math-and-meta

[quick gorge]

They are only readable if you yourself have made them.

[bleak wagon]

yeah

[prisma kraken]

i find the graph visualizations pretty awful for figuring out how to build anything

[quick gorge]

I think you lot might be more helpful with asking for suggestions on a build 😅

[bleak wagon]

I like the flow charts because i can see where stuff goes and plan belts out accordingly

[quick gorge]

What can I add to this?

This is going on SAM nodes
I just want to build it out here

[bleak wagon]

it reminds me of a little of a graphics card lmao

looks nice though

[quick gorge]

This miner is mining crypto

SAM is a coin right?

[bleak wagon]

could be

[quick gorge]

I just feel like the miner needs something else

I have been trying to align these signs with the... these whatever they are

[tawdry blade]

Weird community phenomenon.

[quick gorge]

Justification for it being in this channel is that I'm trying to minmax coolness rating on a miner

[prisma kraken]

how about adding a spoiler? people think they make cars look cooler?

[quick gorge]

Okay, added a spoiler

[prisma kraken]

haha

[tawdry blade]

Cyberpunkesque

[quick gorge]

srsly tho I have no idea where to put a spoiler on this thing, it has a pipe coming out where a spoiler would normally go

[thorny root]

Okay I think I'm very clever now. I've devised the rest of my refinery plan to be demand based auto adjusting. I don't know exactly what my final needs will be now and I've determined that what I have left for options are actually fairly straight forward due to the number of refineries required to use the remaining heavy oil residue. And unbeknownst to me, I've already done half the work in over-building my final plastic ring. A huge amount of the potential final fuel usage was going to be on plastic and to fuel that process I needed to build several diluted packaged fuel refineries. Thankfully they're at the end AND my needs for petroleum coke are low, such that I can basically overbuild refineries here too. What gets produced will go directly to the aluminum refinery, and it can only use as much as it needs. 4920 required, and 41 refineries to do it, but my pattern lets me have up to 72 refineries, so I'll build 72. I will wire them separately such that I can flip on the extra bank with a different switch, essentially disabling that outlet for heavy oil residue. This leaves 2 slots in the ring, because at this circumference I can fit 3 refineries side by side. They will be diluted packaged fuel, and they will tie into the previous ring of diluted packaged fuel, allowing me to join their outputs and direct them elsewhere, should I NOT need 25200 plastic. (I won't, so this is a smart plan.)

I will control what gets turned on automatically allowing the production buffer to be used. I will control the plastic production by controlling its usage. This also dictates fuel availability, and rubber availability.

What I don't use in rubber is available to be made into plastic. What I don't use in plastic leaves that much heavy oil residue for fuel / petroleum coke.

All based on what I siphon off.

Knowing all this now, I will give my alt recipe usage another consideration, trying to prefer rubber so I use less plastic so it costs less fuel, so that leaves me more heavy oil residue for petroleum coke, and what is most likely going to end up as turbodiamonds (the fuel)

[thorny root]

At the very end when I "finalize" my refinery production needs, I can make things tidy by underclocking portions or all of the machines in the last two rings such that their combined outputs match my needs. Then they will all be running at 100% efficiency, if underclocked to do so.

[nova steppe]

Hi, question about gasses,

Usually if I need to transfer 600 of a liquid I use 2 mk2 pipes to avoid the float issue. Is that even an issue for gasses however?

Like do I need to worry about anything if I want to carry 600/min nitrogen gas in one mk2 pipe

[frosty owl]

"Float issue" sounds like a misunderstanding. Overall, moving 600/min by pipe is possible but not so easy. Eg: feeding a single row of machines using one pipe fed 600/min from one side will almost surely see the machines at the end of the piping starving (Fluid piling up at producers); of the input was provided from the middle if that row of machines, the system just might work (there's more techniques one can use too)

Sushi Load Balancing when?

[nova steppe]

I understand what the "float issue" is and that is more of a "skill issue". I just wander if the so called "issue" persists with gasses?

[frosty owl]

I think so, yeah

@oblique hollow might be able to confirm

[vapid gorge]

wdym 'float issue'? do you mean back flow?

and yes, you'll often need to loop long gas pipes

[nova steppe]

Okay now im curious about 2 things,
how would I go about fixing backflow?
and how would I loop the gas pipes? Just connect the end to the start of the manifold?

Because I have a line of 5 pipe junctions pointed up into the ceiling going to 5 refineries, with a pump before the manifold starts

[vapid gorge]

show some overhead images of your layout?

[nova steppe]

Currently on my laptop so cant load up the save but this is a good side view:
the square on the right is a pump

[vapid gorge]

well with gases feeding from below probably doesn't cause any issues

[frosty owl]

Feeding from the middle rather than the side is a immediate improvement, even without loops

[vapid gorge]

that might work with only 5 machines being fed

[nova steppe]

This is a liquid setup tho. the gases will be this without the pump ofc

[vapid gorge]

otherwise looping like this , with the split before the first junction generally does well

[nova steppe]

or does the pump help cancel backflow

[frosty owl]

Don't steal from Cobal, you shameless!

[vapid gorge]

nah, pump can help ameliorate certain stabalisation issues but generally won't fix an issue like this w/o a loop

[oblique hollow]

All fluids are affected, head lift or not does not matter

[nova steppe]

Alright I'll apply the loop in my gas setup. But will this setup with Liquids cause any issues? Just pretend there are 5 more refineries along the line

[vapid gorge]

with liquids? yeah feeding from below can be a serious issue if not handled properly

there are more and less stable layouts to for fluids.

feeding from below is one of the hardest to manage

[nova steppe]

What would you recommend?

[oblique hollow]

Spread your supply out

[vapid gorge]

in general? don't feed from below. Feed from the same level as the input

[nova steppe]

I already built most of my factory like this tho 💀

[vapid gorge]

not just with gasses? with liquids too?

are they all 600 pipes?

[oblique hollow]

If you really must feed from below: use a shit ton of pumps, if needed one for every split

And dont feed it from one end. Feed it from at least 2 ends

[frosty owl]

I don't quite see the issue with bottom-fed systems, ngl...
Some people find them counter-intuitive, but that's true for anything pipe-related...

[nova steppe]

Wait that makes so much sence just pump it from both sides so the backflow doesnt go to supply

[vapid gorge]

ehhhh.... ive seen issues with feeding a loop from both ends feeding from below

this is the most stable loop I've developed

sometimes need a pump before the loop starts

[nova steppe]

Like this, for the liquids then

[oblique hollow]

Bottom feeding is a big dependancy chain.
"A must remain full for B to remain full for C to remain full"

Because head lift is only transmitted through full pipes.

Pumps break that dependancy up

[frosty owl]

Isn't that bottom-fed?

[vapid gorge]

yup, most stable bottom feed loop I've made

I've managed 600 pipes with it fairly easily? but I'd still recommend doing less than 600 with bottom feeds just to not have to push things to the limit

[frosty owl]

Isn't that like every other loop? I don't get your point

[nova steppe]

is the loop at the start a one time or replicated for longer manifolds?

[vapid gorge]

one time

[nova steppe]

Yeah none of my liquids are exactly 600

[oblique hollow]

Another tip: Use mk 1 pipes for all connections to machines.
The main pipe can stay mk 2, but the pipes into machines should be mk 1

[nova steppe]

Yep done that already :)

[vapid gorge]

it's very similar to the other loop yes, but it seemed that keeping the feed pipe above the input pipe, and the U bend to the side helped

[frosty owl]

Huh

[vapid gorge]

the previous loop I shared seemed to have more resilience than other systems I've set up in a different layout? I haven't kept meticulous notes on it though

[nova steppe]

Then I have one final question before im gonna work hard instead of hardly work. Are valves one way gates?

[vapid gorge]

... technicallly, yes

[nova steppe]

Like would they stop backflow

in a bottom-feed

[vapid gorge]

technically, no

[nova steppe]

https://tenor.com/view/animation-disney-movies-the-emperors-new-groove-kronk-gif-14836333

[vapid gorge]

fluid hitting the front of a valve can act as a knock back to the fluid behind

so while individual fluid units don't cross over, it can cause flow to go backwards

which is the important part

if you flood a system, use the bottom feed loop, and put a powered pump before the loop splits, it'll probably work at 600.

[nova steppe]

so this would cause issues

[vapid gorge]

generally valves will either do nothing or cause issues. If you're going to put valves on thtere don't put a limiter on them

[nova steppe]

alright, if im correct. Unpowered pumps are way better at acting like 1-way gates

[vapid gorge]

power them. They'll kill headlift otherwise

and it basically costs no power.
but you shouldn't need them within the manifold

[nova steppe]

Just at the start feeding in from both sides preferably

[vapid gorge]

look, can it work from both sides? probably. Not my first go to though

there's a lot of things with pipes that can work.

whether you can get them to work consistently is another matter

[nova steppe]

I'll try this setup with a pump at the start

[vapid gorge]

especially with bottom feeding

[nova steppe]

and see if it works

[vapid gorge]

honestly though, if you have issues with bottom feeding and using loops, I'd split the pipes in two, shorter manifolds with lower flows are inherently more stable

Don't get me wrong, I love bottom feeding

all these run at or near 600, but you have to put a lot of effort into it

[oblique hollow]

With care, valves can be useful.

The main limitation with them is that the input pipe needs to be full and that once you start dictating flow and limiting the directions fluids can flow, you inevitably end up having to micromanage it all

[nova steppe]

Okay one final thing. im dealing with recycling in my aluminum. Both these have recycled water should I just pop down a pump instead of a valve?

[oblique hollow]

So not using too many valves is important

[vapid gorge]

where would you put them and why though, even with the pipes are full?

[oblique hollow]

Do you think the valve even gives you a benefit here?

[nova steppe]

I thought so yeah... Pipe mechanics confuse me

I thought it was just a throughput limiter + 1 way gate

[oblique hollow]

Best case to put them is directly on the pipe leading intonthe machine and limiting it to the exact demand.

As you know, machines consume fluid in big batches. This means the pipe acelerates to max speed for a moment.

A valve only lets through a constant and limited amount

But there it really depends. Does the system suffer from that issue?

in bottom feeding, i'd argue yes

[vapid gorge]

the other main thought on waste water is just don't mix them in the first place. Example alum layouts

these are clockings of a set up using sloppy electrode recipes combos I've done

blue is fresh, red is waste

[nova steppe]

Yes that is a gold-standard I have been following

[vapid gorge]

it's just so reliable

[nova steppe]

I just need a little bit of both wastes to fill up a loose refinery

[vapid gorge]

why?

[oblique hollow]

It is, but it doesnt let fluid flow through unimpeded just until it hits the flow rate limit

It is pressure sensitive.

What that means is that the pipe on the valve input must be full for the valve to output as much as you set the limit for.

If the input pipe is only 50% full, it will only output 50% of the limit

[nova steppe]

It will eventually fill up tho due to the excess amount of water in the input pipe

[vapid gorge]

the alumina? why? just clock a group of refineries to use only fresh, and another group the waste

[nova steppe]

Alright I wanted to visualize it well enough so I cooked up something in draw.io

This is basically what I got rn, it isn't turned on since im still building the factory so I can change the logistics of it pretty easily

[wind spade]

what's the "75/min"?

[nova steppe]

a valve

[wind spade]

I wouldn't recommend using valves

[nova steppe]

what would u recommend then
-# my goat greeny

[wind spade]

though if it works for you, go for it 🤷 valves are often known to be a bit buggy unintuitive though and not really needed

have set of machines that make 150 water and connect it separately

[nova steppe]

So splitting the refineries in the middle to one that outputs 150 and one that outputs 60 that splits to the other pipes

with one way pumps

so no backflow

[wind spade]

you don't really even need to "split to other pipes"

imo make three groups of reinferies (on the left on your image), and each group will go to one group on the right

for example like this

(obviously with proper clocking and maybe different machine counts)

[nova steppe]

Right, since I already have the design layed I might overclock the Aluminum scraps on the top and bottom, and underclock the middle two

thanks again

[wind spade]

np

in general, keeping things separate is an easy way to do things in SF, rather than trying to "split perfect amounts"

[thorn bane]

@frosty owl my newest sushi design
finally able to do what i wanted with priority mergers
principal idea is to have an always flowing sushi belt created by merging 9 lines
those lines get fed by the end of the sushi belt using smart splitters
new items are fed to the sushi belt using priority mergers so the loop always has prio
i added some buffers so the loop always runs but they are mostly empty unless its starting up
a nice thing about it is that the priority merger stops the feeding belt so it can be oversupplied (full belt sushi)

[frosty owl]

This also makes for a neat example to link in convos ^^ (videos are great for sushi, pictures can be confusing)

[thorn bane]

bro i just cant stop looking at flowing sushi belts
i swear its like crack

[silent folio]

My most advanced blueprint. 1.875 Supercomputers all in one blueprint. I did not think it was possible!

[waxen condor]

Don't ask me what this is but i hope it works

[tawdry blade]

Sushimixer I guess

[crimson moat]

They're known by many (incorrectly) as being buggy, because of stuff that was broken in alpha years ago. Some people kept repeating the old info without testing to see if it was still valid, so it persisted as a legend of misinformation long beyond its time

[pastel obsidian]

I still feel the same way about pipe floor holes sadly

[wind spade]

my info is based on up-to-date testing from 0.8/1.0

yes, there has been some improvement in 1.0, which still doesn't change the fact that valves are not recommended due to them being pointless in majority of conventional builds

and if you have a thing that either does nothing relevant or in some cases breaks systems, then it's kinda pointless to build that thing 🤷

[crimson moat]

The specific claim that i am debunking is

** "valves are often known to be a bit buggy though" **

It is factually incorrect, and to the best of my knowledge hasn't been accurate since alpha

[wind zinc]

last time i checked the only problem is if the input pipe isnt full
then it wont work properly

[wind spade]

if pipe is full, why do you need a valve? the system will naturally take only what it needs

[crimson moat]

Not really a bug so much as an intentional but possibly misunderstood mode of operation / poor communication

For limiting flow they are % limiters, not absolute value limiters. That behavior has some desirable and some undesirable impacts

[wind zinc]

thats cool cuz the game's UI never tells you that it goes off of %

[crimson moat]

Yeah, it should (or you should be able to choose)

[wind spade]

well I guess the "buggy" there stands for "not working as you expect" rather than "not working as intended"
true, bad choice of wording, but it's usually enough to convey the message of not needing valves

[deft lichen]

buggy implies they aren't working as intended by the developers, but they are
"unintuitive" would fit better

[wind spade]

if someone asks for clearer explanation, I post this: #math-and-meta message

which should be correct, apart from the 255 possible values, which (not yet officially confirmed) was fixed in 1.0/1.1, not sure which

[crimson moat]

Was fixed at or before launch of the game

[wind spade]

and I thought you know that I don't really think that valves are bugged in the sense of "not working as intended", given that I'm one of the people that often say how fluid system behaves as intended

[wind spade]

yeah well that message is from 0.8

[crimson moat]

I know, just don't want newbies to get the wrong impression from stuff that is being said here today

[wind spade]

I may update the old message to clarify the fix in 1.0

[deft lichen]

quasi-confirmed by McGalleon but a definitive source would be great

[wind spade]

though afaik it still wasn't officially confirmed?

all we have is file headers, where the property type changed, but we don't know if it's handled properly in code?

(unless there have been new info on this which I'm not aware about)

[crimson moat]

stuff generally doesn't get officially confirmed or unconfirmed, we just know that it's fixed because we tested it and it works as expected, contrary to the old wiki documentation

the test case on their explicitly does work now, when the wiki used it as an example of something which didn't work

[wind spade]

well many bug fixes get at least mentioned in patch notes (and I'd say that only having 255 values would be considered a bug)

[crimson moat]

Yeah, just not everything

[wind spade]

fixed the message above, now gonna fix the old message so that it no longer lists 255 values

also fixed

should now adhere to your standards 😛

[fierce ruin]

Could someone help me make a 6 to 13 belt balancer

[wind spade]

first question for any balancer is: "why do you need it?"

[thorn bane]

split 6 to 12
overflow splitter all 6 to 13th

[fierce ruin]

5*1200+228 to 13 400-500 belts

[wind spade]

route each belt to machines that need the amount that is on the belt

go the easy way, don't complicate it for no reason 😛

[fierce ruin]

I don't really understand what you mean

[wind spade]

if you have 1200 on a belt, build machines that need 1200 in total, and hook that belt to those machines (use a manifold)

repeat that for every belt and you've solved the problem 🙂

[fierce ruin]

I have already built the machines in systems that use 400-500/min

[wind spade]

well that's kinda what you shouldn't have done 🤷

but anyway, just hook the belt to one set of machines, overflow to another set, etc etc. Add more resources if the belt runs out. No need for a balancer 🙂

[fierce ruin]

That makes sense, i understand now. Thank you

[restive sparrow]

I really need to get into doing this more (clocking and sectioning to match production/consumption rates).

Not doing that is what creates the spaghet I always end up with lol

[wind spade]

clock speed is most powerful logistic tool we have

[restive sparrow]

I always tend to make big manifolds for all of the machine's making material X which doesn't lend itself well to this style. I'll have to do some thinking about modifying my approach for sure.

[wind spade]

yeah that then leads to "how do I make 47:223 balancer" 😄

[thorn bane]

dont actually need to underlock them for this to work

[wind spade]

no, you can priority merge/split, ofc.

but imo clock speed groups are better, as your builds become more independent 😛

[frosty owl]

I know, right?!

[thorn bane]

no i mean do the exact thing you are talking about
just dont underlock

machine will just balance based on output or input

[wind spade]

sure but then you'll have yellow lights

[frosty owl]

It comes with pros and cons

[thorn bane]

yup
but i wouldnt call not having yellow lights "our most logistic tool"

[wind spade]

not having yellow lights is great debugging tool

if you build things such that yellow lights should never happen, then when you see yellow light, you know there's a problem in your build

[restive sparrow]

The one place I used the blocking/clocking technique and loved it was with a blueprint of 8 refineries piped and clocked for the sloppy aluminum/electrode scrap recycle loop.

[thorn bane]

so do you if its not on the last machine

[frosty owl]

Not having yellow lights can be quite the chore in cablework though, while allowing for yellow lights also allows to rely on backpressure to regulate factories (something some might appreciate)

[wind spade]

or second to last or other if they have uneven clock speeds or recipes, or when you have mix of multiple machine types, or manifolds from different sides, or ...

you'd have to check on case by case basis

[restive sparrow]

Greeny, I did have a question for you. For systems that rely on a feedback loop like aluminum water recycling, what can you do to stabilize a system that is oscillating shortages? I feel like the classic "underclock and let it fill" approach doesn't quite work because of the feedback loop.

[wind spade]

separate machines that handle the byproduct looped back

[restive sparrow]

I'm using separate machines for recycled and fresh.

[wind spade]

fresh product goes to one set of machines, looped back byproduct goes into a different set of machines

if everything is clocked correctly, it will fill after some time and work just fine

[restive sparrow]

Nothing you can do to pre-fill machines like the underclocking strategy used elsewhere I'm assuming?

pre-fill the pipes I mean.

[wind spade]

you can prefill them in any way tbh

because if it's separate, it shouldn't be possible to deadlock

[restive sparrow]

I just remembered that I'm slightly underfeeding Bauxite so that may be the only issue. I'll readdress once I get the bauxite supply up.

It just generally got me onto the idea of the underclocking to prefill pipes doesn't really work in a feedback loop.

[thorn bane]

use a smart splitter to prioritize recycled water and limit bauxite like so (480+120 = 600)
this way it works even with empty pipes

[wind spade]

why the bauxite limitation?

[thorn bane]

so the bauxite goes to the recycled refineries first and not to the fresh refineries, via the smart splitter

this way the bauxite regulates how much fresh water is added to the system
which is gonna be all if its just starting up and the pipes are empty

[wind spade]

so, you're providing more bauxite than the system needs?

[thorn bane]

no

[wind spade]

then why should it matter? if system needs 480 bauxite, why is the limitation needed?

[thorn bane]

because the machines arent underclocked
if they were the whole design wouldnt work

[wind spade]

if they were, then the design I suggested would work 😛

[thorn bane]

well yeah this is just that but better since it can handle too little or too much water

[wind spade]

and why my suggestion wouldn't?

(not to mention that water input will be fixed anyway, it's coming from extractors)

[thorn bane]

hm i guess both handle too little water

[torn plaza]

i temporarily sloop machines to fill buffers of later machines

[wind spade]

yeah, and "too much water" will be handled by machine just not accepting that much (which isn't a problem for fresh water, and won't really happen for byproduct water, as there's not much possibility to make "too much")

[thorn bane]

well i dont trust it to not happen in the byproduct loop
and this only costs a smart splitter so 🤷

[restive sparrow]

Oh that's clever!

[thorn bane]

just like i dont trust mixing recycled and fresh water

[wind spade]

if it happens in byproduct loop, it doesn't matter either - because the machines will process it

the machines are designed to consume the amount of water that the setup can produce, if for some reason you produce more than that, then you've built it wrong 😄

[thorn bane]

but it gets refilled
the amount will stay constant

[wind spade]

again, if you somehow make more water than your max, there's something wrong in your setup 😛

[thorn bane]

unless its too much and some refineries shut off

[wind spade]

wdym "too much"

how can you make more water than the max you can make? 😄

[thorn bane]

idk stuff happens

[wind spade]

apart from new ingame bug, I can't see that happening

[waxen condor]

it should hopefully make four belts with same numbers to 6 equal outputs with also a overflow Belt

[thorn bane]

for the same reason that you dont mix recycled and fresh water

[wind spade]

again, how could you make more than the max possible amount you can make?

[thorn bane]

bugs, outputs fill, weird fluid dynamics blocking stuff

[wind spade]

bugs can affect your system the same way as mine, that's not an argument
outputs full stop the cycle, but don't deadlock it
weird fluid dynamics could brick your system as well 🤷

[thorn bane]

well no since mine is water negative if theres too much in the system
im actually not sure about outputs filling

[wind spade]

"is negative if there's too much"
yeah, assuming no weird bugs... same as mine

obviously for any system you can't account for every possible bug, so the system should be stable in a stable version of the game

full output of some alu product just stops the production, which is fine, since that means no more production of water

[thorn bane]

idk its the same reason to not mix fluids
just less chances to go wrong

[wind spade]

but mixing fluids can actually go wrong in a stable version of the game

[thorn bane]

well the machine will still buffer 500 scrap which is enough to fill the pipes on small setups

a tank might fix that

[thorn bane]

how

[wind spade]

imagine you stop production of everything, fresh water fills the pipes fully, byproduct water has nowhere to go

[thorn bane]

well just overflow sink

[crimson moat]

Smart splitter. Feed everything to the recyclers first, then overflow to the fresh water refineries. It will throttle gracefully.

If you allocate equally or the other way around, it won't

[wind spade]

the water? waste of packages

[torn plaza]

or concrete

[thorn bane]

no the alu

[wind spade]

the "stopped production" can be any reason. Truck bringing bauxite stuck behind something, accidentally dismantled wire for bauxite mine, train delayed or something...

doesn't necessary mean "output full", can mean "input empty"

[restive sparrow]

I ended up in this scenario by pure happenstance. The fresh water machine is at the end of the manifold. It doesn't deadlock, it just oscillates yellows around the recycle loop. I guess this could be considered graceful throttling. I'll only worry if this happens when I'm providing enough bauxite.

[crimson moat]

Yup

[vapid gorge]

it's much simpler to just underclock some consumers until it's flooded.
slooping can also create extra by product in some situations and slooping a pipe or belt that's already full just clogs things up.

[restive sparrow]

That gets complicated if the consumers are also the generators that feed the machine you're trying to fill. That's what I was talking about.

IE: Trying to fill the recycled water into those machines that use recycled water and trying to fill the alumina solution in the machines that create the recycled water.

[vast inlet]

is this working right?

every time i check on it it says something different under the power consumption

[waxen condor]

it depends on the amount of outputs you put down ( i think but i am not the SF God so i don't know )

[thorn bane]

really? wiki disagrees

[waxen condor]

I think but i am not the SF God so i don't know ;D

that's the only possible reason for the power cons. to lower or go higher ecapt for de-/increasing the overclocking

[vast inlet]

sat for dunno how long watching to see if the light turns yellow and dont think it has... but maybe my eyes are bad

[thorn bane]

power also checks out 0.8^1.321928 * 150=111.7

[thorn bane]

what exactly is it saying? at 80% it should be 111.7

[vast inlet]

the picture is what it shows

[thorn bane]

well you said it says different numbers?
are you changing the clockspeed?

[vast inlet]

the little discription under the power changes... working sorta what u mean sorta ?

[thorn bane]

ooooh
dont think that actually means anything
its just flavor text afaik

[vast inlet]

ok ty that shhit been bothing me just wanna know u working or not 😅

[waxen condor]

I can't anymore! This is Killing me!

[prisma kraken]

you know, after you smelt the scrap, it would take half as many belts?

[pastel obsidian]

Faster belts would help too

Something nice about seeing so many times moving all at once

[slow crater]

So Im building baby's first rotor factory and hitting some questions. I get the impression the 'best' rotor recipe is steel rotor, but what about for the copper wire? I like the output rate of default wire recipe a lot better but read somewhere that copper is rarer and should be used a bit more scarcely. I've got copper rotor and iron wire for reference.

[wind spade]

there's no "best" recipe, and in fact steel rotor is kinda expensive

as for copper, there's more than enough copper

[pastel obsidian]

Personally use what you have available, unless you plan to use every resource on the map. it's not really a problem most people will face

[thorny root]

I've decided they're water refineries from now on. That's what it uses most of. Not oil, or anything else. Water.

[slow crater]

Yeah thats kinda why I put best in quotes. I'm just being particularly caution about resource use cause im building off a main bus

[wind spade]

(I wouldn't recommend a main bus)

[pastel obsidian]

Oh nice, you can always belt in more resources

The cost of placing belts is a one time cost

[slow crater]

Yeah thats the plan. I guess Ill just go with the recipe I prefer for now and worry about problems later XD

[pastel obsidian]

A bus is really good at understanding how all the different productions work together and is flexible

[slow crater]

I just think it looks neat

[quick gorge]

I am a bot, and this action was performed automatically, please contact Nat.co@ficsit if you have any questions or concerns

[slow crater]

good bot

[quick gorge]

[vapid gorge]

1 generators are the final step - should not be feeding anything

2 underclock the final step, everything behind it will get flooded

[thorny root]

So... the wire size for the machines stops making sense at around 20MW of usage. Assuming 3 phase and 13.8kv, the conductor to insulation ratio, and the apparent size of the wire bundle (0.1m)

At higher voltages, there isn't an insulation in the world that will keep that from shorting.

So if we trust the power readout and the wire size we have to make a few bold assumptions about how the grid works... Like how many conductors are in use, the addition of an assumed invisible perfect ground, and possibly even advanced power transmission techniques using EM field harnessing.

Grok did the math.

Yet again my theorycrafting rides the line between math and meta and design and architecture... Forever doomed to not fully belong to either =/

[restive sparrow]

All 6 of the machines on the bottom of this graph are both consumers and generators in a loop. Underclock the scrap machines and you will starve the sloppy machines, not fill them. This is the problem I've been talking about. I don't see how underclocking anything in this system helps fill the machines. I can see how slooping one or two of them temporarily might. Maybe I'm missing something fundamental here.

Ultimately it's a system that should probably work itself out without any underclocking shenanigans, given enough time, so I probably shouldn't be worried about it.

[vapid gorge]

oh you don't want to flood this system at all

you fill the first one with fresh water and let it go

don't try flooding it

[restive sparrow]

This is the conclusion I was coming to. Thanks for confirmation.

[vapid gorge]

By product cycling like this I think is the fluid systems you don’t want to flood.

Besides maybe the primary input. The fresh water in this case

[fallow siren]

separating fresh and byproduct water into their own system proven to be unbreakable system

and its easier to do with clocking

[wintry jewel]

does anyone know how much fuel drones consume per km?
im asking so i can see how much extra plutonium fuel rods i need to produce for the drones

[frosty owl]

The Drone ports tell you the fuel consumption while in operation. You can use that to compare

[upbeat summit]

want to produce rubber and power as efficiently i could, with these available recipies, currently have 240 m3 crude oil available

can overclock the oil extractors tho

any suggestions as welcome

[vapid gorge]

you've got to prioritize rubber or power, can't maximise both

in general my recommendation is never mix power and production honestly.

go find a more distant oil node, make power there

[upbeat summit]

hmmm i currently have 5.8MW of power

[vapid gorge]

easy to import distant power

[upbeat summit]

so i think that might be enough to carry me through phase 3

hopefully

[vapid gorge]

to phase 3? yes.
through? maybe not

is 5.8 what you have spare, or total?

[upbeat summit]

this is my current setup that i will try to impliment

5.8 in total

[vapid gorge]

well that is 2.4. do you ahve that spare? plus whatever pumps, train, other vehicles needs?

that also doesn't inclue any over clocking, if you like over clocking add 1/3 more

[upbeat summit]

i'm currently building my setup as this location

so maybe i could use this area for production and focus on power in that other forest area that has oil

[vapid gorge]

sounds solid

[upbeat summit]

my initial plan was to just generate enough power to help me get through phase 3, and then when i unlock rocket fuel / etc, i could have a proper solid plan on efficiently generting power

[vapid gorge]

eh, what you could do now is find the diluted packaged fuel recipe and make a nice enough power plant to get you to nuclear power

much better deal than Rf

[upbeat summit]

more hard drive hunting yaaay

[vapid gorge]

worth it. heavy oil residue + diluted fuel recipes is a massive power output

plus it's much more oil efficient to make rubber and plastic that way too

goes from 3 oil to 2 plastic to like... 3:9?

[upbeat summit]

you're referring to this thing?

[vapid gorge]

theres 2 recipes that come at different points in the game. Both do the trick

[vapid gorge]

easiest way to use it is to have a packager -> refinery -> unpackager then loop it directly back

[upbeat summit]

ok

[opaque quartz]

It depends on the distance but ime a single plutonium rod will power a drone for like 100 round trips. My current nuke plant makes 3.2 plutonium rods per minute and I’m only burning 3 of them in NPPs. The last .2/min is for drones and whatever is left after that gets sunk

[fathom warren]

What is your opinion on using quartz and oil in electronics production (silicon high speed connectors, plastic ai limiters, isulated crystal oscilators and crystal computers) to save caterium and other resources?

[wind spade]

if you want to save caterium and have quartz, why not?

all recipes are good if you like them

[fathom warren]

is caterium usefull later in the game, or should I rather save the quartz and oil?

[wind spade]

practically you won't run on anything unless you do some crazy stuff

so just use whatever you have on hand and want to use

usefulness of raw resources heavily depends on used recipes anyway

[prisma kraken]

it really depends on the mix of recipes you use in the phase 4 components to determine which of the 'rare' resources become tight. the way i typically build, i find that i usually end up with a surplus of caterium and short on quartz

[fathom warren]

so it would be better to rather use caterium?

[prisma kraken]

at the end of the day though, aluminum always ends up being the limiting factor and everything else kind of gets limited because you are horse-trading to maximize aluminum

unless you're goal of production is really lofty, you won't run out of any resources

by lofty i mean that a 60/min delivery goal of the phase 4 parts stretches the map limits with a 'max nuclear' build along side of it

you just won't end up coming close to map limits unless your target is that high

something else i'll just make note of is that in phase5, you gain the ability to convert ores into other types of ores at the expense of SAM, caterium can be kind of useful for that

[wind spade]

use whatever you want 🤷

[prisma kraken]

i think more of what you should be thinking about is the time it takes to build the infrastructure to marshall resources to where you need them

[fathom warren]

What is the objectivly best way to make aluminium

[prisma kraken]

resource distribution on the map, while a little bit lumpy is pretty even; you'll find a good deal of each resource in most biomes. it is ok to say 'i'm dedicating the X in this biome to making something'. If you just set aside one of the starting biomes (dune desert, rocky desert, grass fields, north forest) as a do-not-touch-until-phase-5 sort of area resource-wise, you'll be find

there's 2: sloppy+electro or instant scrap both methods refined with pure alum ingot

both methods give a 1:1 ore->ingot conversion

you can do better than that if you add silica to make the ingots with the default recipe

[fathom warren]

alumina solutin isn´t good because of smaller amount of ingots made?

[prisma kraken]

the problem is that there isn't enough silica available to do that with all the bauxite in the world

sloppy alumina just makes more solution without silica byproduct. the default recipe isn't very favored because the byproduct, while able to be used in the ingot step is a pain to handle and doesn't increase yield as much as you'd get with sloppy+electro

i don't have the numbers handy, but iirc, the conversion rate with that ends up being 5:6 if you use every trick available

[fathom warren]

Why not? There is 13500 silica availible

[wind spade]

none, because there's no way to "objectively" analyse recipe paths, without specifying what you want to optimise for

[prisma kraken]

what i generally do is plan on sloppy+electro for all the baux in the world and then if convenient add some silica to boost the ingot amount for a few prod lines

a lot of quartz for me goes to oscillators & hsc's

having built it in a lot of different ways in different playthroughs, i'll just say that logistically getting silica to bauxite is a PITA

[fathom warren]

so, the smaller maount of aluminium ingots is worth not having to deal with silica?

[wind spade]

that's for you to decide

some people like to not deal with silica, some like to make most from their bauxite 🤷

[torn plaza]

very few people bother importing silica to make more aluminium

sloppy/pure is a well loved combo

[wind spade]

tbh most people I've talked to do import silica

[prisma kraken]

i think the better way to go is to use pure ingot until you need the additional yield

[wind spade]

well, to each their own 🤷

[prisma kraken]

fyi, i just found that you can place a good number of water extractors in this stream. might be kind of useful for someone processing the canyon oil:

[somber folio]

What are the numbers for the max yield nuclear setup again?

[vapid gorge]

Quartz/Caterium

there's chains of recipes to make everything and nearly avoiding using any crystal in the whole game

same thing exists for caterium. You could do a massive factory that makes everything and use little of it.

it's just up to you

Alumina - like everything else with recipes there's no single objective best. Unless you define parameters for 'best'

what's most popular though? probably sloppy + electrode. Since most people do pure ingots so as to not have to haul loads of silica around, and you can move a bit of oil instead of a lot of coal.
but coal is generally closer to bauxite in the first place. So like greeny mentioned, you ahve to decide if the trade offs are good for you

[oblique hollow]

Huh.... I didnt know Ficsonium can be deleted with your trash can, but it cannot be sunk

[thorn bane]

using 320 quartz to safe 150 bauxite just isnt worth it

[versed violet]

unless you maxed all the nodes and still want MOAR alu, then yes.

[frail wadi]

Is there a chain of recipes to make Fuel out of Heavy Oil Residue with a 1 to 1 ratio? I'm failing to find one right now and yes I know Diluted Fuel exists and is actually the best Fuel recipe.

[versed violet]

there is diluted fuel (blenders) and diluted packaged fuel (refinery). depending what tier you are on.

[vapid gorge]

!wikisearch fuel

[brisk shore]

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

[vapid gorge]

these are your options

[thorn bane]

i mean residual is 3:2 diluted is 3:6 so a mix works
but like why

[vapid gorge]

if you used the HOR from making plastic, then make diluted fuel, you'd have 3 oil to 2.666666 fuel?

[deft lichen]

...why doesn't the "as byproduct" label have a border...

[hushed trellis]

sorry i eated the border

[vapid gorge]

should it?

[frail wadi]

the reason why is simple. I wanted something similiar to this just for recycled plastic without the recycled rubber loop

[vapid gorge]

When I make plastic, I sink it immediately. I just use the HOR

I, somehow, might hate that planner more than modeler

also https://www.satisfactorytools.com/1.0/production?share=rQvCyeahnXFSfriuiked

[deft lichen]

there (missing CSS variable, sorry to interrupt)

[vapid gorge]

ah here it is w/o the loop https://www.satisfactorytools.com/1.0/production?share=weyXlOdLPer0RUIlh9Vd

[thorn bane]

ah i see
hm ye dont think thats easily doable without weird mixing of diluted and residual
btw you can do this
its more complicated but doesnt actually waste fuel, if you dont wanna do the loop

[restive sparrow]

I'm running into issues trying to get materials delivered to an from particular platforms. I have suppliers that have stations with many platforms of the same material. Then at the consumption end I have a station with several platforms expecting to accept one material each, of different materials.

In a situation like this how can I make the train ony pickup/dropoff to platform 3 as an example? I don't want to have empty platforms because I might need a different train to come through and deliver from/to platform 1 or 2 or whatever.

Is it typical to make a train that has engine cars as placeholders where you don't want to load/unload(The train equivalent to an empty platform I guess???), or is there a better solution?

A more specific example: I have an aluminum casings factory that is loading up 4 platforms with casings. I have a computer factory that has a station expecting casings on platform 3. I have a nuclear plant station expecting casings on platform 2. I was imagining this as a simple loop of pickup, dropoff at computers, dropoff at nuke site, but this is getting casings in too many platforms at the computer factory.

I'm thinking the only solution (using the current number of stations, because I don't have a ton of room for more) would be to split this up into 2 trains. One going to computers that is Engine, Engine, Engine, Freight and one going to nuke plant that is Engine, Engine, Freight.

I think this would probably work, but it "feels" dumb.

[frosty owl]

I think this is the most common solution. Anything else would probably require a lot of tinkering, if at all possible
BTW, you can also use Fluid cars
( @restive sparrow as I forgot to reply and some time has passed)

[restive sparrow]

Fluid Cars? I thought there was only 1 kind of freight car and the difference was only in platforms? (I'm at work so I can't check)

[near smelt]

is there a recipe that allows me to make heavy modular frames using just from iron and concrete?

cause i've got like 1600 iron and 900 limestone per minute near a factory im not using and i still need to automate those

[wind spade]

You can put those in some online tool and see what it tells you

[undone vessel]

pretty sure there's a recipe that uses concrete encased beams frames and pipes just use iron pipe and encased industrial pipe

[wind spade]

Though generally I'd suggest picking a location based on what you need, rather than trying to force usage of some nodes you have nearby

[ashen stirrup]

https://www.satisfactorytools.com/1.0/production?share=Bg3OFLMkuIwvB0rUyU5s

You can also disable water under "Items, Input" if you don't want to deal with that

[restive sparrow]

I'm also curious what toplogy changes you'd recommend making (at least for better future builds). I currently have a bunch of sources of materials with one single station per material. Naturally the consumer locations end up needing a ton of different materials so this results in cramming different materials into the same station (cause who has room for 12 stations at one location?).

Should I be splitting things up differently? Organizing my source stations differently? Being ok with sushi mixing stuff at my consumer stations? I'd consider having mixed platforms and sorting with smart splitters but I'd be concerned that overproduction of one material would render a different material incapable of making an unload.

[bleak wagon]

is it possible to make a belt compressor that goes from 13 belts to 10 full and 1 overflow? Im trying to do it rn and I feel like i keep doing something wrong

oh actually i think i figured out what i did wrong

nvm im still confused

[ashen stirrup]

It is mathematically possible

but a real pain

[bleak wagon]

yeah I can tell

[ashen stirrup]

It will be much easier in 1.1 fwiw

[bleak wagon]

Im on 1.1 atm

[ashen stirrup]

I think you just use priority splitters and priority mergers then

[river night]

a waterfall merging scheme with priority mergers should be able to do it, but its still going to be ugly

[bleak wagon]

yeah I found a post talking about it, and priority mergers seem to be necessary

[bleak wagon]

I have a lot of area under the floor

[ashen stirrup]

Well, strictly speaking, they aren't necessary

but they make it a lot easier

[wind spade]

Just connect each belt to the machines that will eat that much

[bleak wagon]

No. And my stated reasoning is going to be "because I want to"

[bleak wagon]

I figured out how to do it and it doesnt look bad either

I couldve made it with no clipping, but I value it being small

as far as small goes for this monstrosity

[near smelt]

load balancer?

[bleak wagon]

its a belt compressor that makes 13 belts into 10 full and 1 overflow, Im not sure it works for all cases, but i know it works for my specific case right now and I doubt im ever going to need this again

testing it for the first time now

its actually working first try

[tulip kraken]

Hey there. Im working on a Big Coal Powerplant. I need 36 Water Extractors for everything. I Know that I dont need to worry about headlift if i build a watertower thats higer then every other pipes. My question is, How do i do that? can i Conect all 36 Water Extractors to a single pipe that gose up the watertower? do i need to split the 36 Extractors into more Pipes? I never made a Watertower but i need Like a million Pipeline pumps if i dont use one. Can someone tell me how i can do this?

[fallow siren]

mk1 pipe can only hold 300/min water, 600/min with mk2

and no, dont use water tower

for decorations? sure
for practical use? just dont

pump is easier, and it barely use any power

[tulip kraken]

Oh okay? i thougt they were a good way to get rid of pumpes

[fallow siren]

heres my small coal plant pipe system

[restive sparrow]

Water towers are generally planned as being higher than where you're building so you'd likely need more pumps, just in a different locale.

[fallow siren]

simple solution? build them at water level so you dont need to worry about headlifts

this place is northen west of rocky desert, one of the perfect location for coal plants

[amber umbra]

The sensible solution is to build the coal generators on the same height as the water extractors letting you omit pumps. Or build the coal gens above the water extractors requiring one pump per pipeline. Generally the pipe, pump managing is much easier if you build on discrete foundation levels.

[fallow siren]

crater lake also a nice choice with 3 pure coal nodes and plenty of flat space if you chop all the trees

but personally i wont touch that area until i unlocked at least mk4 belts

[prisma kraken]

'small'

also, i tend to agree that building coal at water level is a really good way to go about it. it just eliminates one possible source of problem in something that you want 100% reliable

[prisma kraken]

i did my intro plant at the same spot

you know, it would be nice if there were a 'chain delete' function that deleted everything touching what you have selected

[fallow siren]

multiple filter for mass dismantle would be good

[prisma kraken]

i'm thinking specifically for deleting a big slab of foundation, but probably could be much more powerful of a concept

[fallow siren]

i think theres a mod that has area dismantle, which let you do dismanlte certain area in once without having to hover to them one by one

[prisma kraken]

i'm sure there is

bp delete does wonders though already

[radiant lance]

hi yall. it's time for the daily valve mention. after seeing this post (https://www.reddit.com/r/SatisfactoryGame/comments/1k7t6kq/valves_lie/) I'm attempting to reproduce this person's valve testing on experimental branch, and I can't reproduce their signed byte issues

following their reproduction steps, I have a setup of two rocket fuel unpackagers feeding another two packagers. the two unpackagers are set to produce a total of 60/m rocket fuel, and the packagers use the same total. the valve is set at 60, which should mean that the actual flow rate would be 25*(300/127) = 59.055118, but I'm letting this run and it's not giving me any stalling whatsoever

after several minutes, all number ranges for the packagers inputting fluid are the exact same, but with valve inaccuracy, they should've gone down by .95/min

[prisma kraken]

without digging into it all and understanding exactly what the test is and what you are trying to reproduce, valves are notoriously inexact and if you are trying to replicate a precision issue with their rate limiting, often small decimal inaccuracies take an hour or two to accumulate to the point where they're measurable

[radiant lance]

if the OP's testing is correct (which I don't doubt it was in the past, but I'm unsure about experimental), a valve saying 60/m should only be supplying 59.05/m to the machines, which take 60, to my knowledge

which is such a significant difference that it should be visible in only a few minutes, since it would be missing 1m3 of fluid per minute

[prisma kraken]

out of curiosity, when was the post written?

[radiant lance]

10 hours ago

[prisma kraken]

i thought perhaps it was posted before 1.0, which i think had a small fix-up to valves, but idk. I know better than to care. valves for me are a decorative piece

[radiant lance]

yeah, I'm just going a little bit insane because I don't think anybody properly knows if valves are accurate right now or not

[fallow siren]

theres really no use for valves unless youre going for a very specific pipe system

[radiant lance]

if anybody on stable branch can reproduce the setup I posted here, I'd be very very interested in the results

I can only assume the OP's on stable, but I can't reproduce their inaccuracy issues at all

[prisma kraken]

the last real treatment of the subject that I read, which may be outdated, but sounds as if it is still correct is that the valve flow adjuster is an 8 bit integer, so the flow can only be adjusted in increments of 300 or 600 divided by 256

[radiant lance]

yeah at this point I just want to know out of pure curiousity, it doesn't fix valves being F tier

[prisma kraken]

if you absolutely need to control the flow rate for 'reasons', i'd suggest packaging up the pipe's full volume, splitting it exactly on belts and then unpackaging what's needed for each destination

i think about the only time i've resorted to that is when making use of the game's water wells which are pretty icky to use, lol

[fallow siren]

a lot of work, but can actually be a nice challenge for fun

[radiant lance]

I saw a post about packaging aluminium waste water to use prio mergers for water recycling and it's such a funny idea that I need to do it

[prisma kraken]

i needed to split the water from the well near the nitro god well between aluminum's fresh water and a pure copper line, and make use of ALL of the water available, if you could suggest a better way of doing that, i'm all ears 😄

it honestly ended up being the least problematic thing in that factory complex, so yeah, extra work, but eliminated a source of problems

the other option was to drag some cross country pipes from the titan forest river through the large trees and under my train rails, and that's why i did what i did

[radiant lance]

redoing testing with mk.2 pipes, packagers that both produce and eat at 120/min, and a valve at 120 (which with inaccuracy should result in 50*(300/127) = 118.1/min flow)
and results are the same: after several minutes of testing, all number ranges in packagers are the same (one of the packagers starts at 23.7 fuel in it and finishes with 25.6), despite the valve claiming it's only flowing at 118.1/min, and after initial stabilisation, a power pole hooked up to all 4 packagers shows zero idling

setting the valve manually to 118.1 has visible starvation happening after only seconds

[vapid gorge]

why not just clock the packagers? machine consumption rate is a more reliable 'valve' than valves

[prisma kraken]

i think he's just having fun experimenting for science's sake 🙂

[radiant lance]

yeah, I don't care at all about actual factory usage of this right now

[opaque quartz]

Jeff Goldblum is over in the corner mumbling about whether we could or whether we should

[radiant lance]

I just want definitive proof of valves being broken or not

[radiant lance]

(after setting the valve back to 120, there has been no more starvation in the machines)

[vapid gorge]

valves aren't broken, they just have mechanics that aren't readily visible and their functions are less reliable in most cases than other options

[radiant lance]

by broken I specifically mean the common knowledge that their flow limiting is inaccurate and only has 128 possible values

in my testing both now and a couple weeks ago, valves in experimental branch are not showing any signs of this inaccuracy

[vapid gorge]

iirc I heard that was resolved some time ago

[radiant lance]

to my knowledge that conversation and any info stemming from it may have been from the testing I did, and I don't want to accidentally have everyone's info on valves somehow coming from a loop-de-loop of sources

[tawdry blade]

I wonder where that comes from lol

area delete would also be nice

[prisma kraken]

you know, i wasn't thinking at all about dsp belts

[thorn bane]

interesting
obligatory @crimson moat tag
unfortunately i just cant get myself to test it because i just dont care enough about it
even if they are precise i would still never use them
maybe if im really bored

i wonder if its also different because rocket fuel is a gas

[radiant lance]

doubtful, my testing with water from a few weeks ago had the same results

my working theory is that it's fixed on experimental but not stable

[thorn bane]

" Flow Testing:
over/underclock the Fluid-Input packagers to ingest the fluid rate you want.
set the valve to the same setting.

If the Fluid-Input packagers keep stalling, increase the valve setting by 0.1 to 4.7 depending on how exact you want to be. "
could this be fucked because machines clock to 4 decimals? so its not because of the valve but because of the machine?

[jovial scroll]

if someone here can code id love a custom recipy calculator for satifactory lmao

[frosty owl]

Sorry, I missed all notifications.
About fluid cars: you can load fluid on a car once, and just leave the fluid container "stuck" on that car so that it won't interact with solid freights.
About splitting differently: I think it all comes down to your own preferences about complexity and throughput aviable. If you need a lot of throughput, keeping things as they are can make it difficult to reach very high throughput without being careful about train timings and so on. If the throughput is low enough, sushi can be a possibility; so long as you make sure that everything that is loaded into a sushi freight can be taken away completely, there should be no issues at all (ie: the sushi can't stop flowing ||wirhout extra-special care||)

[cinder silo]

I have unintentionally left myself with another monster build, all this for my power station byproduct, what have I done!

I need to squash the base of the foundry building into this space.

[oblique hollow]

Maybe it was you i heard it from.
I then went ahead and tested it on my own, also reached the conclusion "yes they work" and have been rectifying my statements regarding them.

So in a way, it is a loop de loop. But a reinforcing one

[crimson moat]

i highly recommend not doing that with the CC

[oblique hollow]

good lord why is the CC steel recipe like that

[crimson moat]

If it was 48 foundries it would be pretty fine lol but it's 480

[harsh schooner]

no alternate recipes?

[oblique hollow]

This is with alts

Compacted Coal

[crimson moat]

basically the compacted steel ingot is sacrificing a huge amount of build time, space and sulphur to save iron. Which is the last thing on the planet you typically need to do

[cinder silo]

The CC is there as a byproduct of the power station, I really don't want to just burn it via coal or sink it, I'm not exactly short of power.

[crimson moat]

honestly i would rather sink it and move on and maybe come back later lol

this would take you hours

[oblique hollow]

It technically also conserves coal.

You COULD be funny and transmute sulfur to coal or vice versa

Power shard the hell out of those foundries

then its only 240 or 192

[cinder silo]

True, I have a lot of slugs lying around after I farmed all the sloops yet again, I can just sloop a constructor and spam shards.

240 feels a lot less painful than 480.

[cinder silo]

16 rows of 15 machines, divided across 4 floors looks like the plan, this will take a while.

[oblique hollow]

BP to the rescue!

[cinder silo]

I'm manually laying because the machines don't fit perfectly onto a foundation, there is a slight offset.

[crimson moat]

you can kinda do that with bp too, but a bunch of machines at a time

[cinder silo]

True, though it isn't hard to just hold ctrl and spam them once machine #1 is in place.

[restive sparrow]

The same is true of the BPs. The biggest pain here is going to be connecting all of the logistics. Even without autoconnect I'd rather belt together 8 or 12 blueprints than place lifts, holes, mergers, splitters, and belts for absolutely everything

[slow crater]

So my foundry manifold is not working and I'm not entirely clear on why. Mk2 belts going in from Mk2 miners with each foundry underclocked to take 40/min. Coal elevators are mk1 and any belts after the first splitter are mk1. The resources in the second machine are still decaying faster than it can be refilled. I've tried skipping the warmup period by manual filling but no luck.

Oh and obviously plenty of power

[thorn bane]

are the feeding iron and coal belts flowing or stuttering

[analog meteor]

btw low power cant slow down production. the grid will either be running or fuse blown

[slow crater]

Stuttering seemingly when the first machine fills up?

[thorn bane]

well if it looks like its flowing slowly then theres probably a mk1 belt somwhere that should be mk2
if its almost empty then its the miner

[slow crater]

I mean the lines are completely full, like its not intermittent so im not sure its a miner or belt issue. Its just like they seem to freeze up for a sec when the first is full. I wish I had a screen recorder lol

[thorn bane]

since youre consuming 120/min they should always flow
so check to see which segment makes it not constantly flow

[slow crater]

yeah, ill keep digging

[thorn bane]

also are the foundries full on either iron or coal? that would rule out that belt so you can focus on the other one

[slow crater]

Do you agree the belts after the first splitter dont need to be higher than mk1? And no, its decaying too fast on both resources

[thorn bane]

well its splitting it 120->80:40
the 40 can be mk1 the 80 has to be mk2

so the belt to the foundries can be mk1 but the other one not

[slow crater]

Does it not split 60:60?

oh wait I see what you mean

maybe thats the issue?

[thorn bane]

youre trying to feed 2 foundries at 40/min so 80/min which doesnt fit on a mk1

[thorn bane]

yup upgrade all to mk2 (up until the splitter that splits 40:40 to last and second last foundry)

[slow crater]

Gotcha yeah I see the math now

Cue beautiful mind equations

[lunar pollen]

what website is that?

[crimson moat]

satisfactory tools

[lunar pollen]

k thx

[crimson moat]

https://www.satisfactorytools.com/1.0/

[ashen stirrup]

Current project. The 600 iron ingots go into a blueprint to make the motors.

[radiant lance]

I'm glad you've reached the same results! I've been writing up a post on the sub putting forward my testing and evidence, and hopefully it can serve as a proper "yes, valve flow limiting works now" post for people

was your testing on the experimental branch, or stable?

I asked a friend to test on stable branch yesterday and their results showed valves being accurate on stable too, and while I don't want to conclude based on just one result, it does seem like both branches have them fixed

[quick gorge]

continue testing pioneers

[radiant lance]

🫡

[thorn bane]

ill do my own test if i get a cake at the end

[oblique hollow]

both actually. All over the place.

I verified it in 1.0 and in EXP

[radiant lance]

fantastic

[oblique hollow]

I noticed the bevahiour change on 1.0 launch already, but couldnt quite make out what it was

all i noticed was that some valves were more accurate, but the display still showed the usual (limited) flow values

[radiant lance]

is it okay if I write in the post that you've verified the changes too? it may lend credence to the data for people knowing that the author of the pipeline manual has also tested it

[oblique hollow]

But after these tests im now conviced, as well as considering other previous tests, that all flow rate displays are inherently only "sort of accurate"

[radiant lance]

yeah flow rate displays are fucked

[oblique hollow]

They calculate averages over time after all

so its kinda the belt throughput counter situation, but much, MUCH older

As far back as Update 3 / 4 old i would say as pipes had always had flow rate displays

and when you split pipes, you always had some slightly odd values at times

like a mk 2 split into 2 yields 301/min or 302/min

[radiant lance]

matter creation...

[oblique hollow]

its easy to assume that, but there is never any buildup

so i always accounted those things as "flow rate display is inaccurate"

[radiant lance]

agreed

[oblique hollow]

On valves, it seems to be the same now.
Limit is accurate, display is inaccurate.
display seems to still be in 8 bit in a way (256 values for flow rate)

[radiant lance]

from the valve testing I've done, the rate displays on valves and pipes are a signed byte, but the actual real flow is not

so simply yet another instance of don't trust pipes ever

[oblique hollow]

well, you can trust them, but only within a certain degree of accuracy

Like, if you look at the pipe flow rate display, you rarely encounter these issues except for off-by-one errors, usually

on valves, its trickier as they decided to show a decimal there for some reason

so they gave them more accuracy than they can actually display

all other pipe buildings use no decimals for flow rate

so its very much misleading on the valve

[radiant lance]

for sure

[harsh schooner]

just no use valve

[oblique hollow]

Also: sure

Using valves is now doable, but it has the same restrictions as per usual

1. if you split off of a junction, you probably need valve son both exits because valves are still pressure sensitive
2. using valves too much can interrupt the balancing of pipe flow and content and make flow suck more than if you didnt use any valves
3. Feedback loops in aluminum are still subject to failure with valves because its an unstable system design to merge byproduct and fresh water.
   (it can work, but its a fragile equilibrium)

[crimson moat]

Also had 603 from a 600 pipe

but it would stutter if it was actually over 600, and didnt

[oblique hollow]

yeah it tends to show that on buffers

[crimson moat]

there is just a half percent or so error

[oblique hollow]

602/min usually

[radiant lance]

I also noticed in my testing acc that the valve had issues moving fluid from a full pipe to an empty pipe, but when both pipes were full, there were no issues

did you notice anything like that?

[oblique hollow]

No, none at all.
full to empty worked fine

was it vertical?

[radiant lance]

nope, no verticality

funky

[oblique hollow]

i could not verify that at all.
no issue like that ever popped up for me

[radiant lance]

I'll just assume I got smth else wrong and changed both the pipe fullness and fixed a number being off in my testing at the same time

i very well could've had incorrect input/output rates causing starvation

[oblique hollow]

all of those are possible

[viral sparrow]

@upbeat summit example with 300 crude oil
crude oil -> heavy oil residue (alt) (400 heavy oil residue)
heavy oil residue + packaged water -> diluted packaged fuel (alt)
unpackaged fuel -> fuel generators (800 fuel, 10GW)

Vs crude oil -> fuel
300 crude -> 200 fuel (4x lower than with packaged)

crude oil -> heavy oil residue -> fuel
300 crude -> 400 hor -> 266.66 fuel

couldnt get the whole setup in one screenshot but there is packagers around the d.p.f refineries to make the water + unpackage the fuel

[upbeat summit]

that's insane, thanks a lot

[crimson moat]

You can also step the same fuel up to rocket fuel later by adding additional ingredients

[viral sparrow]

you can just fill the packaged water packagers directly with the empty canisters

no need to automate canisters is what i mean

[crimson moat]

Fuel + compacted coal = turbofuel, + nitric acid = rocket fuel

or

fuel + HOR + sulphur + coal = turbofuel, + nitric acid = rocket fuel

[viral sparrow]

also, theres an alt recipe in phase 4 that skips the packaging stuff and just takes hor and water to make the same amount of fuel, its a godsend of a recipe

[crimson moat]

Is good but i wouldn't rip everything up if i already did packaged in p3

[viral sparrow]

yeah definitely dont get rid of what you already made

[restive sparrow]

7.2 Crystal Oscillators/m
21.3333 Heatsinks/m
36.8 Control Rods/m
all to supply a 300 Uranium/m -> 7.2 Uranium Fuel Rods/m factory. It feels a bit silly that the input and output stations take up more space than the manufacturing elements. I need to figure out how to get comfortable with sushi freight platforms 😦

This is the final supply site for the nuke factory. This marks 1 week to the day from when I finished my notes on how I would break up the sites and where they would go.

[prisma kraken]

i think probably the best way to contend with that all is to plan on the first floor of a factory being a train station and the rest stacked atop

[restive sparrow]

The remaining build at what I refer to as the "Exclusion Zone" where radioactive materials will be handled, manufactured, and burned. I'm trying to figure if it's worth my time to try to use the water output from the non-fissile blenders as part of the supply for my nuke plants or just say fuggit and package it up and sink it.

[restive sparrow]

That's a good idea. Or cases like this where the rail is coming in elevated I could have put the stations as the top floor. Just kinda hard to plan out. The factory ended up taking shape around the station locations.

I guess it coulda been a similar game of "build giant floating platform, build stations, build factory" just started from higher up and building the factory below 🤷

[prisma kraken]

yeah, i think probably the rule is to plan the train first and then do the rest of the factory around it

my current WIP

[restive sparrow]

Looking good. I like the angled water extractor platforms

[prisma kraken]

thanks, i have no idea of what the form of the final production will look like (its going to be a huge complex), but i'm settling on something kinda pleasant for the recycled oil loops so far

the end product is going to be a 600/min MF factory

[restive sparrow]

Whatcha think about my water predicament? Package it or try to manage it going into nuclear power plants?

[prisma kraken]

idk, water for nuclear is always a challenge

[restive sparrow]

I am seriously leaning towards package and dump

[prisma kraken]

i mean, you can support 2 reactors at 250% off of a 1200 line of water

it doesn't seem like a bad way of going for the power plant side of things, but extracting it and packaging it is still going to be a headache

[restive sparrow]

It just stinks that the water to be managed from the blenders is a pretty low amount so mixing it into a nuke plant that's running at 250 would be tricky. Otherwise the plant will just run off two extractors at 250% piped directly in.

[restive sparrow]

It sure will be. I'm having to add another intake platform to bring in plastic now.

[prisma kraken]

how big of a power plant are you building?

[restive sparrow]

I'm maxing out one impure uranium node (300 uranium -> 7.2 rods)

I think thats 90GW. This is my first nuke plant so dipping a big toe in the proverbial pool.

[prisma kraken]

so that's 36 reactors

[restive sparrow]

I plan on overclocking the reactors

[prisma kraken]

what are the rates on ufr burn?

.2/min or .5 at 250%?

[restive sparrow]

I'll be building 15 reactors at 240% (just for a symmetry design thing). I was planning on feeding them each 600 water (even though they only use 576). I guess I could just do the math to feed them the blender water and clock my extractors appropriately since I"m not even using nice round numbers anymore

[prisma kraken]

i would suggest 240%, yeah

that means .4/min per reactor, and that is 7.2/min / .4/min = 18 reactors

[restive sparrow]

.4 would be 200%

[prisma kraken]

i would highly suggest you go with 16 or 18 reactors instead of 15

derp. i'm mathing poorly

[remote flame]

I was just about to say that is exactly what I do, 2x extractors at 250% to drive a 250% reactor (0.5 rods/min).

This isn't so much related but some inspiration / a different approach I used in one of my playthroughs;

#design-and-architecture message

Final finish: #design-and-architecture message and #screenshots message

Uses a mk3 blueprint to completely self-contain 2 Uranium Fuel Rods underneath a reactor (1 without sloops), which can feed 4 (or 2 unslooped) 250% reactors. The inspiration part is the water extractors go in the base plate the reactors sit on top of wherever the factory isn't. So it makes a very tidy finish 🙂 depends where your nearest source of water Is / where you want the reactors to be! Training water is painful, but I suppose long pipes just as much so too haha

[prisma kraken]

you kind of really want to balanced-split the fuel for the reactors and get rid of that 5 in the reactor count

[restive sparrow]

Kinda toying with the idea of shaping the plant like a ☢️ which is why I was thinking 15 (divisible by 3)

[restive sparrow]

The goal being balancing instead of manifolding for radioactivity mitigation?

[prisma kraken]

well, not having the plant be radioactive is a benefit, but more than that, not having to wait 8 hrs for a manifold to fill is much more of a +1

[restive sparrow]

I guess 18 reactors is still divisible by 3, but is also easily balanced (2 way split, 3 way split, 3 way split)

[prisma kraken]

yeah, 18 is 2x3x3

[restive sparrow]

So maybe I go with 3 arms of 6 reactors

I haven't really done a circular build before...should be interesting

[prisma kraken]

btw, 36 at 100% = 16 at 225%

dunno how attached you are to the multiple of 3

[restive sparrow]

Why would I care about hitting 225?

Not saying I shouldn't I just don't get the significance

[prisma kraken]

i'm just saying it is another option

[restive sparrow]

Ah

Should be able to do any number relatively cleanly if you go around clocking things right

[prisma kraken]

or you could go with less reactors and use some rods as drone/vehicle fuel

[hushed trellis]

chat do i build a 20/m smart plating factory

[spiral geode]

Drone Fueling Question: Anyone have a best guess on how many batteries per minute would be needed to try and use only drones for low quantity item throughput for the late game?

[prisma kraken]

i think min/max are 4/9 per trip with batteries

[spiral geode]

I suppose 480/minute would suffice for all needs until a more efficient fuel can be provided like Uranium then?

Perhaps going simpler and doing 240 even?

[oblique hollow]

you likely will never need more than 240/min batteries for every drone you ever wanna build

and even 240/min seems extreme

drones do not need a lot of them

the actual amount they need per trip (not per minute)
is 4 batteries minimum + 1 for every km they fly

[spiral geode]

That makes sense, plus batteries only other use is super computers.

[oblique hollow]

and even that is an alt

otherwise, batteries would have no use other than drones

[spiral geode]

Appreciate the quick responses. @oblique hollow @prisma kraken

[prisma kraken]

do keep in mind that drones can now take packaged fuel(s) as well

if you can tolerate longer round trip times as a result of using the worse fuels, it is a pretty good option

[crimson moat]

rocket fuel outperforms batteries, and you need it for jetpack (and maybe power) anyway

[torn plaza]

yeah I made 200 batteries because that seemed like a nice, round number, didn't need that many machines to make, and used no more than 600 of anything

looking at how many batteries my ports use, uh. I way overdid it lmao

[wind spade]

can use them for other vehicles as well

[torn plaza]

trains every day for me

[prisma kraken]

yeah, honestly if you're making batteries, they do very nicely as vehicle fuel

[prisma kraken]

it can be beneficial if you're tight on aluminum to use a fuel that doesn't require any

[cinder silo]

I discovered to my cost that mergiing 3x 400 onto a mk6/1200 belt will eventually back the system up, mergers just don't manage to seamlessly mix the streams resulting in less than 1200 downstream, and a backup at the output end.

[prisma kraken]

yeah, 1200 belts are a little lossy

to be fair, back in update 7 & 8, i was seeing some loss on 780 belts as well

i think probably you could see similar with the mk4 belts if you really got lucky catching them being naughty

[cinder silo]

I had a shed load of tests on the 780s before fixes went in, I'll stop using the 1200s at capacity for byproducts, a backup causes chaos.

The resin system was replaced with two 780s.

[steady yacht]

I dont know if this is the right channel for this question but does anyone know what is the best interchange type for trains? Have been experimenting with roundabouts but had always problems with stuck trains. Would like to have some kind of u-turn possibility integrated so trains could just turn around quickly.

[vapid gorge]

roudabouts are the worst for throughput
but realistically you'll never have hundreds of trains either to make it a problem. It just sounds like you have some bad roundabouts

[marsh nacelle]

If I have 90 of something a minute and 30 of the same thing, how can I merge them and then split them into equal 60 per min outputs? I can’t just merge them into one and split that into two

[vapid gorge]

why can't you?

you have at least mk2 belts if you have 90pm going somewhere

[marsh nacelle]

Cuz I’m trying to make a schem that will do the same thing for higher and higher volumes and eventually I’ll be limited by belt ipm

[vapid gorge]

split both in 2 and merge one of each then

or you could do the smart thing and clock machines as you need them

if you need X ppm going to point A? put X ppm on a belt

[wind spade]

something like turbine interchange is great (but as was said above, usually not needed)

U-turn is pointless

[marsh nacelle]

Oh can’t believe I didn’t think of that

[vapid gorge]

what? just making the numbers you need to put on a belt?

yes that does seem silly

[wind spade]

usually the question I ask first when given these questions is - why not just use what you have on the belt (so hook 90 to machines that need 90 in total, etc.)

[marsh nacelle]

I’m planning on making a huge ass factory that needs 1200 ingots per min and I won’t be able to just rely on belts to deliver those evenly to all the smelters so I’m gonna do a huge buffer of all the different ppm miners and evenly split those into 4 separate smelting schems

[marsh nacelle]

No the splitting thing

[wind spade]

or just do a manifold

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

[vapid gorge]

I mean, that's the worst option at hand, so I'm not surprised

[marsh nacelle]

No that limits me to the ppm of the first belt unfortunately, manifold for each of the four outputs of the splitter thing

[vapid gorge]

if you're doing that you don't need to do any of hte previous. Just split the belt 2 times and you'll have 4 belts

[smoky aurora]

build 4 / 300 lanes 🤔

[wind spade]

well you make manifold out of each belt

but you don't have to "balance" the belts

[marsh nacelle]

That’s what I mean

[vapid gorge]

but also why can't you 'rely on belts to deliver them' ?

[marsh nacelle]

That’s what I’m doing

[wind spade]

if you have belt with 1200, hook it to a manifold that needs 1200

no reason to do any balancing

[vapid gorge]

or buffering

[wind spade]

or that yeah

[vapid gorge]

like, you can do that, but it doesn't benefit you, and just takes more time and space

[smoky aurora]

get it in via train 2 platfoms ,... each got 2 outputs ,,.. if one lane is full it switches to second lane ,..

if i build __Bigger plants usally i import slightly more as i really need so i dont need to rely on the in time delivery

[vapid gorge]

if you think it's fun to do it? sure go for it

but it sounds like you're trying to solve a problem that doesn't exist

[wind spade]

(or rather - you're trying to solve a problem which was created by yourself picking a weird way to solve things)

[vapid gorge]

or that

[marsh nacelle]

Cuz I’m using a truck network early in before trains and even when I get trains I’m limited to 2x the rpm out of ehatever belt I have which won’t be enough until mk 5 so I need to find a way to combine all the different iron inputs and split them all evenly without bottlenecking on the ppm of the tier belt I’m on

[smoky aurora]

so your bottlenecked with the belts anyway

[wind spade]

no need to "combine and split"

if a train/vehicle/whatever delivers X, hook it to machines that process X

[vapid gorge]

well honsetly you're not likely to get 1200 per platform even with trains

also what greeny said

[wind spade]

whether it's a train or a truck or a belt, it doesn't matter. It's all just "a way" to deliver things, that's capped to X max per minute

it doesn't change anything in terms of planning a build

[marsh nacelle]

It’s one big machine that’s hungry for 1200, trying to make 50 mod frames a min and that’s way more of a logistical nightmare

[wind spade]

what machine eats 1200 of something?

afaik most a machine can need is 975/min (for heavy flexible clocked to max), if not counting biomass processing

[vapid gorge]

anyway, tapping out of this convo, gl with things

[smoky aurora]

i got 3 Groups consuming 1200 each ,.. i got 4 Platforms delivering 4x 1200 ,.. the 4th Lane i use to feed a container ,..

so i feed the container with
1/4
2/4
3/4

so even if the train arrives theres always enough in container to "bridge" the unloading time

[smoky aurora]

50 is a late game taget but defintly not needed in midgame ,.. 20 schould be more than enough

[marsh nacelle]

Obv but I don’t wanna be tearing down my whole factory late game to meet it so I’m setting up machines and networks early on just to plug it all in when it starts being needed

[smoky aurora]

so where the problem ,.. underclock it so your actual feed fits machines need ,..

[vapid gorge]

yeah future proofing is almost always a massive waste of time and 10x the work and effort

[wind spade]

^^^

[marsh nacelle]

???

[wind spade]

the best advice I can get for future-proofing is: don't

[smoky aurora]

yepp i agree

[marsh nacelle]

That sounds like the most massive troll ever lmfao

[wind spade]

just make what you need now, if you need more in the future, build more then. (and no, you don't need to upgrade your factory, just build a new one)

[smoky aurora]

just think of ALT´s
or the better logistic solutions you get late game

[wind spade]

(and no, this is legit)

[vapid gorge]

you have several people with tons of experience in the game telling you this

[smoky aurora]

your whiole ssetup changes if there are alt recipes available

[wind spade]

(kinda related to conversation 10 minutes back, but I'm posting it anyway) this is one of the big problems I see new players run into

you don't ever need to allow "all resources to flow everywhere". If you have X of some resource, hook it to something that consumes X. You don't need to connect it anywhere else, especially not to other places that process the same resource. Because it will always make/consume X (unless it breaks for some reason, but that's not the point anyway), so it doesn't need to be connected anywhere

This kinda goes hand in hand with trying to centralise things. Don't. Build near nodes, build near factories that need that item, whatever. But don't just put all production for one item in one place, because it then becomes a logistical nightmare

[vapid gorge]

build factories from an end point.

[smoky aurora]

i got 1200+ hours ,.. 😄

may you guys have more

[wind spade]

I most likely don't 🤔

[vapid gorge]

no but you have logistic experience

I haven't really played much for a couple years and I still have 3500hrs

[smoky aurora]

😄

and i´m still learning ,...

i love it

[vapid gorge]

I'm stll learning?

mostly fiddly architecture stuff which is unending, but still

[smoky aurora]

i´m me ,..

today i fully understood pathsignals

[marsh nacelle]

But big factory 🥺

[smoky aurora]

i needed to set a new network for my aluplant ,.. and the last tik in my head clicked ,..

[smoky aurora]

Later 😄

[vapid gorge]

yeah, build that from the end point

[wind spade]

the factory is still big... just divided logistically

[vapid gorge]

later. When you are building it

[vapid gorge]

look - you can still do what you want, we aren't your parents, we're just people with experience telling you it's going to be 10x the work and time and probably won't work out how you want it to.

especially since your goals may change

or the path to your goal will be different with different recipes

[wind spade]

when you're early and need to make iron plates for storage, you don't need to build 3600/min because you may need it in the future
you build like 30/min, hook it to storage, and that will last you for rest of the game

next time you need iron plates for a different factory, you calculate how much that factory will need and build that much, e.g. in that factory.

and so on and so on (replace "iron plate" with any other item)

[smoky aurora]

we dont have bad intensions ,.. we just tell you the mistakes we made 😄

[vapid gorge]

like setting up world rail systems before you've built any of your factories. Ugh. What a waste

[wind spade]

honestly I see this exact "mistake" as a big source of burnout, there's like at least one person daily who comes here and complains about burnout, who then turns out burned out because they built tons of extra unneeded stuff in advance

[smoky aurora]

yeah i stoped that too ,.. kinda ,...

i know some frequent places i connect anyway ,.. but hey ,.. the whoile map ,.. ? ,.. nahhhh

[vapid gorge]

I'm glad I built a 9000 plastic/rubber factory 'for future use' because I learned some logistic and planning stuff from it

But I tore it down afterwards because I also learned that's not a good way to do things

[vapid gorge]

yup, burnout central

[wind spade]

centralised burnout

[vapid gorge]

spent at least 80hrs on that plastic/bauxite factory

[smoky aurora]

80 Hours are pretty fast collected ,.. imo 😄

[real fossil]

alternate iron pipe or alternate automated speed wiring?

[wind spade]

Whichever you need/want/like

All recipes are useful in some situations and you can get them all anyway

[real fossil]

cool, thx

i just found out by accident that holding E while having a build piece selected brings up similar parts. Sigh... so much in this game and i love it

[vapid gorge]

if you right click while holding E it'll bring up different materials available for those objects too if available

[real fossil]

well damn

these guys really thought of everything

[vapid gorge]

it's a very good way to make compact hot bars

[prisma kraken]

i favor really large 3 way intersections.

[vapid gorge]

only need 1 item of each general type

[vapid gorge]

you know you can eye drop an obj to build with middle mouse button click right?

[real fossil]

yuppers

[vapid gorge]

oh and that if an object takes multiple click to build, like a pipeline stand, it means there's options to modify the position each click?

like the height of hte stand, or the angle the circle bit tilts to?

[real fossil]

yup. that one took me a while

[prisma kraken]

i've been doing a bit of experimenting with fluid trains and know the wisdom of how to use both fluid outputs of the station into a buffer, but for my need at hand, i've only been looking to move 300/min per car for a short (~5 min rtt), and instead of the more complicated double piping into the buffer, i've been finding this configuration adequate for my volume & trip requirements - I'm just sharing because sometimes you can do things somewhat more simply:

[vapid gorge]

You can do this with belts too if you’re just using a fraction of the through put

It’s just more of a universal solution you can apply having 2 outputs rather than doing work trying to figure out the throughput

[prisma kraken]

yeah, it is pretty basic, but i'm sure there's a few people out there that are just following a yt guide on it all and thing you have to do something complicated

tbh, i was thinking that i'd need to as well, but a quick bit of experimenting showed me that i was so far under capacity on the train route that what's pictured worked fine

i'm not sure why they decided to keep the lockout problem in the game - it just makes building trains more tedious

[vapid gorge]

if 2 outputs to 1 container is 'tedious' this isn't the game for you

also, it's more work trying to figure out if you can, rarely, manage 1 pipe instead of 2

[prisma kraken]

2 pipes aren't tedious, doing it 8 times for each train route is 🙂

[vapid gorge]

still less so than mathing it all out

or makign a bp of the fluid buffer

[warm river]

Am basing my loop on this image, btw

[quick gorge]

Just having "TRU" was getting on my nerves, I didn't even read the context. I just needed put the E on the end...

[restive sparrow]

I feel stupid for even needing to ask this, but on SCIM, what in the world do these numbers mean in the statistics page?

I was thinking "parts made vs parts scheduled to be made (indicating machine downtime) but that doesn't even make sense cause I have plenty of numbers where green exceeds red.

Parts made vs parts consumed also doesn't make sense because I couldn't possibly be consuming more parts than I'm making.

[vapid gorge]

It’s probably the most stable loop you can make

Do some overhead shots of your layout?

[prisma kraken]

it tallies all the machines with specific recipes set, so that random constructor you built somewhere just to look at a recipe appears in the numbers. on top of that, it does a strange comparison for equality of floating point numbers that doesn't always end up indicating that things are balanced

[frosty owl]

Iirc, the "grayish" parts are machines with recipe set, but without power

[frosty owl]

I happened to make a 300 Uranium - > Plutonium Facility too, with all raw inputs delivered (1 station for Uranium Processing, 1 for Plutonium) using... 7 freight platforms 😅
I could have used less if I dared mixing resources per freight in some cases, but I had a few reasons not to and enough space for it. This is the floorplan (Uranium processing floor 1 out of 2 + station)
https://cdn.discordapp.com/attachments/553550313533997057/977507323796914186/1653126239785.jpg?ex=680ea563&is=680d53e3&hm=8485bbc680076ede7d5e49f3453109012050644b59971ddadf4e1ed26315df43&

I'd say it's a decent ratio of freights-to-machines, but I added the screenshot for you to judge ^^
Edit: most machines are heavily overclocked

(The train needed 2 locomotives)

[pastel obsidian]

If you want to do a u turn its best for it to be done in a loop that goes above or below the track,

[quick gorge]

#streams-and-videos message
@frosty owl :)

Thank you @balmy steppe for your restaurant 🫡

10/10 pixel Mario btw ^_^

[true fog]

Dang, do you have an ingame picture of this somewhere? Looks very clean

[frosty owl]

There's a few #screenshots message
And even a video (though it focuses on beltwork rather than machines)! https://youtu.be/RoL8cB_4ev8?si=DqOm9eVuWw7VoVdG

[true fog]

Honestly, it was the beltwork I was the most interested in, so there being a video is just

amazing

[quick gorge]

Hey Sir Sushi, have you ever gone to an actual sushi restaurant?

@frosty owl <3

[frosty owl]

Nope

[quick gorge]

Well I do want to create a sushi belt that looks like an authentic Japanese sushi belt restaurant...

So was hoping you could help one the function and the form 🙃

[jovial scroll]

120 into 90 any ideas?

[wind spade]

what do the numbers mean?

belt speed?

[jovial scroll]

I need to do quite a few belts of 120/belt into 90/belt

[wind spade]

one splitter, machines will regulate it themselves

[frosty owl]

Split in 4 (30x4 or 60+30x2), merge 3 and loop 1 back before the original split

[frosty owl]

I can try... I'm not sure what you have in mind though

[remote flame]

oh as in you want to get 90/min off of the 120/min belt? So the machines will flood and it will distribute correctly over time as greeny said, or if you want to load balance it;
(Diamonds are splitters, square is merger)

[jovial scroll]

issue is that I have like 8 belts of 120

[wind spade]

and what do you want to do with them?

connect each belt to machines that need 120 in total

[jovial scroll]

turn them all into 90 basically lmao

[quick gorge]

This is much more leaning into design over the maths at this point 😅

[wind spade]

but why? just use what you have on the belt

[frosty owl]

You can merge the 30 you left out back onto the original 120

[jovial scroll]

ok ty

[jovial scroll]

🤷 this is a challange yourself factory anyway

[remote flame]

yeah just left it out wasn't sure haha

[jovial scroll]

hopefully this design works

[jovial scroll]

also I am not sure this is going to work for how I put my machines down anyway

[wind spade]

you should put down machines based on what you have on belts, not the other way around 🙂

[jovial scroll]

😭 I wanted to try this out

[wind spade]

then do that, and clock each set to need 120

[jovial scroll]

I may not have access to it yet XD

but yeah good point

[wind spade]

getting over/underclocking is like 2 minute job in MAM and exploration

if you don't want to get the shards, you can build more and underclock

[jovial scroll]

god I wish it was that simple here

[wind spade]

it is

[jovial scroll]

playing modded be like

[jovial scroll]

ik this is modded but still painful lol

[quick gorge]

SF+ updated?

[jovial scroll]

YEPP

[jovial scroll]

😄

[quick gorge]

Neat.
Time to wait for 1.1 to goto stable to even consider starting it

[jovial scroll]

👀 time to play sf+ in the meanwhile

xd

[quick gorge]

No, a pioneer must test and report to FICSIT about any structural problems

[jovial scroll]

woo

finally like 1/4th done lol

[jovial scroll]

I hate doing beltwork on 3x outputs lol

[jovial scroll]

sf+ loves making you do miga projects if u want to even think about doing a decent amount of automation 😭

[quick gorge]

I did play it before 1.0, got to trains before my old computer said "I don't want to anymore"

After 1.0 I got a new computer sooo yeee
But as I said I want to stick to 1.1 and test for the team to make game more gooder

[jovial scroll]

ill prob need to upgrade my cpu to play this game lmao

Im legit running at lowest graphics since my pc cant run satifactory at higher

[jovial scroll]

I need 50 smelters 😭

PAIN

[quick gorge]

I'm about 0.005% into my project so don't worry

[jovial scroll]

whats your project lol

[crimson moat]

if you finish the game you end up with like 1000 smelters

[quick gorge]

But the smelters are in the shape of refineries

[jovial scroll]

im in stage 2 only

[jovial scroll]

huge

what u gonig to use that power for

[quick gorge]

#satisfactory-experimental message also this

[quick gorge]

It's not about the power... it's about the builds that make up the power

🇮🇹 is going to be fun 🙃

[jovial scroll]

plz no 😦

[quick gorge]

?

[jovial scroll]

not the spegeti 😭

[quick gorge]

My nuclear pasta factory is going to be shaped like Farfalle

[jovial scroll]

ah

[quick gorge]

And will be delivered on the macaroni belts and housed within the leaning tower of lasagna

If you don't like that idea then don't ask about Guru's visit part 😏

[restive sparrow]

Hats off for the daily battle I see you fighting. 😅

[vapid gorge]

https://tenor.com/view/honest-word-its-honest-work-it-aint-much-it-aint-much-but-its-honest-work-gif-13763573

[prisma kraken]

i think i may get burned out more by answering such things here than by the game itself

[quick gorge]

Type out the most descriptive version of a answer to FAQs and keep a list of message links, keep a list of links somewhere

"Be more efficient and less empathetic" FICSIT probably

[vapid gorge]

look, greeny has a few of those

[quick gorge]

mhm :)

[balmy steppe]

🫡

[winged knot]

Is it worth accelerating particle accelerators or is it more profitable from the point of view of energy consumption not to accelerate them?

[wind spade]

for any machine it's more energy efficient to build more (and possibly underclock) than to overclock

[cold stump]

I have 2 belts with aluminum ingots, one having 720pm and the other 480pm (1200pm total). I need to split this into 300pm and 900pm and i dont have mark 6 belts. How do i do this?

[prisma kraken]

split each 4-ways and then merge pairs of them

480/4 = 120; 720/4 = 180; 120 + 180 = 300

[winged knot]

I'm shocked I have 650 hours in satisfactori, but I just now noticed that they work like this, although generators work completely differently in creating energy

[wind spade]

I'm talking about machines, not generators

generators scale linearly and the only saved thing there is space

[winged knot]

i understand bro

[cold stump]

ty

[strange fractal]

How do i balance 2 max overclock and 1 100 %Sloppy Alumina taking 1200 Bauxite/m into 4 referinies using electrode aluminium scrap, 3 of them max overclock an 1 running 50% . I cant get the water to work somehow

[opaque quartz]

Split your refineries into two groups, one taking only waste water, the other only taking fresh

!wikisearch alumina+solution

[brisk shore]

https://satisfactory.wiki.gg/index.php?search=alumina+solution

[opaque quartz]

Wiki has the ratios

[oblique hollow]

you put the water the scrap refineries make into alumina refineries first.
Make sure you got enough refineries to use exactly all the water they make.
After that, use fresh water to make alumina in different refineries so all your alumina demand is met

[strange fractal]

Does the wiki set-up also work when i double it with powershards so instead of 600 i do 1200 in? The water in the pipes would be greater then it can handle right?

[opaque quartz]

max mk2 pipe is 600/min. you're gonna be better off making two modules that handle 600 bauxite each

https://www.satisfactorytools.com/1.0/production?share=GLV5y1l0Rgd7BSYHEmZD

clock one sloppy alumina refinery to 90%, it will only require 180 water (fresh only)
clock the other two to 105% each, they will take 210 water each (or 420 water total, waste only)

[near smelt]

Honestly the hardest part for me when it comes to alumina is how you get coal/petroleum coke to the refinery. Although I also build my first aluminum factory in the swamp so the nearest coal is in the cliffs area next to it.

[opaque quartz]

when doing aluminum in red bamboo forest, there is lots of oil in the lake forest area directly at the bottom of the cliffs on the north side

[near smelt]

I think i turned that oil into a fuel power plant. 🫠

If we are thinking of the same oil nodes

[prisma kraken]

also the fracking well overlooking crater lakes

[near smelt]

There's a well there?

How much does it make do you know? I just got the tech so it could be good

[prisma kraken]

yeah right here

iirc, it is 1350 crude/min

aka a lot

[near smelt]

Yeah 2 full mk2 pipes and then a bit left over

[prisma kraken]

its enough to make all the coke you need for all the pink forest bauxite, the rubber you need for heatsinks and then have some leftover

[near smelt]

Yeah I'm just starting the automation for those so that oil spot will be very useful i think. Especially if I fully overclock it

[prisma kraken]

since it's actually up on the same level as the pink forest, it is rather convenient for aluminum

[torn plaza]

so about water recirculation: it's easy enough to split fresh and recycled water for aluminium, but how do you do this for non-fissile uranium?

nevermind, think I figured it out; loop it with sulphuric acid or nitric acid, prefill the water pipe

[oblique hollow]

You can seperate all recipes inti a section that only receives byproduct and one that only receives fresh fluid

[prisma kraken]

the alt recipes for ufr's are also pretty nice, being that they don't need fluids

[torn plaza]

yeah I'm making alt ufrs to maximise production, but plutonium is all default

well, I will when I'm done with phase 4

[strange fractal]

Got it running at 100% efficiency 😁 now gonna scale up

[opaque quartz]

noice

[cinder mango]

what have i got myself into😭

[opaque quartz]

the factory must grow

[cinder mango]

yes

[opaque quartz]

make sure you have an overflow sink at the end of your production line (on the scrap or final ingots), otherwise you can still have water jam up even if you have waste+fresh split

[ashen stirrup]

@wind spade, do you know why the calculator won't give me a plan for plutonium fuel rods?

[wind spade]

did you provide it with a way to get uranium waste? e.g. add it to "input"?

because there's no ingame recipe for uranium waste

[ashen stirrup]

Gotcha, aside from nuclear power plants

and it doesn't have that as a production machine

[wind spade]

those aren't recipes technically

[ashen stirrup]

Thanks

[wind spade]

and since Tools extract data from game directly, it doesn't see it as a recipe

[ashen stirrup]

If I'm mathing this right, it looks like one impure Uranium node can yield 143 GW if I burn Plutonium

[restive sparrow]

That sounds roughly right. I've not done the math on all of it since I'll have to sink the plutonium rods for now, but the uranium alone should be getting you 90GW. An additional 53GW from the plutonium sounds like the right ballpark

[ashen stirrup]

I remember reading somewhere that the amount of plutonium waste is small enough that a storage facility ends up not being unreasonable

[wind spade]

storing any waste is reasonable

[ashen stirrup]

I think where I'm at now is that I just need to estimate how much power I could conceivably consume for the rest of the game, then build some or all of that in nuclear. I don't think I want to build a nuclear plant more than once, though

[mighty ravine]

what do you all think?

[vapid gorge]

that you should use a better planner

one that describes recipes being used

[mighty ravine]

ok

what do you recommend

[vapid gorge]

literally anything else - in particular though https://www.satisfactorytools.com/1.0/production

[mighty ravine]

well i think its broken

[vapid gorge]

refresh?

never seen that happen

[mighty ravine]

still broken

[vapid gorge]

https://www.satisfactorytools.com/

try the main link first.

[mighty ravine]

and its still broken

[vapid gorge]

what browser are you using?

[mighty ravine]

firefox