then yes

[violet turtle]

wow i am awful at math

[foggy tendon]

yall think this is enough? 900/min rocket fuel

(for drones)

[forest solstice]

do dimentional depots stack?

like does making more then one of them, and then shoving 2 conveyors in double the upload?

[valid plume]

it doubles the speed but not the storage size

[tame harbor]

for drones should work

[unique thicket]

i havent even gotten to mixers yet but they still intimidate me

[tame harbor]

they're not that bad

as with everything just hook up all the inputs and outputs

[urban kite]

Blenders are pretty normal yeah. Like it's still 4 inputs at most, so same complexity as the manufacturer. Just it can also do liquid

Actually is there any building in the game other than the elevator that takes more than four inputs

[ashen girder]

Not without mods, nope.

Quantum Encoder is 3 belts + 1 pipe in, 1 belt + 1 pipe out.

[urban kite]

Tragically missing a 3 pipe 1 belt building 😭

[ashen girder]

Not really enough liquids for that I think.

[urban kite]

Also it's funny to me that the packager and refinery have the same input/output.

[valid plume]

what would you even make with 3 liquids? diluted rocket fuel? with the base recipe just add water?

[urban kite]

Id make smoothies

[woeful geode]

Im so lost, since when do I need Copper and SAM to make Quickwire?

[ashen girder]

It's turning the copper into caterium.

For funsies. Disable the converter building.

[unborn ermine]

so far we have 11 fluids

Oh wait I missed a few funny ones

[urban kite]

The tool likes to optimize for a special weighted point value. Disable stuff like Sam input if you don't want it to try using weird chains for "optimal" resource usage

[unborn ermine]

Wiki throwing me off because it "has" some, but only in the recipes

[urban kite]

Rocket fuel best fluid

[valid plume]

you on the wiki.gg one?

[unborn ermine]

Yeah

[urban kite]

Does liquid biofuel even count though

[unborn ermine]

https://satisfactory.wiki.gg/wiki/Category:Fluids

[woeful geode]

YO, thank you. I didnt know there was a building that did that

[ashen girder]

I count 15 there too. 😂

[unborn ermine]

Pages in category "Fluids"
The following 10 pages are in this category, out of 10 total.

[unborn ermine]

Numbers be hard ok

[urban kite]

Are those liquids and not gasses

[ashen girder]

People keep seeing out of date category pages. 😂

[unborn ermine]

huh not messed up AT ALL

[urban kite]

18???

[unborn ermine]

Fluids are gasses AND liquids

[ashen girder]

Click purge here.

[unborn ermine]

nice

[ashen girder]

2 are in chinese for some reason and the "Fluids" page is in the category too.

And yeah, fluids are liquids, gasses, and nongasses.

[valid plume]

there should be a 15 fluid input building to make a smoothee out of all the liquids/gasses

[unborn ermine]

now that I know I was seeing out of date pages... im scared what out of date info I have been refering too

[ashen girder]

Apparently mixing sulfuric and nitric acid is exothermic, that sounds like a fun thing to mix with 4 different kinds of fuels.

[ashen girder]

It's specifically the auto-generated category pages that that happens with. The regular pages don't cache the same way I don't think.

I don't even know what happens if you mix DMR and EPM. 😮

[urban kite]

pretty sure the universe instantaneously collapses

need someone to make a mod that has a building with all 15 fluid inputs that makes a single smoothie in like 15 hours and then if you drink it it deletes the save

[dark bane]

Hey pioneers I think this is a math question, how much more efficient is turbo fuel than regular fuel? 46%~ less production than straight crude oil or HOR (40/min vs 18.75/min)

Currently running 4 fuel gens and 6 coal gens (just unlocked phase 4)

[broken oasis]

Im currently playing satisfactory with a friend. He just sent me this (he's an engineer)

[urban kite]

oh no he's diagramming

[cloud comet]

Tell them there are tools for this

[urban kite]

its too late to save him im sorry

[timber garnet]

what tools do you guys use/ recommend?

...asking for a friend 😄

[vapid gorge]

just get hi mto use tools https://www.satisfactorytools.com/1.0/production?share=E37XD0s18kxbNz77COMB

[vapid gorge]

this one ^

[timber garnet]

thank you!

[vapid gorge]

you can use the bubles as base layouts for factory planning to and break it down further with pen and paper

that's particularly useful for bigger more complex build where it might be convenient to have multiple sections making wire for example since many parts may need it, rather than 1 source

[coral steppe]

Is that calculator not able to deal with fuel? I've tried turbofuel and ionized fuel in the past and it says it can't calculate a result

[vapid gorge]

you're likely not ticking off one of hte receipes you need

[broken oasis]

he said (and I quote) "using the official tool is cheating" lmao

[coral steppe]

Oh, I guess because those are unlocked via hard drives / mam they aren't on by default. They're technically alt recipes

[cloud comet]

Tell them engineering isn't about reinventing the wheel but using already built tools

[vapid gorge]

upgrade your friend's bios

[vapid gorge]

with Turbo fuel it's quite common to forget to tick Compacted coal for example

this is a plan for ionized , I just chose 'all' recipes, though I never actually use 'all' when making a proper plan https://www.satisfactorytools.com/1.0/production?share=HvLwUln2pYSqg0MR9Nd4

[broken oasis]

bahahahaha

[polar wave]

Is there any way to determine how many ticket's I've already pulled from the AWESOME sink without totalling up the cost of everything I've purchased?

I'm trying to see if I want to build a ballistic engine factory in order to get my golden nut or just leave the game running all day tomorrow

[vapid gorge]

you can probably back track how many points the current ticket costs and use the formula on the wiki ?

[cloud comet]

You could figure it out based on the cost of a ticket

[vapid gorge]

but that sounds insane

[polar wave]

Oh that's a good point

[glossy schooner]

do i need more than 1 block signal if i have more than 2 trains on a single rail loop?

per station

[polar wave]

Oh the wiki already calculated the inverse for you, it's literally just punching it in lol

[vapid gorge]

w/o seeing it - probably no? but do you want more than 1 block section? likely

ah some weirdo added that to the wiki?

[glossy schooner]

you can basically imagine it as a circle with 6 stations and 3 trains on it

[vapid gorge]

ON the circle? or off to the side?

[glossy schooner]

on the circle

[vapid gorge]

honestly if you're doing a 1 way loop with stations on the circle it doesn't matter it's gonna be super inneficient anyway and adding more blocks won't do anything

[glossy schooner]

? how is it inefficient

[vapid gorge]

because it means every train has to stop or go through every station
if they aren't needing to stop at a station they'll need to wait behind a train that does

[vapid gorge]

imagine if mcdonalds had their drive through on the main road. How much would that slow people down?

[glossy schooner]

mhm

[polar wave]

yeah usually you want roll on roll off stations if you have one big backbone loop

[glossy schooner]

i should probably add a bypass to my stations shouldn't i

[vapid gorge]

not bypasses, just make hte stations to the side

[polar wave]

My personal choice is to leave the backbone rail alone and have the station be on the side

[vapid gorge]

they have made it so that stations count as 'longer' sections of rail. so trains will generally go around them to the 'shorter' path

[glossy schooner]

i'm not understanding what's being said here

[vapid gorge]

so trains always use the shortest distance between their paths right?

[glossy schooner]

sure

[vapid gorge]

if you have 50m of Station rail, and 50m of normal rail next to each other in a fork of the road, a train will read the station as 100m , and go down the the 50m of normal rail

unless it needs to stop at that station

[sullen mulch]

wasn't aware that there was actually a recipie to replace the need for coal

[glossy schooner]

each train stops at each station for me though

[vapid gorge]

yup, uses more oil per Turbo fuel, but it's nice not having to bring coal in

[vapid gorge]

I also like breaking down oil into 3 parts and then recombining them 😄

[sullen mulch]

yeah, but suppose it's nice having it for the late game now coal can be used for diamonds

[vapid gorge]

well it's also funny cause you can turn it into rocket fuel and get compacted coal back with it 😄

[sullen mulch]

that being said do you have a layout for a min max turbo fuel setup, looking to just do 600 oil p/m now

I'm looking at this but I thought there was a better way to do this to maximise what yoy get from the oil

[sullen mulch]

and yeah, which is quite handy all things considered

[glossy schooner]

i assume this is useful for setups that don't have trains stop at each station?

because each of my 3 trains picks up from each of my 6 stations

[vapid gorge]

honestly, most set ups should be like that.

each extra train that stops at a station lowers that station's throughput

[vapid gorge]

I suppose you could recycle the Compacted coal at the TF plant and have some of the TF produced be base recipe? it'd be fun

[glossy schooner]

my loop is bottlenecked by train speed

[ashen girder]

Oil and HOR can be used for diamonds too.

[sullen mulch]

yeah, it's certainly a more endgame idea to play with, turbo fuel for me here is just going to get me by until I go big on nuclear

[urban kite]

agg i need fuel generators to consume rocket fuel so i can turn byproduct resin into rubber..... but i need rubber to build the generators.galdkfyghnlraktng

[ashen girder]

Bootstraps!

[sullen mulch]

yeah, going to be an interesting experience learning about all the new recipies in the end game at tier 9

[magic island]

I just used default turbofuel since I had already cranked out a ton of regular fuel, and wanted to keep it simple by merging in the byproduct from rocket+ion

I imagine you could whip up a pretty decent two-recipe setup (blend for the initial turbofuel, then use the compacted byproduct for one of the other recipes)

[urban kite]

im using diluted fuel -> nitro rocket fuel because sulfur is tasty

[ashen girder]

That's exactly how to maximize rocket fuel, yeah.

[sullen mulch]

what I reckon then is I'll just stick with that the tool reckons is the way to go and keep the coal for later on, do something mega way later instead

[magic island]

every time I have to make more singularity cells my soul screams

there is no part in the whole game that makes me wish more desperately for an alt recipe. even if it's a harder alt recipe, anything to shake life up a little

[amber jacinth]

Endgame stuff should definitely get some alts... would be nice to see..

[abstract kite]

is there a go to way to handle excess water?

packaged water and dump it to a sink?

[amber jacinth]

The most optimal use is running dedicated machines off the byproduct

[urban kite]

if you have some limestone make wet conc

or put it into pure ingots

[prisma kraken]

wet concrete or coal power - coal power is probably a better option

[vapid gorge]

for aluminium?

[weary lotus]

does anyone have like a spreadsheet for the transfer speed on drones for different fuels and distances

[fringe peak]

Def coal u can’t go wrong with more power

[polar wave]

Hrm, so to get the golden nut I can either let my factory idle for 16 hours or try to build substantially more at a faster rate than I generate coupons 🤔

[light fox]

i have like a 1.5km distance between my plastic/rubber making area and my main factory, should i use trucks or trains (i just unlocked both)

[vapid gorge]

#math-and-meta message

[weary lotus]

with the fuels

[urban kite]

generally both are good. Truck pathing can be a pain, but trains you have build the track so 🤷‍♂️

[prisma kraken]

i don't think anyone has properly scienced it out yet

incidentally, i wouldn't trust theoretical numbers generated by spreadsheet math and instead would rely on empirical measurement - oft times theoretical math is incorrect due to hidden variables or due to game bugs, sunspots, etc

[jovial lichen]

What are the best like 3 (hoping only need 3) alt recipes for HMF designs to make them easier?

[prisma kraken]

well, there's only 3 hmf recipes, of which heavy encased frame is the best of breed

that takes, pipe, encased beam, concrete and mod frames

[jovial lichen]

How about best alt recipes to make them? Like not not just them

[prisma kraken]

so what you're looking at is encased industrial pipe to make the encased beams cheaper

[jovial lichen]

Ah alright

[prisma kraken]

because the recipe is concrete heavy, you're going to want either wet concrete or rubber concrete

and then there's the recursive problem of 'what are the best recipes for making modular frames, of which there are 3 alts...'

that one is more complicated and situational 🙂

[weary lotus]

how long does plut rod last in a dorne

[prisma kraken]

i think the math on that is a little funny, and they work like batteries, where a drone takes an integral number per round trip

i'll most likely be building a table of that data in a week or so

[jovial lichen]

Nah you just giving me the best HMF just saved me sooo much (sorry was looking at how I would build that and cuts everything basically in half) thank you! Was dreading making this factory lol

[prisma kraken]

the recipes are all subjectively good in situations

with hmf, the best in breed really is heavy encased though - it uses less mod frames per HMF, and that makes the bottom half of the iceberg smaller

incidentally, HMF's can now be made without steel thanks to the iron pipe alt

iron pipe uses tons of iron for pipe, but it allows you to make them w/o coal

for mod frame, there's a bunch of different ways to make them, none of which are strictly better than one another. because i decided to get some stout supply of them cooking before unlocking oil, this is the recipe chain i used for 90/min:

[jovial lichen]

So should I use pipes to make the steel frames or beams?

[prisma kraken]

give me a sec

this is kind of the go-to for hmf's

[jovial lichen]

Thank you!

[prisma kraken]

you can do it with wet concrete instead of rubber if you have more than 1200 limestone handy

i pick 45/min as the target because the recipe makes 45/16 hmf's per min (2.8125)

45/min is quite a lot as a production target

i also picked solid steel for steel because coke steel ends up with weird numbers where you need 1620/min of iron and coke, and that ends up being 2 belts of 780 plus one with the remainder and that's just weird to build

(or 18 mk3 belts exactly, which would be a real pita, lol)

no, i'm mispoken, 9 mk3 belts

i really shouldn't do the numbers in my head 😛

i'm still wrong, its 1620/270 = 6

[cerulean vault]

I use the in game calculator way too much. I just love that the devs added it!

[sinful dagger]

ok, using alt recipes i got one that is 4 encased steel beams a minute for 30 limestone, 25 water, 18 iron ore, 9 coal, and 9 sulfur thoughts? or might it be more wothy just going 30 limestone, 25 water, 24 iron ingots, and 24 coal so i dont need to use sulfur

[cerulean vault]

Lot more coal though. Depends on if sulfur is nearby and not already being used.

[stuck ingot]

is satisfactory calculator down?

ok nvm fixed itself

[sinful dagger]

after i just did all the math the hard drive i was scanning just finished with a new steel recipe

[glossy schooner]

surely i can add an electronics factory to my grid right

[ember fractal]

45 HMF is my next project

[violet turtle]

does this mean i need 26 machines and one at 66.667 percent?

[cursive heron]

yes

or 27 at 98.765%
or 11 machines at 242.424%

[night pasture]

I'm not sure how I did it, but while I was shooting things and then reloading while getting attacked from the side by a spider, it reloaded 50 rounds instead of the regular 15 lol

I noticed it after shooting 10 more rounds and it not reloading lol

[kind silo]

https://docs.google.com/spreadsheets/d/1TQoPunziQ92gFOlH65xSzqCiGLXTsky02LCAFG1lsqY
here, have a spreadsheet on lizard doggos

[night pasture]

I've wondered if these odds also depend on the area you get the doggo from. My last playthrough on from the northen forest pure nodes area had a doggo that kept giving me nuclear waste lol

[kind silo]

the doggos got changed on 1.0

[night pasture]

nice

[kind silo]

i gathered lizard doggos from around the map and they give the same items

so, naturally, i scooped as much as i could and made this spreadsheet

each doggo produces about 10k sink points of materials per minute which isnt too bad all things considered

this also speaks not of all the power shards youll end up getting

[cloud comet]

Wait, people actually use lizard doggos to farm items? I thought that was a joke

[kind silo]

you also get 2/3 a power shard per minute if you can get an army of 20 doggos which is pretty awesome

[night pasture]

ya, they were a pretty good source for extra power slugs

[kind silo]

they still are

but now they can give you almost any part you find at crash pods

[night pasture]

not really needed as much anymore now though. There's so many power slugs now, I can use them to reduce the number of generators for power and still have tons left over for reducing factory sizes on top of using them for miners and elevator parts

[kind silo]

i have a save on tier 1/2 with smart plating being crafted and 20 lizard doggos and no factory, and a good amount of progress completed in the mam

its got 76 power shards worth of slugs and ill be letting it be a good starter save file for my friends

i saved before doing all those milestones too

[night pasture]

lol

[kind silo]

but yeah, lizard doggos are op, guys

this game is now satisdoggo

[unborn ermine]

ive spent so much time around my doggos I almost have 5 stacks of fused frames and over 1000 alclad sheets/casings and I only have 6 doggos

[feral breach]

Well because my lizard doggos are turbomotor fanatics and they keep bringing them I unlocked most of the awesome shop quite early 😄

[unborn ermine]

Yeah and thats almost the same as well!
even with opening some drop pods.

[light fox]

how much can a train platform/each train car hold

like if i were to use a train to transport plastic and rubber

[orchid pond]

freight car holds 32 stacks, platform holds 48. so 6400 and 9600 plastic or rubber respectively

throughput is determined by # of trains and travel distance

only use 1 item type per train car for your own sanity

[keen moss]

So...
I have to distribute 120m3 of water between 4 pipes (30 for each) until then OK.
idea 1 (questionable)
just put an extra water pump in the pipe and let the world divide that extra 120 into 4 pipes and sort itself out.
idea 2 (questionable) a weird system that I invented out of my stupidity that divides a water extractor into 4 pipes
Which of these two inventions is the least worse?

[orchid pond]

do idea 1. pipes do not split like belts, so you're gonna be better off with fuller pipes when it comes to water (which is functionally infinite)

if you want to split them into exactly 30x4 you would need to use valves

[keen moss]

I don't even have that advanced technology there

[keen moss]

I forgot that the world of pipes is a dark and lawless place, a place where the blind see fear, the deaf hear despair and I get screwed because in my world of magical unicorns the world works the way I want (and thanks for the help <3)

[orchid pond]

bro is talking like a mercer sphere lmao

(no problem glad to clarify)

[valid plume]

Anybody know the benefit to the alt dark matter crystal (the one that uses time crystals instead of diamonds)?

[next pewter]

Probably higher yield? I mostly use dark matter crisatallisation. One input, slightly lower yield but no dependencies!

[valid plume]

its the same amount of diamonds you just turn them into time crystals first, so Im not sure Ifts its even less power

[next pewter]

Could be usefull in sone edgecases, where u have spare time cristals and want high dark mattwr output?

Doesnt make sense in isolation, but in a factory ecosystem, it might

[valid plume]

yeah theres definatly a use I was just wondering if it was better to use for direct conveserion

it looks like its just a bit more power tho, hard to tell with the variable power from converters/particle accelerators

[vapid gorge]

pls don't do this

this is going to murder your flow

[still elm]

Can someone logically explain how I should understand such separation of iron? Numbers and theory is one thing but where do I separate it.... wut [I only have 30 hrs XD]..

[vast jungle]

you just manifold them?

[vapid gorge]

I mean ignore the right image

and yeah if you have mk3 belts,

1 line with spliters going to constructors

[vapid gorge]

or 1 line and split it 2 ways to two groups of constructors.
it will self balance

[still elm]

so far I've only done single projects such as the same bars or slabs.
I mean, all I have to do is line up the beams first and then the plates and the excess will go where it should ???

wait... thinking after 8 hrs of playing doesn't work

[orchid pond]

sort of. the machines at the front of the manifold (the "line") will fill up because they are being overfed with resources, and eventually the resources will overflow to the machines further down.

the benefit of this is that you don't need to worry about complex load balancers. the downside is that it takes time for the first machines to fill completely, during which the factory will be running at less than 100% efficiency.

the majority of experienced Satisfactory players use manifolds 99% of the time, with the outliers being stuff like nuclear power plants which have extremely slow item production and would take hours upon hours to saturate completely

[vapid gorge]

and even with the slow to saturate - you have lost of time going to build and explore elsewhere where it'll saturate. Or pre fill machines with stacks of items

[vast jungle]

balancers are nice for radioactive stuff where you don't want a full blocked belt of it in the factory ^^

[still elm]

I'm thinking whether to make two floors here in case of possible expansion, one for rods and the other for plates, and also so that it goes up and down evenly [if it makes sense].

[vapid gorge]

I tend to do something similar

1 node per part

1st floor smelters
2nd floor constructors
ect

connect things as needed and then you can expand sideways

[still elm]

Whether to use manifold to row the same machines, but use balance to distribute between different sections [this game is probably so elaborate anyway that my head will explode if I want to do something right without correcting X times].

[still elm]

Hmmm

[vapid gorge]

it's how I get through the tiers and then build very tight and perfect factories with the products

for example this makes everything through to phase 4 space parts xD

it's not efficient but it's fast and you don't have to htink about it much

[still elm]

https://tenor.com/view/hmmm-thinking-confused-gif-14450462

[vapid gorge]

oh yeah, the top layers become a bit of a mess because blenders and manus needs so many inputs but the lower layers are pretty tidy

[still elm]

it's hard for me to do something future-proof because it either becomes inefficient or I need parts that I have to destroy half a factory to start producing. the fear of having to run conveyor belts like spaghetti to move a part to a place where there is no room XD [perfectionism is bad]

whether to use a group of 3 sources to mass produce plates, rods, screws and send this in excess to the 3 warehouses from which it would then be taken to produce at individual locations further items like reinforced sheet metal and modal frames [because all the time I tend to rebuild the factory for what it needs XD

[unborn ermine]

Future proof is smart splitting the ore at the miner
factory stays the same but you can still siphon off ore.

[vapid gorge]

oh yeah, that's why you don't future proof 😄

[frosty owl]

I strongly suggest against using tools to help you through the early game.
It's like having someone hold your bike while the game provided you with training-wheels to help you learn riding it, you're likely to become dependant on that help and unable to solve simple issues (aka play the game) on your own

[frosty owl]

Somersloops are all the future proof one may need

[keen moss]

this what

my idea is 90 in each side... I guess that will work... Right?

[vapid gorge]

trying to load balance fluids

[wintry marlin]

what are u trying to do?

[keen moss]

separate coal

and btw where is my blueprint

[wintry marlin]

whats ur inpuit and expected output? i am too lazz to look for it xd

[wind spade]

--S--S--S--S--S
  |  |  |  |  |

[keen moss]

but thats boring (I'm going to do it because my blueprint doesn't work and I'm too lazy to redo it)

[wintry marlin]

technicallz can be done with 2 seperators, but doesnt look as neat

[keen moss]

My idea is to enter 480 and leave 90 on 4 sides (2 on each side) and the rest (120) will be divided into 8 generators :)

HMMMMM? I LOADED IT AND THERE IS NOTHING OVER THERE? (even spent material)

[wintry marlin]

i mean, 120 is just a 1:4 balancer, which is just 2 splitters in a row, then merge the rest, and make another 1:4 splitter with that

[wintry marlin]

check below

[keen moss]

no way

[wintry marlin]

idk, just guessing

[keen moss]

magic!

[wintry marlin]

i just like to come here and spend my workday here, i dont have enough time to actually play the game

[wintry marlin]

can u walk through where u placed it or is it just invisible?

[keen moss]

walk, build, it's as if it didn't exist

[wintry marlin]

i had an invisible packager the other day, thats why i ask

mhm, then idk

[keen moss]

i have a solution!

kms 😍

[wintry marlin]

or reloading the save

always works

[keen moss]

Ah

didnt work

[wintry marlin]

mhm, try redoing the blueprint

[keen moss]

a guy in the Questions area helped me!

[teal urchin]

using a smart splitter, is there a way to get it to go to the dimensional depot first, then the storage, THEN the sink?

[edgy leaf]

i dont think so, u can use two smart splitters

orrr, just put the depot on top of the storage and then connect them with a conveyor lift

[teal urchin]

tru

[fierce ruin]

Whats the tool on the right image called

[past reef]

For rocket fuel power I should still recycle the compacted coal into rocket fuel right

[unborn ermine]

Or just slap the depot ON the cargo

then feed the depot from the cargo

[past reef]

Depot on cargo and use a higher lift than the input

[unborn ermine]

Yep

[vast jungle]

the only ugly thing about "dim. depot on storage container" is that you need the feed the depot from the output side of the container, which is often the "engineer facing part" of your storage...

[past reef]

But you can do lift on 2nd isc output

To be fair depot under isc might be nicer cause head level

[vast jungle]

if you already have a smart splitter behind the depot with a "sushi belt" running across, you can easily use the smart-splitter output opposite the container for a lift, ending to a depot on top of the container

[indigo lily]

what does the weak signal mean

[deft lichen]

no precise locations

[indigo lily]

any way to make it precise or is it just telling me that they are here

[deft lichen]

just telling you the count

[indigo lily]

ty

[vast jungle]

object scanner

[cloud comet]

Is it normal for me to get this recipe without even having steel production unlocked?

[wind spade]

yes

you got sheet production unlocked

[outer vale]

not particularly useful until you unlock refineries mind, but hey

[vast jungle]

"Pure Copper Sheets"

[fierce elm]

So I have a question about the best way to distribute my reinforced Iron plates. I was producing 10 reinforced Iron plates per minute and wanted to build Modular Frame 2x (need 3 per hour each) and Smart Platings 2x (need 2 per hour each) out of it but I'm not sure what the best way to achieve that is since I cant split 10 into 2/2/3/3 with just splitters. What would be the best way?

[outer vale]

a) manifolds (just split and let it eventually balance itself)
b) produce the reinforced plates as part of those production lines rather than siphoning from elsewhere

you could also do an exact splitter if you're so inclined

[wind spade]

--S--S--S--S
  |  |  |  |

[fierce elm]

Yeah manifold is the one I currently have but was just wandering of another way. To have so much excess needed, especially early, feels kinda bad

[fierce elm]

What is that?

[wind spade]

you don't need excess tho

[wind spade]

also you could merge 6 and 4 RIPs separately

[fierce elm]

But the back machines wont get enough until all the machines before them are full, right?

[wind spade]

yes

[fierce elm]

Thats what I mean

early on need 300 reinforced iron plates as a buffer at a rate of 10 per minute just felt bad

[wind spade]

200*

[fierce elm]

ah so 600 in total?

if each machine can hold 200

[wind spade]

100

and only need two machines to fill

[outer vale]

you can make a 1:5 splitter by making a 1:6 and merging one output line back to the input, as long as the belts aren't close to their limit (which they won't be).
So for your case

• Split two ways
• Split the first half into thirds
• Merge two of those thirds back together to get your 2
• Send the other third all the way back to the start
• The other side's already the 3

[fierce elm]

true

[fierce elm]

Yeah making a 1:5 splitter first and the still merging and splitting is perfect. Prob wont use it in this case but perfect to know for the future

Know each of the steps individually but didnt think about putting them together

[fierce elm]

But wait im a bit confused about your instructions. Is it first to make a 1:5 and then your extra steps?

[outer vale]

no, those are the steps to make your 5 into 2 and 3 split

[wise plaza]

Silly question: What is the throughput of a "Pipeline Junction" ?

I'm trying to balance pipes to get 3 x 600 and 1 x 120 :v

[outer vale]

since you don't need to do the whole 1:5 split just to merge bits together afterwards

[drifting trench]

There is no throughput limit of a junction

its only limited by the pipes going in and out

[wise plaza]

So connecting two pipes with ea ~400 should allow me to balance them out to 600 and 200?

[drifting trench]

should work no problem

[wise plaza]

Thanks!

[drifting trench]

you can have 2 600 in and 2 600 out and it would work fine

[wise plaza]

Gotcha!

[fierce elm]

Ah I was thinking about 10 into 2/2/3/3 and was confused how it works xD but yeah simply just split once before

[cloud comet]

Is bolted iron plate a scam or am I not mathing rn?

The plate input and output is tripled but the screw input is much more than triple (4.16 times)

[urban kite]

sure but it's more space efficient right

[cloud comet]

Effectively costing more materials to save power

[urban kite]

yeah exactly. There's no such thing as a strictly better alternate recipe? They all save something (power, space, resources, etc) at the cost of something else?

[cloud comet]

Idk I've just never seen an alternate recipe specifically meant for saving power at the cost of more materials

[drifting trench]

i mean theres a few that are clearly worth it

[urban kite]

but it saves space too right? since one machine makes more plates / min

[cloud comet]

Space is the least of my worries
I'll prob run out of fps before space

[urban kite]

although i do agree bolted is a bit weird since it doesnt change the recipe just the ratios

[drifting trench]

bolted with steel screws doesnt seem like itd be that bad

[fierce elm]

With an input of 10 I think its a bit more compact to do
split two ways
split both 3 ways
Merge 1 line all the way back
Split 1 line into 2 others to get 2/2/3/3

That should work aswell, right?

[cloud comet]

Isn't steel screws also bad? Like you get more screws out of steel rods unless you use an alt recipe for the steel beams

[drifting trench]

i personally love steel screws just for the output

its what 280 from one constructor?

[urban kite]

i mean if you need a crap ton of screws and dont want to build a whole floor of constructors, steel screws is nice

[vast jungle]

think about what happen when you add a Somerloop and OC the machine!

[drifting trench]

full mk6 lol

[vast jungle]

Yeah, we are screwed, even with mk6

[drifting trench]

hahah

[magic jackal]

Is anyone able to help me understand a problem better.
For the sake of simplicity lets say i have refineries which take up 1200m3 of oil a min.
Could i have 1 pipe which connects them all and have an input and both ends say providing that 1200 (e.g./ 2x 600m3 oil) or do i need to have 2 separate pipes?

[drifting trench]

screwed

should work fine other than 600 full pipe shenanigans

[urban kite]

in theory yes but pipes can be weird and getting full throughput can be hard

[edgy leaf]

nooo satistools is down?

[magic jackal]

Im just trying to understand it a bit better. As with conveyors you would need 2 to provide the 1200 in total.
But with fluids is it correct that the 1 pipe can technically supply the 1200 if i have multiple inputs spread out over the distance?

[drifting trench]

yeah that should be fine

as long as no pipe in specific has more than 600

[urban kite]

i mean conveyors would work like that too with mergers

[drifting trench]

If you want a really good guide on pipes this has everything you need https://www.reddit.com/r/SatisfactoryGame/comments/obqjrd/ficsit_pipeline_plumbing_manual_first_edition/

[vast jungle]

The difference might be that we don't have a splitter/merger with two inputs and two outputs

[urban kite]

industrial container 😉

[drifting trench]

it doesnt do the best job at splitting lol

[verbal lake]

or at all

[urban kite]

neither does the pipeline junctrion 🤷‍♂️

[drifting trench]

but the container just straight up doesnt split

it almost always goes to the bottom

[urban kite]

it splits if it needs to

[vast jungle]

Better use two splitters and two mergers

[raven ivy]

Whats the better recipe alternet basic iron ignot or alternetr steel cast plate

[spark marten]

I just wrote a wall of text describing a problem of mine. But before sending it I just wanted to double check something.
Seems like I forgot to set some fuel generators back to 100% after filling them with fuel.

Thanks anyway 😄

[ashen girder]

Why not both with solid steel?

[ashen girder]

Glad we could help!

[raven ivy]

It's a hardrive i can only choose one

[ashen girder]

Oh, I'd probably pick steel cast then.

Basic iron is nice, but you don't really need more efficient iron for awhile.

[ashen girder]

Also, see this comment: #math-and-meta message

We should really get that comment pinned. 😂

[wind spade]

you can get both anyway

[raven ivy]

Oh Neat ty

[ashen girder]

Yeah, there's enough drives that you can eventually get every alt.

With two to spare, last I saw.

[cursive heron]

I think I just realized why they put hard drives in the awesome shop

[urban kite]

to waste tickets?

[cursive heron]

I got this on my scan but I haven't fully unlocked the tiers

wait no this is going to happen eventually anyway

or maybe?

[urban kite]

if you dont have all the alts and that showed up, then yeah i guess it would be possible to "waste" drives by rescanning them before you unlcok the tiers

[ashen girder]

You have a pool of unlockable recipes at any given time, that's based on what materials you've unlocked so far.

[cursive heron]

no my brain was thinking of something else, end of the day its one drive for one alt anyway

so its never really going to matter

[ashen girder]

And I don't believe you can waste drives.

It didn't let me rescan last time when I was out of them. I imagine it'll probably show the same recipe if you only have one left.

As you unlock new milestones, you can continue scanning hard drives.

Like, you can't get the nuclear alts without the base nuclear recipes.

[urban kite]

fgklsnlgk my iron factory has a byproduct of 213.333333333333333333333333333333333333333 ingots 😭

[ashen girder]

Switch to iron pipes. 😁

[burnt pecan]

awesome sink

[urban kite]

its making 480/min of screws, rods, plates, and wire for the uploaders though

[ashen girder]

Honestly, underclock your ingot production to get rid of it.

Oh, that's going to back up quick enough anyway. 🤣

[urban kite]

id rather die than not fully utilize the nodes

[ashen girder]

Guess you're gonna die then.

[blissful epoch]

use the alts to make some motors from the pure iron, to sink em

[urban kite]

i was hoping there would be enough to do 240 cable as well but id need like 63.33333 more per min

[blissful epoch]

i had 10 refineries making pure iron because i needed 40 iron plates/min, and needed to get rid of 200 water

[ashen girder]

Wet concrete tho.

[urban kite]

right now im just trying to keep my depot full of everything i could possiblyt need

then maybe ill actually build something

[ashen girder]

Honestly though, not the worst idea. 🤣

I tore down one of my basic factories because it was in the way of my view.

[burnt pecan]

i mean shoot i did that too

[ashen girder]

Then I completely ran out of RIPs building my turbo motor/supcom factory.