math-and-meta

[kindred carbon]

Well, not anyone who just likes to build a little big to enjoy the game

Obviously you can break 300GW if you 250% overclock and sloop everything

Don’t worry I got my handy dandy ISC blueprint to provide me with 300 hours of storage

Each

[glad shuttle]

Nice

Personally I just don’t like the idea of storage tho it just makes shit feel like it’s on a ticking time bomb even if it’s gonna be a long ass one

[vapid gorge]

whats going on now? xD

ok I'm now running a steel screw system directly into a sink , gonna let it run for ages @visual yarrow

it's clocked to be 1198 screws pm becaues I was getting repeating numbers, it was the high per cycle output and simple to spin up since it's just being fed extra beams from my starter factory

all the machiens started at teh same time too so there's some stuttering when their production cycle is going

[visual yarrow]

that should be... 240/m per? just to make sure my math is right

[vapid gorge]

260

the 5th machien is doing 137

I'll let it run for a few hours

[visual yarrow]

hmm.

you know what. I wonder if that is actually relevant. I always balance my machines to produce the same amount, instead of underclocking only one machine.

[vapid gorge]

that really wouldn't impact things

at worst, what will happen here, is that ecaues they all started at teh same time is that some machine sill slowly shift their cycle around

but that shouldn't happen either

[violet wharf]

I'm sorry for sparking the bus debate earlier, but I ended up creating a 1600 rubber and 1600 plastic factory with the recycled recipes

It's super messy and I have to make it look better, but now I'm just waiting for the manifolds to fill up so I can see all the smoke stacks together

[vapid gorge]

looks alright 🙂 little worried about hte piping keeping the flow stable but fingers crossed

[kindred carbon]

It definitely won’t hurt to elevate the junctions a little

[violet wharf]

My fingers are also very crossed

[kindred carbon]

You built it a little bit of a weird layout

[violet wharf]

Yeah the layout was not preplanned and ended up very... "organic"

All lights are good and the atmosphere is screaming. Perfect

[vapid gorge]

@visual yarrow last machine in the line if you're curious, emptying it's internal buffer before the cycle

[visual yarrow]

yeah, I don't know if screws are going to have this, given the large stack size.

the other example I was going to use was pure iron ingots

[vapid gorge]

it's still 20% of the stack size, much like the 20 copper ingots of a max 100

[visual yarrow]

well, let me rephrase; stack size and production speed, if that makes sense

[vapid gorge]

I feel 400pm is a heafty production speed.

and 104 is more than 1/4 of it

[visual yarrow]

yeah, but that's kind of my point

the more total machines you have on a manifold, the more of an issue this can be

so a three machine manifold will probably not run into the issue

[vapid gorge]

Give me a recipe for me to try out. I have a starter base that has some of everything going on and I can slap together something

and the 6 machine manifold is still going strong

I know what you mean about the merger going round robbin, but the thing is, if the merger tries to merge an item from an input and noting is there, it doesn't skip that possible input

it just picks another available item being shown.

I think that's why they have internal buffers maybe?

[visual yarrow]

okay, how about 20 pure iron ingot refineries, all set to 85.715%?

and I am only half serious

[vapid gorge]

sure I have water and iron here

[kindred carbon]

This is mine, I just started it up lol

[visual yarrow]

i don't expect you to actually test that on my behalf, but I wanted to give an example that was roughly close to something I remember having issues with in the past

[kindred carbon]

So I’m in the same boat as you

[violet wharf]

Your piping looks way more neat

[kindred carbon]

1800 oil in for 2700 of plastic and rubber

[visual yarrow]

cause it occurred to me that silica had it's stack size increased at some point and so all of my examples aren't precisely the exact way I remember them

[vapid gorge]

and I believe you but I also want to show you this is built on a misinterpretation of what was going on. It's possible you might have been making a tiny mistake each time too but that seems less likely

gonna do as close to a 1200 pure iron manifold after dinner. I'll post images

[visual yarrow]

if it was a misinterpretation, it was still an issue that I was having

i can believe that there exists the possibility that it wasn't what I thought it was

[kindred carbon]

Things are definitely taking their sweet ass time to spool up lol

[vapid gorge]

I mean if you were reading what the issue is wrong, or even if it was a true issue that's a thing 🙂
I wasn't there. I didn't experience it. I know visually belts often don't look like they are working right

[visual yarrow]

but I don't know what it would have been, then, given that I was able to fix those cases by balancing the outputs

[vapid gorge]

it's the correlation vs causation thing again - you may have had a small issue that was also just fixed by rebuilding a bit of belt or something

[visual yarrow]

nah this wasn't an issue with belts. the internal buffer of the producing machines were filling up, and it was always with most of the backlog being further down the chain

[vapid gorge]

there used to be a duplication bug

[visual yarrow]

which... to be honest, would still happen even if the issue was something different

[kindred carbon]

52GW max consumption

[vapid gorge]

but I can't recall what you had to do for it. I don't think it was something you could do accidentally easily

[visual yarrow]

but trust me when I say that I spent a LOT of time debugging these systems, so at the very least, I know it wasn't a simple math error or something

[vapid gorge]

and pre 1.0 there used to be a different much more rare duplication bug.
but it was such a tiny bug that only someone running Single Input Sushi Load Balancing would have noticed that one

and that's like probalby 2 people

[visual yarrow]

remember the old bug where fluids in the invisible buffer between pipe and machine inventory would dissapear?

[vapid gorge]

that old one. shadow fixed in u7 I think

[visual yarrow]

it made it impossible for me to do my completely closed loop instant scrap recycling. was so glad when it was fixed

[vapid gorge]

but yeah I'll run that pure ingot recipe tonight 🙂

[kindred carbon]

How long does a recycling set up take to enter full production usually?

[vapid gorge]

starting totally fresh? it'll depend on the size and how many loops you do

it can take a WHILE

[vapid gorge]

this took a few hours I think

but to be fair it wasn't a clean start. I was flooding each system

and the few hours probably included some tweaking of the system

this was a while a go xD

[fallow siren]

reminds me that i used to do bottom feeding with pipes

[kindred carbon]

All the rubber is being sink

[visual yarrow]

If it matters, every time that I ran into this, it was after going to a base I had been away from, and finding it not running as it "should" have been. So it may have something to do with... uhh, loading? I'm not sure.

[kindred carbon]

Rubber and plastic all going into the sinks now let’s goo

[fallow siren]

how much are you making

[kindred carbon]

And it’s finally 💀

[kindred carbon]

2700 of each

[fallow siren]

using the oil in golden coast?

[kindred carbon]

Oh boy power is already at 33GW

[kindred carbon]

Nope the oil north of the bamboo fields

[fallow siren]

ic

[kindred carbon]

It’s just 2 normal nodes 1 pure, supplemented with the oil wel

1.4m/min points

Very noice

Power is at 37GW

Only got 3GW to spare now lol

Seems to stabilize at 38GW rn

[vapid gorge]

I can have a look at that too 🙂

[jovial jacinth]

Does anyone have a pre-built layout that uses all 2100 Uranium/min? I'm trying nooklr for the first time and trying to wrap my head around which Alts are most efficient, since it seems I need some of that Uranium to combo with the waste, maybe? brain hurts

[vapid gorge]

for most power I'm pretty sure you just ignore the plutonium alts

so just use the uranium ore alts that are more ore efficient

[jovial jacinth]

ok, I'll start there, thnx

[vapid gorge]

this is base recipe for everything except 2 uranium ore alts for the max of 50.4 https://www.satisfactorytools.com/1.0/production?share=KadAFwe6XNunf11ss2lD

you'd prob want todo alts for other components though

[jovial jacinth]

those ore counts are painful...

5800 Caterium is a bit much lol

Pure Caterium isn't much better, 3864 Caterium/Water each

[vapid gorge]

Tempered caterium is good. Just needs a bit of ore.

Alloys for copper and iron. Iron wire

[frigid summit]

im looking to make my first rocket fuel power plant and ill be making 1666.67 rocket fuel, how many generators will I need

[jovial jacinth]

4.16667 / min = 400 Generators at 100%

[frigid summit]

thanks

[brazen cloud]

quick question... im working on a fuel power plant but rn I'm going to make 1066.67 fuel/min... should I make this exact amount or would it be better to make a bit more

[vapid gorge]

in general you'd probably want to roud up the 4th digit of what the plan says.

[slate kiln]

how tf can you cap the Oil production to 240 and it produces 600 ? 😄

or is it 250%

[brazen cloud]

yea that!

[brazen cloud]

sorry I dont get it

[frosty owl]

It depends on what your goal is. Do you want to have all production machines at 100%? To have the Fuel Generators always running? Both?

[brazen cloud]

preferably both 😅

[slate kiln]

@brazen cloud you dont have Diluted Fuel recepie atm ?

[brazen cloud]

nope I need blenders for that right?

[slate kiln]

No

but Yes xD

[brazen cloud]

oh that one, yea I have it but this is way simpler and produces more than enough power for me, I dont tend to build mega projects often

[frosty owl]

To have both you should match the Fuel production with its consumption. Ie: a Fuel output that leaves "clean" numbers for both the Fuel refineries and generators. Otherwise you end up in a situation where one of the two must be slightly off compared to the other (so can't have both running all the time without some extra care)

Sorry, I forgot to reply...

[brazen cloud]

dw

[brazen cloud]

thx a lot!

[wind spade]

The packaged variant is just as efficient

[brazen cloud]

but then I nedd the canisters and a LOT of watter XD so much work

[wind spade]

And contrary to what some people say, is very easy to deal with

[frosty owl]

These kind of situations often end up in having to compromise. Like: if you find a Fuel output that works nicely for the Fuel Generators and refineries, your numbers for Plastic and Rubber will likely not be "clean"

[wind spade]

Eh, they are looped so you just handfeed a few

And water is needed for both in same amount

[brazen cloud]

thats alr I'll mess around with it for a bit

[frosty owl]

Do science!

[brazen cloud]

Ill take a look

[brazen cloud]

Yea nope not doing that XD
handfeed a few: 3200 canister / min

[frosty owl]

Depending on how you build things, 20~50 empty canisters per Packager can suffice ^^
Note: with blueprints, one doesn't need to handfeed more than once, for the very first setup

[brazen cloud]

wait why would I need less canisters?

dont bother answering its ok I'm not doing it XD

[frosty owl]

The canisters aren't consumed. You just need enough to keep the machines running, so smart machine placement and beltwork can require as few as ~10 canisters per machine loop

[brazen cloud]

well thanks anyway

[frosty owl]

Too late :P
And I'm not bothered anyway xD

[brazen cloud]

hahah

thx man

I just dont feel lik placing 60 packagers and 160 fuel gens lol

[wind spade]

20 canisters per refinery

And they are automatically fed since they are in the blueprint

So they are just taken from depot

[heady vine]

What's the easiest recipe to test large assembler blueprint on?

Aka doesn't need large quantity to produce item, small stack size...

[unique cypress]

!wikisearch assembler

[brisk shore]

https://satisfactory.wiki.gg/index.php?search=assembler

[unique cypress]

^ list of all assembler recipes

[heady vine]

ok does anyone has experience of doing this and can share tips?

that link didn't help much.

but I put my eye on portable miner since they take much time to build so at least input betls should fill quickly

but probably not the best overall pick

where did the research on 'best' modular frame go? I remember there was like points comparison for best power and resource consumption

[wind spade]

"Best" recipe depends how you define what is best, so will differ from person to person and from analysis to analysis

[heady vine]

yes, there were calculations done. that's basis upon which you can choose what works best for you... now I can't find it

[wind spade]

I don't know of any analysis that would work so generically

It's always based on author's subjective point of view

[heady vine]

there was like a table with calculations of resources spent, power needed and even space taken by buildings, so at least you could choose what to optimize for right away

[wind spade]

Yeah those are BS

[heady vine]

did it got removed from wiki?

[wind spade]

There was something like that on wiki as well, yeah. Got removed for being misleading at best

[unique cypress]

eh, I liked the resource breakdowns

[wind spade]

The very misleading opinionated resource breakdowns?

[vapid gorge]

that don't cover the most important aspects as to what and when a recipe is good

[unique cypress]

I found them reasonably useful for comparing for example the bauxite cost of different turbo motor recipes.

But they would've been better for me if they used lowest WP recipes for everything upstream instead of making random substitutions

[muted tide]

Every recipe is good in its own way. Sometimes I force myself to use generally not used recipes just to get a factory that is completely different than to what I usually make

[thorn bane]

on its own sure
but make bad choices multiple times in a production chain and you can easily end up with more than twice the amount of required materials

[vapid gorge]

it does depend one which resources you value more.
like maybe you'll spend 3x as much in other things, but only 1/2 on a resources that is more valuable at that moment

[wind spade]

Tools can do that much more reasonably than a static table and will actually include your concrete case instead of just a few selected one from billions of options

[wind spade]

Only relevant when you actually care about resource cost

[visual yarrow]

Wait wait, waiiit.

[charred mist]

is it better (powerwise) to make a diluted fuel factory than a turbofuel factory using diluted fuel?

[visual yarrow]

I heard a faint echo of something about diluted packaged fuel being equi- okay, just...

[charred mist]

powerplant*

[visual yarrow]

quickwire cable

My archnemesis.

[thorn bane]

except that those even out over 30 recipe choices
in the end you are left with needing way more miners

[visual yarrow]

But just to be clear, I am very much in the camp of "WP is inherently flawed and shouldn't be used as more than a guideline at best".

I just used my own deductive reasoning to conclude that quickwire cable was my archnemesis and then it kind of reached the point where it became an injoke and then I stopped being able to tell how actually serious I was about it.

[thorn bane]

and it goes for other things as well
bad choice can lead to 3x the buildings required
or 3x the power required
and having to build 3x the machines or power just leads to people quitting since the progression is too slow

[wind spade]

Not all people have the mindset of "I must use only absolute lowest amount of recipes"

Not to mention that you can't assign WP value to a single recipe, only recipe path

[visual yarrow]

See, here's something that I don't see a lot of people acknowledge. Buildings required and power required are essentially the same restraint.

That's sort of why my own philosophy just boils down to "what set of recipes will create the least amount of actual work I need to do, or at least be the most fun to work with"?

So for that reason, I use things like steel rods -> regular screws, not because of it's efficiency, but because it's a good excuse to make a lot of constructors in one place, which can be fun.

[kindred carbon]

I need blueprint ideas

[unique cypress]

make balancer blueprints

[kindred carbon]

I’ll explain more in detail later
But basically I want a constructor-assembler system that takes Caterium ingot, turns it into quickwire and feeds into assembler
This will be used for AI limiters and circuit boards, I’m hoping that I can avoid running 12 belts of quickwire this way

Using Caterium circuit boards and Plastic AI limiters

[latent swallow]

im training resources from the swamp to the desert, to maintain 1200/m throughput should i split 1 pure node into 2 freight cars? already have 2 trains going and throughput is lacking will a third even help?

[burnt folio]

so once I unlock Mk6 belts and Mk3 miners my steel factory will be producing 2700 ingots/m

how much of that should be Steel ingot/pipe/encased beams?

[wind spade]

that can vary from 0 to 2700

honestly just stop at ingots (though personally I wouldn't even make those). When a new factory has demand for one of the resources you mentioned, dedicate that amount to that factory

[burnt folio]

I'd love for all of my steel production to be in the same place, should I just add more machine in that area when I need more of that material somewhere? ( I'll be transporting with trains )

[unique cypress]

depends. 1200/min/platform is entirely possible with mk5 and mk6 belts. but it's not guaranteed. you need the right number of trains docking to the station, depending on the distance between the stations.

if you're using default stop settings, you can have too little or too many trains.

if you use "depart when empty/full" adding extra trains to the route will at worst do nothing

without more details, about your setup, it's impossible to know what's your issue exactly

[wind spade]

yeah just add machines when there's demand

[latent swallow]

ive found the issue and it was a belting mistake... thank you tho

im making 2200 copper powder/m and ive slowly lost my sanity

[spare osprey]

So currently i'm producing 200m^3/min of fuel. Should i utilize all of it and make sure i dont stop production of plastic and rubber or is there another way to go around it? I kinda want to save some for jetpack etc

[unique cypress]

sink the overflow

or use different recipes so you don't have excess

[spare osprey]

Meaning i have to package it

[unique cypress]

you'll have to package it for the jetpack anyway

so just overflow that and burn the rest

[latent swallow]

my game lags on mass deconstruction, is this the old entity limit issues prevelant in pre 1.0 or just quality settingss needing to be lowered?

[kindred carbon]

A mk3 miner can take like 10 minutes to fill a 4 car train so I think he is fine

[latent swallow]

it was a belting problem i didnt pick up the trains are working well, my brain has just melted after building and connecting 380 refineries for copper...

[unique cypress]

they said they had 1 wagon tho

[sharp mist]

wich is the best altertative reinforced iron plate?

[heady vine]

yet you can run out of sulfur, it's not that hard

[bitter grail]

Whichever one suits your needs and you can make with what you have available

[sharp mist]

i mean screws are so annoying, so stiched?

[latent swallow]

personally stitched and iron wire

[unique cypress]

stitched or adhered

[kindred carbon]

The easiest thing to solve was to just add another wagon

[latent swallow]

the issue i had was solved lol, it was bad belting but that would have been my next call

[wind spade]

screws are fine 🤷

[latent swallow]

is fluid "sloshing" within pipes still a thing with how fluid works?

[wind spade]

yeah, always was

[latent swallow]

did they fix mk2 pipes not running at full cap by any chance?

[wind spade]

there was never any "bug" with them

[latent swallow]

for a long time mk2 pipes would occasionally not transport 600 fluid/m

[wind spade]

that's build issue, not pipe issu

[latent swallow]

if pipes arent running at the correct capacity the fix is buffers before?

[unique cypress]

I have never ran into a situation where a buffer helped

[latent swallow]

ive got 600 water going into 60 refineries using 10 each and towards the end of 2 sections water is just not making it, headlift is not the problem either, bit stumped tbh

[unique cypress]

Build the main pipe a few meters above the refinery inputs. Rarely had issues when doing that

[latent swallow]

yeah ive done that, surely using a blueprint cant be the problem when 2/3 is working fine?

[unique cypress]

Are the junctions level or tilted 45/90° down?

[latent swallow]

level, the pipe goes down at an angle

[tardy bough]

This shit just popped up on my social media as soon as I woke up and used the restroom 10 days before my calculus 2 exam. 😔

[latent swallow]

the manifold has had more than adequette time to saturate too

[unique cypress]

I noticed some of my old blueprints with level junctions behaved weirdly recently but those with the junctions tilted 45° down (towards the refinery) work fone

[latent swallow]

its weird too because of the 380 refineries ive connected this group of 60 is the only one playing up

[unique cypress]

Level junctions work like a splitter manifold, tilted work like a smart splitter manifold

[unique cypress]

You can always try flushing that pipe

Multiple times

[latent swallow]

yeah true might have to resort to that not a bad shout

[unique cypress]

That gets rid of sloshing so as long as it doesn't start again, you'll be fine

[kindred carbon]

#off-topic-general

[dreamy umbra]

burn the overflow

[thorn spoke]

When using a manifold, should I make the splitters go in one direction or two?

[oblique hollow]

there is no rule / guideline for that

do it however you want / need

[dreamy umbra]

exactly,there is no drawbacks if you have enough conveyor capacity of course

[burnt folio]

wait so the "Alternate: Fine Black Powder" just saves a ton of coal without using any more sulfur?

[split sierra]

i guess

consumes marginally more energy

which is negligeable since you need it in microscopic quantities anyway

[unique cypress]

Huh, last I checked, I thought it used more sulfur. Did they change it?

[heady vine]

please remind me where do i ask greeny about tools

https://www.satisfactorytools.com/1.0/production?share=ELNL9Numw7H1uDctmW9q
I have some weird shenanigans

[unique cypress]

You can probably @ them here, or ask in the Tools Discord server. Or rather, check the #faq there first

But if you're talking about those 0x and -0x boxes, they can just be ignored

And any change in the plan can make them appear/change/disappear

I'm assuming they're just floating point errors that are just too big to get rounded to 0

[heady vine]

oh, found it

[queen slate]

Same sulfur, 1/3 less coal. Seems like they changed this recipe.

Oh and in any case, there are Somersloops, and later - T9 conversion stuff.

[burnt folio]

where do ya all make your black/smokelss powder? was thinking of making it in the lake forest due to the oil, sulfur and coal there

[glacial shadow]

dunno what to take

[unique cypress]

Encased frame 100%. The only good HMF recipe. I never build HMFs without it

[humble dirge]

someone able to explain what I have done wrong?
my train that heads straight will come in fine however wont leave the station behind me

purple line has a path signal , ahead of it is a block and to the right is a block
trains turning right have no issue coming in or out

the train has 3 carts on it if that makes any difference

[unique cypress]

Probably because the block signal at the start of cyan (forward) is red.

You should have block signals every 100ish meters on long straightaways

[humble dirge]

yeah i got blocks roughly that amount

or atleast should...

[humble dirge]

blues are path and green are blocks
train travels into the main station on the right hand lane

[dawn quartz]

is there any way to get a specific alt recipie
ive been gambling for one but have only been getting awful alternates

[wind spade]

Whichever you like more

[wind spade]

Eh, pretty much all alts are useful

[humble dirge]

figured it out and feel dumb, station on the right was facing the wrong way...

[unique cypress]

Either cheat or save scum

[candid steppe]

why doesnt the junction turn the other way?

[humble dirge]

Not sure what you mean i dont have that bottom line connecting to the right side of the image’s station as those trains will never visit it

[light cosmos]

I made this goddamn thing with the exact amount of resources needed but for whatever reason the Iron Pipe constructor is going idle because there is no consistent supply of ingot everytime it finishes a craft, I have the 4 smelters (manifold) merging in one MK3 belt that split 2 times, I have 4 constructors stacked in pairs (the second image) on top of eachother, using a lift with a splitter attached to it to split between top and bottom constructor, the pipe one is "sharing" the supply with 1 wire constructor and the other 2 wire constructor are sharing their own supply belt, all 3 wire construtor are constantly with 100 ingots in the input, which means any other ingot shared with the pipe constructor would go to it (unless I am completely understading splitting wrong) and yet it's not enough

the belts with the ingots coming from the smelters are under the foundation

[thorn trail]

the belt/conveyor feeding the iron pipe constructor is at least a mk2, right? If it is only a mk1 that would be your bottleneck

[light cosmos]

yea, it's mk3 too

[thorn trail]

are the input buffers on all 3 wire constructors full and are the belts feeding them backing up?

[light cosmos]

the right one are only wires, they're baking up not 100% of the time since it sometimes consumes when they craft, but like 99% of the time, the left are 1 wire and 1 pipe, the wire constructor has 100 ingot in the input pretty much the whole time

[near zodiac]

So, if I understood correctly, to calculate the max theoretical throughput of a train station is the amount/min that's being "fed" to the station * the time it takes to do a full cycle till you get back there?

[unique cypress]

It's not that simple

[thorn trail]

that part sounds odd, because if you are splitting the 113 ingots out of the smelters by 2, then that should be ~56 ingots going into each pair of constructors, which would mean that the wire only pair should be filling up very quickly since they only consume 22, while the mixed pair of constructors should be starving until the wire only pair fills up

[unique cypress]

This is my train calc

[light cosmos]

it is backing up like this, but obviously sometimes it moves

[thorn trail]

the conveyor lifts are also mk3 correct?

[light cosmos]

the mixed one has the wire constructor "full" with ingot and the pipe is starving

yea, I changed everything to mk3 when I unlocked

[thorn trail]

are the smelters all running at 100%? are they backing up at all? Check the smelter output manifold for a bottleneck as well

if they are actually outputting 113 ingots then they should be backing up if the constructors are not consuming all of those ingots

[light cosmos]

one of them was underclocked since it's not 120 ingots, but then I changed them all to 100% to see if it would help, it's not backing up, it feels like it's disappearing somewhere wtf

oh

I found the issue, jesus fucking christ

2 of them were unpowered for some reason, probably a wire broke somewhere and I didn't notice

where is the factorio horn when you need it

[thorn trail]

excellent, glad you found the source of the problem

[light cosmos]

thanks for the help

[near zodiac]

so I understood wrongly

[near zodiac]

explain

[unique cypress]

There are 2 things that limit how many items per minute you can transfer via a station:

How fast you can get the items to it (a train will drop off up to 32 stacks every time it comes by)

How fast you can take items out of the station

The transfer rate will be whichever one of the 2 is the limit in that particular situation (whichever one is lower)

The main complication is that the belts connected to the platforms stop for 27 seconds every time a train docks

The top 4 functions are just different stop settings and belt configurations.

Most of the mess is because of desmos' variable handling and I wanted to minimise the number of symbols I had to type (I typed this on mobile lol)

[near zodiac]

so I need to calculate the item input speed, which I assume is 3200 * stack size / belt speed
And the time it takes to fill, which according to the website, is stack size * 32 / belt speed * 2 + 0.4513

wait shit they're the same thing

I am confused and sleep deprived

[unique cypress]

It's probably better explained on the wiki

[unique cypress]

!wikisearch tutorial:train+throughput

[brisk shore]

https://satisfactory.wiki.gg/index.php?search=tutorial:train+throughput

[near zodiac]

that's the one I was looking at lol

[unique cypress]

Not sure if it explains how having multiple trains works out

[near zodiac]

so I'll need to calculate the RtD of each station

[unique cypress]

Measure, not calculate

[near zodiac]

hm that's kinda what I was gonna do

[near zodiac]

sorry am stupid

[unique cypress]

Make sure you measure the entire route. Choose a station, and measure the time it takes between the departure honk and another departure honk after the train has driven its entire timetable

Not departure to arrival or something

[near zodiac]

the honk is from the same station, right? After it has done a full loop

[unique cypress]

I mean you can choose any other point in the timetable, but I like using the honk.

[near zodiac]

IS THIS THE FUCKING DIJKSTRA ALGORITHM

[unique cypress]

Also, the loading/unloading takes 27 seconds, so if you use just a locomotive to measure the time, you'll have to add 27 seconds for each stop

Plus a little extra for the train being heavier than just a locomotive

[near zodiac]

yeah, I'll also be sure to consider traffic in case I decide to use more than one train

[unique cypress]

Plus a little extra to account for traffic...

Well, with multiple trains driving the same route, you might want to consider using "depart when empty/full". Without it, the trains might bunch up and you'll lose most of the benefits of using multiple trains.

Or you might fall on the left side of the graph if you add too many trains and you'll actually lose throughput

With depart when empty/full, adding extra trains will at worst do absolutely nothing and just extend the queue behind the station

[near zodiac]

makes sense

when I started playing and saw people using notebooks, stickers, even boards to write down stuff I thought
"Heh it can't be that bad."
Now I understand them.

[unique cypress]

I am a digital kinda guy, but I still use 2 different production calculators, a map, and my own train and drone calcualtors

[near zodiac]

yeah I have those stickers that come with windows

[vapid gorge]

only if you pick every sulfur recipe in the game. For the most part it's main uses are

1. nuclear

that's about it. You need if for the nuclear chain. Ever since they murdered batteries you can bypass that lump. Everything else is niche. And if you burn yourself out on rocket fuel thats your own fault

but it's the same with any of the resources. If you purposefully pick all the caterium recipes you're gonna run out. Same with crystal, hell even oil. the only one that's kinda hard to do that is limestone and iron.

@visual yarrow got that pure iron rig going. Gonna let it run for a while

[heady vine]

eh, turbofuel? batteries?

murdered batteries? what did i miss?

[vapid gorge]

batteries have barely any use post 1.0, unless you're still doing drones with them

and tf is entirely skipable and not needed en masse in any way

[heady vine]

I mean, drones got to fly somehow

[vapid gorge]

drones can use other fuels now

[heady vine]

I have to roleplay them having to use proper energy source

[amber umbra]

Drones don't really eat that much fuel if you set them up properly. Need a lot of drone routes to even eat 1 sulfur patch.

[vapid gorge]

it used to be a thing where drones were a bit of a deal cause you gotta set up that battery infrastructure, but now....

[vapid gorge]

I can respect that 🙂 but even a huge fleet of drones isn't going to put a big dent in the world sulfur supply

[vapid gorge]

@visual yarrow ok spent an hour away from it, zero ingots buffered in teh machines, all running smooth

[tidal swan]

hello any alternative for splitter? its so slow

[vapid gorge]

splitters have infinite throughput

[tidal swan]

its stopping

oh its merger

[vapid gorge]

then your belts either doesn't have the throughput or youre not consuming all the parts

or both

this is not a splitter issue

or merger

[tidal swan]

hmm how do i consume all the parts?

[vapid gorge]

have machines that consume them or send them to a sink

but if you're doing this and you don't have an overflow to a sink? yeah your line is going to get clogged

[tidal swan]

do i need 2 storage thhen?

[vapid gorge]

you can do mixed item belts easily, but you have to build for it

[tidal swan]

wat is a sink? theres such thing?

[vapid gorge]

have you unlocked smart splitters yet?

[tidal swan]

nope

im still phase 1

[vapid gorge]

ok then doing mixed belts for you is essentially impossible.

keep it to one item type per belt.

[tidal swan]

ok thank you

[vapid gorge]

and pay attention to the numbers you're sending down the belt 🙂

I'd probably wait to do mixed belts until you have a good handle on the basics

[tidal swan]

its not stopping now. ty bro

[vapid gorge]

@lyric trail so what exactly are you trying to build?

[lyric trail]

I been waitn..

[vapid gorge]

how many super comps pm and which recipes? 🙂

[lyric trail]

Planned on making 10/min, and everything else beforehand was going to be 100/min AFTER the supercomputers were produced

All caterium recipies

Can't remember super recipe though, mabye normal?

[vapid gorge]

so 100 of each and every single other part involved?

[lyric trail]

10 super computers

[vapid gorge]

"and everything else beforehand was going to be 100/min AFTER the supercomputers were produced"

[lyric trail]

100 limiters, high-speed connectors, computers

[vapid gorge]

so +100 circuit boards, +100 high speed connect...

[lyric trail]

All but circut boards

Didn't care for them

I have all the resources and everything, but couldn't build it for the life of me

Idk y

[vapid gorge]

ok yeah no wonder you're havingb issues

[lyric trail]

I ended up with like 16 belts with different amounts of quick wire on each one lmao

[vapid gorge]

gimme a min

are you using the recycled plastic recipes? I would hope so

[lyric trail]

And it's impossible to merge them into specific amounts vertically into 6 different systems where each system takes over 780 per min

[lyric trail]

Yes, made on another plant

[vapid gorge]

you sure you're making enough of each?

[lyric trail]

It's the only thing made off site

Yep

[vapid gorge]

ok that means I'll need exactly which recipes you're using.
plastic AI limiter?

[lyric trail]

I think so?

I got off so I can't see

Yes it is

[vapid gorge]

jsut for hte super computers, does this look right? https://www.satisfactorytools.com/1.0/production?share=5tehIqRshRLzChn4p50L

[lyric trail]

Not wuote

Give me a sec to type all details

Plastic ai limited, silicone high-speed connectors, caterium computer, supercomputer, pure caterium, recycled plastic, caterium circut board,

Had like 7000 quick wire I think

[vapid gorge]

this is just the 10 super computers. And I did alloy iron and copper because it's simple and useful

ok so just changed the HSCs, should be good https://www.satisfactorytools.com/1.0/production?share=RfPgOa5VRfLsSq6Iz39k

[lyric trail]

Are you suggesting that or smthnM

[vapid gorge]

I figured I'd chuck it in to save on copper. But I can take out the alloys if you like

if you don't have the iron near

[lyric trail]

Iron for what?

Don't need iron I don't think

[vapid gorge]

never mind, I'll axe it 🙂 copper alloy creates more copper in exchange for iron

[lyric trail]

Sorry I'm confused

[vapid gorge]

ok this should be the final look of just the super computers https://www.satisfactorytools.com/1.0/production?share=5BuPsWuo6o5p2V2gSlbg

[lyric trail]

I was never using copper or iron

[vapid gorge]

oh woops one sec forgot something

oh weird, I did do caterium computer, it just didn't wap to it

[lyric trail]

Caterium computer and circut board

[vapid gorge]

yeah usually the planner swaps to them w/o having to turn off the base recipes, soz

ok THIS should be it https://www.satisfactorytools.com/1.0/production?share=a1BRqTSqoM1Irs0HFhRZ

[lyric trail]

I'm using 128 refineries, 128 Constructors to make the quick wire, then everything else is just plastic I thinksies

[vapid gorge]

so just to confrim, that last link is what your super computer chain looks like right?

[lyric trail]

Essentially yes

[vapid gorge]

ok and is it 100 of everything else? like wire and cable and ingots?

[lyric trail]

Ok no lmao

100 AI Limiters/min 100 high-speed connectors?/min 100 computers/min

All as spare for sink, storage, and other factories

[vapid gorge]

not circuit boards?

[lyric trail]

No

[vapid gorge]

ok give me a min

HSCs https://www.satisfactorytools.com/1.0/production?share=4vUJ6HHJAbsr4BE17Gu2

AI
https://www.satisfactorytools.com/1.0/production?share=gzSBLW6EoqVFAE0yWuTx

Comps
https://www.satisfactorytools.com/1.0/production?share=gzSBLW6EoqVFAE0yWuTx

siphon off what you need from these 3 to put the last factory down

for Supers

using satisfactory tools will make your life much easier than a spread sheet

but you also made this really hard on yourself having many end points for hte factory

[lyric trail]

I tried to use this earlier today but it confused me more

Cause it's like spaghetti

[vapid gorge]

it takes a bit of practice but it's really worth it

[lyric trail]

Ight thanks 👍

[vapid gorge]

make these 3 different factories 🙂 independent of each other

[lyric trail]

"Kisses"

[near zodiac]

Alright so, to calculate the throughput of a fluid station is the same as the regular freight?

[unique cypress]

Basically. Assume stack size is 50 and use pipe speed instead of belt speed.

And you can't really fully use "depart when empty/full" because fluids can't be balanced

[near zodiac]

What if I'm filling with both fluids and materials in one station? Would it work?

[vapid gorge]

you'd have to calculate each car

different stack sizes and different amount each can carry, and probably different parts per min

[near zodiac]

Yes I am aware

[crimson moat]

that's a pretty bad myth - not only can they be balanced, but the biggest problems with liquid flow come from the fact that they're actively balancing themselves so hard that it breaks when people are trying to force imbalanced flow.

if you put a Y junction you will get an essentially perfect 50/50 split every time, they'll actively balance to do that and you cannot turn it off. It's just inherant to the pressure/flow system.

[unique cypress]

I literally had to switch to packaging my nitrogen because the platforms wouldn't drain at the same rate when it was a gas, even though they were all connected to the same system

[crimson moat]

got a save file?

[unique cypress]

So if you have a method to ensure fluid platforms fill/empty at the exact same rate, please share

[crimson moat]

They've always done that for me, with <1% variance

but i dont use em a lot

[unique cypress]

I switched to packaged nitrogen so no, I no longer have fluid platforms on it

And besides, it's a U8 save. The nuke plant the nitrogen is supplied to will stop working soon after loading the save

[crimson moat]

ok lemme know if you make one and it doesn't work and i'l take a look

only thing off the top of my head is that you might have to make the pipes equal length (actually capacity)

[unique cypress]

I won't because I have since learned that packaged fluids are more dense and more efficient to transport.

Well, unless I'm moving less than 600/min, because then the imbalance doesn't matter

[mossy ibex]

can someone talk to me about pipeline loopbacks? I think I mostly have a handle on fluids now, but I'm not confident I know where to add buffers and loopbacks. I'm at the stage of knowledge that makes me want to add them "just in case" but I want to be at the stage where I can reason clearly about it. I think it mostly matters at or near the pipeline capacity, right?

[unique cypress]

Don't do them

[thorn bane]

correct
and the more uneven the split the earlier it is required
like 600->10 machines is 600->540/60
where 600->30 machine is 600->/580/20
so the bottom one is more prone to needing to be split in 2
also feeding from below/from above seems to affect it as in feeding from below is worse so you need to split earlier but i havent looked into that a lot since i never bottom feed anyway
also just a friendly reminder that all the "loopback" does it split the pipe into 2x 300/min

from my experience its use it for >500 to be save, most work for 550 (assuming "normal" builds) but like use at your own risk
low throughput stuff like non-oc turbofuel gens prolly even as low as 450

[oblique hollow]

a loopback does something similar to an "even" split. Thats the trick to why adding loops works

[mossy ibex]

when you say "earlier" do you mean doing a loop earlier in the run, as well as later in the run (eg at the end)?

[oblique hollow]

always at the beginning

the pipe with the highest flow rate is the one that needs "protection" from junction sloshing

which, in basically ever pipe network, is the first pipe. The main input

whatever method you use, the point is that the very first junction that has to split the pipe with the highest flow rate is the junction that can ruin flow

and thats the junction where the loop must begin at

or where you must apply an even flow rate split first

[thorn bane]

no i mean earlier as in at 500/min already not 550/min

[mossy ibex]

hm, ok I'm not completely sure I get it, but let's say you were feeding 900 fuel down 3 mk1 pipes, and your goal was to feed 40 refineries--so 3 pipes at capacity, feed by exactly enough machine to produce at capacity, with exactly enough machines to consume it. I think this is where dragons are, right? So let's imagine the producing and consuming setup are totally separate, the only connection between them is the 3 full pipes.

In my current state of knowledge, I'm going to let the producing machines mostly fill up a pipe to capacity, and since there is inevitable math shit about what they produce and what pipes can hold, I'm going to junction the three full pipes together right before they exit the producing area. I'm unsure whether also connecting the three pipes at the back of their respective lines would be helpful.

So now on the consuming side I have 40 machines that need all the fuel in 3 pipes. I make 2 rows of 20 machines, I send one full pipe down each row. I also split 1 full pipe into a second pipe for each row, so now each row has 2 pipes, one of which is half capacity. These pipes are junctioned in front of every machine. At the very end of this line 2 pipes of the row connect across to the matching 2 pipes of the other row.

Is this all fine? I suspect not

ie, the math works fine naively, but I suspect shenanigans

[oblique hollow]

mk 1 suffer less from this.
but anyway: you just need to have a junction that splits each mk 1 pipe 3 ways. not just one way into a machine and then down to the next junction

but... honestly, just draw a paint image of what you mean by "junction the 3 pipes together"

because to me it sounds like you are trying to do something pointless

[mossy ibex]

ok one sec

[mossy ibex]

in that diagram there are only 2 machines consuming on either side instead of 20, but that's the idea

[thorn bane]

that seems odd
just do this?
if you want a loopback then just connect the last to the first in each row

[mossy ibex]

eh, I guess if you used 42 refineries and underclocked them yeah

[thorn bane]

ye

[mossy ibex]

but this is more the principle, I'm trying to understand the dynamics here, not really solve a specific problem

[oblique hollow]

if you can, dont interconnect systems

its not good practice

[thorn bane]

imo its way simpler to keep pipe systems independent at <600 total

[mossy ibex]

"system" here means pipes? Like don't connect the pipes if I can help it?

[oblique hollow]

system = pipe network. Everything thats a junction, pipe, buffer or pump

[thorn bane]

your pic is 1 system
mine is 3 separate systems

[oblique hollow]

one example is the classic 3 water extractors to 8 coal generators all hooked up to a feed pipe network

thats one "system" . one pipe network

if you now build 16 generatos and 6 extractors and all join them together, you turned 2 seperate systems into 1 interconnected system

when it wasnt ever needed

[crimson moat]

that's why i like that simple design which is just 4 pipes and a junction

[oblique hollow]

Granted, yes, theres some level of flavor as to where you draw the line

you could reduce every single pipe network in the game to 1 to 1 connections with 0 junctions, probably

but generally, I dont merge systems where i already have a maxed out pipeline

[thorn bane]

meanwhile i have 800 alu solution in 1 pipe loop 🙃
its my rule so i can break it

[mossy ibex]

well I do have a thing in my save where I have 4 full pipes of HOR, going to a diluted packaged fuel place, and none of the pipes are connected, just the refineries exactly filling their pipe, then other refineries exactly consuming them, and no connection

but I was never sure if that helpedor hurt or made any difference

[oblique hollow]

keeping it seperates makes troubleshooting easier

you fix one system, done, next one

[crimson moat]

courtesy of tinker we have 108,000 water on one system x3

[oblique hollow]

tinker that maniac

[thorn bane]

wait what?

[oblique hollow]

their friend is called tinker

and with a name like that, yep, that pipe network checks out.
tinkered out to the max

[amber umbra]

@mossy ibex My personal pipe rule after verifying the 600 fluid/minute pipe “feature” is:

1. Never exceed 550 fluid/(minute * pipe) for mk2 pipes
2. Use simple, flat manifolds without loops
3. Don’t use buffers except for intermittent supply like trains
4. Don’t use valves

Just sharing an alternate way to handle the situation.

[mossy ibex]

thanks everyone

I wish they had reworked pipes, honestly

[thorn bane]

same
but its not easy to do

[wind spade]

Eh, I feel that they are fine

[mossy ibex]

I feel for them, I've actually been there. Programmed something bug free that worked, but it was just so weird and confusing I had to scrap it and start over with something simpler so it was possible to use

[amber umbra]

Yea is what it is. If they could easily tweak it I’m sure they would.

[mossy ibex]

yeah no, it would be a rewrite that would put existing saves in shambles

the time to do it was 1.0 if ever

this is our lives

[amber umbra]

Interesting seeing Factorio redo their fluid system in 2.0.

Distance based system with infinite throughput inside a given fluid “area”. Pretty different.

But yea, imo that speaks to fluids being tricky to code into games like this if their solution is basically “yea we’re going to just not bother with realism simulation stuff”.

[mossy ibex]

having thought about how I would implement it for about 10 seconds now, I can confirm it seems dicey

[thorn bane]

relevant dev stream:
https://www.youtube.com/watch?v=COJV4Ubn1Ac&t=191s

[mossy ibex]

one thing I'd say though is that valves just need a rework

I get how and why they work, but it's bleh, they should just change them

the limit should be absolute, not proportional to fill

[amber umbra]

Mhm. Not going to get much discussion here though.

[wind spade]

honestly - that wouldn't change the fact they are pointless to use

[cerulean stratus]

Don't try to have fun with fluids, essentially

[amber umbra]

Working fluids are fun fluids to me 😆

[cerulean stratus]

Valves not working properly are especially funny

[oblique hollow]

idk what you mean. Just have a full pipe and it works fine

the thing about the valve is that it essentially is almost exactly like a pipe with a different flow limit cap

if you were to remove the valve and just have a pipe with adjustable flow speed.... you'd get the valve again.
Adjustable flow speed pipe would act exactly like the valve does right now

(minus the preventing fluids from moving back the other way)

[wind spade]

valves work properly though...

[cerulean stratus]

I heard they only work properly if the pipes are full

[wind spade]

that depends what you call "work properly"

[oblique hollow]

thats not a problem / bug you know

[wind spade]

they work how they are supposed to work

[oblique hollow]

its a feature

[wind spade]

the "problem" with them is people expecting them to work differently

[oblique hollow]

pipes only get max flow when full, valves only get "max flow" when full
same deal

[cerulean stratus]

So if I want to do 250 per min here and 150 there, do valves help in that?

[oblique hollow]

if your input is 400 total, 250 valve here, 150 valve there

the input pipe will / has to fill

theres no other option for it.
It cant empty at 400/min unless full because the valves wont let through more until the pipe is full

[cerulean stratus]

So what problem would it give me?

[oblique hollow]

none

what are you trying to ask here is what i wanna know

[wind spade]

generally the solution is "let fluids sort out itself"

[cerulean stratus]

Put some in a train, put other thing in a truck

[oblique hollow]

train: use buffer tanks after valve
truck: packager does its thing

[cerulean stratus]

Thing is, buffer means manifolds will take long to balance and I don't want that

[wind spade]

I don't think the situation requires valve anyway

[cerulean stratus]

Hence why I was thinking of valves

[wind spade]

you want full pipes no matter what

[oblique hollow]

you need buffers for trains or else you can just forget using trains

[wind spade]

so you want the manifolds to fill up

[cerulean stratus]

Do I really need it full? And not 2/3 full

[oblique hollow]

the train station enters lockdown for 27 seconds during which you need to buffer the flow out of the valve

if you take a valve directly into the station, you lose out on throughput for 27 seconds

thats also why you dont take a belt directly into a station if that belt is full

[wind spade]

the more full, the better pipes work

[unique cypress]

I have literally never wanted anything like that

[wind spade]

you most likely don't need things 100% full, but it also can't hurt to just fill up

[wind spade]

and also you can just have two groups of machines, one making 250 and one 150

[unique cypress]

I'd say it's best to start pipes empty and let them fill up naturally. Then they're as empty as possible while maintaining throughput

It works well for me

[wind spade]

well filling the pipes makes it easier to find issues

[unique cypress]

It makes issues show up faster. Doesn't really make them easier to find

[oblique hollow]

literally just this. whats there to manifold

[wind spade]

well that's what I mean

[oblique hollow]

finding issues faster is a good thing isnt it. Means you can fix it faster

[unique cypress]

I'm pretty sure that pre-filling pipes can cause sloshing if not filled completely, so I'm generally against it. And it has never fixed any issues for me, so I don't really see any reason to do it either. I can wait

[oblique hollow]

General reminder that "im fine with it / i can wait" is something specific to everyone.
Others might NOT be ok with that

but the pre-filling discussion honestly just misses the point of the original question

you need full pipes to get full valve throughput. Not half full and not 2/3rds full

Not taking into account verticality differences

[thorn bane]

@eager girder
seems to be 8:7 split of the crystals

[eager girder]

yeah

with 15 crystals/minute, that's 3 machines making shards at 60 each, and 140/min to crystallisation and 40/min to trap

so I put two machines straight into crystallisation and one machine into a junction, pipeline mk2 to trap and mk1 to crystallisation with valve

and also underclocked both machines exactly

should be Good Enough I hope

[thorn bane]

not a fan of valves
id build 2.33 encoders for going into crystallization
and 0.67 encoders going into trap
that way you get your 140/min and 40/min directly

[eager girder]

that is a better idea

now realising I made the totally normal supergenius move of adding the fourth converter but not plugging in any of the inputs

[cerulean stratus]

thoughts on quartz?

I generally only see it used for crystal oscillators

[unique cypress]

Oscillators, RCUs, Silicon High-Speed Connectors, and Infused Uranium Cells for me

well, the last 2 use silica, not crystals. idk which one you're asking about

[oblique hollow]

its pretty pink

[queen slate]

It's great with Uranium! Also I prefer it in HS-connectors and Shotgun Rebars.
I also liked it in Circuit Boards for one of my setups.

And how could I forget about Signs!

[near zodiac]

So, I have a railway system that takes around 5:35 minutes to do a full cycle from each station (not exactly 5:35 from each, just around that). Is it good enough if I calculate the throughput considering 6 minutes?

[humble dirge]

thats like one of the esential recipes to grab right?

[wind spade]

there's nothing like "essential recipe"

all recipes are good in certain situations and bad in certain situations, also depends on user preference and such

as for this recipe, it doesn't really save any resources, so I personally don't really use it, but you can do whatever you want 😉

[humble dirge]

i remember there being certain alt recipes that are commonly used that makes life much easier
i remember there being a screw one that skips out the need of creating rods which im guessing that is it

[unique cypress]

Not making screws in the first place usually makes life much easier

And the only essential recipe is heavy encased frame lol

[humble dirge]

true

[near zodiac]

hey Kyo, if my railway system goes at around 5:35 minutes each trip, would it be safe to calculate the throughput considering 6 minutes?

[thorn trail]

the holy trinity of hor/recycled rubber/recycled plastic are not essential?

[unique cypress]

I would calculate for 5:30, 5:45 and 6:00. If all 3 provide sufficient throughput, you're probably fine

Assuming the time includes all the loading/unloading

[near zodiac]

hmmm, maybe 5:45 and 6:00 then

[unique cypress]

I measure my time without items, and then add 30s for each stop because I usually don't have items to transport yet
Idk how you did it

[near zodiac]

my measurements already include the stopping times

[unique cypress]

Without D(P)F, that combo is kinda meh. And the only amazing thing about it is the oil efficiency. If you're not making large factories, you don't really need it.

HEF, on the other hand, makes all HMF factories half the size and iron by itself

If I could only choose one recipe, HEF would be it

[cerulean stratus]

honestly I tried HFF and Hef beats it

you need to do so fewer modular frames

[wind spade]

"commonly used" yes
but not "better"

[unique cypress]

I'd say HEF is closest to being objectively the best recipe for HMFs. The only reason it isn't is because HFF saves you limestone at the cost of oil. Theoretically, there could be someone who wants that

[cerulean stratus]

it doesn't even need coal

[unique cypress]

I meant oil...

It's 1am

[cerulean stratus]

honestly I just did it once to try it

10 hff per min
And I needed 50 modular frames, that was the hardest part

[near zodiac]

Alright so, considering 6 minutes:
The time for one pipeline to fill a fluid station is 4 minutes (2400/600). If I consider that, I would have to halve the pipeline in order to get the station full in 6 minutes and that would essentially double the amount of stations I would have to use. So scrap that.
I will then have to put two trains in the railway, which would theoretically bring one train to each station every 3 minutes. 3 minutes is 80% of 4 minutes, so the station would be 80% filled when the train gets there. To get the remaining 20%, I would have to get an extra pipe on the fluid station bringing 120m³ of oil.
Trains are harder than I thought.

And this is only for oil. I still have nitrogen gas, sulfur and coal to account for.

[quick gorge]

A friend sent me this.
I have no words so now you must suffer

[wind spade]

ah yeah, the "balancer" that won't work as a balancer because pipes are bidirectional 🙂

[vapid gorge]

just tell them pipes don't work that way 😛

[quick gorge]

I don't want to burst their bubble... that is within their pipes apprently

[vapid gorge]

buuuuurst it. for the sake of their factory

[quick gorge]

Every pioneer needs to learn, the guy has plenty of hours

[vapid gorge]

and that's how you get weirdos going 'damn mk2 pipe bugs' 😛

[quick gorge]

https://tenor.com/view/ryan-reynolds-ugh-fine-gif-12676450205444971845

Wave hi chat, I linked them the message :3

[vapid gorge]

👋

you did good, you probably saved them from becoming a terrible reddit weirdo

[quick gorge]

So ... like 95% of reddit users?

[vapid gorge]

yes

But especially the satisfactory reddit too.

[quick gorge]

The Satis reddit is... well.
I post something that took hours of effort .. no interaction
I post zero effort... 200 silly internet points.

[thick agate]

i have been summoned

bidirectionality or not, the bottleneck in this whole debacle is the train not being able to transport the nitrogen gas fast enougvh lol

[vapid gorge]

oh. that's even worse. You're putting the gas directly into the platforms?

[thick agate]

yerp

oh well thats a problem for future me

[vapid gorge]

not gonna cut it. Even if you're buffering it properly, gas hates buffers

package it first, it'll save you space and logistics at the same time

[quick gorge]

https://tenor.com/view/colts-indianapolis-colts-dashawn-dashawndontplay-comebackcolts-gif-12298488741851068567

[vapid gorge]

It sounds like it's a problem for now and past you too tbh

an 'all time frame issue'

[thick agate]

now and past me roll for a persuasion saving throw and roll a nat 20
they shant be bothered right now and will go to sleep
tomorrow me will have fun

[quick gorge]

This is going to end well

[vapid gorge]

well you can put gas directly on trains if you only really need like 1/2 or 2/3 the throughput

[unique cypress]

Package it. You'll get 5-7.5 times more throughput

[quick gorge]

That's for awake beans. Not when it's 3am

[thick agate]

the fuzed frames "only" need 500 gas/min but it hurts to have the 1650/min node just pump nothing for the rest so yeah, tomorrow me will have fun

the balancer shit was for a on site packaging facility in the swamp, right next to the bauxite

[quick gorge]

I am personally making he Nitric acid on site before putting on a fluid wagon.
Also I WANT a fluid wagon because I want to use at least one of every thing in the game and fluid trains are one of those

[thick agate]

and fluid wagons look cool!

[wind spade]

why not just hook each pipe to one platform? no need for balance lol

[quick gorge]

Hey have they fixed pipeline floor holes? . .. I cannot remember 👀

[thick agate]

because i use 6 packagers so 2 packagers per pipe but i didnt remember what platform had what input so i wanted to balance the 4 outputs into 3 pipes

[vapid gorge]

use the packager's clocking to 'balance' it out. save yourself some grief

[thick agate]

i can bury the grief in a pit of tar, alongside my dreams of cool fluid wagons

[unique cypress]

Could've balanced the items instead

[thick agate]

i'll probably just keep the BS i build here because i like how it looks and figure out a way to send the empty canisters to the gas node station

[quick gorge]

Best of luck Beansy <3

I'll be playing in the design channel

[thick agate]

may the power the pipe be with me

[opaque quartz]

Past me is an asshole
Future me deserves what’s coming to him

[quick gorge]

Truer words have never been spoken

[vapid gorge]

@visual yarrow alrighty. 2 hrs away from factory, running at near 1200pm, every machine clocked to like 50%

[silent hawk]

Wait people actually do non spaghetti?

I thought it was a joke

[vapid gorge]

it's more likely than you think

[split sierra]

minor gripe that i dont understand, why does the intersection start tweaking if i place the circled block signal? (works fine otherwise)

[wind spade]

wdym by "tweaking"

[split sierra]

wont work and doesnt seem to recognize that block signal (by the color of the rails at least) despite snapping correctly

[wind spade]

rebuild the section

also why do you have signals so close to each other?

[split sierra]

the suspended pillar section is a blueprinted part that i use to grid the world with rail auto connect, i started builing the intersection a bit further out so that it wouldnt clip

i solved the issue by just moving back the signal that was causing the problem, i suppose there has to be a distance greater than 8 meters between 2 block signals to not cause it

[thick agate]

honestly, wrangling 100+ refineries is a lot easier if you do grids 😅

[split sierra]

funnily enough, the same thing happened here. Looks like the bottom line is that block signals do not like being placed close to each other on intersections

[thick agate]

what error does the signal spit out when it starts going tewaky

[split sierra]

everything loops into itself

its the first time i've had this issue with trains

[thick agate]

aaah my favourite one (sarcasm)
I usually got that fixed by disassembling and rebuilding rail segments and signals until it no longer has something to complain about
Havent fully figured it out yet

[split sierra]

nono that wasnt sarcasm somehow

[unique cypress]

Whenever the color of the rail before and after a signal is the same, something is wrong. Usually either the rails are too close, or a switch/connection is glitched and needs to be rebuilt

They fucked something up in 1.1, the glitched switches were pretty rare before

[candid steppe]

hear me out

[unique cypress]

no

why?

[candid steppe]

free plastic

residue be gone

[unique cypress]

it's not free if it costs oil

[candid steppe]

yea i know

[unique cypress]

and 210 plastic from 155 oil isn't fully efficient

you could get 465

[candid steppe]

how

the normal way gets you 104

[unique cypress]

https://www.satisfactorytools.com/1.0/production?share=0LLQxgbj4KymFuyj61mx

and that's why you don't use modeler

you have to know what you want

Tools automatically optimizes for resources

[opaque pebble]

Modeler's useful when I want to plan out the actual number of machines, I guess

but I still feel it gets cluttered with too many lines

[wind spade]

tools give you actual number of machines

[opaque pebble]

More like machine count + exact clock speed for each machine, an edge case

or I haven't poked around tools enough

[wind spade]

what difference is there between box with "10 machines" written on it vs 10 boxes? except that the first one is easier to count

[jovial jacinth]

Exact clock speed is a bit of a stretch. When the things tells you to use 4 machines at 100 and 1 at 50....I would hope most people don't follow that

[unique cypress]

tools gives you the same information as modeler (other than UC/OC support)

[wind spade]

why not?

[jovial jacinth]

to be fair, that is an accurate clock speed usage, but anyone who plans stuff out would do 5 at 90

[unique cypress]

yeah, idk why that's included
clock speed distribution is very customizable

[wind spade]

there's no reasonable difference between those two options

[unique cypress]

other than power consumtion. but that's minimal

[opaque pebble]

Some people like that granularity I guess
I've tried it and I couldn't stick with it (though the dealbreaker was that modeler couldn't handle resource loops)

[candid steppe]

this only gives me 165

[opaque pebble]

Every time I tried to touch my recycled rubber/plastic loop modeler croaks

[unique cypress]

I'm still not sure why you even included it tho. everyone can figure out their own clock speed spread

[unique cypress]

check your tabs

[unique cypress]

because that's what I sent

[wind spade]

the amount of questions "what does it mean 4.25 machines" was very big

also plan was to include other possible spreads

[unique cypress]

sigh...

why are people like this...

[wind spade]

not all people have the same way of thinking

[unique cypress]

idk, to me it's obvious that 4.25 machines means 4.25 times more that one

[wind spade]

well some people don't know of existence of over/underclocking

[plain rivet]

modeler is still nice when you want to split the block of 34.38383838 machines into a block of 10 at one speed a block ot 15 at another speed and a block of 15 at another speed

[unique cypress]

you don't even need clock speed to achieve fractional output

[candid steppe]

yea but 500 water

[wind spade]

I know, but not many people know that and for some that's not acceptable

I see no reason to gatekeep information that the tool can easily show just because you can calculate it manually

[unique cypress]

1. water's basically free
2. nearly all oil has water nearby

[unique cypress]

true, but imo the choice of UC just one machine is kinda weird
I feel like more people UC all equally rather than one

[wind spade]

based on feedback, most people using tools (and giving feedback) indeed UC just one machine

[unique cypress]

weirdos lol. equal UC at least saves some power

[wind spade]

equal UC requires more effort

and is easier to work with when extending manifolds (which I personally wouldn't do, but many people do)

[candid steppe]

the who point was that you just had oil

[unique cypress]

that's wasting tons of oil tho

[opaque pebble]

Also not like you wouldn't have water extractors by the time you get to oil

[mossy ibex]

pure recipes are all good, right? In terms of production efficiency per time and inputs, if you don't care about power or complexity? Are any not so good?

[unique cypress]

pure aluminium has tradeoffs vs default

the other pure recipes are all good if you don't mind refinery spam

[wind spade]

if you mean "good" as "save raw resources", then kinda yeah

[crimson moat]

and sloops, and organisation (tools can only merge every instance of a certain recipe into 1 node, so it can create a huge mess of overlapping lines and ambiguity as scale increases)

[plain rivet]

pure aluminum is sort of a misleading name compared to the others really

[wind spade]

well, it's kinda... pure

[plain rivet]

oh. that doesn't mean anything to me.

i just know all the others are add water at this step and get more. whereas aluminum is take silicon out and get less

[wind spade]

well pure = just given material

so technically none of the other recipes are "pure"

[unique cypress]

i think the default could be just scrap, and the pure would be water, silica and scrap in a blender to get more

[honest ridge]

Should I post a diagram here of the electrical distribution system I'm planning? There's not much in the way of numbers but more just to provide some idea of what I wanna pull off, or is that better sent to the design channel?

[amber umbra]

@honest ridge Here is fine.

[honest ridge]

Basically, all intersite nodes are connected to each other with redundant connections.

Ideally, every zone should be able to hold their own if cut off, or if one grid goes down, the remaining nodes should take the load of the power plant(s) that go down.

If we have a major problem and more than one goes down, central backup takes over while I panic and figure out what happened and try to fix it before the whole system fails.

[amber umbra]

Idk electrical stuff you just all connect together. Then make sure you always have excess power available. So you’re always competing with the simplicity of that style setup.

So you’re well within the “I’m intentionally wanting a fun over engineered setup”.

[honest ridge]

I have an unhealthy habit of overcomplicating things-

[mossy ibex]

you're winning then!

[plain rivet]

add a mod that gives all machines a very small chance to fail so you can see your system in action

never been so excited about a catastrophe

[jovial jacinth]

Does this accomplish something that Priority Power Switches don't cover in terms of emergency power issues?

[honest ridge]

I'm not sure, I haven't gotten that far yet, I'm only Post Tier-4 (I haven't researched anything in Tier 5 yet)

And I haven't messed with priority switches.

[amber umbra]

That, uh, seems like an issue.

[honest ridge]

But then again this goes across the entire map when I inevitably expand to that level

[mossy ibex]

you should probably look into that before proceeding

[amber umbra]

In games like this, there are specific game mechanics and items that the player can interact with in a given game system. For power there’s a given set of tools, mechanics. Anything you’re designing is inherently grounded by those items, mechanics.

So a big picture concept like “I want blackout, brownout protection for my power grid” needs to start with exploring the game tools for that.

[honest ridge]

So my entire idea is just basically made useless by a single priority power switch.

[mossy ibex]

plausibly, haha

[amber umbra]

Idk maybe. But definitely play with small scale tests to see.

[dusky dust]

Yeah, I'm not quite sure if the solution you'd proposed up there is really feasible with the built-in game mechanics

[unique cypress]

I mean I find even priority switches useless. just make sure your production is above consumption and you're fine

[dusky dust]

Priority Power Switches make it pretty trivial to always leave your power generators up and running, even if you exceed capacity on the network as a whole; just gotta make sure that all the power-generation support machinery is sitting behind a PPS which is configured in Level 1

And yeah, the "best" solution for power is to just keep an eye on your power graphs, and if your "max consumption" ever gets to within a third or a quarter of your "production" line, then it's time to build more power. But I understand the desire to build backups. :)

In terms of more "active" backup systems, the usual thing is to set up a bunch of Power Storage devices, charge 'em up, and then disconnect them from the grid altogether. They'll stay charged indefinitely, and can be hooked up to things to jumpstart 'em later if you need be

[honest ridge]

Because I already have the first node building set up with 16 batteries charged and ready.

[dusky dust]

Build enough to power your power-generation machines for a few hours and you'd be able to kickstart systems that way (though you'd want to make sure the Power Storages only get connected to the exact things you're looking to kickstart)

Though putting in PPSes is a good way to ensure that even if you trip your grid, your power-generation machinery itself never goes offline

My own usual advice is to just not play with fire; ensure you've always got a healthy power overhead. That way you'll never end up tripping your grid in the first place. :)

But even then, building redundancies can be fun

[amber umbra]

PPSes?

[burnt hill]

Though I think everyone agrees that Satisfactory is best played however you want to / find fun 😄

[dusky dust]

Priority Power Switches

[honest ridge]

Just trying to also use building the node stuff as a starting point for factory rebuilding because the way I'm set up now I've practically reached the ceiling for how much it's gonna help me, as if I try to go into Tier 5 progression at all at this point, it's gonna get even more messy and unusable

Which means I need to tear literally everything down and rebuild it.

[amber umbra]

@dusky dust Ah. Afaik it would be PPSs for the plural acronym.

[dusky dust]

Nah, I'll stand by es as the proper suffix, whether it's the acronym or not. If the thing you're pluralizing ends with an s, then es is the appropriate suffix, and "switches" itself uses es too. :)

</grammartime>

@honest ridge On the other hand, of course, there's always the possibility of building out things like this, should you be so inclined: #screenshots message :)

[dusky dust]

Nah, never any reason to tear anything down unless you really want to. It's a huge map; you can let your existing stuff continue to do its thing and then build elsewhere

[honest ridge]

God, that looks ready for a horror game (in a good way)

[dusky dust]

You can always add more power generation into your power grid. Spin up some new coal/oil/nuclear/etc whenever you like!

It's definitely tempting to try and get clever with power stuff, but in the end the simplest solution is generally just "build moar." :D

[split sierra]

quick question

lets say i'm planning bauxite refining, default recipes,
if i bottled the byproduct water, as well as input one, and used the new priority mergers

would that work?

[dusky dust]

There's a lot of ways to make it work, but nothing (IMO) is more simple than just using the recycled water in a different set of machines than the "fresh" water

For instance: #satisfactory-experimental message

[versed violet]

Do you plan on having separate power plant per factory/zone?
How about centralized power plant that is split into multiple grids?
Three-phase power system 😉
Because I do have a mod blueprint for that https://satisfactory-calculator.com/en/blueprints/index/index/idU/49577/user/Tomtores

[dusky dust]

Oh, I suppose this is math-and-meta; I can post pics. This is a small vanilla-recipe aluminum setup; 2x refineries taking in fresh water, 1x getting recycled. Probably underclocked as necessary so that the numbers all match up

[unique cypress]

yes, that's exactly what I did in my 1.1 run. But I do not really recommend it, the packagers take up a lot of space

[dusky dust]

re: simplicity, I suppose I must acknowledge that doing it that way does require doing a bit of math to know how to clock the refineries

[split sierra]

i'm planning on processing all the bauxite in the map this time

[dusky dust]

The important part for simplicity's sake is just not actually combining the fresh/recycled pipes. Keep 'em totally separate.

[split sierra]

so i wanted to create modules that capped the overclock at 125 and pipes at 300

[unique cypress]

Power shards are gonna be your friend if you do that lmao

[split sierra]

so whenever i get mk6 belts i can upgrade and fully overclock so my processing capacity goes from 6000 to 12000

[versed violet]

Been there, done that, a lot of fun in red forest. Will you be using the superior electrode recipe?

[split sierra]

(the remaining umpure node will be ignored because the numbers are nicer this way)

[split sierra]

i was thinking about it because it barely uses any oil at the end of the day

at the cost of (relatively) more quartz and slightly less uniform numbers

[unique cypress]

also, if you're doing all bauxite on them map, do not fucking use default recipes lmao

the silica management is gonna be insane

[kindred carbon]

If you like pain you can gather 4800 quartz and use the quartz purification recipe to get lots of quartz for pink diamonds and lots of silica for aluminum

[split sierra]

it's like half the quartz on the map (the numbers will be doubled once i hit endgame)

[kindred carbon]

Use sloppy alumina

[unique cypress]

that's what it looks like with alts

[split sierra]

i was trying to figure out the single array and i realized why you left the canister routing untouched xD

satisfactory modeler does not like 2 priority merger / splitters

[unique cypress]

you can actually connect them normally, but they're just unnecessary lines

[arctic raft]

Hello everyone! I'm struggling to understand fuel a little bit, i currently have 5x5 Refinerys pumping out 200 Fuel per minute, going down to 10 fuel generators that uses up 20 per minute. But they are not working at full capacity, does anyone know how that could be?

My pipes are MK2

my pump is pumping 600 oil per minute with 200%

on a pure node

[kindred carbon]

Is it just one or two at the end?

[arctic raft]

So both sides are seperated, so it's 5 refineries pumping out fuel for 10 fuel generators

5 on each side

[kindred carbon]

I’m talking about the fuel generators

Are the ones turning on and off at the end of the line?

[unique cypress]

are some of the refineries full of fuel?

[arctic raft]

I see where we're getting at.. it's a "length issue" maybe

You are right bookie

it's the end of the lines ones

[kindred carbon]

If there are no yellow lights at the refineries it means the manifold just needs more time

[arctic raft]

But it's weird to me, cause the other side is working 100% with ALL of them being full.

But the one side is struggling

where they are identical

[kindred carbon]

That’s just caused by the manifold

You can overclock one the of the refineries on the side that’s struggling for a couple of minutes to get it going quicker

[arctic raft]

Aaaah i see, yeah the next to last two are slowly filling up

I can do that yeah!

[kindred carbon]

The other side is full probably because you built it first so it had more time to get going

[oblique hollow]

its usually just a matter of waiting

[unique cypress]

or a mistake

did you check what and where the issue is exactly?

[arctic raft]

Yeah it was the end of the line ones that were struggling

[kindred carbon]

Wait there is one refinery that’s not connected to power I think

[arctic raft]

the next to last 2 are filling up now slowly

[unique cypress]

but why?

[arctic raft]

Yeah it's weird to me too, the math should be correct

and all the pipes are connected

[kindred carbon]

Could be a bug with the visualization software but you should double check anyway

[oblique hollow]

why no external flow indicators?

[kindred carbon]

Take a picture of the refinery side too

[unique cypress]

I'm asking you to find the issue. the fact that the generators are struggling doesn't say anything about what the issue is

could be literally anything

[arctic raft]

Yeah i literally can't find the issue, the math should work out, 5 refineries pumping out oil at 40 per minute.. 10 fuel generators eating 20 per minute

405=200
2010=200

so that should be correct, right?

[oblique hollow]

yep. so far so good

thing is: are the refineries all at 100% productivity?
(not clock rate)

[arctic raft]

Hmmmmmmmmmmmmmmmmmmmmmmm

I see, i see what's happening..

Why the heckers is it doing that.

[oblique hollow]

probably oil

[unique cypress]

I still don't because you cut off the i/o buffers

[oblique hollow]

send a bigger image

so we can see the entire refinery menu

[kindred carbon]

Ts so dumb I give up lol

[kindred carbon]

It’s an issue with the oil extractor

[oblique hollow]

probably the oil, yes

[kindred carbon]

Check its clock

[oblique hollow]

probably not getting full 600/min throughput because thats a classic problem

[oblique hollow]

checks out

[kindred carbon]

Oh he is using a 600 pipe

[oblique hollow]

how did you split the oil from the extractor up

[kindred carbon]

You need to elevate the pipe junctions

[unique cypress]

the extractor's at 100% tho

[oblique hollow]

ok but there is still a buildup

[kindred carbon]

The efficiency meter is unreliable right?

[oblique hollow]

it can be at 100% until its full of 200m³ oil

[unique cypress]

could be a constant 10

[oblique hollow]

IF its constant its fine

but if it isnt, thats gonna be an issue

[arctic raft]

Could this be the issue then?

That im not feeding it properly

[unique cypress]

I mean I'm going off the uptime meter

[kindred carbon]

You need to elevate the junction that are connected to the refineries

[arctic raft]

I see i see, do the split above the refineries basically

[oblique hollow]

it could have been running at 595/min for like 10 minutes or so

[kindred carbon]

So take the junction, line it up with the input, press h and nudge it up

It’s best to elevate a junction by 5 meters at least if you’re doing flow very close to max throughput

[oblique hollow]

further: dont use mk 2 pipe if you dont need more than 300/min flow
you split the pipe perfectly in half here, so use mk 1 pipes after the junction

[unique cypress]

why?

[oblique hollow]

junctions prefer it

[arctic raft]

Well i think it's sorting itself out, it's going up

Yup, the last two machines are working 100%. At least i've learnt something. That it takes a while

But thank you guys, this is my first playthrough and the first time im playing with oil, hella fun.

[regal brook]

So i know drones take a long time to unload, do they take long enough that they cant actually fill a belt even if drones are constantly qued to unload? (i am thinking this would be stack size dependent, but then at what threshold would it be impractical?)

[unique cypress]

the takeoff and landing animations take 53s each iirc. so drones can drop off at most 9 stacks every 106 seconds at a single drone port

[regal brook]

so in other words, no, they cant fill a belt

[dusky dust]

Dependent on stack size and belt speed, yeah; I've not mathed it out myself. I'm assuming with a low-stack item and mk6 belts you'd be unlikely to be able to saturate the belt

[unique cypress]

it caps out at 509/min for items stackable to 100. so you can fill a mk4

with stack size 500, you can fill a mk6

[regal brook]

i guess belts across the map it is then

[unique cypress]

trains?

[regal brook]

take just as long to build, if not longer

[unique cypress]

depends how much you want to transport

2 rails can handle trains that can move 100s of 1000s of items per minute

for a single belt, they're not worth it
but for 10, they are

[regal brook]

sure, but i can build any number of belts in the same time

because of course im just going to autoconnect a blueprint and stretch it out max length... which happens to be the same method as building the double rails

more materials, but materials are infinite

[unique cypress]

trains are expandable with little extra effort, doubling belt throughput after the fact requires double the work

[dusky dust]

Yeah, the real strength of rails is that once you have the initial rail laid out, that rail can end up supporting a staggering amount of throughput

[regal brook]

i think the issue here is that i dislike doing more than one material per train

and stations all have to made manually

[unique cypress]

that's what I do too

it minimizes traffic

[regal brook]

which adds a lot of time, when i can just build a dozen or more belts at once and have more throughput than i will ever need

even though i only have t5s, i can just make the bank bigger

[unique cypress]

true, but you have to plan belts beforehand to minimize effort. trains are expandable

[regal brook]

expandable only helps if you anticipate ever expanding though

i'll be done playing long before i need more

still, picked up some useful things here. thanks

[noble timber]

I swear I spend more time making spreadsheets for this game than actually playing it

[wind spade]

you could just use existing tools and save time 😛

[noble timber]

I am a creature of habit 😂

Plus I actually enjoy making sheets so there's that

[amber umbra]

If it works it works

[noble timber]

I have about 7 years work of spreadsheets on my onedrive

Actually 6

[heady vine]

if it's not Eve - it's not enough spreadsheet

[feral dock]

not sure if this is the best channel, but I'm thinking too hard about my blueprints. I've got some tileable blueprints with logistics stored below machines, and I'm debating some details on the logistics routing. I realized quickly that having all the I/O on the same "plane" is painful, as I have to do belt shenanigans to route things around, so I re-did some with an expanded logistics floor (two walls high, probably not enough). Each I/O belt is on it's own plane now, but there's a detail I can't decide on: stack all the I/O in the middle, or split it out? Two Foundry approaches as examples:

centered approach is mirrorable without having to redo the blueprint, and leaves all the negative space at the edges. If I do that consistently I have plenty of room between "lanes" to do stuff. But the center is really crowded. Thoughts?

(this is what I'm doing when I should be building my computer factory)

[vapid gorge]

you could do sushi belts with smart splitters and drastically redue the logistics?

you could even put the outputs back onto the sushi belt to wherever it's going next

[feral dock]

I kinda prefer the "out and back" layout, so I can just add another copy of the blueprint to the end to extend without changing anything else

[vapid gorge]

I'm not sure why you couldn't do that with a sushi belt

[feral dock]

putting both input and output on the same belt would have it going the same direction, unless I'm misunderstanding?

[vapid gorge]

oh sure but that's not a forced thing of sushi belts

you can have a sushi going in warsd and right next to it, on the same plane, an output belt going back

[feral dock]

also I don't have programmable splitters yet, if it matters.

[vapid gorge]

just need smart

there's ... nearly nothing you ever need a prog for

extremely niche

[feral dock]

good to knwo

I think the question behind the question is that my logistics get ugly quickly when I'm routing between different blocks, and the verticality I'm playing with is attempting to mitigate that. I also lean heavily on smart splitters outside between these "blocks" (i.e - saturate this lane, then overflow to the next lane if they both share an input)

[vapid gorge]

part of this sort of planning is making room for 'logistic cores'. Paths designed for logistics

[feral dock]

yeah, I underestimate the space needed

I also feel like I'm at a transition point in this regard. (Phase 3)

[vapid gorge]

for example, I've got several floors of machines like this in one factory and most of the exchanges happen like this

[vapid gorge]

I seriously wouldn't stress too much about it.

imo tiers 1-9 are a tutorial that preps you to do your own sandbox projects.

[feral dock]

fair enough, I think I'm getting insight into how I approach my work (software eng) in this game. 😄

[honest ridge]

If that floor is coated concrete I need that

[vapid gorge]

and this is pre lumen

[feral dock]

thats a nice curve

[vapid gorge]

thanks 😄

[feral dock]

hashtag goals

okay now to spend too much time playing with sushi belts 😄

[vapid gorge]

they are super convenient and arn't very hard. You do end up with shorter manifolds overall ofc but that's easily managed and still a lot less belt space

[feral dock]

yeah I'm primarily concerned with belt capacity

my constructor blueprint already uses Mk4 belts just so I don't have to worry about screws, and I don't have the steel screw alt yet.

[vapid gorge]

I'd make the seet screw akt a priority then

but belt capacity isn't a huge issue, look at it like this

say you had mk5 belts and planned a manufacturer manifold using 1400 total ppm.
you could just use single item belts and do it one long manifold with 4 belts

or you could do two 1/2 length manifolds with 1 belt each.
about 1/4 the total belt usage, same overall machines

[feral dock]

I think this gets back to the logistics between "blocks"

which then allows for a meaningful decision between one long-cat manifold, vs multiple shorter manifolds

maybe the long-cat is my way of avoiding reckoning with my broader logistics

[vapid gorge]

well you do need to plan the numbers a bit more with sushi. but overall I feel it's less work

[feral dock]

and not really more work for the low-level stuff like foundries, where the proportions are easy

[vapid gorge]

yeah foundry sushi can be real good too 😄

I made this a while back.
takes 36.6 ppm to make 10.6 frames pm

comes in at the bottom, works up, then spits it back on the primary belt

a prototype

[feral dock]

hah my 10/m modular frame factory is... much bigger than that. Bolted frame?

[amber umbra]

To the original question, design wise satisfactory is very unconstrained due to the 3D thing. Picking a way to limit, order that design space is really needed.

Picking an item like computers or HMF is a good complexity tier to demo things.

[vapid gorge]

and plate.
it did receive steel beams, plastic and steel ingots though

so there was more production before hand

[feral dock]

ahh, mine takes iron ingots as input

[vapid gorge]

this was pre 1.0 where they had Steel Coated Plates. Miss that recipe so much. Steel and plastic to spit out an unholy number of plates from 1 machine

[amber umbra]

What tools you using to do the ratio math, planning? That can matter a lot for how you plan factory layouts? Planners like Satisfactory Tools naturally give a 2D layout, so keeping things rather 2D in your layout is pretty natural. Like 2D with multiple stacked.

[feral dock]

Yeah, as I'm finishing phase 3 I feel like I'm hitting an inflection point where I'm being pushed to have intermediate factories for things, and now I'm thinking about where to put those factories and how to maximize the resources available.

it's a fun challenge without a clean "correct" answer.

esp with petrol entering the equation

[vapid gorge]

with planning and judicious use of alt recipes you can do things up to super computers in one spot pretty easily

hell, if you want you can just do everything in the east swamp

and basically not worry about logistics and hub factories xD

[feral dock]

yeah, I'm doing quite a bit starting in grass fields, just expanding up the west coast for oil. But that damn coal....

[vapid gorge]

what about it?

[feral dock]

it's not convenient 😄