math-and-meta

[fierce ruin]

I would rather give up technology than intentionally have yellow lights.

[signal nimbus]

@oblique hollow Re: Flow stabilizer for multiple pipes
I know that design doesn't, but I had another (similar) design in mind that I was wondering if it would work for two pipes. Fluid buffer with a junction in front of and behind it, each connecting to two different pipes. In theory that would cut down on the floor space needed and still allow the pipes to be stabilized a bit, even if it's a bit less stable than the normal version.

[rain sky]

So I'm trying to think of a solution for having 2 belts coming from the same freight station that receives two different resources that I then want to split into two containers and I also want an overflow in there somewhere before they land in the containers.

It's easy with one smart splitter, but I don't want to bottle neck the two belts.

Nor do I want to lose resources from dumping stuff in overflow.

Current setup.

[signal nimbus]

A built example of what I'm talking about, though this is more of a 2 buffers, 3 outputs variant.

[oblique hollow]

That turns the stabilizer into a compensator + pipe balancer

[signal nimbus]

Okay, gotcha.

[oblique hollow]

Its not ensured that it will actually be stable

[signal nimbus]

Oh?

[oblique hollow]

Multi- output can be a bit unpredictable

Gotta try and see i guess

[signal nimbus]

I see... think it was described as "sloshing" in the guide?

[oblique hollow]

Yes, possibly

[signal nimbus]

Gotcha. I'll test it out, see how it works.

[oblique hollow]

Sloshing is a rapid change in flow amount and/or direction

Precisely what you dont want from a stabilizer

[signal nimbus]

Yeah. So, results are in. The two outer outputs are varying between nearly 300 and less than 100, maybe about 80-90. Don't have the tech yet for valves, but I have a feeling those might be a good idea. System's probably only working smoothly atm because I pre-filled it.

Oh, and the middle output reads 0.

[near zenith]

in case anyone was curious, you can fit ~82 water extractors on the lake on the mountains west of the dune desert

[signal nimbus]

@fierce ruin Gonna ask the question here because it's a meta question. I know why I went for those coal nodes instead of the easier to get 3x normal nodes, but why'd you do it?

[fierce ruin]

There was a pure one
and I did not know they were there. before using the map for the first time...

oh and because of the 12 power bugs I found, 3 yellow, 9 blue.

[signal nimbus]

Gotcha. I skipped over them so I could grab the slightly further coal first, and then later on pick up those for either more power or steel. Be warned, if you haven't been there yet, I think there's an alpha spitter or two in the area.

[versed leaf]

if a mk3 belt is recieving coal at max speed (270) then I can do 18 coal gens right?

[signal nimbus]

Yep

[frosty owl]

If you ignore the issue with belts at max throughput, yes

[dusty isle]

I have a weird issue... Coal Generators... What is the "maximum" split chain on a manifold line before the final generator just doesn't get enough coal to permanently run?

On, lets say, a 300/m line

[oblique hollow]

Naturally 300/15

So 20

[dusty isle]

But it doesn't work...

[frosty owl]

If you talk "permanently" (in the long run), it's not about the number of splits, just about how much coal/min each connected input takes
If your generators want more than 300/min, you can't sustain more

[oblique hollow]

Do you have mk 4 belts

[dusty isle]

Yes. Have Mk4 belts. The problem is that, for some reason, it just doesn't seem to fill up the early generators and by the time you get to the later ones in the chain, they just don't seem to get enough coal to stay permanently on.

[oblique hollow]

Turn them all off

And wait till they are full

[dusty isle]

They don't accept coal unless they are on

[frosty owl]

Manifolds rely on (most of) the involved buffers being full, so they take a while to reach max efficiency. Machines consuming the items while the belts try to fill them up naturally makes this even longer

[oblique hollow]

They do

Coal generators, like any gen, accept fuel on standby

[dusty isle]

I'll have to double check... I haven't had success with this

[short holly]

It takes a while if you're using 100%of the resource/belt. Can also load it up manually to give it a kick. Also, use mk1 lifts/injection to help spread the coal out.

[frosty owl]

Or balance

[dusty isle]

Done all this... Left it for a week... Haven't seen an improvement... and my maths is right for the incoming coal.

[dusty isle]

Space is a premium... Busy looking at reworking it all

[short holly]

Hm, You may have an error or wrong belt somewhere. It shouldn't take a week to get steadying don't think

How did you get to 300 coal? Likely, Either you don't actually have 300 coal or there is a section of wrong mk on the belt.

[dusty isle]

I know what the problem is... My work has been reworked by another player on my server... Trying to "make it better" and just overcomplicated it

[short holly]

Ah, classic. Good luck with the fix!

[dusty isle]

480 in, 32 coal generators. I had a Mk4 belt split into 4, each feeding 8 generators. Math works out... What they tried to do... I'm not sure. Looks like they have load balancers and mid-bank splits to try balance everything out... I can't figure out what the hell they were trying to do

[frosty owl]

That actually sounds like a nice challenge...
For someone to hop into my save and try overcomplicating things @proven prawn Wanna try? 😆

[dusty isle]

"3 x 480 Belts, split into 8 x 180 belts, each belt feeds 12 generators"

This is what he was trying to do... WHY! split 3 into 8...

[crystal charm]

does the calculator site work on ficsmas stuff?

i just realised i need stars for vehicle skins

so i want to set up a line for it

[frosty owl]

Also, maxed out belts (often) end up in Less-than-max throughput, so.... :P

[short holly]

Lmao just manifold for coal tbh unless you're there for the challenge that just isn't worth the effort imo

[dusty isle]

This IS a manifold

[short holly]

.. then why the balancing

[dusty isle]

I don't know!

[short holly]

Brilliant

[dusty isle]

Let me rephrase... It WAS a manifold. Now it is a nightmare... I understand what he was trying to do now... But holy hell, round-about-way to do it

Water is at least not as complicated... Even if you don't get perfect splits, it still works as long as you don't exceed input or pipe size

[frosty owl]

Btw, the guide on the sushi split is up~ ( @thorn bane @proven prawn feedback much appreciated)
https://www.reddit.com/r/SatisfactoryGame/comments/roomxk/balanced_sushi_split_how_to_properly_split_mixed/?utm_medium=android_app&utm_source=share

Up next, in order: "Saturated sushi belts" (if @thorn bane allows me toake such a post 😁), "How to feed multi-input machines with one belt" and finally "Sushi belts, smart and programmable splitters: A comprehensive guide" that will take the longest to make... That's that plan for now

[dusty isle]

So it isn't a case of if you have more than 15 in a line, the first 15 would simply consume the 15 that the 16th one would need and by extension the ones after it going down...

[dusty isle]

That looks like fun! I'll look into that when I start messing with manufacturers

[frosty owl]

"Fun" is relative, but it certanely is fun to me! 😁 😆
Feel free to at me for any doubt if you'll have any, I quite enjoy sushi talks ^^

[short holly]

Hmm, interesting. My nuclear sushi belt (in via drone) just splits into a manifold for each thing via buffer and overflow. Could be useful for the current expansion in working on

[frosty owl]

That does sound like an appropriate scenario to try it
Nuclear manifolds can take so long to fill up...

[fierce ruin]

Prefeed them

[cinder silo]

I moved away from nuclear manifold because of the feed time, no way I want thousands of fuel rods just sitting there while some stations sit idle 😛

[fierce ruin]

<extremely Saruman voice>
TENS OF THOUSANDS

[cinder silo]

Once my system is onlined, for the third time lol, there should be zero (0) fuel rods sitting around on belts.

[wintry aurora]

What’s the plutonium rod production you guys are using? Just want a baseline idea as I want to make a steel production area and I can make a huge amount just off of two pure iron and twi pure copper nodes.

[cinder silo]

My setup, at least once I finally take heat sinks down to the final stage, should be 5.25 plutonium rods per minute.

Using standard plutonium recipes with the aim to get rid of waste though.

I'm gonna wake up the nukes again and see how well my infrastructure works at full whack, not 17% following 90 minutes of operation like last time.

Lmao, reloading my game, I hit the on switch for all the water extractors and my game crashed,

[wintry aurora]

Greenys calculator: ERROR! DOES NOT COMPUTE!

[cinder silo]

Worked fine when I did it.

[wintry aurora]

Maybe somethings wrong with it, I tried just maximizing and it still gives an error.

https://www.satisfactorytools.com/production?share=KgydXSA95DJjkk9VeSbf Does this work for you @cinder silo ?

[cinder silo]

Odd, yeah it isn't calculating off the output alone, Glad I kept my screen.

[wintry aurora]

Tried a hard refresh with ctrl+F5

[cinder silo]

Holy shit, my entire nuclear station onlined in five minutes, not several days!

[wintry aurora]

@wind spade Somethings wrong with the calculator, trying to do one plutonium rod/m yeilds no results: https://www.satisfactorytools.com/production?share=keSkf4Eq1iH7Wz7ijMfj All recipes are selected, there's no restrictions, and I even tried using maximize. Taromani did it previously with 5.25/m

[fierce ruin]

Do you have the Uranium Waste manually plugged in?

[cinder silo]

1050 waste in my case.

[wintry aurora]

No, I didn't realize that was needed.

Same problem on the U4 page.

[fierce ruin]

Well the system doesn't know how much you're making.. so you have to plug it

[near zenith]

yea, the ur waste needs to be manually input, it's irksome and probably a hinderance to many trying to do a plrod calc

[wintry aurora]

Thing is, I don't know how much uranium I'd need, thus the waste.

[cinder silo]

That explains that, I added the waste as an input when I did it because I was calculating my nuclear chain in smaller chunks, but my setup uses all of the uranium on the map.

[near zenith]

600 ur turns into either 300 waste on regular reactors, or 720 waste on 250% oc'd reactors

per min*

using the alt recipes for ur rods and cells

[cinder silo]

I need to reload my game again, suddenly it's chugging? like wtf it hasn't done that like ever.

[wintry aurora]

I'd need a lot less steel, like a third of what I was thinking of.

[cinder silo]

I gotta fix the heat sink gap but so far it looks like the whole place actually works, and that nuclear load balancer solved the power station radiation.

[wintry aurora]

Using any uranium alts? I know you need to use no plutonium alts if you're just sinking them.

[cinder silo]

Not using alts, the whole reprocessing station is there to turn 1050 waste in to 5.25 plutonium rods for sinking.

[wintry aurora]

'k

Surprisingly little iron is actually needed for the whole thing.

[cinder silo]

How much fuel are you producing?

[wintry aurora]

I maximized t off of the 1060 waste number you gave, https://www.satisfactorytools.com/production?share=u4gLOJKKFeSJLWuoKbXn

Probably just the configuration of using nearly all other alts.

[cinder silo]

1050 😄 , I can't get an extra ten with defaults 🙂

[wintry aurora]

*1050

[cinder silo]

Numbers look right, I'm not using some of those alts,

[wintry aurora]

Yea, I just wanted to throw in a baseline.

[cinder silo]

I've also overclocked some areas of reprocessing to reduce the size & complexity of it, 9 blenders instead of 21, same for the plutonium fuel rod manufacturers.

[wintry aurora]

That wouldn't change the inputs though.

[cinder silo]

This place should be solidly jammed up in about three minutes, it's been going full tilt without the heat sinks in the final part haha

[wintry aurora]

Making a steel ingot production area and then shipping the ingots via train elsewhere would be doable, right?

[cinder silo]

Ficsit might not waste but my power station definitely does 🤣

Provided you match up the throughputs the trains should be fine moving ingots, I'm just not good at trains personally.

Watching 1050 waste zoom by every minute is a bit disconcerting.

[wintry aurora]

I thought you were lackig the heat sinks?

[cinder silo]

I am, the system isn't completely backed up yet.

[wintry aurora]

Sounds like a successful test, unless you want it to back up.

[cinder silo]

The main test was on the massive load balancer I installed in the roof of the power station, the startup time was 5 minutes from complete shutdown to all 100 reactors going.

When I do the proper startup after the last step is completed, I'll pre feed 200 rods in to the system and let it go wild.

[wintry aurora]

Won't you die with the rods in inventory?

[cinder silo]

I won't be holding them, those will be belt fed in to a can up to 200, I'm so not handling that personally.

[wintry aurora]

Point there.

[cinder silo]

It is strange seeing a nuclear power station as basically a radiation free zone when going full tilt, the build is working 😄

It's backed up almost totally now, waste barrels are piling up near the reactors now.

I just checked over the backed up system, I seem to have a problem in the acid refinery.

[versed violet]

make an ISC buffer? 40 should be enough for small backups.

[cinder silo]

The facility itself is tuned to require ever so slightly more waste than I produce, the acid refinery is being a problem, it jammed up completely. the vip connection where water is fed in didn't seem to do it's thing.

I was backing the place up on purpose but this test firing just highlighted an issue I didn't anticipate.

[wintry aurora]

Is it possible to have an overflow for the acid? If that's the problem. Sounds like the water is the issue.

[cinder silo]

Definitely the water loop.

[wintry aurora]

Screenshot of the VIP? Maybe you built it weirdly.

[cinder silo]

Let me rip up the floor and i'll cap it

[wintry aurora]

Although it's a pretty simple thing to build and seems like a hard thing to screw up.

I haven't done complex VIPs yet.

[cinder silo]

Ignore the floor hole btw, that is being clipped through not part of the pipeway, I put it there just for looks.

[wintry aurora]

Remove the valve? Might not be needed?

[versed violet]

how about running some refineries on recycled water, or wet concrete, if undocumented feature does not work?

[wintry aurora]

The VIP is your valve here.

[cinder silo]

There are a grand total of three refineries here.

[wintry aurora]

Running an overflow might be helpful, yea.

[fierce ruin]

That is EXTREMELY blue.

[wintry aurora]

Because water.

[fierce ruin]

EXTREMELY

[cinder silo]

Underneath the general acid, the feed going in to sulphuric is all returned from the blenders next door handling non fissile uranium.

That's why the vip, to pre charge the pipes for the sulphur refinery, and hopefully not have it jam up.

[fierce ruin]

Wait if blue = water then what is gray?

[cinder silo]

nitrogen in grey, nitric acid in white.

[fierce ruin]

There's a white?

[cinder silo]

Well off white, but yeah, the two sets of pipes are differently coloured.

[fierce ruin]

This is why I just use the color the pipes give me. 🤷‍♂️

And don't paint them 🤷‍♂️

[cinder silo]

Last time I tried that I had the wrong pipes going to the wrong places.

[fierce ruin]

Sounds like you forgot to measure twice before you cut 😉

[cinder silo]

This place has half a machine too much capacity, I left it as is to prevent backups here,

For the refinery though, the problem is definitely my implementation of the vip, I need to rip it out and go again,

[fierce ruin]

Where is your caution: wet floor sign???

[cinder silo]

Sod the wet floor, I think rad haz in this little back up is more important 😄

[wintry aurora]

Heh, floor does look wet.

[cinder silo]

I need to reload now though, I have the power stations roof to fix since I ripped that out to build the load balancer, and I'd better get started on those heat sinks.

[fierce ruin]

If you sod the wet floor that would be quite effective as the grass would grow pretty well given all the water.

[cinder silo]

If I can't fix the refinery as is, I'll rip the loop out, build another water house + feed and sink wet concrete and be done with it.

The hard part tbh was getting 100 nukes to start, and keep the belts totally free from fuel rods forever, that's a solved problem 😄

[cinder silo]

In a can ready to be injected on startup via belt 😄

[supple belfry]

Any thoughts on transporting nitrogen gas using a lengthy pipe versus rail?

[frosty owl]

Nitrogen behaves "weirdly' with trains (and buffers in general), so just watch out for that. Generally speaking, adding valves can help tons

[fierce ruin]

Package.

[wintry aurora]

Package and drone transport would be the way to go I say.

[fierce ruin]

I always either pipeline or pack and ship.
Location dependent.

[supple belfry]

Great. Onward to battery production. I had been considering a drone route, but was a bit worried about throughput. Looking at the recipes involving nitrogen, it seems that shouldn’t be too much of a concern yet.

[wind spade]

Nah, it's the same no matter the clock speed

[near zenith]

? 36 reactors produces 300 waste per minute using 14.4 ur-rods per minute, but if you crank them to 250%, they churn through those same 14.4 rods and produce 720 waste per minute, which is still maintaining 50 waste per rod, but just making it faster

[wind spade]

They have increased rod consumption by the same amount as they have increased production

[near zenith]

never decreased the ur rod production per reactor number, the 300 waste would come from some lower number

[wind spade]

Not to mention that 250% clocked gen only produces 200% power and 200% waste and consumes 200% rods

But there's no way a gen eats the same amount of rods when overclocked

[near zenith]

all i know is i'm getting the 300 waste/min from the wiki on outputs from 1 fully tapped ur node, idk what proceducre they actually used to get there. the 720 waste from 14.4 ur rods from 600 ur in 36 oc to 250% nuc reactors is what i'm getting both literally in game right now and from your calc to begin with

so the 300 is probably the ishy number

[wind spade]

Point is that if you clock the gens, they would need more than 14.4 rods

If they need 14.4 rods/min at 100%, then they need 28.8 rods/min at 250%

[near zenith]

safely ignore every number i mentioned that wasn't in relation to 250

never bothered to downcalc from there to see what a normal reactor setup would need

[wind spade]

If 14.4/720 is for 250%, then 100% is 7.2/360

[near zenith]

wiki is off in its expected waste numbers then iirc

[wind spade]

That depends on alt recipes used and stuff like that

[near zenith]

yea, it might be based on defaults only, they never actually specified

[wintry aurora]

2,000Mj is how many MW? 2?

[wind spade]

MJ can't be converted to MW

1 MW = 1 MJ/s

[wintry aurora]

Anyways, was just wondering how much I'd produce with turbofuel vs just fuel. I need more power before I do much of anything else and am considering whether to use just diluted fuel or go with turbo blend.

[wind spade]

TF is 2000 MJ, fuel is 750 MJ

IMO turbofuel is only worth if you don't go nuclear later

[thorn bane]

turboblend > fuel anyday
its just way easier to bild

[wintry aurora]

I'd still get the same amount of power if I used 100 plants turbofuel as with 100 plants using regular fuel, right?

[thorn bane]

yes but TF plans only need 4.5 TF/min while normal ones need 12 Fuel/min

[wintry aurora]

I don't know when I'll do nuclear, maybe someday. Not in the immediate future, no.

[thorn bane]

awesome
feel free to make a post about saturated sushi xD

[proven prawn]

Im thinking of doing design work with saturated sushi belts, as basically it would be manifold belts I commonly, only difference would be single belt for production and smart splitters to put the materials where they are needed, but of course I would have to do some more logistic such as figuring out how to recycle where materials where possible, and still would likely need a overflow for sinking as I would be slightly over producing on all production lines but would aim to try to reuse as much of the overflow as possible. Though I think I might aim in the for balanced sushi belts for efficient, with basically very exact production numbers, it would be like your design with the rod plant but use both machine inputs so no one production line can clog up any system, my aim would be with such system to design so no matter what happens, the entire system can be restarted with retweaking the raw material inputs, issue with the rod plant designed in that form is, you have to flush the entire system once its out of sync, so my design goal would be in all of these cases to prevent that from being necessary. But beyond that the basic design goals would be the same.

[still trout]

factorio fun game

[frosty owl]

You're starting to see how the different approaches can compliment one another
Focusing on one specific belting method over others doesn't often lead to the most efficient result after all 😆 (eg: ignoring balancers because "they're always bulky" or avoiding manifolds because "they always take forever to reach max efficiency")

[proven prawn]

yeah for sure, in short I'm definitely going to try to combine approaches more in the future where more complimentary aspects can be achieved

[cinder silo]

Even I learned to use some balancers, even if it is to control the almighty Raz-Haz 🤣

[proven prawn]

Even manifolds technically can't really be done without some form of balancer, even if its a simple as using splitters to belts down different paths, otherwise making single immensely long lines of single line machines would just be so ugly the thought of it makes me want to uninstall the game

[cinder silo]

I made those long lines, a build style I'm aiming to revisit when I re develop my starter base to something more modern, with more splits to parallel manifolds instead of the long drawn out ones.

I'm still making good the structural damage I did when I ripped out the attic in my nuclear power station to build the largest load balancer I probably ever will.

The belts in green are all a part of the same load balancer, the whole thing is too big to be seen in it's entirety, but there is a map of this 1-100 in #screenshots

I've spent the last three hours reinstating the roof space I tore down, most of the beltways will be behind glass or in the middle of steel frameworks when I'm done again!

[frosty owl]

Truth be told, I never made anything bigger than a 270->244 balancer... And that one I never even used for long 😅 😆
#math-and-meta message

[cinder silo]

1-100 a bit much then ? 😛

[frosty owl]

In my standards, yes 😆

[cinder silo]

Its usage case is entirely unique though, equal distribution of nuclear fuel.

[proven prawn]

your designs looking amazing, you have alot of skill🤩

[frosty owl]

Ofc, I just prefer to avoid the splitting (for the same result) by "just" merging the correct amounts instead ^^
Less beltwork overall

[cinder silo]

I was lucky the original splitters from the old manifold were able to be reused, the balancer would have been somewhat more laborious otherwise.

Laying down new conventional factories since the floor paints were added has been a joy, those simple markings make the shop floor look fantastic, and its just simple lines & numbers.

[frosty owl]

Everyone in this channel likes numbers anyway

[cinder silo]

I still have a vip junction to rip out and rebuild at the general acid refinery, blasted thing choked up during the reactor & reprocessing facility test.

[proven prawn]

Sometims I wonder if vips junctions are in reality a bug exploit and thats why it seems like in some cases they don't work as expected.

[cinder silo]

Possibly, there is still a chance down clocking one of the water extractors could work, since both parts of the refinery are linked, fluid loss should be self correcting.

[frosty owl]

If I won't have any VIP issues in my setup, I'll be coming for blood from anyone insinuating the VIP doesn't work if not for use errors
(I think that's the case atm, but haven't troubleshooted much yet)

[cinder silo]

In my next save I have a hell of a lot of learned lessons to apply 😄

[proven prawn]

Just theorizing, I haven't proven vip junctions don't work in normal use cases myself😅

[cinder silo]

If vips did get broken, there would be a lot of annoyed aluminium refinery owners swapping to wet concrete sinks.

[frosty owl]

I totally haven't redesigned my nuclear factory ~3 times just because I felt I would "change too much if I could remake it" (which doesn't make me feel satisfied for a building ai want to be "final") 😆

[cloud swan]

how do I sustain a 480 item production line? I got a perfect match but somehow it`s running short

[cinder silo]

Heh true, mine has had two rebuilds and it has only been onlined four times just for tests, the glass beltways above the power station rows have been reinstated, I just need to fix the centre now 🙂

[thorn bane]

my design would still work 🙂

[cinder silo]

Can't say I've seen pipework quite like that before.

[frosty owl]

If you're trying to feed it by relying on a full mk4 belt, it might be a throughput issue

[cinder silo]

I thought the throughput thing only really affected me and my 10km mk5's ?

[thorn bane]

if his last machine is only running 95% instead of 100% sure
but im pretty sure its unrelated to that

[frosty owl]

Any belt can sustain its max throughout for a single belt segment. Any belt-belt connection makes the items delay a bit (there are workarounds) @cloud swan

[thorn bane]

again unless your machines are running at 98% its not that bug

[cinder silo]

My belts are so long though, merging the sections merely crashes the game out.

[frosty owl]

I'm not assuming, just listing a possibility (one that many don't know about yet :/)

[thorn bane]

999/1000 its not that but just a mistake somwhere...

[cloud swan]

mk4 on the main line and mk2 on the shorter ones

[cinder silo]

How many sections in the 480 before the machine feed?

[thorn bane]

are all the machines filled up?

[frosty owl]

If that mk4 tries to carry 480/min for more than one belt segment, you're not getting 480/min inside the first splitter of that manifold

[thorn bane]

yes youre getting 478
throughput loss is incredibly low on <mk5 belts

[cinder silo]

Hmm, my bauxite refinery and it's six mk5 belts that are 4km long will need a more local bauxite belt injected at the refinery to unscrew the throughput loss.

[cloud swan]

1-6 are filles 7-10 are on par and 11-12 almost empty

[thorn bane]

then you need to wait until all but the last 2 are full
manifolds just take time to fill up and wont runn at 100% until then
if youre impatient you can handfill them

[frosty owl]

||Don't do gambling ||

[frosty owl]

But... The last 2 will keep on starving then

[thorn bane]

its like the occams razor thingy
the simplest explenation (he just didnt wait long enough) is the best explenation
and not a bug that doesnt affect 99.9% of players

[cinder silo]

Oh god I was only guessing at the 10km belt, it's actually slightly longer than that!

[thorn bane]

you know theres trains right? xD

or if you dont like those id make 2 belts with 390

[cinder silo]

Those belts going the 10km thankfully only need to move 350 each respectively, quartz & my last surviving sulphur mine.

[rich zenith]

I have 2 refineries that are producing fuel from the basic recipe of Heavy oil residue to fuel. The refineries produce their load, and then even if they have enough heavy oil to go through the process again, they remain idle for like 5 seconds and then start over. My two thoughts are that there isnt anywhere for the produced fuel to go, or that its just a visual glitch but has anyone encountered similar problems?

[cinder silo]

What are you doing with the byproducts?

[rich zenith]

This problem stems from a fuel power plant im finishing that has 6 fuel generators that im trying to keep filled constantly but they are teetering between having enough and not enough fuel

My byproducts are being sucked out in pipes to the fuel generators, so im curious if there is a way for me to speed up the flow rate in my pipes

call it flux or flow rate idk

[cinder silo]

hit standby on one of the fuel gens and let the refineries fill the buffers, restart it and see if the issue persists, its quite possible you only just meet the needs of them and every time it dips just 1 cubic metre below, a gen flames out.

[shrewd yacht]

is the satisfactory calculator production planner messed up?

get way different results than using greenys tools

[cinder silo]

I don't know, I stopped using the calculator planner when I found out about the tools.

[near zenith]

they might have different base assumptions of what to prioritize

[wintry aurora]

The SCIM one? I've never really used it, tried it and didn't like the method of presentation. Not sure if it is up to date or not, think I heard that it's not up to date on a few things, not sure what.

[shrewd yacht]

ahh ok

[wintry aurora]

This is likely the case.

[shrewd yacht]

I've only used that site for the map

the total use of resources to make some RCUs was insane

[wintry aurora]

Did you accidentially maximise or something? It could also be alts....

[shrewd yacht]

all alts enabled and 10 RCUs per min

calls for 970 crude per min

and 266 bauxite

at greenys tools its about 62 crude and 192 bauxite

greeny does spend 144 quartz vs 42 on the other

[wintry aurora]

How many buildings? More or less?

If it's about the same, then something is certainly wierd with the math.

[shrewd yacht]

no its way more on the other

[fierce ruin]

Weights are subjective so different calcs will do things differently based on how the program is told to weigh things.

[shrewd yacht]

the calculator site uses 69 refineries alone while the total number of buildings at greeny is 64

[wintry aurora]

50.63 crude with all alts actually. Maybe you have an alt or two off.

[shrewd yacht]

ahh hmmm

well then its even more crazy as I'm down to 56 total buildings

wonder if I might have disabled the blender recipes and forget to turn them back on

[wintry aurora]

Total buildings is also 56, percentages probably just got rearranged between the two.

[shrewd yacht]

was 64 before I reselected all alts

didn't bother to scroll in the list to see what I had disabled

what is the most efficient item to sink when considering energy cost and complexity?

realistically that is

[cinder silo]

I'm not sure what the best is, I've been overflowing super computers for ages, and uranium fuel rods for weeks while my final nuclear cycle stage is being built.

[shrewd yacht]

way back before the tier changes I had a a factory with 4 manufacturers making turbo motors that would sink

this time I set up a few refineries making concrete at 50% speed

they are pretty efficient at lower speeds

my most expensive sink right now is the overflow of alu ingots 🙂

[thorn bane]

not sure about enegery cost since nuclear makes power virtually free anyway, and complexity since its subjective but building count wise id recommend Thermal Propulsion Rockets or Turbo Pressure Motors
#math-and-meta message

[shrewd yacht]

trying to understand the awesome sink wiki table

seems the sink value of ingredients is higher than the product it makes for a lot of them

[thorn bane]

btw https://www.satisfactory-planner.net/?f=v3_U5,40000000000000000000003ffffffffffffffffff7fffffffffff,points:per_minute:10,,70380:100:0|28860:243:0|52860:133:0|30900:227:0|11040:637:0|10500:670:0|6840:1028:0|11700:601:0|9780:719:0|2100:3351:0|12000:586:0|0:1:1,0,0:1:0:0
has input coupon points and settings minimize for power efficiency and you can select complexity

[thorn bane]

its the otherway around
for most default recipes the output is 2x the points of the sum of the inputs

[wintry aurora]

There's a bunch that have high numbers and invalid due to having stuff with no sink points, fluids mostly.

[thorn bane]

Rotor + Stator is 140+240 but motor is 760 = 2 * 380

[shrewd yacht]

I was thinking maybe I should sink the alu plates or casing instead, but both recipes with copper combination makes the resources have more sink value?

[wintry aurora]

The highest that isn't labelled as invalid is silicon circuit board.

But that one doesn't have much sink points in itself.

[shrewd yacht]

alclad casing is listed as 2680 resource value while the casing is valued at 393

[wintry aurora]

Great for early/mid game before points get reallu high,

Highest improvement ratio is what I'm looking at.

[shrewd yacht]

the table is way to narrow on the page as well

annoying having to scroll sideways all the time

oh there is an expand button

[wintry aurora]

Super state supercomputer is good,

You can't expand horizontially, you have to zoom out. Blame the sites layout for that one.

nvm, zooming out only works to a point.

[shrewd yacht]

you can

3 buttons on the top left

expand, content and edit

expand will fit to the width of your screen

[wintry aurora]

Oh those buttons.

[shrewd yacht]

yep

not good for reading, but nice when looking at tables like that

I'll probably overflow and sink everything once I get my power sorted

right now I can use twice my generator capacity if its all on 🙂

obviously done full belt setups of things for later that will not run 100% yet anyway

finally got RCUs going again so I can make my first blenders ever 😄

[thorn bane]

wait wtf
tool is suggesting sinking packaged water with the coated iron canister recipe
intersting

[wintry aurora]

Which tool?

[thorn bane]

the one i linked before
https://www.satisfactory-planner.net/?f=v3_U5,7fffcffff9fffbff3fff67fffff9ffedfff7ffffeefffa7ffffff,points:per_minute:10,,70380:100:0|28860:243:0|52860:133:0|30900:227:0|11040:637:0|10500:670:0|6840:1028:0|11700:601:0|9780:719:0|2100:3351:0|12000:586:0|0:1:1,0,1:0:0:1

[wintry aurora]

Greenys tools does silly things with wastewater sometimes too.

[thorn bane]

no but i think its actually good for awesomepoints

7 packaged water is 1 AI limiter but you make 60/min instead of 5/min

it has alot of steps though so its not simple it just uses few buildings

[shrewd yacht]

geeny should recycle the water from the scrap production back to the solution

those would normally be nearby

[thorn bane]

its like super random where it goes it wouldnt look at it

[shrewd yacht]

have to isolate each step in its own calculation

[thorn bane]

ah yes iodine infused filters ofcourse
2718 points per 🙂

[wintry aurora]

Same....

[patent briar]

So I mentioned this in #satisfactory but it might be more appropriate here.

I am discovering a minor flaw in my train system and I need to figure out the cleanest way to handle fixing it. And I think the solution is I have to basically load balance everything in /and/ out

Or to some degree, design my factories crosswise like this:

But as I move up to 3, or even 4 input recipes, I will encounter even worse issues with this problem

This factory above worked specifically because I happened to place my HOR "crosswise" with my Water

[wintry aurora]

Is this a station load balancer?

[patent briar]

Maybe? You have the inputs feeding back into the outputs?

That may actually just "soft" lock your station if you try and use "wait until full"

[wintry aurora]

Of different platforms. Just want to know what a balancer looks like.

[patent briar]

A balancer evenly distributes outputs across belts or pipes. Pipes inherently are already balanced

For example in that pic above, lets say the inputs were 600/min on the left and 300/min on the right.

The outputs would be 450/min each, balanced

[thorn bane]

this makes it so the train loads evenly

[wintry aurora]

I thought you were talking about something coming out of or into the station and wasn't sure.

Or connecting platforms to each other.

[patent briar]

The issue I am realizing is, this pic may make it more clear, this is my HOR factory here, Oil in the left, Residue out the top, and HOR out the right side okay?

Now, I have my trains all set to "Wait until full" and "wait until empty" to make everything deterministic

[wintry aurora]

I thought liquid would balance itself?

[patent briar]

but... I am realizing that also shoots me in the foot here

Looking at that factory above, what would happen if:

Residue
[Full][Full][Empty][Empty]

HOR
[Empty][Empty][Full][Full]

Occured?

[midnight wolf]

Hello, I returned to the game after a long hiatus (I played a lot from Update 1 to Update 3, then stopped because of the continuous game changes to recepies and ratios that could break the factory in the new update), from what I remember the maximum amount of turbomotors in U3 was 156/m (limiting factor was 7800 bauxite/m), and from further researches I found that in U4 the new maximum was 812.89 turbomotors/min. Is there a new maximum now in U5, or is it still the same? Thanks a lot in advance!

[shrewd yacht]

not sure if we measure in turbo motors anymore 😛

[thorn bane]

U5 hasnt changed the recipes at all so they are still the same
however U4 introduced items that are made from turbomotors, Thermal Propulsion Rockets which are more point efficient

[midnight wolf]

https://tenor.com/view/interesting-batman-very-interesting-thinking-gif-16421084

[wintry aurora]

Is this a rhetorical question? Reads like one. I'd guess it'd probably deadlock because it'll wait until the empty ones are full, which never do.

[patent briar]

Exactly, so I have to load balance every input and output to prevent these lockups... I think

[thorn bane]

oh also assembly director systems are a thing and are even more resource efficein t than thermal propulsion rockets (but use ALOT more machines to build)

[midnight wolf]

Thanks a lot! Gonna see if someone has already done the math in order to calculate the maximum achievable

[wintry aurora]

Probably needs some logic saying 'if x doesn't fill in y time, move on'.

[patent briar]

Nah that will just punt the problem down the line unfortunately

[shrewd yacht]

all stops really need some kind of max wait before moving on to avoid trains waiting for nothing forever

[wintry aurora]

To what?

[thorn bane]

if youre interested https://www.reddit.com/r/satisfactory/comments/ph878m/i_have_calculated_theoretical_maximum_of_coupon/

#math-and-meta message

[wintry aurora]

That's the idea, though said more clearly.

[patent briar]

Nah, thats sort of a crutch and not ideally the solution to a 100% throughput setup

[shrewd yacht]

trying to balance the whole system is going to be an insane task

[patent briar]

well fluids are easy to balance

freight though is a bit more annoying but I have a solid 8:8 balancer array I have designed that only takes a couple minutes to construct

[wintry aurora]

More of a failsafe I think.

[thorn bane]

if you only do wait until full and not wait until empty it would still work just at 50% capacity

well its only an issue if HOR is full which means its not 100% throughput

[thorn bane]

got a screenshot?

[patent briar]

https://imgur.com/a/Ms9vbsU

[thorn bane]

thats not a balancer?

[patent briar]

It definitely is

Oh wait hold on

the thing didnt post the whole album

[shrewd yacht]

a balancer would balance any input to all outputs

[patent briar]

https://imgur.com/a/Ms9vbsU

There, I hate how discord takes a whole album and just gives the first pic

[thorn bane]

wtf

[shrewd yacht]

jeeze

[patent briar]

Its easy to throw together because its mostly just stacked, so you can quickly slap down the towers of 8 splitters or mergers

Then you just sew the towers together which you can do pretty fast

[wintry aurora]

That's a really complex looking one.

[patent briar]

8:8 haha

its also tileable, you can make it any n:n

[shrewd yacht]

🙄 I'll take Factorio balancing any day over that monstrosity

[patent briar]

I dont think a factorio 8:8 balancer would be any less of a monster

[shrewd yacht]

its not so bad actually

[patent briar]

8:8 balancers require 64 crossovers

[thorn bane]

3d actually helps alot damn

[wintry aurora]

It does have an advanntage of being vertical.

[patent briar]

yeah keep in mind a large chunk of the space taken up isnt actually part of the balancer

This shows the actual footprint of the balancer itself:

[shrewd yacht]

https://imgur.com/5DGsCAF

I think thats the proper one for factorio

there are so many variations that do different things

do keep in mind that Factorio have left and right side on belts as well

[thorn bane]

could probably build that in satisfactory with heavy clipping/spaghetti
but then again why are you using balancers if output priority is a thing

[shrewd yacht]

can do that system now with the foundation floor holes

[wintry aurora]

Then what's the big stacked thing if that's not part of the balancer?

[patent briar]

which one?

Right also I recall why factorio splitters and ours arent the same

[wintry aurora]

https://i.imgur.com/M6c5DyN.png This big stacked complex.

[patent briar]

Factorio Splitters are also Mergers at the same time, and behave as an inherent balancer if you have 2 belts in and 2 belts out

[patent briar]

that is also in the other pic

Thats part of it, theres basically 2 "Steps" to our balancers in satis

[thorn bane]

well 2 mergers and 2 splitters like the 2 by 2 balancer is a factorio splitter
and if you use 2 smart splitters with overflow to 1 merger and any to the other you get a splitter with output priority allowing you to balance a bus like factorio in satisfactory

[patent briar]

You need to first split your inputs 1:N

Then merge N:1 but inverse

[wintry aurora]

I thought you meant it wasn't part of it, I misunderstood then from the image.

[patent briar]

thats why factorio balancers are a lot cleaner than satis ones, their splitters are also mergers

[wintry aurora]

Plus sushi belts. Though it doesn't really have a failsafe in the same way as we do.

[thorn bane]

i actually started playing factorio recently and found out that the design i use to ballance my bus has been used in factorio for years xD

[shrewd yacht]

priority was not a thing in factorio for a very long time

simplified things a lot

[thorn bane]

2018 xD

but yes with that a main bus is actually super good in satisfactory imo

[shrewd yacht]

main bus was always nice, but you had to do some more complicated load balancing

you should try with angels+bobs mods... you'll get lost easily in the insane complexity

[wicked tinsel]

https://questions.satisfactorygame.com/post/60621724aa0ba107e325529a feel free to upvte in the meantime

[fringe pawn]

Any items that I should add to the list? https://www.satisfactorytools.com/production?share=pUWXA9uaBgdSgNaEk6cB

I abandoned my plan for an all default factory, over 200 constructors just for iron rods and screws for comparable output to the above.

[thorn bane]

i always just do the 4 space elevator parts and use the ingredients they require for building supplies by just splitting off a part for storage
i think id make more alclad sheets but not sure, rest looks fine

[fringe pawn]

I stopped where I did because I'm right at the limit of a single pure bauxite node.

[thorn bane]

ah

[fringe pawn]

That could be worth looking into, though.

I could just make fewer batteries, for instance. I doubt I'll get close to that many drones.

[thorn bane]

recipe wise caterium wire seems a bit odd id use fused wire
also maybe try out automated speed wiring to reduce cable amount but thats super subjective

[fringe pawn]

Caterium wire is strictly a building count reduction, resource efficiency isn't really a concern.

[thorn bane]

hm i feel like fused wire is less buildings if you factor in the smelters since caterium is 15 but copper is 30 (or 100 if you do copper alloy)

[supple belfry]

Signal coming right out of a station, on the junction between the station and track. Kosher or no?

[fringe pawn]

Playing around with how that looks. For now I like that copper only needs to go to powder and sheets. But that thinking may change once I get to an actual layout. Power efficiency is also a factor, and I'm not overdoing it too much. I have plenty of headroom to overclock constructors, but overclocking higher tier machines gets trickier.

[thorn bane]

ah i see

[fringe pawn]

I would consider using pure caterium ingot or other non-default ingot recipes if that were the difference between using one node and two nodes. Or something like that. Copper alloy is a tougher choice, though. That's a small building count, and eating the power cost of 250% clocked foundries probably wouldn't be a big deal there.

[thorn bane]

ye copper alloy is pretty awesome especially with how many iron nodes there are on the map

[fringe pawn]

I'll have some power savings from underclocking manufacturers and particle accelerators too. But I also know I'm going to be tempted to overclock a lot of the oil processing.

[thorn bane]

are you planning on doing nuclear eventually? cause with that i dont think it matters what you overclock xD

[fringe pawn]

No, this is intended to be a self contained factory on its own grid. Possibly even at a single location. It's under 1,000 buildings, so bringing everything to one place should be doable, even on my modest hardware (7700HQ/1060 MaxQ).

I might get fancy and use geyser power for the power infrastructure.

[thorn bane]

ye i did something fimilar with 4:4:1:1 SE parts in 1 place
#screenshots message
definitely alot of fun

and then you just build 30 of them

[fringe pawn]

I picked 3/3/.75/.75 because it still finishes the final load in just under 24 hours (about 22 and a half).

There is something novel about perhaps instead using 4 250% nuclear power plants instead of the 120- turbofuel gens, though.

[versed violet]

If you lower that by 1/3, you will have exactly one accelerator, and a bit less fractions on rest

[thorn bane]

ye you can see my 7 dedicated nuclear power plants in the back in my screenshot

[versed violet]

It will take around 36h to make all the space parts, but you are going to spend 100h building and designing this thing.

[fringe pawn]

36h is still a very respectable finish time, ultimately everyone's coffee cup looks the same 😄

[thorn bane]

problem is once its done youre kinda just waiting 36h cause theres nothing left to build xD

[versed violet]

if you have nothing to build, you must be doing something wrong

[thorn bane]

thats why in U5 i did all but ADS in 4:1:1 ratio and then build alot of ADS (20) while waiting for it to fill up

[thorn bane]

if you pre plan a project like this thats selfcontained in all manners including power its kinda hard to add stuff once its done

[fringe pawn]

One thing is for sure, I don't think I'll be doing anything close to my 20/20/5/5 factory again.

[thorn bane]

ye 20 ads was waaayyy too much for me xD

i might make a playthrough with a bunch of TPRs though they seem fun to make

[fringe pawn]

The other thing I notice about nuclear power is that it tends to make all my numbers go haywire. If I were to do 8 nuclear plants at 100%, I'd need 5.333 electro rods and 1.6 oscillators per minute, and a lot of the numbers further back in the production chain get wonky as well.

[fringe pawn]

Though I suppose there's an argument to be made for having the waste disposal capability for completeness.

[near zenith]

see, those are reasonable people times, at least you're not being dumb like me, cause i'm making a 20:20:5:5, to finish the thing in 200 minutes

we won't talk about the fact it might take me 2 weeks to build this nonse

but that's what winter break is for

ahh, scrolling up a bit i see red has done this

[fringe pawn]

I feel like once you cross 4:4:1:1 you pass the sweet spot where you actually start losing time overall, in terms of completing the final delivery

[near zenith]

did you survive with your sanity?

[fringe pawn]

Yeah, it's powered by 300 uranium. I don't regret it, but I'd also never do it again. Like I said, everyone's gold cup looks the same in the end.

[near zenith]

meh, what else am i going to do in this first world i've slapped together. I got inefficient starter factories, mk1 miners, bad railroads

this is the coup de grace of the learning world

[thorn bane]

thats why i started doing builds for awesomepoints instead (like making 20 ads/min)

[fringe pawn]

I don't even use trains at all. They feel like a time sink overall.

[near zenith]

but.... choo

[fringe pawn]

If I ever incorporate trains, it will be purely for looks. Otherwise the world just isn't big enough and resources aren't spread out enough to really justify their use IMO.

If you look at my current plan, that isn't really a lot of nodes you need to bring together. Only 1 or 2 nodes of every type of resource required. Except copper, which needs 1900. So why use trains? Even on my 20/20/5/5 build, I've posted pictures of its map layout and asked where people would use trains, and I get a lot of 'mehs' as a response.

Here's the 20/20/5/5 factory, btw: #screenshots message

[near zenith]

yea, i'm doing a self contained dune desert factory, the only train is baux from the eastern edge of the red forest

i'm going to be very interested to do a full factory tour when i'm done here, seeing how all my factories improved over the course of play

[thorn bane]

i had a train to dune desert for 2200 steel pipes and a train to gold cost for ~3000 rubber+plastic

[fringe pawn]

Looking at how nuclear power would change a 3/3/.75/.75 factory, I'm inclined to stick with turbofuel. Even though it's higher building count, all the belting and piping for nuclear power doesn't seem easier.

The other thing working against nuclear power is that in a real game, you'll already have existing power infrastructure before you get there. I'm thinking about tweaking my factory to make it feel more progression based as well, possibly even incorporating coal power.

Of course future particle accelerator recipes may drastically change power needs, but that's for another playthrough.

[frosty owl]

...why are you publicizing duplicates?!

[midnight wolf]

Same, belts all the way

[fierce ruin]

trains look cool

[tidal thorn]

any good guides with all input output amounts?

[fluid cave]

Wdym?

[wicked tinsel]

the other version is classic QA site overkill that will never get implemented 😦

[wintry aurora]

All recipe inputs and outputs on a single page I guess?

[frosty owl]

@cinder silo What you said isn't wrong, but I think @thorn bane's point wanted to be that it's generally not worth MAXIMIZING uranium processing if your not using all the resources on the map too (lots of work for power you'll most likely not use)

[cinder silo]

True, In my case I'm essentially replacing a 133.2Gw turbo fuel plant, and then wanting to rebuild my main base on a larger scale, followed by automating the final space elevator parts, having 262Gw is plenty enough, I can even dismantle my legacy coal station too.

I do need to iron out an issue in my general acid refinery before swapping exclusively to my nukes though.

[thorn bane]

i actually did a playthrough in U4 only using 1/29.6 of resources (~3%) to make 4 ADS 1 TPR 1 pasta 4 magnetic thingy and had enough power by using 71 uranium (2100/29.6)
you just get sooo much power out of it its insane

[cinder silo]

The coal thats going to the turbo fuel AND my legacy coal station can all be moved over to solid steel.

My nukes only generate 262,500mw, I won't need anything else power wise, like ever, in addition it's clean burning thanks to plutonium sink.

[thorn bane]

thats not "plenty enough" thats 20x as much as you need xD

[cinder silo]

I'm currently using nearly 70,000mw right now, and I don't have the automated space elevator nor the expanded steel production yet.

[thorn bane]

well thats max consumption
unless youre sinking lots of stuff and wasting power then the consumption should be quite a bit lower

since its impossible to build stuff faster than 70GW of machines are producing

[cinder silo]

If anything consumption will be significantly higher when I hit a bunch of on-switches across the nuclear processing site.

Having a power source that I know I'll never need to revisit though can only be a good thing, I can focus on building everything else, and likely significantly reduce the objects in my game by decommissioning and dismantling my older power stations (especially the spire coast installation)

[livid leaf]

This is the big brain channel that I can't be a part as of rn XD

[reef belfry]

F for my fingers, just hooked up 260 assemblers, no breaks

modular frames, 520/min output

[thorn bane]

thats alot of frames

[patent briar]

What is the setup to properly load balance 4 mk2 pipes? Do you just loop connect em?

I presume just this isnt enough right?

A   B   C   D
||  ||  ||  ||
||==||==||==||
||  ||  ||  ||

Do I also need to make a fourth pipe that connects A<->D?

[oblique notch]

Don't load balance. Just make sure the pipes are full

Plus, one mk1 pipe can't actually support 8 generators (6 and a bit extra)

(Mk1 pipe has max flow rate of 300m^3 and a coal gen requirs 45m^3 a minute.)

Or was it 50. Hmm. Been a bit

It's 45. 8 gens needs 360 water which is 3 extractors at 100% but a mk1 pipe can't support that.

If you aren't strapped for space I'd do 1 underclocked to 90 extractor directly commented to two coal gens.

Or 3 set to 90m^3 to 6 gens, if you want to get the most out out of the pipe

[patent briar]

Not an option, these are pipes coming out of train wagons

[oblique notch]

Huh?

[patent briar]

has nothing to do with generators, these are pipes of Heavy Oil Residue and other things

Fluid Wagon stuff, so it /actually/ needs to be load balanced

[oblique notch]

Oh. I thought you were asking about coal

Then overclock/underclock your refineries

[patent briar]

Nah, this is for Plastic/Rubber stuff, and Diluted Fuel, and other stuff

[oblique notch]

It just becomes 100x easier the

[patent briar]

Not good enough, it doesnt fix the wagon overload/underload problems

[oblique notch]

Buffers and look in thr Piple

[patent briar]

The pipes /actually/ need to be load balanced because I am using trains to move them

Buffers are also not the solution

[oblique notch]

Piple buffers, not liquid storage

[patent briar]

I did a whole write up on this earlier but basically the problem lies in the behavior of multi-wagon "Wait until Full" / "Wait until Empty" trains

specifically, they wait until all their wagons are empty/full

And when using multiple trains you can run into a shitty "block" problem where you lock up a factory if its inputs and outputs are both to train stations

[thorn bane]

i dont think you should do wait until empty

[oblique notch]

So basically you need 3 things

1. storage for the fluid as it comes off the train, enough to store the fluids so the machines will run constantly despite the big gaps of fluid input due to train travel time

2. extra storage so you aren't having to time the trains to be 100% perfect (an impossible task given the mk2 loss and the load buffer loss)

3. overflow valves on the storage and so.e way to handle that overflow into something sinkable

At least that's how I would divide up the problem.

You plan foe your trains to bring more than you need for the interval between trains - as close as you can of course, but still at least 5% more than you

Then using some of the gravity assisted overflow designs in thr Piple you let the excess in the last storage spin off into some refinery or two that can produce something sinkhole.

Anything else would probably just be an exercise in frustration (ie trying to be exact)

[thorn bane]

have you tried that setup? im curious if it works

[patent briar]

It shouldnt no? You would get this wouldnt you?

A   B   C   D
0   0   0  600
||  ||  ||  ||
||==||==||==||
||  ||  ||  ||
75  75  150 300

[thorn bane]

well that wouldnt be an issue for the deadlock since it just wont consume HOR on the last 2 pipes
also im pretty sure it would split more evenly since its a liquid (the exact split is pretty random though and depends on how the fluid is flowing at that moment)

[patent briar]

well that wouldnt be an issue for the deadlock since it just wont consume HOR on the last 2 pipes
Thats precisely what causes the deadlock

The issue is the way the "Wait until Full" and "Wait until empty" behavior is

the train waits until all wagons are empty/full, which means if you have a hiccup in production that causes uneven distribution, it "locks" the whole factory

[thorn bane]

again dont do wait until empty
but anyway it would still unload though

[patent briar]

Consider this scenario:

Oil In, 4 wagons
[Full][Full][Empty][Empty]

HOR Out, 4 wagons
[Full][Full][Empty][Empty]

again dont do wait until empty
This would remove the determinism of the production line and make trains no longer have predictable throughput and ruin everything

[thorn bane]

yes if they are connected like you said the first 2 refineries wont produce so they wont take oil meaning the right ones would get oil

[patent briar]

Well if my outputs and inputs are load balanced it works, but what happens in that setup if they arent load balanced and just standard 8x manifolds?

[thorn bane]

aslong as you have wait until full it is perfectly predictable

[patent briar]

Only if each train only unloads once per load, or loads once per load, and you are draining faster or equal to the throughput of the train

As soon as you are trying to split your resources to multiple outputs, it wont work

Or well, you can theoretically still do splits but you would be forced to create "Splitter" depots, where you have 1 input station that splits to 2 output stations, and then 2 trains become 3 trains

[thorn bane]

?
"does this load balancer work"
"if they arent load balanced it wont work"

[patent briar]

Im highlighting the "what locks things up if I dont load balance my stuff" problem

[thorn bane]

so load balance them?

[patent briar]

Right so, Im trying to find info on a "proper" load balancer that actually works

I guess theoretically I can just slap valves on em

[thorn bane]

how exactly are you handeling multiple stations?
cause i did go to source station -> go to dest 1 unload ->go to dest 2 unload
and because it will just unload how much is missing in dest1 it will have left over to bring to dest 2

[patent briar]

You usually need minimum 2 trains for high throughput stations no matter what

So you can just send Train A to output station A, and Train B to output station B

[thorn bane]

i just did 2 trains with the same path
if train 1 just docked in dest 1 that means that train 2 will bring everything to dest 2 since it didnt unload much in dest 1

[patent briar]

Especially for fluids, at close to max throughput a platform fills up in just over 1 minute, so you need to have a train almost always racked up ready to go

[thorn bane]

yes you just have multiple trains with the same path

[patent briar]

Every train that visits a station takes a big chunk out of its throughput

You want to get as close to 1 train per station

[thorn bane]

you can have 10 trains and it works
because they just wait in line until they are full

[patent briar]

No that takes massive chunks out of your throughput

[thorn bane]

for the output? yes but you cant have max throughput there anyway cause you have 2 stations
for the input? no since they will just wait until full so its always the same time

[patent briar]

For example, using Mk5 belts and 50 stack size items, every train you have visit the platform chops off over 100 items/min per belt

[thorn bane]

but its not more trains/min
they just wait

[patent briar]

The issue is your ouputs, you have 2 trains now dropping off non full loads into the depots

You never want a non full train dropping off at a depot, you just hamstrung your throughput

[thorn bane]

yes if you have 580 input going to 2 stations you CANT have 580 going to both

[patent briar]

I suppose its still within the max throughput of the output stations, because a double blocked station still has >50% max threshold of a single blocked station

[oblique notch]

this seems backwards. The whole idea of trains is that you can increase thruput by simply adding more trains.

[patent briar]

More trains increase throughput but lower max throughput

[oblique notch]

Im ... not sure how you arrive at that.

[patent briar]

Trains block the inputs when docked

[thorn bane]

more trains/min at a station reduces throughput
more trains in a system doesnt if you have wait until full

[oblique notch]

the inputs to the stations?

[patent briar]

The more time the station spends being blocked, the lower its max throughput becomes

[soft scarab]

If you have a train unloading 100% of the time you have 0 throughput

[patent briar]

yes, inputs and outputs are blocked when a train is docked

So you can compute the "true" max throughput of a perfect platform by assuming absolute minimal dock time to achieve "just in time" docking, which is easy to do by simply using the "wait until full" / "wait until empty" option

When you set a train to "wait until full" it will pull up and "pseudo" dock, but doesnt block input/output, and waits until:

Wait until full: The platforms have enough material to fully fill all wagons

Wait until empty: The platforms are empty enough to empty all wagons

[thorn bane]

wait until empty just makes 0 sense to me

[patent briar]

Then when it crosses that threshold then it truly docks and performs the load/unload

[patent briar]

I think you may be able to get away with not using it, but Im crunching the numbers and not sure

[soft scarab]

What if some platforms are set to load and some are set to unload?

[patent briar]

You can simply just imagine each platform is "ready" or "not ready" based on what you set it to

[thorn bane]

you cant unload faster than you load so it will always be bouned by the "wait until full" time

[patent briar]

Huh? What makes you say that? Loading and unloading take basically the same amount of time

[thorn bane]

exactly you cant unload faster

[soft scarab]

wait until empty makes more sense if you are having a train deliver multiple station’s worth of outputs to one location

[patent briar]

I think that will however just propogate the "soft lock" problem to be the previous / next factories

In my prior example, the train would leave but now its wagons are:

[Full][full][empty][empty]

And when it arrives to load, it wont pull a full load, so that factory will have uneven draw, and that input wagon will have uneven draw...

[thorn bane]

well you have to make a choice both work but not both at the same time
do you have trains waiting at inputs until full and have a dedicated train for each output
or do you have trains waiting at outputs until empty and have a dedicated train for each input

[patent briar]

It would just propogate the issue all over and you'd have a hell of a time isolating where the actual issue is originating

[thorn bane]

yes thats why you load balance

[patent briar]

You would know "Well somewhere at some factory, an issue is happening on line #3 and/or #4""

Right thus thats the conclusion I came to, load balancing the inputs/outputs should make everything go smooth and accomodate for any hiccups

Theres another fix btw, but its tricky to do if you have >2 input items

[thorn bane]

its just that production at 50% is better that a complete deadlock

[patent briar]

This is the other theoretical solution that sort of "just works"

Its a pseudo "load balancer" that basically distributes the loads inversely, so if you have uneven input it still evens the output out without having to load balance, the machines themselves are the load balancing logic

But unfortunately once I go past 2 inputs, it gets a lot tougher

for 4 inputs I can "sushi" the trains theoretically

But if I need to use the bottom section for input, I have to load balance that one

[ionic torrent]

with trains, i use a caching method using the mk2 storage with 1 input and 2 outputs into the station. It seems to work most of the time for me. you are only limited to 1 belt of items/cart.

[fringe pawn]

Has anyone ever tested the claims that manifolds don't work? Maybe something like 300 iron into 10 smelters, with a balanced and manfiold setup sending the ingots to a sink? Then have an overflow splitter to catch backed up ore (if any) and put it in a container?

I'll set that up now before turning on a movie.

[fringe pawn]

Here's how I laid out the 10 way balancer. I'll be plugging both miners in at the same time once I double check that recipes are set.

[cinder silo]

That 10 way balancer looks way more managable than my 100 way balancer 😄

[fringe pawn]

Starting it up now

[cinder silo]

The 100 way balancer! , It has only one use case scenario, feeding nukes their fuel without the need to have thousands of fuel rods sitting on the map because of how manifolds work.

[fringe pawn]

I actually do balance my nukes, but by way of having 1 manufacturer per nuke plant, and those are fed with a manfiold.

[cinder silo]

I didn't think to set up the manufacturers in such a manner, that 100 way was the result of a massive retrofit, it's essentially eleven 5 way balancers, with a lot of two way splitters connecting them , and the ends to the nukes.

After testing, 100 fuel rods in, each nuke gets 1 😄 , no deviation.

Suffice to say after dealing with radiation mitigation, I got a new appreciation for load balancers under specific circumstances, even though I primarily manifold.

[fringe pawn]

I embrace radiation

#screenshots message

I've got thousands of spare fuel rods stored just in case.

[cinder silo]

I aim to only have like 3 at most in a nuke at a time, though I am building a massive emergency power system in the event something goes catastrophically wrong,, four towers with 20x20x20 power stores fed by geothermals (and initially charged off the main grid.

[fringe pawn]

They actually came in handy when playing early U5 deleted my primary uranium node.

[cinder silo]

I had a uranium node and several bauxite nodes decide to ghost on me after U5 dropped, the mine was right there but it really wasn't,

It will take me a while to place 32000 power stores haha.

I still haven't even picked a site for the monsterous backup yet, away from my nuclear for sure, my game stutters like a bitch when I'm there for some reason and half of it is offline 😦

Worse still, when I watched the process monitor, my cpu actually drops during the game freezes, like why?

[fringe pawn]

Highlighted machine is third from the end and just filled its hopper to 100. So if the system is going to back up and lose efficiency compared to the balancer, it should be relatively soon.

[cinder silo]

You having manifold issues?

[fringe pawn]

Nope. I was just browsing Reddit and saw some weird beliefs about both manifolds and balancers.

[sinful knoll]

Aw gross you can make that last one 90 degree bend

[fringe pawn]

That'd interrupt the test at this point.

For the time being I don't see any indication the last 2 machines are backing up. No ore in either overflow container. Plugging in a movie (Dredd). 😛

[sinful knoll]

:p

[cinder silo]

Some people can be die-hards defending their positions, manifold vs balancer will be a debate going on long after these tests are done.

I normally make a complete hash of balancers myself, It's only handling radioactive materials that caused me to revisit and redouble my efforts with balancers to cut radiation.

[sinful knoll]

Balancer for nuke, manifold for everything else

[cinder silo]

That's what I'm doing right now, though in some situations when I can make it compact, I will run balancers elsewhere for quick startup.

[sinful knoll]

Since it takes so damn long for nuclear to catch up

[cinder silo]

In addition to the long nuclear startup, one end of a manifold ends up more radioactive than chernobyl, and the other end is starved,

[sinful knoll]

😂

[cinder silo]

As such, this was formerly fed with a manifold, two belts splitting 50 ways, is now fed by a 1-100 balancer, and the time for startup difference was immense.

[carmine stump]

Do you put a pure node with mk3 miner oc‘ed to 780 on a single belt or is it better to split it right at the miner?

[sinful knoll]

Still outs 780 max either case

Something something throughput is broken and you will never get 780

[carmine stump]

How much of the 780 throughput is „lost“? Does it matter or just ignore it and it‘ll be fixed at some time?

[sinful knoll]

I'm hoping the latter

I'm ignoring it too :x

[fringe pawn]

Load balanced setup and manifold setup both still working perfectly. Machine levels have been steady for over an hour.

[frosty owl]

Link?

[fringe pawn]

There's a recent thread with a meme image. I'd go dig it up but I'm pretty sure I remember you have multiple posts in it already. Unless you have a different Reddit name.

[frosty owl]

Ah that one, gotcha

[fringe pawn]

Nothing stood out from it that we both haven't seen claimed before

[frosty owl]

I didn't see any particularly blatantly wrong info thrown around (other than maybe 1 claim), but I haven't kept up with the convo much 😅 😆

I don't interrupt if people shit on balancing or manifolding, as long as they don't fake data

[fringe pawn]

I see one person who I recall has already been downvoted into oblivion, claiming they saw manifolds 'desaturate.' I've also seen people claim that feeding a split back into a source doesn't work for creating balancers, but I don't think I saw that one in there.

[frosty owl]

Classics
"My math works out, it's the game that is wrong!"
"Did you try checking for MK1 belts?"
"... My math works out!"

[fringe pawn]

"But on Youtube...."

[frosty owl]

Bruh, I got told it's too "bold" to assume people know the meaning of something "just because the wiki has an entire page dedicated to it" 😆
Apparently YT was the first result for their search 😬

This is actually quite misterious to me... Why so many people "miss" the wiki and other official resources ending up on (eg) sites with old versions of the map and/or recipes or werid YT videos...
Doesn't the wiki show up high enough on the results or something 🤔

[fringe pawn]

Google probably favors video content

The balancer page looks pretty good, some effort was put into it

[frosty owl]

The balancer page on the wiki?

[frosty owl]

What if we include short videos in the wiki...

[fringe pawn]

That'd probably be useful, even besides SEO. And there are good arguments to write the pages with SEO in mind so that people aren't directed to wrong info.

[frosty owl]

SEO?

Satisfactory Entry Only?

[fringe pawn]

Search engine optimization. In other words, writing content so that it appears higher in search results.

There's a whole industry behind doing that

Wikis by their nature should be pretty high in results, as long as they're formatted well and have proper page titles. The last step is having key terms in the text body as much as possible.

[frosty owl]

Yeah, I have an idea of what that is, I'm just bad with acronyms (and too lazy to look it up 😅)

[fringe pawn]

I used to write SEO content. It's ruining the internet in a lot of ways

Garbage sites like eHow were just the beginning, it's only getting worse.

I can't bring myself to stare at the setup, but I'm pretty sure it's been 100% green lights once the manifold filled, too.

[frosty owl]

Internet is still very much a Far West, first come first served (as in, the first to find an exploit rips the benefits)
||Autocorrect trying to correct "rips" in "RIPs" ||

[frosty owl]

Use the power draw to judge the efficiency exactly. You might miss a light changing color, but you can't miss a dip in the graph ^^

Well, as long as you check the graph frequently enough 😅

[fringe pawn]

What's the time frame on the graph?

[frosty owl]

~30s iirc. Maybe 1 minute

[fringe pawn]

I suppose it doesn't matter. Now that it's been running perfectly stable with no changes to hopper amounts for hours, that doesn't seem possible. I've paid particular attention to the last 2 manifold machines, and their hoppers have never been in danger of hitting 0 any time I've checked. Generally right on 2, +/- 1

Actually, not even +/- 1 now that I'm looking again, it's generally on 2, occasionally ticking up to 3 very briefly like a clock. If it hovered at 1 at all I'd be nervous.

On both the last 2 smelters in the manifold and all 10 of the balancer, all hoppers seem to have a predictable rhythm where they briefly tick up by 1 item. The first 8 in the manifold stay at 100, but you can sometimes catch a frame of them being at 99.

[frosty owl]

Speaking of frames (funfact): smart splitters actually output on overflow when the overflow stack reaches 4. You can see it for an instant with the dismantle tool

[fringe pawn]

So the internal buffer needs to fill, then it will disgorge the entire buffer?\

[wintry aurora]

I’ve seen it overflow smaller amounts, so, I guess just overflows beyond that.

[frosty owl]

No, it just outputs the 4th item in buffer (dunno if FIFO or LIFO), so that the buffer is always at 3 (but you can still catch a glimpse of the moment the buffer reaches 4 xD)

[wintry aurora]

I thought the buffer was 9 or something?

Or that’s a separate internal storage?

Although I don’t think it’s quite that big….

[frosty owl]

To clarify, I used a smart splitter to send possible overflow to a container, testing to see if a belt sustained 780/min. If the belt doesn't, you notice the splitter slowly piling up items on inventory before finally reaching 4 and outputting 1 item on the overflow belt

Funny enough, if the belt MANAGES to carry 780/min, the splitter may still pile up items up to 2/3, but I've never seen it output overflow and sometimes the items in the splitter go down to 1....

[frosty owl]

That might be the TOTAL buffer (eg: you can cram 9 items inside a merger, but the overflow option works on a 3-items buffer)

[wintry aurora]

I see.

[oblique notch]

are people still going on about this "bug"?

[sinful knoll]

Yes

[cinder silo]

For as long as it remains unresolved of course people are going to "go on about this bug"

[oblique notch]

🤦‍♂️

I dont even have the enrgy this morning to deal with that. never mind.

[cinder silo]

Good job because if you aimed to white knight a known issue, you would have been disregarded anyhow.

[oblique notch]

White knight?

what the fuck are you on about?

never mind. dont answer that. Im out.

[sinful knoll]

Bye~~

[cinder silo]

Bye bye now!

[wintry aurora]

Something something chess, I dunno

[cinder silo]

With his reaction after the going on about "the belt bug" meh, I can see he would have defended leaving that to the hilt with no movement, I cba with thick headed people.

[wintry aurora]

I’d call that being a Don Quixote maybe.

Although that’s more ‘wild goose chase’ than ‘imma die on this hill’

[cinder silo]

The Don Quixote is more eloquent and would go over most peoples head 😄

[wintry aurora]

True, it’s a little obscure too.

[cinder silo]

😄

Hmm, after I started to move my hub, space elevator & storage, I have a horrible feeling the new site is liable to be included in the spire coast changes.

[thorn bane]

imagine not restarting when they change that

[cinder silo]

I'll restart when I feel I've done what I want to in my existing save that's been going since update 3.

[thorn bane]

ye that usually takes the time of experimental for me xD

[cinder silo]

I do hop over to experimental and see what changes do what, aside from some hilarious bugs and weirdness, the new toys added a ton of value to my existing save 🙂

Being able to draw on the factory floor to quickly imagine a layout has been awesome, and it makes for a cool looking factory when completed to boot.

[thorn bane]

update launches on eperimental: hard core gaming 14h per day for 1 month
update launches on EA: burnt out for a second playthrough so just backseat gaming in discord for 3 months
good stuff 🙂

[cinder silo]

I have burned out a few times, then I hammer grand theft auto 5, diablo 2/3 or stellaris for a bit, only to come back because things need built ❤️

The moment I realise I built mergers down the centre line of a manifolds inputs.

[sinful knoll]

And the white knight is talking backwards and the red queen is off with her head

Remember what the doorman said, feed your head

[lofty edge]

Will 27x32 foundations be big enough for 89 fuel generators and the necessary pipes and power connections?

[near zenith]

general rule of thumb, always build it bigger than you think you need

[sinful knoll]

Isn't a fuel gen like 2x2 already

3x3

[lofty edge]

2.5x2.5. 10 in a line = 25 (+2 for waking around). Each row is two lines, 2.5 + 2.5 (+ 1 for pipes) = 6. 6 x 5 rows (90 machines, I will leave one spot empty) = 30, + 2 for walking around = 32.

[sinful knoll]

Math, not even once

[wind spade]

SF launches in 2018: backseat gaming since then and making mediocre tools

[oblique hollow]

mediocre but still better than average xd

[wintry aurora]

Thought here, if I split a MK5 belt immediately (as immediately as you can ingame) after a miner doing 780, will the belt bug still happen on the segment that is between the miner and splitter?

[wind spade]

IIRC it won't

[wintry aurora]

'k

I do need to split it anyway, and I have an idea actually for mixed belts as I'm doing https://www.satisfactorytools.com/production?share=dYIbZ7pp9MjNwzbvethW I don't know for sure what I'll do after steel ingots, was just exploring the numbers there.

[frosty owl]

Stfu, the improvements on the stats for machines and such are actually awesome (I just happened to discover them yesterday)

I already spent 2 hours in the tool just because of how well they let one analyze the different solutions

[frosty owl]

Welp, you know who to ping for help xD (also, I confirm splitting a belt allows you to get 780/min from a miner. Just don't try moving 780/min from ONE belt to ANOTHER belt, as belt-belt connections cause the issue)

The throughputs involved don't seem very favourable for mixing though, unless you're planning to split the productions in smaller chunks

[wintry aurora]

I'll have to split it some way or other given the size of it anyway.

[frosty owl]

If the game had some useful machine stats... I can't think of anything atm (unless they fix the efficiency display or add a power graph for single machines too)

I'd be much more interested in something similar but for belts or smort/poggers

Knowing sometimes requires to take cumbersome notes that could be avoided (plus adding some nice infos) by having what you mentioned

Imagine having to go to your notes whenever you move from one programmable splitter to a new one. Sounds like needing to look up your notes to know what factory you're walking in because nobody added signs since "if you don't know which building is which, you should've planned better"

[pure haven]

what would be the best ratio between production and power (turbo fuel) with oil? (i've got 8100m3/min)

[placid cipher]

I have to ask... I am making a stupid large Pure Ore Plant and i did some math just now... Does it make sense even to try and wash ores if i can't even build as many refineries as there is materials from nodes coming anyway? Should I just make is all smelters and use the normal recipe?

[frosty owl]

That's making assumptions that I don't share.
I dislike taking notes whatsoever. Even worse if they're out of the game.
I wanna show off factories and I might want to change the design later on. I don't assume I'll never rework it, so ways to show info in game are much appreciated. Also, troubleshooting

Might be appealing to you, but doesn't cover my preferences

With poggers and such, there's plenty of ways to change a design without needing to rebuild much, thus making a complete rebuild way too much

I use them, but the least I can

Wdym?

[wintry aurora]

What's the link to that yet another satisfactory planner?

I want to see if it has overclock controls or something.

Nvm, found Zyras post from a few days ago.

lol!

[near zenith]

does anyone have numbers/a spreadsheet handy which has maximum theoretical throughput for a truckstop relative to distance travelled?

i know they load/unload at 120 stacks per minute, and i dont think they suffer from the load/unload lagtime of trains

but before i dive into average speed calcs, wondering if someone has already done the math

soooooo, like exactly what i asked for

good contributions to the discussion

👍

relative to distance traveled

delightfully unhelpful, yet technically correct, you remain on form sir

you focus on the specifics and ignore the context

[frosty owl]

Reminder of the rules about being respectful, polite and helpful~

If one doesn't feel like they can abide, silence is always an option ^^

[near zenith]

yea, i'm stepping out, i'm saltier than i was expecting, and i should go find out why

night

[fringe pawn]

Well that was an interesting read O_o

[sinful knoll]

Math, not even once

[wintry aurora]

Is this kind of behavior normal for sushi manifolds? Everything else behind this one is filled up to 100 units each, but this one isn't filling up completely. It actually seems stable, though flickering around 9 to 11 with the copper, occasionally dips into 8 for a miisecond, but always recovers to the previous range.

Also, I've set up a feedback loop since there is a bit of leakage and keeps it in the cycle. Going to see if I can modulate that with programmable splitters though.

I've been watching that for a while and I'm 99.9999999999 sure that it's stable, but I'd feel better if it was filled up.

Oh, I thought it did.

Maybe I wasn't paying attention.

and I know that, that's the purpose here.

Setting those last two to overflow did the trick though.

[fringe pawn]

After my experiment with iron ingots not long ago I'm actually quite pleased with the game's precision. Even if it's not perfect over the course of hundreds of hours of runtime, solids are close enough that I would be comfortable saying people shouldn't worry about it.

That makes sense. I've yet to attempt sushi so I can't comment more than that.

Oh dead lord 😬

Way to go voice to text 🤣

[wintry aurora]

Yea, without a copy paste function, I don't see myself doing this on a large scale.

[fringe pawn]

If they do implement splitter copy paste it'll be interesting to see if sushi manifolds become quick enough to be an obvious choice. That is a lot of cut down belting for four ingredient manufacturers.

[wintry aurora]

Very cut down belting.

[frosty owl]

🍿

[crystal charm]

what's the most fuel efficient method of production

if you don't have turbo fuel

i.e oil to fuel? or oil to HOR and then to fuel?

[wind spade]

Oil - alt hor - diluted fuel

[crystal charm]

i don't have any oil/fuel alt recipes yet

not sure why, just bad luck i guess

[cinder silo]

Ahh nuts, I just realised when I do start demolition of my turbo fuel station, that's going kill my fabric>filter plant.

[oblique hollow]

Im surprised the wiki doesnt list how long machines idle

When output is full or not enough input

[cinder silo]

Two more little structures to go, then I gotta go back to the acid refinery and find out why it clogged during my last nuke test.

[native marten]

what exactly is a sushi manifold?

I don't think I've ever heard that term before

[cinder silo]

mixed belt that is in a manifold configuration.

[native marten]

oh so the belt contains more than one kind of item and uses smart splitters

I'd considered using those but realized I'd want sinks at the end of the line to prevent clogging and reconsidered until I get to a much later game stage

[cinder silo]

Some of my more complex lines could do with being converted to sushi, I just haven't figured out how to implement it without some inputs being heavier than others and blocking the whole thing.

[native marten]

that was exactly my hesitation as well

if you have a balanced line like Steel or the alt steel recipe that uses iron ingots instead of ore, the ratio remains the same on both lines so it should technically work cleanly at all times

anything that's top-heavy like something that requires 1 reinforced iron plate but 12 steel beams, well...

[cinder silo]

I love the floor holes for conveyor lifts, that's let me tidy up building intakes & outputs in a manner I couldn't match with the wall holes.

[frosty owl]

I think the thought of having to change one's design to accommodate for (even if temporary) a sink connection intimidates many

[cinder silo]

The first place I'll try it then will be at my starter base when I get the rebuild underway 🙂

Products like computers/super computers & hmf will be ideal trials, and overflows can go to the warehouse (then a sink when that's full)

[spice hare]

Hey can anyone tell me how to efficiently make the modular frames

their recipe and the way they get made in the assembler is just total ass

i aint making them by hand thats lame

i figured out i could get 3 reinforced plate factories

which would be 15 per minute which could last for 5 assemblers cause 15:3=5

and then i would need 60 iron rods per

a whole factory makes me feeds into one assembler which is making 5 per minute

look i only have like 40h in this game 🤣

yea

my definition of factory

is enough for one assembler

im not using them for a recipe

its for a milestone

a few milestones actually

as much as possible

its just i cant make a factory without any excess if i have 120 iron ore per minute

yea but something i can pull off

i dont even have tier 6 unlocked

jesus thats so many merging and splitting

much*

52.5 a minute

wdym

with how limited merging and spliting is, which is either x2, x3, and /2, /3

my problem is how do i split 120 iron ingots into 52.5

my splitter can divide by 2 or 3

thats 60

oh

well great i only got to know this now

never thought about it

anyway thanks a lot

and btw whats that site you used called?

farts agressively

[sinful knoll]

Uhhh what just

Nvm ..

[native marten]

Scalti's newest videos showed me what was possible with those things, and I haven't looked back since. MIND BLOWING.

[native marten]

that ? that's NOTHING. I've got one factory generating six crystal oscillators per minute. I'll let you do the math on that one. I can share photos if I'm allowed to paste an imgur link

this will give you somewhat of an idea of the belting and splitting and merging going on in the layer below this one (noting in addition the 3 columns in the back delivering parts to and from the layer above it (screws down, cables and reinforced iron plates up)

I didn't want to go full ham with it in spite of the fact that I was farming up 2 pure iron and 1 pure copper node with overclocked mk2 miners, from the nearby desert, underneath the mountain through the long cavern near the quartz nodes, because this dam thing took me long enough to figure out at the time. I can always put together another of these right next to it with the overflow from the first factory

you can juuuust barely see the iron and copper overflow belts protruding from the left side of the building in that photo -- this factory smelts ALL the iron and copper passing through it, so technically I could just mirror every floor above the smelter section to double this output, and would probably STILL have leftover iron and copper. This also doesn't even account for the fact there's an alternate recipe for oscillators I haven't yet discovered, so I didn't want to go nuts with this one just yet

that's what I was figuring -- I just wanted to get a self-contained starter factory going, that was producing more than 2 at a time

I still haven't fully automated heavy modular frames, high speed connectors, computers, or circuit boards yet

[native marten]

I just unlocked one of the alternate recipes that will let me triple my modular frame output so I'm considering finally getting around to rebuilding that segment scalti-style and then starting a new HMF factory between it and my steel production near the caterium node in the grasslands

how so?

that one caterium node is near enough iron and copper and coal that I've got caterium wire, ai limiters, steel beams and pipes, stators, and encased industrial beams all coming from the same spot and I'm not even using all the iron nodes yet

I don't have them all coming from the same BUILDING mind you

[willow igloo]

Iron and copper will never run out, you'll be limited by caterium, bauxite and quartz before you even strain on iron

[native marten]

no of course not

but I had fun putting it all together without miles of conveyors connecting everything in disparate locations

the only real bottleneck right now is coal until I unlock tier 3 mining

or, (blegh to me) route coal in from further away

[thorn bane]

you cant realy say what youre limited on without knowing what alts you use since they swap around the requirements alot

[native marten]

ugh I am not looking forward to scouring the spider-infested bamboo forest for hard drives

I need a fricken flamethrower

we have the technology

[soft scarab]

I really enjoy hoverpack + gun for clearing out infestations

[native marten]

yeah I haven't -quite- unlocked the hoverpack yet but am close

JUST managed to scour enough stray computers from crashed pods that I got the train unlocked but haven't even BEGUN to think about how I might utilize one yet

[static venture]

Trains are helpful for transporting items made in bulk to other locations to be combined with other items to make something else

Have you started anything with oil yet?

[oblique hollow]

Oh yeah stray computers wont get you actual trains

Only the unlock at best

[soft scarab]

I eventually caved and just bought alclad sheets from the awesome shop so I could unlock + build it. 💯% worth it

[native marten]

yeah thats what I used them for, since I kept finding them

[oblique hollow]

Now you need more

[native marten]

yeah I have the two normal crude nodes on the west coast dedicated to an even split of plastic and rubber, with the results going to an awesomesink for now, and the rest used to make petroleum coke fueling another 16 (could be 24, I think, but I didn't go there yet) coal generators and overflow of the coke also going to awesomesink

I haven't yet set up power generation on the two pure nodes over there

[native marten]

it has dawned on me that no matter how much of a thing I have, there will always be the need for MOAR

[static venture]

Trains will be helpful for getting that plastic and rubber to other production areas to create higher tiered items 🙂

[native marten]

well those oil fields are right nearby a caterium node, thankfully (just down the beach a bit) and slightly further along are two lovely iron and one copper node I set up some prep for so I can send those towards the oilfields, to make circuit boards and computers with there, and then maybe train the lot of it back to the hub

I'm mildly annoyed I can't make steel from the petrocoal somehow (alt recipe or something) but oh well

but I did find the steel beams -> screws recipe which is magnificently efficient, so I got that goin' for me, which is nice.

[somber fable]

I got this recipe and used it for all of 2 things. I always just aim to never need to produce screws in a production line

[native marten]

well, that particular recipe is hugely performant on making heavy modular frames efficiently

[somber fable]

ah yeah I use the heavy encased frame recipe in my save

concrete + encased beam + steel pipes + modular frame. https://www.satisfactory-planner.net/?f=v3_U5,4000000100088000090101bffffffffffffffffff7fbfffffffff,heavy_modular_frame:maximize:20,,480:100:0|480:243:0|52860:133:0|30900:227:0|11040:637:0|10500:670:0|6840:1028:0|11700:601:0|9780:719:0|2100:3351:0|12000:586:0|0:1:1,0,1000:1:0:0

[native marten]

heck the enhanced steel screws recipe even makes Heavy Flexible Frames feasible

[wet yacht]

Does anyone have the nice picture handy of extracting pure oil into Fuel Power Plants? I don't think I ever saved it 😦

[terse umbra]

is it impossible to get a 100% efficient caterium iron ingot factory without under/overclocking?

nvm it is

[native marten]

just do the math and factor it out, multiplying by however many you need, and you can also switch things up by adding pure/normal/impure to get the balance right

[wide gate]

guys is there a gallery or website where there is a decent collection of meta blueprints/ popular layouts for stuff?

[wintry aurora]

Why is NF steel coal 1 showing a distance of three times the others when they are all exactly the same length (+/- a little because inexact driving)?

Given that the consumption is roughly the same as the others, I'm inclined to believe it's some wierd display error.

[final glen]

I think I'm recreating the grain belt in my game ...

need a track system and station design tutorial

[native marten]

what do you generally need more of? plastic or rubber?

[native marten]

so, this seems like a decent plan, given that I'm already producing 100 rubber and 100 plastic / minute with my current setup and just burning the petrocoal in generators and sending everything to awesome sink at the moment anyway

I just got to find one more hard drive that can unlock the caterium computers recipe

but this is WAY better production with less resource usage than I expected, and that friendly caterium and copper node(s) just down the beach from the crude make this eminently possible

[sinful knoll]

Nilaus and xclipse

[final glen]

game performance wise, is it better to build a giga factory or spread out and rely on logistics?

[sinful knoll]

The latter so not many obj on screen at once, no?

[final glen]

true, but the train system is getting really really complex

it's currently a push-pull system on a monorail, such a hard problem

9 stations

and, if I have one location with 5 inputs, and 3 outputs, is it better to make a 5 and 3 freight bay 2 station design, ot a 8 station single bay design

this way, stations are any to any

purpose being - scalability

[soft scarab]

Try trucks

[soft scarab]

And double tracking

[final glen]

trucks no good. fuel issue and distance across entire map

[soft scarab]

Oh yeah for longer distances trains for sure

[stuck iron]

Double tracks is niiiiice

[final glen]

yeah, double tracking, I was afraid of that

[soft scarab]

There’s a reason it’s the standard on real freight lines

Only low-traffic / long distances between stations work for sidings and I don’t think satisfactory supports sidings/passing lanes well

Though I haven’t tested it much with U5

[final glen]

okies, gonna need to convert fomr push-pull to loop system. - track design solution . but what about train station design ?

like any station can send or receive from any station. best to make multiple single bay stations at each plant ?

[soft scarab]

Do you really need any station can send or receive from any station?

Why not set up a single (or set of) transfer stations as intermediaries

[final glen]

yes. because as I create more iron ore mini factories, I'd like to add statsions easily to the main factories aroundt he map

[soft scarab]

Hub-and-spoke model instead of many-to-many

[final glen]

oooo.... transfer stations

feels like that would be more complex... I'll need to thikn on it

actually no. I can control via transfer stations and train route zones

instead of having one major loop, having 8 loops to and from the transfer station might be better. adding more trains for more throughput - nice !

woohoo! thanks a bunch

[soft scarab]

You’re welcome! Glad I could help 🙂

[final glen]

and I think I'll need a two tier transfer station. each zone having it's own tansfer station, and then a central one - to help break down the troughput limits

and suddenly, I'm playing railroad tycoon, lol

[soft scarab]

Yup 😄

[soft scarab]

The transfer “station(s)” can be done any number of ways. For each central station, you could have one receiving train station and several outgoing stations (eg connected with belts) each customized to their destination factory, letting you keep the push-pull for local/factory-specific trains. You could have global transfer stations sorted by material type (eg iron in one place, copper somewhere else) or production chain (eg one network for frames, one for nuclear or whatever). Lots of combination options

Drones can also be a helpful addition for smaller production items, if you’re not already (planning on) including them

[final glen]

cntral transfer station would be choking point... the transfer stations in each zone coule be any to any - that might work

I'm afraid that each transfer station would be one tran station per item - since there's no ordering sytem in satifactory

like, telling the freight bay to load the car with 1/3 ore, 1/2 plastic, 1/6 fuel - etc

[soft scarab]

Depending how frequently you have trains coming in (since frequency reduces throughput due to the freeze on train load/unload), smart splitters may do it for inputs

[native marten]

what no sushi stations? sacrilege

[final glen]

splitters are too slow, even if I double them up.

[soft scarab]

Just have dedicated platforms for each of those items

[final glen]

yeah, dedicated platforms - like bulk logistics

[native marten]

splitters are only limited by belt speed so far as I can tell

[final glen]

ok, zoned track system with dedicated single station per item at each factory. the station will have to act like the ordering system

true - but the train make the round trip faster than the splitter can dump it

[soft scarab]

Then you should be using trucks instead

[native marten]

can you alter the schedule to happen less frequently in that case?

[soft scarab]

Yeah or add wait times

[final glen]

the load/unload rule wasn't working in U5,, but I haven't given it a fully shake

honestly, I just hate trucks

[soft scarab]

Hahaha. I used to hate them too but have come to appreciate them

I am enjoying the challenge of building out an additional packaged fuel logistics system

And I really appreciate how space-efficient they are (I’m working in the northern forest and as an added challenge trying not to cut down all the trees)

[final glen]

ok, gonna go scketch out the new rail system

[oblique notch]

heh. This games permutations are going to be friggin crazzy to figure out:

Hint. this is just paths to a basic resource, not permutations of different combinations to make this...

{
    "item": "Nuclear Pasta",
    "totalPaths": 11826,
    "allPaths": [
        "[Standard: Nuclear Pasta] (Copper Dust + Pressure Conversion Cube) -> Copper Dust[Standard: Copper Powder] (Copper Ingot) -> Copper Ingot[Alternate: Copper Alloy Ingot]  -> Resource Node (Copper Ore + Iron Ore)",
        "[Standard: Nuclear Pasta] (Copper Dust + Pressure Conversion Cube) -> Copper Dust[Standard: Copper Powder] (Copper Ingot) -> Copper Ingot[Alternate: Pure Copper Ingot]  -> Resource Node (Copper Ore + Water)",
        "[Standard: Nuclear Pasta] (Copper Dust + Pressure Conversion Cube) -> Copper Dust[Standard: Copper Powder] (Copper Ingot) -> Copper Ingot[Standard: Copper Ingot]  -> Resource Node (Copper Ore)",

...

actually permutations for Nuclear Pasta are super easy to figure out - there are 2 resources, and only 3 ways to make Copper Dust so... 11823*3 permutations on different possible recipe paths to make Nuclear Pasta. Most of which are totally not viable lol

[fierce ruin]

oh yeah. epic ratios

[proper karma]

Should I have pumps down this pipe?

[native marten]

so, make my day, I go running around atop Red Mountain, dodging hellspawn looking for hard drives, and find one, go to put it in the MAM and LO! there's already a drive in it with research completed having the exact recipe I was hoping to get after researching several more drives. so much win 💯

[native marten]

nah that's horizontal - if stuff isn't getting to the end maybe run a parallel pipe above it and loop it in at the terminus of both

[proper karma]

Alright

[native marten]

either that or you don't have enough head lift going on at the beginning

[native marten]

right now I'm mostly enjoying the fact that I can turn alien organs into truly staggering amounts of biocoal with a single constructor, and use that to fuel the tractors and trucks

[soft scarab]

I built a black powder factory on top of some sulfur deposits. Leaves & wood go in, explosives comes out 💥

[frosty owl]

I feel like this HUB-and-spoke idea wouldn't reduce much the numbers of trains and stations @final glen would be using if they just tried their many-to-many approach...
Generally speaking, I think one should keep the "any-to-any" approach only for some transportations, doing that for any kind of transport and volume you may need can end up in a (lag-wise) terribly convoluted train network
TLDR: Imo, trying to use a "many-to-many" approach, with or without hubs, rises the number of trains needed way too much for little gain, best to keep that approach for some specific kind of items or volumes of items

@vast jungle might be able to add some good info on this, if the convo interests him ^^

[pallid cobalt]

hey guys just a quick one, im running 40x refineries for turbofuel, both compacted coal and the fuel itself are in manifold style setups and the belts are at the right speeds as they move down the line - is 40x machines too many for a manifold?

the numbers are right throughout so.. in theory..

[frosty owl]

As long as you're not trying to exceed your belt throughput anywhere, you're good ^^

[pallid cobalt]

great 😌 i let the belts and pipes saturate the machines fully bit by bit as well with some buffers before production

[frosty owl]

That's generally referred to as "priming" your manifold of belts or pipes ^^
Great to avoid possibly huge waiting times before the factory reaches full efficiency

[fringe pawn]

The fluid loss bug remains an issue for fuel generators. As long as you have some power storages you'll be fine. Or are overproducing turbofuel.

[neat kettle]

i read somewhere 3 water pumps/generators are enough for 8 coal plants, how do I go dividing it fit all 8?

[frosty owl]

1 pipe merging the 3, each end of the pipe can feed 4 gens

Some layout examples can be found in the wiki

!wikisearch coal generator

[shadow prairie]

https://satisfactory.fandom.com/wiki/Special:Search?query=coal generator

[soft scarab]

Yeah it’s not trying to reduce the number of trains and stations, but rather simplify the system design - they were originally (from my understanding) trying to come up with a universal station layout for every factory/outpost that would allow any train to go to any other station. Hub-and-spoke allows one-to-many; it takes more stations but the stations can be smaller and more custom for the specific factory/outpost. It isn’t really a “many-to-many” approach because the hub is a centralizing point - it’s “many-to-one” and “one-to-many”

Could even just be a bunch of “one-to-one”s that are linked at the hub via belts

[frosty owl]

Ah, makes sense 👍

[oblique notch]

hub and spoke is what the big shipping companies (Fedex/UPS) use for their distribution networks, by the way

actually too a multi tiered hub and spoke - with big primary hubs like (for FedEx) Chicago and Memphis, that spoke to small terminals and smaller hubs (Indianapolis, Atlanta) and then to smaller stations beyond that

[cinder silo]

I found logistics very quickly gets away from you in this game.

[soft scarab]

I thought this game is logistics 😂

[cinder silo]

Hah, might well be, moving stuff seems to take more time than building 🙂

Before you know it, your train station is the size of an entire biome.

Oh crap, I was so concerned about getting the heat sinks made, I never planned on how to get them down to the plutonium fuel rod factory that's half a dozen buildings over and a level down, past the criss cross of belts.

[frosty owl]

That depends a lot on how complex your buildings are (how many processing steps they include). As in: the less complex the factories, the heavier the load on long-range logistics

[soft scarab]

It’s also what passenger airlines use - there’s a reason you’re always transferring in Denver/chicago/Texas/Atlanta (in the US), and it’s not because planes can’t fly across the country without refueling

[frosty owl]

Imagine having it all in one place instead

[soft scarab]

Or just your local transit network. Cars are pretty unique in their any-to-any functionality, but that’s also because we mandate universal design access stations for automobiles (ie parking) at virtually every destination & origin point

[cinder silo]

The trouble with my fairly ad-hoc nuclear on a large scale with each building having it's own job, that isn't as straight forward as it would seem.

[soft scarab]

Trucks!

[cinder silo]

No room.

[frosty owl]

Easy, divide the production in chunks that each takes care of the entire chain 😆
(Sorry for the messy pic, but gives an idea of the distance between my sinks production and rods production)

[cinder silo]

More zoomed in at least 😛

[soft scarab]

This doesn’t work?

[frosty owl]

I'm using the only SCIM screen I could find 😅