math-and-meta

[wind spade]

then make 5 inputs instead of 4 and do 1:1

underclock them to 80%

[proven prawn]

you can by essentially having 4 rows of manifolds with the desired outputs, and you can still manifold it all by having splitters and splitting off and essentially having a manifold array

[frosty owl]

Or go the "machine efficient" way and overclock the consumers so they're 4 machines at 125%

[proven prawn]

there is no can't with manifolds, as long as it stays within the belt limits, but you'll run into those limits regardless if you use manifolds or balancers, so it doesn't solve the problem any better or worse.

[north cloak]

hahaha well i am getting there right now

[proven prawn]

i used many types of belt designs, balancers, manifolds, sushi balancers, sushi manifolds, etc, the sushi ones can be kind of fun to play with what i like about this game is there isn't a one solution option

[frosty owl]

Even better, you can create problems while trying to solve one!

[mint sedge]

You can split the diff and have a manifold fed in both ends, or split in the middle and feed left and right.
Depending on the length of the manifold that can get you closer to even split

[north cloak]

here is the balancer lol

[frosty owl]

||Well, we guessed right. It does take up more space than one more input machine (or some power shards) ||

[north cloak]

lol

Especially when you refuse to clip

[frosty owl]

The no-clip is appreciated though

[north cloak]

haha

choas

[ashen mantle]

This is going to be the most densely packed factory in my world

All of this

Needs to fit in there with the exception of the silica and the circuit boards

[tropic hawk]

Yeah no.

That's not gonna fit on one level

[remote flame]

Out of curiosity, by using the alt recipes: Pure Copper, Pure Caterium, Fused quick wire, Caterium circuit boards, Caterium Computers and just standard rubber and plastic recipes with all heavy oil being turned into diluted fuel, is there much room for improvement in terms of resource node efficiency?

Edit: I've been trying to avoid turbofuel to keep a sulfur and coal node free for other factories

[wind spade]

Add alt hor, recycled plastic and rubber

[ornate shoal]

stamdard plastic ia not efficient for sure

[vapid gorge]

100% alt recipes for hor, diluted fuel, plastic/rubber. It's 100s% more efficient

[lapis jay]

Ah okay thanks

[ashen mantle]

For rubber and plastic, you should make a feedback loop with an overflow that is being fed with a ton of fuel, made with diluted fuel from HOR

Resin can be used for fabric or whatever you want

There should be no plastic or rubber being directly made

Just have recycled plastic, split, half goes out, half goes into recycled rubber, split, half goes out, half goes back into the recycled plastic

Every other recipe should be just fine

[remote flame]

Cheers, I figured there would be just that area for improvement

[median heath]

Yeah, switch to Crystal Comps.

Bringing in Quartz lowers cost of all other resources.

[remote flame]

I've got the factory set up to make 80 rubber, 80 plastic, 40 computers, 180 quickwire, 40 packed fuel per minute which is in tern, driving a 23,000MW fuel generator facility

[vast orbit]

Hey can i ask how did u make this?

[ashen mantle]

https://www.satisfactorytools.com/

[median heath]

I was going to give them the Adobe Photoshop link

[ashen mantle]

[vast orbit]

Thanks

[ashen mantle]

https://media.discordapp.net/attachments/553550313533997057/977021526173941820/Screenshot20220519-21343200000.png

https://media.discordapp.net/attachments/553550313533997057/977034769705627698/Screenshot20220519-22265900000.png

This factory is so unbelievably dense

All of these buildings in here except the refineries

Throwing the refineries in an auxiliary building next door

[native crow]

55 manufacturers is a lot of floor and belt space

[median heath]

Not when you mix your belts 🙃

Saves a considerable amount of space.

[native crow]

Still a large sq footage but that can help

[ashen mantle]

Mixing belts causes me too much pain, especially since none of these inputs make nice numbers

[median heath]

How does it cause pain?

[ashen mantle]

It always clogs and runs into issues

[median heath]

Sushi Manifolds are incapable of clogging.

[native crow]

Smart splitters and a scum dump wouldn't clog

[ashen mantle]

Then I must be doing something wrong with my implementation

[median heath]

"Scum dump"???

[median heath]

Items to inputs, overflow sent onward.

[native crow]

"Awesome sink", same thing

[median heath]

I mean there is the designed building with intentional mechanics by the devs, then there is you implying that the devs made a intentional mechanic a dirty part of the game that should be avoid...

[ashen mantle]

I had some items, most notably quickwire, build up at the end and push back to a splitter and slow it down, and if it locks out the other items it'll outright stop

[median heath]

So not the same...

[median heath]

How does something push back if overflow is "sent onward"?

[ashen mantle]

Do you dump it into a sink at the end? I've just routed it into the final manufacturer on the end

[median heath]

Sinking is an option, yes.
But where you choose to send things is up to you.

But overflow has to be sent onward. Not given the opportunity to back up.

Sinking is the simplest. But storage works, routing to other lines works.

If you have the math and send the exact amount needed, you only need to sink during spool-up of the manifold and afterwards everything gets consumed properly within it.

[ashen mantle]

If you have any input that's slightly higher than the manifold needs, then you need to sink it or it'll clog, right?

[median heath]

Failsafe Sinks are a novelty for when the game decides to fuck something up if you properly do the numbers.

[median heath]

Again, sinking is ONE option.
What you do with overflow is up to you as long as you send it onward...

[cinder silo]

I personally send overflows to stores, THEN a sink if that's full.

[native crow]

Looks like you just have an issue with my term, not intent

[ashen mantle]

Well I can loop it back into the input, which will still just increase past the manifold capacity, I can sink it, what else do I have?

[median heath]

Yes... "scum" has so many positive connotations across language...

[median heath]

Storage.
Routing to other lines.
As I also said previously.

[native crow]

It wasn't intended as positive. But if you are dumping cheap low end parts into it later in the game they are a waste. It isn't the sink that is bad, it's the wasted parts

[median heath]

So back to "implying that the devs made an intentional mechanic a dirty part of the game that should be avoided"

👍

Free points = waste. 👍

[native crow]

That was your statement not mine

[median heath]

That's your position because you just restated it.

You're talking about an intended mechanic and reiterating you don't have a positive viewpoint on it.

[native crow]

I implied inefficient wasted parts are bad, and we have a tool to remove those so the system otherwise works

[cinder silo]

Waste is meaningless since everything is infinite, it isn't like resource nodes die off.

[median heath]

Points = Points.

[native crow]

Cool, then why are you in this channel? This is for meta

[median heath]

😂😂😂😂😂😂

[cinder silo]

🤦‍♂️

[native crow]

No, because you can sink cat ore or make it into wire. Or other more advanced parts, so sometimes points don't just equal points

[median heath]

@cinder silo I believe I understand.

Meta = You're not allowed to sink anything that isn't a TPR or ADS.

[cinder silo]

Yeah well I'll sink whatever the hell I want, such as plutonium fuel rods 😛

[median heath]

Not allowed.

Not meta.

[native crow]

sigh

Typical strawman

[median heath]

Telling me to get out of an entire channel because I'm in favor of using the AWESOME Sink as a mechanic is... 😂

[dull bolt]

But you can use plutonium fuel rods for more fuel!
And then you can get rid of the waste by yeeting doggos off cliffs.

[median heath]

Yeeting doggos is definitely meta.

[dull bolt]

leaves the meta channel

[cinder silo]

The math isn't in my favour yeeting doggos 😛

[median heath]

Not entirely sure what the top recipe is about.

Seismic seems to have been turned into Pulse.

More uses for Crystal!

Not sure if name-change or new item?

@oblique hollow

[oblique hollow]

gunpowder renamed?

[median heath]

Black Powder is current name.

[oblique hollow]

ik ik

[median heath]

Using Oscillators and Crystal for ammos.
My baby Quartz is going to shine!

Nuke Nob needs to use uranium. Plz. 😭

Personally I would make it use Encased Uranium Cells instead of just raw Uranium 🤷‍♂️

But we'll see what they do.

[gloomy rock]

What is crystal mainly used for now? I’ve never had to use it yet

[oblique hollow]

silica

[median heath]

Oscillators and Silica.

[oblique hollow]

and oscillators

[gloomy rock]

Ah I see

[median heath]

Oscillator alt recipes are great. You bring in Quartz and reduce all other material costs.

Crystal Computer FTW.

[gloomy rock]

I gotta get more stuff done in the mam I haven’t done much in that

[oblique hollow]

@median heath how TF did you miss this

ACTUAL USE FOR TURBOFUEL

[median heath]

👀

[median heath]

Must have been earlier in the vid than the part I clipped.

[oblique hollow]

default ammo

[median heath]

@wind spade

[median heath]

Damn this is cheap AF

[oblique hollow]

as it should be

[median heath]

Still

Stun Rebar + Basher = Game EZ

[median heath]

@oblique hollow wait, so if we use Turbo Blend with Diluted Fuel for that ammo recipe... would that mean we're making Diluted Ammo???

@dull bolt

[oblique hollow]

technically yes

also... consider the following:
Turbo Heavy Biocoal Ammo

[median heath]

Diluted Packaged Ammo

[dull bolt]

If diluted ammo gets added into the game...

[median heath]

Heavy Ammo lol

[oblique hollow]

#math-and-meta message

[median heath]

U6 = Diluted Ammo

[proven prawn]

well we are getting 4 body slots in u6, so anything is possible now🤔

[dense stream]

diluted ammo is just shitty surplus

[sullen perch]

Turbo ammo is just ammo with lube

[median heath]

DiLUBEted Ammo?

[glad vigil]

Finally ammo that isn’t comically overcomicated to make

[glad vigil]

Pocket nukes

No way

[robust acorn]

hey I'm wondering if this is the right channel for help with fluid dynamics?

[soft scarab]

#old-questions-and-help might be more appropriate if you’re having issues with an actual setup, this channel is usually more for higher level theoretical stuff

[robust acorn]

awesome thank you

[copper sage]

load balancing is pretty good. Its always fun and challenging to figure out. But when it comes to load balancing fluids...now thats interesting. Is it worth load balancing fluids? I haven't personally heard much discussion on this, and its something I've often thought about it. For example, take the 6 oil nodes at the most southeastern portion of the map, what is commonly I think referred to as the blue crater lake area. 3 Pure nodes, 2 normal nodes, and 1 impure node. Some people say that using full 600 m3 fluid flow in a mk2 pipe is a little sketchy. it works but then under cutting it a little might be best. Others say mixing pipes is bad or can cause problems (glitchiness or bugs). Well regardless of these issues, it still is an interesting question for balancing the flow of liquids to multiple pipelines. Sure junctions provide splits in a way, and valves direct flow, but can we truly balance liquids accurately and efficiently? Or is there going to be some kind of offset range that's acceptable to allow over or under. OR is there no math to it? just get as much fluid as possible?

[soft scarab]

What

[median heath]

Stop thinking about pipes in terms of belts.

[burnt wraith]

in order to load balance fluids, you have to make each machine the same distance from the source

so you could really only do powers of 2 and your pipe system would look like a sports bracket

basically there's no way to accurately and efficiently balance with pipes like you can with belts

[proven prawn]

It could eventually be worth but until varies issue's with pipes are fixed like the 5m3 load bug, mk2 pipes not able to deliver 600m3 a minute that are listed etc, there is already enough of a margin of error already that proper load balancing is most likely more trouble than its worth until the issues of inaccuracy within the existing system are resolved.

[copper sage]

i got a math issue i need solving. 85 heavy oil residue refineries, each refinery 40 m3 per minute. I need to split these refineries into both blenders and other refineries. Blenders that make alternate recipe turbofuel, while i'm using other blenders to make the diluted fuel for the turbofuel blenders. Than i also need refineries that make petroleum coke which requires heavy oil residue as well as the diluted fuel blenders AND the turbofuel blenders. the math is rough and my brain hurts lol.

85 refineries at 40m3 Heavy Oil Residue
Don't know how many blenders i will need for turbofuel and diluted fuel and how many refineries i will need for petroleum coke.
It takes 1 heavy oil residue refinery to supply 1 petroleum coke refinery
Turbofuel blender requires:

• 15m3 per minute of fuel
• 30 m3 per minute of Heavy oil residue
• 22.5 per minute of sulfur (i should have this covered, no issue there)
• 22.5 per minute petroleum coke

Diluted Fuel blender requires:

• 50 m3 per minute heavy oil residue
• 100 m3 per minute water

Petroleum coke refinery requires:
-40 heavy oil residue

As far as I can gather, with my 85 refineries, i can use 1 refinery to supply 4 blenders to meet the 40 + 10 requirement for 4 blenders.

Trying to figure out rest of math...any help is appreciated. so i guess the question is how many of each would i need to be efficient.

[soft scarab]

Have you tried https://www.satisfactorytools.com/production?

[copper sage]

that looks way too confusing

[tropic hawk]

It's really not. You put in the end product you want, and it splits out best way to do it. You can fiddle with what it has to work with whether it be inputs or recipes, and it will automatically optimize things for youm

[oblique hollow]

with some slight mathing around i can at least tell you the following right now: heavy oil to turbofuel is 1 to 1 with diluted fuel and turbo blend fuel

so if you make 40 hor, that will equal 40 turbo fuel in the end. you can use that info by setting up a blender that makes 40 fuel and then you just need to look at how much HOR and coke and fuel it needs

[arctic mauve]

this is weird. i tried it and it says stuff like this

how am i supposed to split it up like that

[oblique hollow]

overflow

also, these are 7 constructors

im sure you can group them in a way that you get 2 seperate outputs

one of 60, one of 45

[frosty cloud]

isnt it possible to limit the output with conveyer belts?

[arctic mauve]

oh

[oblique hollow]

the simplest solution is definitely to just group 3 constructors and 4 constructors

[wind spade]

You can't (and shouldn't) really balance fluids, they don't work like that

[wind spade]

That's up to you really, there's tons of possibilities on how to connect factories

[arctic mauve]

well im still figuring out how to make stuff look nice

[oblique hollow]

the eternal struggle of satis players

[wind spade]

Probably best to first learn how to make it functional

[oblique hollow]

once you make functional stuff you can experiment with the prettiness part

[arctic mauve]

oh alright

this is how i made it look for now it does some iron stuffv

[remote ice]

you kinda can. A pipe joint is technically a perfect 2-2 balancer iirc

i contemplated making a grid of those for balancing the oil lines in the crater lake, until i realised that was stupid and i just connected them with pipe junctions in a line

[arctic mauve]

is there a way i can tell it "i have this much whats the best i can make out of it"

or do i just have to try

[vapid kernel]

I think Items, Input tab on the left

[oblique hollow]

item and input tab, there you can set the ores you can extract or items you have available.

but the site always needs to know WHAT you want to make from that

there is no calc that tells you "you could make all these different items right now"

[arctic mauve]

with the stuff i have i cant make it run at 100% but meh whatever

[oblique hollow]

thats how it rolls

[frosty owl]

Imo, the whole thing boils down to: pipes can't be be balanced precisely enough to account for all the bugs they have.

||Neither can belts, but the bugs creep in much slower in that case, making issues much less noticeable||

[frosty owl]

When the game decides to F something up: always, as long as splitters are involved and one saves the game

[wind spade]

No, because backflow exists

[frosty owl]

And can be prevented. Similarly to how items backing up on a belt "ruin a split"

[median heath]

One day people will finally stop thinking about pipes in terms of belts...

[vapid gorge]

I never thought you'd be an optimist .

[median heath]

Ouch.

[muted crypt]

the same day we get the crab boss thing

So it will happen eventually.... just not for a while

[crude coyote]

one day we get base defense? 😛

we are already getting all these fancy ammo and boom-booms

I wish maybe there were more ways of hack-and-slash

like xeno-flaming-sword

or xeno-double-handed-axe

or xeno-nuclear-basher

just kidding, I hate the combat aspect of Satisfactory

[cinder silo]

I'm happy enough with more ammunition in a magazine and no beacons for it.

[versed violet]

Not sure where to put this, but for reference - the game still runs on intel HD 4600 ( t540p with i5-4330M processor).
With all video set to lowest I could load into the main menu/slideshow, but game would crash when loading map. Reducing the resolution from HD to 1280x720 allowed me to load in. 5 fps is still perfectly playable, yes?

[cinder silo]

5fps? 🤮

[versed violet]

[hands over a bucket]
Just checking that it works. So if anyone else has problems with getting it to run, its likely not a specs problem, unless you have older machine than my web-browsing laptop.

[median heath]

Never.

Different Xenobasher variants sounds interesting though..

[cinder silo]

I took to abusing power poles, the hover pack and the rifle, unless they're adding any tougher fauna to the game I doubt anything else apart from the million snowballs I have will be needed.

[versed violet]

I wonder if any of the new nobelisks offer killing fauna without damaging flora

[median heath]

Baseline basher destroys everything so idk why people hover...

[versed violet]

Enemy can't hit you, if they can't reach you. (except the flameys)

[cinder silo]

I do try to avoid nuking the flora, not only does it make the landscape look poorer for it (unless building there) it impacts on save time.

[median heath]

It's also, imo, dumb gameplay that people can do it and I preferred the days before we had hoverpack.
But I'd be happier if they just altered it so that flying everywhere was so annoying it wasn't viable.

[cinder silo]

Even those are trivialised just by sidestepping and shooting.

[versed violet]

There is a mod for undoing the damage, if you wanted 😉

[versed violet]

Not with spideys/cats

[cinder silo]

I tend to avoid mods due to their innate ability to break spectacularly with the slightest update.

[median heath]

Take the nuclear radiation thing, based on how much is in one spot.
Make the hoverpack work on energy but the amount is based on how many full production buildings are clustered. -- so you can float around your factories but out in the wild it doesn't operate.

[cinder silo]

I walk power poles just to make it operate, mostly just so I don't have to build massive concrete ramps to deal with terrain while exploring, being able to skytrooper the fauna is merely a side effect.

[median heath]

I don't like using it for either purpose, but I know removal isn't going to happen so I'm trying to think within the system. 🤷‍♂️

[cinder silo]

I was surprised when I found turbo fuel wasn't usable for transport/jetpack.

[median heath]

Transport?

Works in all vehicles.

[versed violet]

There are QualityOfLife mods that do not add any extra objects and do not affect the save or affect it minimally.

[cinder silo]

Been a while since I tried it, my old turbo power station had a little siphon station to provide fuel that was a giant waste of time.

[versed violet]

Speaking of jetpack - the latest video didn't say about using different fuel types?

[cinder silo]

Something I no doubt missed, I'll re-watch it.

[median heath]

Didn't say anything specifically, no.
But the fuel on the UI was labeled as a type of fuel for the first time... ever. So it's possible/likely? 👀

[soft scarab]

New fuel for chainsaws when??

[median heath]

Also possible if they did alter the jetpack.

I assume they are saving that announcement for another video if true.

[soft scarab]

Need my nuclear powered chainsaw

[median heath]

More uses for Plut Rods.

[soft scarab]

Exactly

[median heath]

Use Paint-powered chainsaw to gather more flowers 😉

[soft scarab]

That actually would be super helpful, instead of running around a field of flowers pressing e all the time like an idiot

[median heath]

Oh yeah. Why would you ever hand-pick flowers?

They are in massive fields for a reason.

[late orchid]

I'm trying to decide how to pull together the three pure caterium nodes in the southwest area of the map to supply a factory making electromagnetic control rods, with the factory being located centrally between the three nodes where I have a train line running past. Would you truck these in, make train stations, or just use long conveyor belts?

[median heath]

Depends on whom you ask.

[late orchid]

I am interested in hearing a variety of opinions, which is why I asked. Maybe there is something I haven't thought about, or someone has a good idea (or has done it before)

[median heath]

You listed every method other than drones.
So you know your options 🤷‍♂️

[uncut sigil]

If you already have a trainline, and it's all going to the same place, I'd use that.

[median heath]

If you already have a trainline through an area and you haven't decided where you're going to build in said area... I have questions.

[late orchid]

I laid a main train loop around the map while collecting hard drives and exploring.

[tropic hawk]

Define 'long'

[median heath]

Ew.

[late orchid]

Perhaps then you might offer more tangible insights?

[median heath]

Build YOUR factory, not someone else's.

You know your logistics options, pick the one that works best for how YOU want to build. 🤷‍♂️

This has always been my stance.

[tropic hawk]

I can help if you give me an actual distance instead of 'long'.

[late orchid]

I feel like there is a difference between asking someone to build a factory for you, and asking for the experience and insights of others when trying to make a decision.

[tropic hawk]

Some people don't like global train networks. I think I know why, and I disagree with them, but they can do their own thing.

[median heath]

I never said I am against global networks.

[tropic hawk]

Sorry, misremembering who said what. It was greeny sho was against those

[late orchid]

Hmm, okay. I feel like the belts are getting too long when they are running straight for 100+ foundations, that's when I start asking if I should be using a truck or train.

[median heath]

You basically said "hey I want to build here and here are 3 equally viable options to do logistics - pick for me"
Because you know they are equally viable options. And 3 different people could all tell you different answers to the question of which to pick.

Meaning at the end of the day the question of which to pick circles back to where it was in the beginning - you having to make the choice for yourself.
So I find the whole point of the question somewhat moot tbh 🤷‍♂️

[median heath]

I'm against making them just for the sake of making them -- which is what most people do.
If you make them with purpose and the rationality behind the logistics is there they are completely fine.

[tropic hawk]

IIRC, in TotalXclipse's logistics guide, he claimed that once you cross the 1 km mark, trains get higher throughput than belts for cheaper cost. Trucks are a bit unstable and you would have to figure out both routes and fueling and for trains you have to lay the tracks for them. Belts are typically the simplest and usually the easiest, but trains are usually better. Trucks are the 'funner' option as you are being more unique.

[tropic hawk]

Understandable

[late orchid]

Thank you. This is helpful. I am tempted by the trucks because they seem fun, but the trains feel like the right solution given what I've already built

[tropic hawk]

You are welcome. I would recommend watching all the guides done by him because they are amazing and he explains why things do or don't work. 10/10 would recommend

[median heath]

"Seem like fun"?
Did you get to trains without ever using trucks?

[mystic moon]

Lol

[tropic hawk]

Since you do think there is a use for global train networks, what do you think is the 'best' way to set it up (please explain also why you think it is the best)

[median heath]

It needs to be organic.
You need to have built things around the map and then connected them via train.

People instead seem to be building trains around the map and thing trying to figure out what they want to build where.

[versed violet]

Sounds backwards. You go to scim, pick a couple candidate spots with respources, survey them via explorer, then build a train line there, then build factory. Correct?

[median heath]

No.

How do you accurately connect a train to a factory that doesn't exist?

[versed violet]

I do not. I build train station, then build factory.
If one of factory buildings need more space, I move the train station.

[median heath]

Aren't you the one who can't add an extra car to fix logistics sometimes because "no space"?

😁

[glad vigil]

Real chads add extra train every time the logistics fail

[median heath]

That actually lowers your throughput so... no.

[versed violet]

Not if roundtrip time is so long the station fills.

[median heath]

Longer trip times get a single train platform closer to 1560 without it ever actually being able to reach.
Adding another train lowers that max value.

[versed violet]

there is no point in it having a max value, if it runs out of items to unload

[proven prawn]

Cheaper cost than free beyond initial cost infrastructure I find that impossible, nothing is cheaper in long costs than belts.

[glad vigil]

1560 items per minute is impossible for train platforms because 20 second pause exists

[median heath]

27.08s

[versed violet]

its absolutely possible...
... if a train never stops at said platform 🤣

[proven prawn]

But then the throughput is 0 because no resources

[versed violet]

Yeah, I think somebody already said that above.
here -> #math-and-meta message
Oh, wait, thats me 😂

[proven prawn]

Now if only platforms wouldnt lock to unload and could multi tasks at once, but I guess multi thread programming is beyond css programming skills because they don't have the talent

[median heath]

I would hate that change.

[versed violet]

pretty sure thats a business limitation (from Mark) not a dev limitation.

also, from a software dev perpsective: it is funny to watch all 3 types of stations have completely different implementation of cargo loading.
tuck station - loads stuff - simple.
Train station - loads, but with a pause?
drone station - has separate load and unload inventories. Doesnt' pause (?)

[tropic hawk]

As in construction costs

[tropic hawk]

Two platforms for one resource?

[proven prawn]

They put all of the multi thread programming work into truck stations and drone stations but we're like why do the same thing we did again with the other two leave it as is

[river pagoda]

So Im assuming this wont have too many issues with headlift?

[soft scarab]

It depends. Which way are the pipes going? What are they coming from?

If it is coming out of a building it needs a pump to traverse more than 10 meters of vertical uphill distance

Flowing downhill is no problem

[river pagoda]

Yeah it makes sense now. The coal generators fire up but than run out of water but I Have pumps staged throughout the pipeline

[soft scarab]

Sounds like insufficient water extractors instead of a head lift issue

[river pagoda]

How many though? I read 2 per 4 coal generators

[soft scarab]

The normal ratio is 3 extractors for 8 coal gens so 2:4 will work too

As long as you’re not trying to supply 8 gens off of on mk1 pipe?

[river pagoda]

Nope just 4

Pipes fill up and all gens run for a bit but than pumps shut off and so does the water obviously

[soft scarab]

Make sure your first pumps are near water level and that none of them say more than 20 head lift

The water extractors only provide 10 meters (2.5 walls) so they have to be within that height

[river pagoda]

Looks like .3 is the highest I have atm

[soft scarab]

.3?

[river pagoda]

Yeah pumps were just a meter shy of where they needed to be

[river pagoda]

Thanks for the help

[remote ice]

IIIIIIIIIII'VE DONE IT. Behold:

all creature spawns on the map. Note: slightly variable, position may vary near the spawnpoint between worlds (will still be pretty close), and spawn count can vary for about 4 of those spawnpoints (no more than about half a dozen creatures difference altogether)

total number of lizard doggos is either 118 or 119, depending on that random spawn. I'm not actually sure if it varies between worlds or between games loads or even game versions. I know it's determined as soon as a map is loaded, at least.

EDIT, READ THIS: A issue with these maps caused them to not display something like a third of all creatures, I resolved it, see updated version here: #math-and-meta message

and here's a map with only lizard doggos

as far as i'm aware i'm the first person to actually map these out, and for lizard doggos at least this is pretty useful

time to maximize that power slug production

[vapid gorge]

Let me quickly post in the dev suggestions forum to shift all the spawns brb

[remote ice]

jokes on you, i already have the toolchain set up to do this. I can map all the spawns again in maybe 5 minutes of work

now, if you go and alter a bunch of the entity paths, that would be a pain

[vapid kernel]

So useful. I have 70 Doggo's so far. Thank you for sharing

[wind spade]

Inb4 devs remove slugs from doggos

[fringe pawn]

I think it would be fantastic to remove the click grind of doggo farming by creating a way to automate power shards in T8 or something.

Some absurdly hungry particle accelerator recipe that demonstrates that you've already 'solved' power use as a problem.

[wind spade]

No, have limited shards

[fringe pawn]

I'm not opposed to limited shards

[wind spade]

With linear clocking, unlimited shards are super weird

[fringe pawn]

Personally linear clocks won't change how I use shards if they remain unlimited.

[wind spade]

Practically it would end up being "always good", since there would be no disadvantage

[fringe pawn]

You either spend time gathering shards or you spend time building more buildings. There's no easy way to say which is more difficult to any particular person. People who struggle to make enough components for blenders and refineries might prefer finding slugs. People who have no such supply issue might have no interest in the repetitive clickfest of doggos or combing the map.

[wind spade]

I meamt with automated slugs

[fringe pawn]

Ah, I see. Which is true. But it also elongates the overall factory setup process. WHich might be undesirable pacing depending on how slowly they're produced.

[glad vigil]

That makes power shards from absurdly high amounts of biomass

[vapid kernel]

@remote ice It appears that there is even more Doggo's then that map is showing. Roughly 7 to 10 more in the Northern Forest

[remote ice]

Hmm. Um, some of them might be overlapping? That's the main thing i can think of

[vapid kernel]

Doesn't look like it. many are close to the cliffs

The rest of the map seems accurate. just in that Biome

[remote ice]

i'll test some things and see if i can figure it out

[remote ice]

ah, yep, okay there's some sort of complexity to how loading creature spawns works, i'll try to fix it

[vapid kernel]

Nice one

[remote ice]

also, what initial spawn did you choose for your world?

just trying to see if that's a factor involved or not

[vapid kernel]

Rocky Desert

[remote ice]

ok i believe i found the issue, the game was using the same filename for different spawners in different folders so some stuff got overwritten. I think that this is the complete map this time:

doggo only:

i'll do some more checking tomorrow to be sure, though

but i'm fairly sure it was just a file issue with how i was doing it.

total creature count is as follows:
{'crabspawner': 516, 'charger': 972, 'alphacharger': 106, 'spitter': 439, 'alphaspitter': 181, 'plasmaspitter': 102, 'stinger': 409, 'alphastinger': 124, 'gasstinger': 25, 'bird': 623, 'what': 171, 'doggo': 132}

[vapid kernel]

Looks great now. Yay. Thanks again for sharing 😁

[remote ice]

np

[frosty owl]

That's part of the point imo: making OC more easily accessible ALSO as a way to get better FPS for the same production goal

[frosty owl]

The SF Subreddit would probably appreciate the content too 👍

[oblique hollow]

It somehow always boils down to:

1. "NOOOO don't make OC better it will be the default thing to do!"
   => Why is that bad?
2. "Shards are unlimited and therefore it is a no-brainer to Overclock"
   => Why are shards unlimited?
3. "Doggo farms"

So in sum its: have non-linear OC and endless shards or have linear OC and limited shards.
I prefer the limited option because its a much more interesting decision-making thing. And it ACTUALLY reduces manchine count, while non-linear OC leads to an inflate of power production thats harder to account for

oh yea ig. @frosty owl and @wind spade

[frosty owl]

I don't see any big issue in having linear OC and endless shards. Getting slugs is much faster when exploring than Doggo-farming anyway (until you need to travel far, ofc). Doggo-farming requires time and patience, I think it's unfair to think of the products of that as stuff that one gets with "no effort at all"

[frosty owl]

Ofc there's a limit and getting slugs becomes harder the more you get close to it, but in the beginning you get quite an amount (>100 shards before it's a bit annoying?) for relatively little effort. Conversely, getting the same amount with doggos require prep-work (getting the doggos) and patience (checking on them regularly enough). Also surely more time (the time trade-off depends on slugs remainig in the wild, doggos aviable, yadda yadda)

The shard amount is not the point, but if I wanted just 200 shards, I would probably just harvest the slugs. Would take less time than taming >10 doggos 🤷‍♂️ (ofc, that's based on how long I take on doing these things)
If I wanted >200 though....

[woven elbow]

hey

i have a problem about flow rates

im exactly sure that i calculated everything right and build pipes and pumps but still got problems about flow rate

is there anything important i should know about valves?

[median heath]

Doing fluids in exact amounts almost never ends well.

Valves don't do precise amounts, just close to what you have set - and only when the pipe is FULL.

Also for the record about previous conversation: shard amount is the point.

No. Kill them all.

[woven elbow]

I'm thinking about recording amount of water in the industrial fluid buffer as a video, then getting data from this video and creating a amount vs time graph. If i get the slope, then maybe i can find the exact value for the valve

[frosty owl]

You should probably at @oblique hollow for details on it

[dense cave]

3/4 of my factory is over the spire coast area 😦

[uncut sigil]

My space elevator and all the endgame parts are in the canyon. I feel you. I hope I can finish it before the update happens (or that they leave the canyon undamaged).

[visual grail]

that's unfortunate, but you can use the same site that made that map to move your base. also... devs have been warning since U5 released not to build there

[tropic hawk]

SCIM has the ability to move buildings. Upload the map to it, shift base to new area, download updated save, and you are golden

[dense cave]

thats a lot of moving :S

when is U6 released?

maybe I can see the damage first, then roll the game/server back to an older version

[tropic hawk]

Early June I think...

[uncut sigil]

Experimental hopefully June. Actual release will be longer.

[dense cave]

perhaps my main factory is tall enough to avoid any conflicts

but for sure ill have to move my oil processing center

both of them aparently

[tropic hawk]

What I believe is that the terrain will change in ways which makes maintenance hard, so you can go back and fix things

[oblique hollow]

i dunno really what your setup is and what you expect to find there

[woven elbow]

i thought if i create graphs of storaged water versus time, then maybe i can understand how valves work. But even i have some ideas about how to get data, i dont know how to make them work

but anyway, forget about it. I'll find an easier way

[oblique hollow]

valves are essentially ratio limiters if you dont put an equal number of limits on every pipe

example: one pipe with a valve limited to 150 and a mk 1 pipe at limit 300

if you have an input flow of 300, this results in a flow of 100 through the valve and 200 through the pipe with limit 300

[oblique hollow]

in any case, follow this page

this page describes their exact behaviour

and also has that example i mentioned

[brisk nebula]

is this a bug or woks as it should?

[oblique hollow]

thats intended

[median heath]

^

[oblique hollow]

valves are pressure based

they dont simply cut off flow rate above the limit

[brisk nebula]

😢 okay 😄

[median heath]

People need to stop thinking about pipes in terms of belts 😭

[brisk nebula]

belts are strange on dedicated server. visually they stop without a reason and a second later they flow again

[median heath]

You say "without reason"

I say "Multiplayer Broken. Plz Fix."

[brisk nebula]

okay but first plsss better auto save without laags

[median heath]

How would you accomplish that?

Game has to gather all the information it is saving and update the file.
The more you build the more info it has to grab for that update.

That takes processing power to do.
Which means you're going to have lag.

[brisk nebula]

hmm. When i look at Minecraft. I can't remember to have the same problem. Maybe it need a complete reimplementation of the save mechanism

[median heath]

This isn't Minecraft 🙃

[brisk nebula]

don't know. Maybe storing all builded items in a database could be a solution

store all information in memory and backup with an interval to database

[tropic hawk]

Minecraft actually has a rigid coordinate system and saves every block in the world.

[brisk nebula]

I got no contact with game development :/

[wind spade]

Minecraft saves stuff as you modify it instead of global save every X minutes

[brisk nebula]

maybe this could also be a solution?

[wind spade]

Not really

[tropic hawk]

Not in my experience where it has a global save every X minutes and whenever you save and quit.

[wind spade]

Unloading a chunk saves it

[tropic hawk]

Autocorrect is roasting me today...

[tropic hawk]

Huh. LSNED.

[brisk nebula]

i will hope that auto save will get a rework before 1.0. 😄 Hope dies last

[tropic hawk]

Known bug. Try welding your belt.

[wind spade]

In SF the whole world has to be loaded as everything works even if you're far away

[mystic moon]

@brisk nebula, the other difference is the size of the save, it's a nit larger than with minecraft

[tropic hawk]

File size, yes. World size, no.

[mystic moon]

The reason that it has to stop the game while it saves is because if belts and machines kept running, you would end up with duplicate items and all sort of other problems as well

[remote ice]

in general, having something change while a computer tries to copy it is a recipe for disaster.

Even windows itself has a sort of 'lockout' system that prevents more than one thing from trying to use or modify a file at once, although its subtle enough that you usually don't notice it.

[mystic moon]

Ever tried to move a file and gotten "this file is open in another program"? That's what that is

[remote ice]

for satisfactory, imagine if you... lets say, deconstructed a train+track after the train saved but before the track saved. You'd end up with a train floating in midair if you loaded the save.

at least we're not factorio in that regard. They went all-in for deterministic gameplay, so they arguably have all-around worse performance, even though satisfactory is 3d, because they have to do a lot of important stuff without multithreading

clever optimization can only do so much to alleviate the fact that you can only utilize something like a fourth of modern processor architecture.

[tropic hawk]

<nods like I understand all of this>

[remote ice]

basically, the satisfactory devs just wanted to make things consistent enough for players to make sense of it and get rid of bugs and stuff

the factorio devs went way further, and made stuff so consistent that you could go back to the start of a 200h save, and if you pressed all your keys the same way you did for the first playthrough, the second playthrough would be exactly the same. Enemies would move the same, 'random' events would be the same, everything

this came at a cost, so much so that satisfactory arguably runs better despite being a way more complicated 3d game.

[wind spade]

how do you know that SF is not deterministic? 🤔

[remote ice]

... it probably isn't? I don't know for sure admittedly, but they have proper multithreading that can take up all of your CPU (pretty hard to have full-fledged multithreading and determinism), and they run on a pre-existing game engine - and most game engines are not designed to be deterministic between different hardware.

It might be deterministic if you started the same game at the exact same time on the same hardware, etc, but then many programs are like that. But on the other hand, i wouldn't be surprised Unreal feeds it's pRNG from standard sources like cpu thermal noise, making it truly non-deterministic.

keep in mind, factorio uses a custom game engine. Something as simple as using floating point math (like Unreal and most other engines) can break determinism between different machines.

[cloud swan]

is it more efficient to transport raw quartz or quartz crystal?

[ashen mantle]

What happened here?

I loaded my save and this is what happens

Power shuts off instantly even though I have plenty of capacity

[burnt wraith]

I thought it was a bit odd that so many other games can autosave without freezing, but I guess Satisfactory save files are big so maybe that's it

[wicked tinsel]

you can clearly see non-determinism in multiplayer, its online-sync game

as to whenever simulation itself is deterministic, it probably isnt by virtue of no attention paid to this

since there is no deterministic multiplayer, there is no value in having deterministic simulation

[tropic hawk]

Deposits are randomized

[median heath]

That's what my bank thinks every time they look at my statements.

[wicked tinsel]

randomization doesnt affect determinism in any way tho

[remote ice]

that's a pretty succint way to put it - you have to design for determinism, and the dev team has no apparent reason to do so

[tropic hawk]

xD

[boreal cypress]

you don have capacity to supply full consumption

or you dont supply your generators enough for full load?

[ashen mantle]

It works fine all the time, I have a massive power storage system, it's only when I load my save or like 20% of the time when I delete a railroad track

[ashen mantle]

Nope, nuclear

[median heath]

What is a dyslexic Satisfactory player's favorite power source?

||Unclear||

[tropic hawk]

I would escort you to pun jail, but I am barred from entry.

[cloud swan]

so, I had a plan to maximize rubber and plastic production and it came up with this diagram where it wants me to cycle them among the recycled alternate recipe, dos this actually work?

[burnt wraith]

yes

[cloud swan]

but it also wants my output to be on the residual rubber instead of the recycled rubber

[burnt wraith]

exactly

[cloud swan]

so does that mean that residual rubber is better than recycled rubber?

should I just put all the rubber to be turned into recycled plastic then cycle them over and over?

[burnt wraith]

no it just means it's trying to use up all your polymer resin

[cloud swan]

oh

[magic island]

recycled rubber/plastic basically converts fuel into either. so it's a way of correcting your final plubber:combustible ratio

[cloud swan]

I forgot to add the petrolium production, so it looks more like this at the end

[brisk nebula]

I understand the problem but the implement solution for this is bad. You can only play without lags that are not announced on servers when you build less items. This solution can not be the final version for 1.0. I'm sure that other games that need to save a bunch have a better solution. I don't think that the actual solution be considered as the final solution and we are all happy with it

[proven prawn]

I'm not going to try to pretend I understand the whole problem and possible solution sets, so I'm just going to assume what we have is as good as they can do for now and be happy with

[oblique hollow]

the thing is autosaves are supposed to be announced

and in mp only the host sees it

[elfin wyvern]

And on Dedi nobody sees it 😩

[wind spade]

Given that they are randomised at the start of the game, that doesn't really break determinism

[remote ice]

it has a full fledged physics engine, those tend to be a bit of a hotbed of non-determinism

[wind spade]

I'm not saying it is deterministic. I'm saying that random doesn't always mean non-deterministic

[remote ice]

fair

pseudo-RNG algorithms are deterministic with the same seed by definition, though often they're seeded from stuff like system time or even thermal noise so they're random in practice

[stark bronze]

No one generated seeds from cursor movement

[sullen mulch]

ok right, I did not think to ask this question before I started my nuclear build, but, can you use 40 reactors at 100% efficiency or is there not enough uranium

[vapid kernel]

252 Reactors for the whole map

[sullen mulch]

oh wow ok I might be able to double up my setup

[vapid kernel]

Just gotta use the correct recipe's to get that many and still sink plutonium

[sullen mulch]

right yeah I am going to have to take a moment to figure out the recipes once I have the reactors and infrastructure built

[vapid kernel]

https://www.satisfactorytools.com/production?share=pLhPEC89S4H3GHgnxOhP

[sullen mulch]

i'm basically diving in head first to the shallow end of the radioactive pool here

what fuel do the reactors actually need haha

i'll go plug that in once I find out how much fuel p/m the reactors need

[vapid kernel]

Already worked out in that link

[sullen mulch]

oh wow awesome

wow the numbers get ugly

[vapid kernel]

haha yeah

[sullen mulch]

hahaha god this is going to be a nightmare but it will be so satisfying when it's done

[vapid kernel]

https://www.satisfactorytools.com/production?share=TfA2tcu5s1ujyQuKyKts
https://www.satisfactorytools.com/production?share=befXraJ1psyQtk1t7hLd
Same as above but separately shown for each

[sullen mulch]

thats really handy thanks

[vapid kernel]

Also we may or may not get more Uranium in the future

And the Beacons will get removed after U6

[sullen mulch]

ahhh yeah i just saw that

I was going to say I would wait but didn't CS say they weren't going to mess with the nuclear recipes for the time being

[vapid kernel]

Correct. Not until after U6

[sullen mulch]

right well that is ok because this will take me atleast 1 update to complete haha

so I reckon I might wait until U7 to do everything production wise and just build all of the architecture to go with it

but atleast I can set up the watewr for the reactors in advance

[vapid kernel]

Yeah. I'm making all my buildings expandable

[sullen mulch]

right now I am planning to use the entire dune desert

if I go up to 80 reactors that platform doubles length wise

and already needs to double in width

and looking at the wiki I can see why the numbers get so scary

[ornate shoal]

@median heath are you familiar with such idea of sushi belt limiter?

[median heath]

Seems like more of @frosty owl 's area.

You'd never need something like this in the manifolds I do.

[ornate shoal]

hmm, ok

theoretically, this setup allows me to change the belt speed from 60 to 120 with 2/min increments, so basically what some people have been asking for

[oak atlas]

this is actually very interesting, is it possible to tune it from partially to fully clogged belt?

I'm thinking of using it for a system like this:

1. Plant A contain resources to produce parts (denote B) or consumables (denote C).
2. Set plant A to produce some B & C

3. C has no resource sink. as C approach container limit the belt slowly clog

[ornate shoal]

the main input belt needs to be clogged, otherwise the system doesn't work

and it can throttle down from 100% to 50%, but no lower than that

and for faster belts you would need probably different setups

but it's very quirky to make and requires lots of calculating, for which i'm too braindead

for setup on the picture, belt loop that silica runs on can contain exactly 60 items, including the merger buffers, that made calculation easier, so the lenght of the loop is actually relevant here.

[oak atlas]

so it's less math heavy to use a packager-unpackager to tune for the exact number, I'll just need a relay system to activate the belt limiter.

[wind spade]

the 2/min per silica seems pretty arbitrary, isn't that dependent on the loop part length?

[ornate shoal]

yes it is, the length of the loop is precisely measured in the picture, once you move the merger/splitter the math doesn't add up anymore 😄

[median heath]

People keep asking for belt tuning but it really isn't needed because of how standard manifolds work.

Same reason prio mergers aren't needed.

People just keep wanting to play this game like the other one instead of playing it like it is.

[ornate shoal]

but people want to use other stuff as well, not just standard manifolds, sometimes

[median heath]

🤷‍♂️

[wind spade]

may be a good point to add to one of the signs, otherwise people can build it differently and get different results (not that it matters anyway, as Sev pointed out)

[median heath]

I get that.
But as you just showed there are ways to get what you want within the current system.

People don't expirement with the system though. They just cry about not having things from the other system.

[oak atlas]

I cried when I spent the weekend doing the 8-to-8 balancer layout in 3D, then realize I can just use the methods mentioned above & some overflow splitter to achieve what I wanted

[median heath]

Aye.

[ornate shoal]

yes, i should have specified that distance between merger/splitter centers is 29 meters

[median heath]

Infinite resource nodes changes how things should be done compared to games that have finite resource nodes. 🤷‍♂️

[wind spade]

or just connect each 1 belt to 1 output 😛

[oak atlas]

That's a lot of belt when you're dealing with 1 GW nuclear plant raw material processing

[median heath]

Only 1 GW?

[wind spade]

then you're not doing 8:8 balancer, but 8:1 merger into 1:8 splitter? 🤔

[oak atlas]

no, there's 8 belt unbalanced at close to mk5 limit, I'm trying to balance them before sending into further processing

[wind spade]

why not put each belt into X machines further down the line? (X being the amount to exactly eat the whole belt)

[oak atlas]

my dumb ass back then decided to build it perfectly for symmetry, and it's too late to scrap it when I realized that

[median heath]

It's never too late to scrap something.

100% materials refund.

[oak atlas]

but....but....that's mah beAuTiFouL FacToRy !

[wind spade]

oh the classic "building the factory before considering the logistics" 😛

[median heath]

Indeed.

[median heath]

You're saying the second, better edition would be less beautiful?

[oak atlas]

Gambler fallacy, I already invested so much into it might as well try to retrofit it

[median heath]

People seriously need to learn when to let go 😭

It's like having kids.
First one is never your best work. 😉

[oak atlas]

and because of how poorly I design it the retrofit turn out to be more work than building from scrap XD

[oak atlas]

wait WTF

[median heath]

😂

[oak atlas]

btw, since the ability to add signs come out I'm playing around with power poles to build some sort of power control panel at hub so I can easily decide what to turn on or off for remote sites.

It does feel like my actual job looking at a SCADA system, but just on/off control

[median heath]

Why turn things off?

[wind spade]

inb4 "you're building power wrong"

[tropic hawk]

<Fuse breaking>

[median heath]

Probable.

[tropic hawk]

The heck was your old apostrophe?

[oak atlas]

multi usage, sometimes I don't need that much power so I'm better off turning off fuel plant and let overflow splitter send them into plastic/rubber/computer line

of course, with some pipe elevation built in as well to act as overflow

[wind spade]

so your factories aren't running at 100% by default?

[median heath]

Oh God... are you a yellow-light player? 😬

[oak atlas]

always design to maximize resources node to parts, but designed with ability to turn off stuff when it can be redirected

a more modular approach to things

[wind spade]

so... your factories don't run at 100% 🤷‍♂️

[oak atlas]

it's 100% when all on

[median heath]

Which means they don't run at 100%.

Because you turn them off...

[wind spade]

if you turn one factory off, where do the extra resources go, if everything else is already at 100%?

[oak atlas]

usually overflow to a secondary priority items or consumables factory

the belt work is crazy though

[wind spade]

so that factory isn't running at 100%

[median heath]

I will never understand how some people play. 🤷‍♂️
Not saying the approach is wrong or bad, just something I don't understand.

[oak atlas]

it's secondary on my list so I don't really care

[wind spade]

well, you said

it's 100% when all on
which just isn't true then 🤷‍♂️ I was just confused because you were saying different things, so wanted to clarify

[oak atlas]

*primary factory are always 100% when all on

so I have a use case like this:

I have limited nodes and too lazy to build another train network.
the factory run like this:

Example:

Primary:
plate/rod/screw/etc. [Resource sink on/off by control panel, with configurable production rate in modules, each modules connect to on/off for production rate control]

Secondary:
make space elevator parts/consumables [Controllable factory, on/off by control panels]

So depends on whether I want to direct all the resource to primary or secondary I can use on/off control on it, when either of them is on it is always 100% efficient.
So say I have 5 modules in primary, I can make a split of resources like 100/0, 80/20, 60/40,...etc.

TLDR: I hate train so much that I designed a system to not use trains

[wind spade]

I would just build every factory near nodes it needs, so there's no primary/secondary, no need to turn stuff off and no complex logistic for overflowing between factories

[median heath]

The TLDR to me is "I choose to not use more nodes" 🤷‍♂️

Also it is refreshing to meet someone who hates trains in contrast to all the train fetishists.

[oak atlas]

I'm a drone fetishist though

[median heath]

So tap more nodes and drone things in..

[oak atlas]

The battery consumption is insane, I already have around 100 drones on my map

[wind spade]

belts do not consume anything

[oak atlas]

no thank spaget man I will not join your cult

[wind spade]

belt != spaghetti

[median heath]

Trucks 🤷‍♂️

[oak atlas]

I hate trucks more than trains, I'm glad those truck AIs didn't get deployed on a Tesla

[median heath]

👍

[wind spade]

if you hate majority of transport options, then maybe building near nodes would be better 😛

[oak atlas]

I always maximize the nodes I find to 250%,
so when I'm producing 100 motor per minutes as excess byproducts of my nuclear plant, the definition of "near" becomes vague.
The same plant could be pretty far away end to end

Man I need a life

[ornate shoal]

ok, i have made my best attempt to boil the math of my belt limiter down to the point it makes sense to me and this is the result

[frosty owl]

Yep, seems to work well enough...
Though I'm not sure how accurate the system would be on the long run, I'm inclined to think the system can self-balance itself.
Item duplication could ruin it quite drastically though :/

Certanely, a very interesting contraption 💯

[ornate shoal]

i've been messing around with it for few hours and it works well if math is done correctly

ty senpai

[wind spade]

item duplication happens on save load

[frosty owl]

Aye. Once a duplicated silica is on the belt, the setup would be ruined 😞

[oblique hollow]

huh what

[frosty owl]

https://www.reddit.com/r/SatisfactoryGame/comments/uqrjfj/how_loadbalancing_for_solids_eventually_always/?utm_source=share&utm_medium=web2x&context=3

Imo, it would be be great if we knew for sure wether this (and the fluid load loss) will be addressed or not...
Contraptions like the one just posted by @ornate shoal, one-input feeds for multi-input machines, fluid loops and funny contraptions... There's a ton of content that would be pretty much useless (and so would be creating/researching more such content) if such issues are just bound to stay

[oblique hollow]

ah funny.

Btw i have an equation for uneven belt mk splitting

[wind spade]

they are already useless

[oblique hollow]

i figured that stuff out long ago. same for valve ratios

its all rather simple and kind of a "duh" moment once you understand it

[frosty owl]

I'm inclined to think the results can change depending on FPS...

[oblique hollow]

hmmm not really, it just depends on input speed and output speeds

[frosty owl]

Can you elaborate?

[oblique hollow]

the rest is of course up to general belt rules

aka if you use a 5000 segment mk 5 you of course have issues

heres an example

now of course the question is: whats the expected output on A and B?

[wind spade]

wouldn't that be 75/75 due to internal splitter inventory? 🤔

or nvm

[oblique hollow]

both belts have a capacity way above 150/min

though its not 50/50 output (if i remember right)

[wind spade]

120 > 150 confirmed

[oblique hollow]

ah

oh well 120> 150/2

[wind spade]

https://tenor.com/view/doug-maclean-mac-kyper-kypreos-nick-kypreos-gif-22264825

[oblique hollow]

so best case its 50/50

but its not really 50/50 if you test it

[wind spade]

would be slightly more on mk3 side I'd say

[oblique hollow]

it should be more like
B = 46.1538/ min
A = 103.8461/ min

mk 4 and mk 2 belt should be nicer

[median heath]

Why would there be this much disparity?

[oblique hollow]

so if A is a mk 4 belt:
A = 120
B = 30

that should be easily tested

[oblique hollow]

ratio of 270 to 120

[median heath]

Splitters care?

[oblique hollow]

or rather: ratio of 120 or 270 to 390

[oblique hollow]

test what i just wrote with mk 4 and mk 2

[median heath]

At work so cannot.
But didn't think splitters cared. Thought they just cycled outputs and deviated only if output wasn't fast enough to keep up or was backed up.

[oblique hollow]

lemme test it out real quick

[ornate shoal]

the problem is that for example at 150 items/min, the belt you are describing is not full, it has some gaps and some filled sections, so the split is variable there

[median heath]

Like the first image you shared should be 50/50 because both can handle 75/min output.

[oblique hollow]

its precisely because you are moving 150/min on a 270/min belt

the speed matters

[ornate shoal]

so basically on a 150 items /min belt you have some sections that are 270/min and some sections that are 0

and once 270/min section reaches the splitter the rules change, the 120 belt cannot service that section at full speed

[oblique hollow]

in that case the easies test is to have 150 evenly spaced to see if that matters, buuut i guess thats a bit too much

[wind spade]

it would most likely differ based on spacing between items

[ornate shoal]

if you place items on the splitter evenly, then you should get even splits in youre example

[median heath]

If you hook a miner to it at 150 then you have 150 evenly spaced.

If so, should be an equal 50/50 split on the first image.

If it is not even then, yes I can see the mk3 having higher because 120 < 135

[ornate shoal]

yes, the miners should put out items 1 at a time

[wicked tinsel]

the problem likely stems from the fact that slower belt is slower, not that splitter misbehaves

it tries to push new item, but slow belt is still in "item hasnt moved far away enough to fit new item", so it instead goes to other belt

[oblique hollow]

hm.... i might actually be misremembering

and the condition was that one output belt is slower than the 50/50 split of course

for pipes i know that this ratio rule always applies, but for belts not i think

[lime karma]

I realized my oil refinery was using a pure node, wasting a bunch a oil per minute

[burnt wraith]

I just tested it with a mk2 belt carrying alternating plates and rods
split to a mk1 belt and mk2 belt

[lime karma]

Two normal one pure

[oblique hollow]

happens. you are not forced to use nodes completely

thats not what efficiency is about

[lime karma]

I know, this is actually good news for me because I was hoping to get some more plastic produced

[oblique hollow]

efficiency is about making the most out of the resource you actually do need

[oblique hollow]

ok so... the odd splitting rule doesnt apply to normal belt mk mixing. it handles exactly as expected

if you do a 50/50 and the belts can handle it, you get 50/50

buuuut i think im now remembering when this weird stuff happens

[median heath]

It's because we figured out their just cycle outputs. M/L/R iirc.
So as long as the output can take something when it is q'd to do so, it will.

Have to move shit faster than an output can handle, like having pockets of 270 on that mk3 which would create pockets of the mk2 not being able to handle 135.

[oblique hollow]

i knew i wasnt crazy

it does happen as i expect it

but ONLY if the input belt speed is higher than the sum of output belt speeds

[median heath]

Yeah, that's working the way we know they would.

[oblique hollow]

there is no backing up here of course

[median heath]

Aye.

[oblique hollow]

the output belts of the splitter never get blocked

so now, belt ratio rules apply

which leads to what i wrote

its necessary of course for the input speed to be fast enough to supply both belts at once

what your actual input rate is doesnt matter

... as long as its smaller than the sum of the outputs

if you use mk 5 belts, you can do ratio math for a lot of combinations.
the limit of this is mk 3 + mk 4 belt i think

[burnt wraith]

if the input is 750, how would it not get backed up trying to output half of that to a mk2 belt?

[oblique hollow]

refer to what i wrote above.

750 is not smaller than 120 + 480

so this doesnt apply

your limit for this setup is 600

which leads to saturated belts

750 would work with mk 3 + mk 4 outputs

which once again is the saturation point

mk 2 + mk 4 belt is a 20/80 splitter, with limit 600/min

mk 3 + mk 4 should be a 36/64 splitter, with limit 750/min

[burnt wraith]

So in order for it to split unevenly, the input has to be over half the smaller belt's capacity. And in order to not get bottlenecked it has to be under the sum of the belt's capacities.

[oblique hollow]

lets see, lemme run a test with only 50/min going in

should be 10/min to 40/min

alright, that seems to work?

[oblique hollow]

it seems only the second statement apllies

50/min on a mk 5 going into a splitter with mk 4 and mk 2, does a 20%/80% split as expected

the actual rules are:

1. the input BELT speed needs to be greater than the sum of the output belts
2. the input ITEM RATE needs to be no greater than the sum of the output belts

so the smallest ratio splitter is mk 2 and mk 1 output with mk 3 input

[median heath]

Is that another way of saying output belt speeds need to be equal to or greater than (1/Number of Outputs) * Input Speed?

[oblique hollow]

output belt speed can be whatever you want for the ratios

[median heath]

Depends on what you want?

[oblique hollow]

if you want a 20/80 split, or 36/64. or 33/66

[median heath]

If you want an even split then you need to make sure (1/NO) * Input >= Input

[oblique hollow]

just have to make sure (Sum of Output Speeds) < Input Belt Speed

we are kinda getting into tricky territory with words now

whats input to you? belt speed? or actual item rate?

[median heath]

Oh you're intentionally making ratio splitters.

Item rate.

[oblique hollow]

yes. do you try to avoid them in your math? xd

[median heath]

Sushi Manifolds don't care about ratios, just total amount. 🙃

[oblique hollow]

if you DONT want weird ratio stuff happening..... then
Input Belt Tier <= Sum of Output Belt Tier

that way you get normal output math.

so 1/2 or 1/3

above that, you get belt tier ratios

[median heath]

Aye.

[oblique hollow]

i could make a list of all the ratios you can achieve

[burnt wraith]

it split 50/50

60 input on mk5

[median heath]

So if (1/NO) * Input >= Input then you get even speed.
Beyond that you go into ratio splits.

[proven prawn]

i think i started to hate belt math when i realized i couldn't actually get deterministic behavior for a perfectly balanced system, i hate belts so much

[oblique hollow]

does not align with my output

those weird splitter/merger combos are rate limiters. precisely, 30/min limiters
the upper one must be 40 because it keeps overflowing

and the bottom one is slower than the 30/min from the upper limiter

so it must be slower than 30

[burnt wraith]

I think the difference is using 100% of the belt 25% of the time vs using 25% of the belt 100% of the time

[oblique hollow]

maybe....

this is where batch size gets important i guess

[burnt wraith]

basically your items have to be running along the belt fast enough for it to skip the lower rated belt

[oblique hollow]

aye, else its a 50/50 situation, most likely

oooor your test needs to run longer and eventually, some cable will land on the other side

[burnt wraith]

I made it a closed loop and I'll let it run

[oblique hollow]

weird.... i tried to emulate what you have by limiting the output speed of each screw constructor to a mk 1 belt and merging that onto a mk 5

now i get a 60/40 split with mk 4 and mk 2

instead of 80/20

[boreal cypress]

dont forget, that splitter/merger have an inventory of 10
did you account that too?

[burnt wraith]

as far as I know it works how Sev said; the inequality comes from the splitter trying to send output to a belt that can't handle it

[oblique hollow]

yes, they dont store it really right now

none of them

[boreal cypress]

maybe its the same like an ICS which have 2 outputs but dont distrubute it evenly

[oblique hollow]

the actual split right now seems to be 0.3666/0.6333

[oblique hollow]

its more even than that, ISC is not as nice as this

now wtf kind of ratio is 36.6% to 63.3%

[proven prawn]

well

[oblique hollow]

... or is it actually 0.36/0.64?

which is weird cuz thats the mk 4 and mk 3 ratio

but i have mk 4 and mk 2

[proven prawn]

us trying to understand belts when all of the code is frame dependent and is those inherently unpredictable

[frosty owl]

Iirc they output clockwise

[median heath]

M/R/L

Had to search for it.

So yes. Lol. Clockwise.

[glad vigil]

Me when train shows that it transports 478 items per minute while being fed from full mk 4 belt

And items don’t get stuck anywhere

[thin carbon]

what is the best way to get power from oil?

these are my available recipes

[glad vigil]

The first thing you do is turn all crude oil into heavy oil residue

Using alt recipe

[median heath]

The standard answer is "go nuclear".

If dedicated to oil:
Simplest is just build Diluted Fuel plant.

[glad vigil]

Second is diluted water fuel

That’s basically the current meta of all oil processing

[thin carbon]

i need power now and i dont have acid set up nor uranium nor have i found uranium nodes +no ideea how to set it up

So i get my node i overclock it then do math to see how much heavy oil i get

ok that is some math and building to be done

[uncut sigil]

If you can be bothered with the coal and sulphur, Turbo blend fuel is the best one.

[thin carbon]

how much of both?

[median heath]

Just do Diluted Fuel.

It is the simplest setup for what you need.

[proven prawn]

well on the other side, because eventually, at some point the recipes for nuclear fuel rods are going to definitely change, wont be until a later date past u6, it might be a better idea not to invest to much into a nuclear setup right now anyways, because the production lines are going to be changed later on, unless you really need the power now and are prepared to do the production line changes later, specifically because at some point beacons is going to be removed from the alternative fuel rod recipe, but we don't a definite time frame on when this is going to happen.

[ornate shoal]

why do you think nuclear recipes are gonna change past the bracon one?

[proven prawn]

because its been confirmed multiple times by the devs this is going to happen at some point

[median heath]

They did say even recently that before 1.0 they will do a sweeping pass on all recipes in the game.

[thin carbon]

so i have a pure crude node that i max out

so 180*2.5 so 450/min

i turn that into diluted fuel somehow and feed that to the generators?

[oblique hollow]

pure nodes are 600 at max

not 450

[thin carbon]

ops so pure is X2 and impure is 1/2 my mistake

[oblique hollow]

600 crude oil = 1600 fuel, if you use heavy oil and diluted fuel alt

[thin carbon]

ok so i add 1600 (maybe 1550) here https://www.satisfactorytools.com/production ?

[wind spade]

pure is x4 and normal is x2

[thin carbon]

ok

20 refineries and 16 blenders + x turbines , blenders will be hard but it is just a matter or time

how much power can i get from the 1600 fuel ?

[burnt wraith]

20,000MW

[wind spade]

53.3 GW according to my calculations

with diluted fuel and alt HOR

[thin carbon]

HOR?

[burnt wraith]

eh

[wind spade]

heavy oil residue

[burnt wraith]

wait how

[thin carbon]

20GW is great but i am tempted about the 53GW

[wind spade]

nvm, I can't read

1600 fuel, not 1600 oil

[thin carbon]

ok so 20GW then

[burnt wraith]

Fuel Generators use 12 fuel and make 150 MW
1600 * (12/150)

[thin carbon]

ok so this is the best way ok time to start building

[wind spade]

you can get 44.4 GW with turbofuel, but TF is pointless to do IMO

[thin carbon]

TF?

[burnt wraith]

turbofuel

[thin carbon]

is it that hard?

[burnt wraith]

the amount of time it takes to set up is time you could spend on Nuclear instead

[proven prawn]

tf is going to be great in u6, finally a real practical use with turbo ammo, infact that is probably going to be the one and only reason i setup a turbofuel plant to make that spicey new ammo, etc that requires turbo fuel

[thin carbon]

600 oil would make 800 turbofuel but it costs S and polymer out is the same

[proven prawn]

i like that polymer resin is even useful, you can even make fabric or rubber/plastic with it

[thin carbon]

here we go

[soft scarab]

I was just setting up my 198 fuel gens this weekend and decided to use the polymer resin to make rubber & fabric as the feeder inputs for my future iodine filters when I set up nuclear

[frosty owl]

@oblique hollow would you happen to know if the fluid load-bug doesn't apply if machines are (1) idle (full output or lack of input)
Or (2) not powered?

[oblique hollow]

applies to idle machines too

as for not powered, i think too

[frosty owl]

👍

[analog loom]

I currently have mk1 everything except mk2 belts. currently i have access to 3 unused pure iron nodes. i would like a factory to produce reinforced plates and maybe rotors too. I can pull up to 120 wire/min and 120 screws/min from nearby factories as well. Given this info, what would be the best choice from my 3 recipes in the #screenshots and should i do both rotors and plates, or just one or the other

[old sparrow]

should 80 plates, 300 screws, and 90 rods a minute be a good ratio for production?

[soft scarab]

Depends what you’re producing. Some alternate recipes completely eliminate the need for screws

But generally yes make whatever you want in whatever quantities

And if you need more you can almost always find a way to make more

[old sparrow]

at the moment im building up to making heavy modular frames if that helps, thank you :D

[tropic hawk]

Only if you need a butt ton of screws. I would recommend 30/min unless you are handcrafting with them a lot.

[old sparrow]

ok, now the copper. How does 240 wire and 90 sheets a minute sound?

[tropic hawk]

Again, depends on use case and capacity.
If you use a lot of wire relative to sheets, yes. I would personally get some cable running, but that's just me.

[old sparrow]

thank you for reminding me about cables! :D

[tropic hawk]

You are welcome. I have run out way to many times because I like straight wires so I make a lot of poles. I would also recommend about 4 SC capacity for concrete with a constructor feeding into it. Plenty for building unless you make a LOT of platforms

[old sparrow]

sorry, SC capacity?

[tropic hawk]

Storage container

[old sparrow]

ah, thank you :D i appreciate the help!

[tropic hawk]

My pleasure. If you have any questions, feel free to ask me.

[analog loom]

I currently have an excess of 120 screws/min, what do i put them towards

[tropic hawk]

Storage for hand crafting. Personally I only siphoned 60 screws/min, sink excess siphoned, and send the rest to making MFs and RIPs until I unlock stitched iron plates.

[analog loom]

I already have another 120 for handcrafting 😅

I’ll just belt them to the rotor and RIP factory for now

[ashen mantle]

My assembly director system and supercomputer factory is finally online, after a day and a half of slowly rolling on all the systems and half a week of building

And some belt fuckery because quickwire is a pain in the ass

[median heath]

QW is Screws 2.0

[ashen mantle]

QW is worse than screws because you can't remove it with alternate recipes, it is the alternate recipe solution

[median heath]

Hence the 2.0 🙃

Also imagine automating Project Parts...

[ashen mantle]

I consider it a fun challenge

I automate everything I can, batch processing is a pain when all my time is spent building

Though part of me wanted to make a manual batch factory just as a concept

[median heath]

Batch?

[boreal cypress]

I like Quickwire

[ashen mantle]

Rail car of material x, rail car of material y, unloads into a mini factory, loads back in to another rail car, everything is segmented but very scalar and allows manual control of what items are being produced

Completely pointless, but it would be fun to try to optimize

[median heath]

I'll take your word for it because I can't wrap my head around that description.

[ashen mantle]

Essentially each step of a process is isolated into it's own factory

Nothing flows automatically

[median heath]

I'm with you until you say nothing flows automatically

Because my outposts are similarly specialized, but continually flowing.

[ashen mantle]

I take a ton of items, they get one step of processing, then load back into a vehicle, that vehicle can take them to, say, a high speed connector factory. Obviously takes large buffers, but really if you wanted to and made some very capable rail infrastructure, I guess you could automate it

[median heath]

Doing logistics is what makes outposting fun.

[ashen mantle]

Again, useless, but I think it would be fun to try to set up

Logistics is my favorite part of this game

[median heath]

Also half of my stuff is moved via truck, not train.

[ashen mantle]

I've been moving to trucks a lot more, they're surprisingly practical for a lot of use cases

[median heath]

I've been saying that since update 2...

[ashen mantle]

I'll admit I used trains and long conveyor belts for way too long, but trucks are great

And I don't need to worry about signals! It's great!

Though I do suck at driving and always take a few tries to record a good path without nicking a rail or similar

I repurposed my small three node rubber factory purely into a packaged fuel refinery and hooked it up to my rail network

Fuel is distributed to each station by rail

I've been getting rid of my short rail lines and been having long distance high capacity rail lines unload into trucks for last mile delivery

Been greatly reducing congestion on my lines

Also giving me a lot of free space that used to be dedicated to rail stations

[glad vigil]

Railroads are satisfying to build though

[fierce ruin]

i like it more to build a train track on the whole world

[mystic flower]

Hello, does the type of conveyor belts affect the ratio of the spliter? Like If I need to split 40 to 20+20 - and first output of 20 is Belt 1 and second output is Belt 4 - the ration will be still 20 - 20 ?

[burnt wraith]

The splitter will try to send one item to each output one at a time, so it will always split evenly across all 3 unless there is no belt attached to the output or the belt it’s outputting to is not empty

So if you split 780 between mk1 and mk5, you’ll get 60/720
But if you split 120 between mk1 and mk5 you’ll get 60/60

[mystic flower]

Ok thanks - this is little bit more difficult I have expect 😄

[wind spade]

screws are pointless to produce since they are not used for building. Unless you want to produce a few to have them for unlocking milestones, but those need like 4k in total and then you don't need them anymore

[wind spade]

in your case yes, it will be 20/20. But usually people don't care about exact ratios, as you can just use a manifold (row of splitters) and it will "balance" itself after some time

[vapid gorge]

It was a while ago you posted but I like dedicating 1 pure iron node worth of iron to each basic part as you tier up. You need a lot of parts.

[fierce ruin]

How do I show that every group of odd order is solvable?

[vapid gorge]

Take a 1st or 2nd year university unit?

[zinc eagle]

what is the best method to calculate production if i dont have the best stuff available yet?

[wind spade]

that depends what you mean by best

[glad vigil]

By hand using pen, paper and calculator

[wind spade]

or online tools linked in pins or #welcome

[zinc eagle]

okidoki

i was checking already tge pins

ill look into better when ill have free time

[tropic hawk]

Is that a self plug?

[tropic hawk]

Easiet: online calculators for it.
Most utilitarian: pencil, graph paper, and calculator

[craggy forge]

aight so, can someone help me calcuate how much turbofuel i can produce with 780pm Coal & Sulfur, and 1200m3 pm Crude oil?

and not to forget i have Turbo blend fuel alternative, i also got heavy oil residue alt + diluted fuel in the blender

[vapid estuary]

the sulfur is the limit. you can get about 10% more turbofuel if you find more sulfur.

also that produces a lot of polymer resin

so, plan for that

[craggy forge]

thanks, i have no idea how to use that planner thing

[vapid estuary]

https://www.satisfactorytools.com/production?share=l17S533i77RgSBzW50WZ

the above is a link to the planner for your situation

if you're on a real computer, the left column has production, items, and recipes sections. In Recipes I turned on all the alts. Under items, I set your resource limits

Under production I asked for max turbofuel, then there we are.

[craggy forge]

👍

kind of concerned how im gonna set up my turbofuel plant, i did the math and i need 367 fuel generators

and if im not wrong, i can make 52 gigawatts out of 1560 turbofuel per minute

[vapid estuary]

seems overkill to me, but that's not a real reason to not do something in this game

[craggy forge]

i mean its always better to have more power than having blown fuse

[vapid estuary]

i would base your planning around the capacity of a mk2 pipe - 600. 1560 / 600 rounded up is 3, so your factory will need 3 'wings' with a 1/3rd of the refineries feeding 1/3rd of the generators

you can bring up each third as your power demands go up, you might hit nuclear well before you build the third third

[craggy forge]

oh yeah im also concerned about another thing

is my potato pc even gonna be able to run it when its completely finished?

[vapid estuary]

let us know!

i run on an i3 10100 and it's ok

the gfx load for individual items on belts is what usually kills people i think

[craggy forge]

i plan on building crude oil pipeline anways, the conveyors shouldnt be a problem since my sulfur node will be really close + my polymer resin will be sinked straight away or used for other things

im just worried about the object amount

367 fuel generators, for sure lots of foundations, all the things needed for turbofuel

atleast gonna be a thousand objects or more

[vapid estuary]

unless you have a 80286 you'll be fine

[wind spade]

what if I'm on imaginary computer 🤔

[vapid estuary]

i've no idea how your site renders on a phone or tablet. but yah

[wind spade]

horribly, because I've done pretty much no optimisation 😄

[uncut sigil]

It's not terrible on the ipad (at least U4, didn't have a U5 tab link handy). At least for viewing.

[wind spade]

u4/u5 should be similar in terms of layout

[uncut sigil]

It doesn't matter too much tbh. If I am planning I want SCIM open as well, and that seems to not like mobile without huge memory issues.

[frosty owl]

||Shameless ||

[versed violet]

Do resource deposits (the rocky things, not miner things) respawn after game load? Or after map update?
trying to decide if to clean them up yet.

[median heath]

Map update I believe. Same as foliage.

[versed violet]

ok, no point vacuuming before hannah gets to it then

[median heath]

Only like... 4 more of those maximum though 😉

I really wish you could those.

Would make cleanup faster and not have random ores in my inventory.

[versed violet]

Only if it doesnt destroy foilage

[median heath]

I'm mainly thinking the ones in the deserts that stick out as massive eyesores.

[versed violet]

S.I.N.K. Project stage 2 complete - all uranium and quartz nodes have been claimed for processing!
Next: exploit all bauxite on the mapp.

[cedar mica]

Sinking raw resources or doing a few steps?

[versed violet]

Currently sinking what I'm not using, this is S.I.N.K. Project, not Process-Project 😉

[cedar mica]

Feels that 1 step processing, still fills that requirement.

[versed violet]

will throw down some constructors for quartz once all nodes are tapped, but it suffices for now. points are points

[native oak]

I am looking at what I need to automate heavy modular frames. It looks like I need to automate the regular modular frames and the encased industrial beams to be able to make the first heavy frames at a crafting bench.

[mystic moon]

I wouldn't bother handcrafted any but what's absolutely necessary

They're very slow

[burnt wraith]

hand craft or go exploring and grab a ton from crash sites

I do most of my exploring for the playthrough while the space elevator parts are being created for package 2

[native oak]

I just need enough to build 10 for a manufacturer

[burnt wraith]

if you grab enough computers and heavy frames you can unlock Mk4 belts early

[sharp canyon]

Would this work to split 300 and 180 coal for my generators and a small factory? (selecting Coal on the mk1 and mk2 and Overflow on the mk4)

Miner (480 coal) -mk4-> Smart Splitter -mk2-> 120 coal |
| -mk3-> Merger -mk3-> Coal Gen. (180 coal)
-mk1-> 60 coal |
-mk4-> overflow (300 coal)?

[burnt wraith]

I think it would work without a smart splitter

it would work with only 1 splitter

the 240 coal would fill up the generators and after that they would only accept 180, pushing the remaining 60 into the other output

[sharp canyon]

Thing is, I got a truck transporting the coal to the generators, so that would mess it up every time it empties the station. That's why I thought I'd do it like that.

[burnt wraith]

yeah a normal splitter would try to send 160 on each belt, but since the mk2 only takes 120 and the mk1 only takes 60, it would overflow the remaining 140 along with the 160 on the mk4

[sharp canyon]

What I was worried about with just a normal splitter was for the mk4 to maybe take more than 300 and kill my power. But I take it the way it works is an equal split until overflow and then it does another equal split on the remaining belts and so on (in my case, just the one mk4)?

[burnt wraith]

it tries to send an item through the middle output, then the right output, then the left output, then the middle output, etc. If it can't send out an item (e.g. there is no belt or the belt already has an item on it) then it goes to the next output.

so exactly what you said, it splits evenly until one belt is filled up, then splits evenly until the next is full, then sends all into the remaining

[sharp canyon]

I see. I'll give it a try with a normal splitter when I get to it and see how it works. Thanks!

[burnt wraith]

you're welcome! happy building!

[frosty owl]

@burnt wraith @sharp canyon I'll TLDR a conversation from yesterday for you:
When splitting a belt with a normal splitter, only SOME situations where the outputs are of different MKs actually result in an even split, it's quite common to see what @sharp canyon was worried about: the speedier belt getting more items than it should.
Using a smart splitter is the best option for this particular example imo. Also, take care of making sure the maxed out MK2 belt is only 1 segment long (see "b2b issue")

[wind spade]

split to same mks should be equal though, right?

[frosty owl]

Correct. I thought it was implied

[wind spade]

I wasn't part of the yesterday's conversation (at least not part of the whole convo) so I checked just to be sure 🤷‍♂️

[frosty owl]

I think it was like... In the premise 😅

[oblique hollow]

funky belt stuff

[proven prawn]

i had no idea how unreliable belts actually were until just today😭

[wind spade]

they are reliable if you use them correctly 🤷‍♂️

[proven prawn]

nope https://www.reddit.com/r/SatisfactoryGame/comments/uqrjfj/how_loadbalancing_for_solids_eventually_always/?utm_source=share&utm_medium=web2x&context=3 they duplicate items and cause systems to fail

[wind spade]

"fail" as in "your belts will be full, but still working"

[proven prawn]

apparently item belt duplication is a thing and my designs are doomed😭

[wind spade]

the "fail" here is that load-balanced setups will fill

but it doesn't hurt efficiency at all

[proven prawn]

no because my nuclear waste will be duplicate and back up and all products produced from it

nuclear shut downs, insert more doom here

[wind spade]

you should have more waste processing than you need anyway, for the occasional lizard doggo waste and for security in case something happens to the waste processing setup

but other than that, it's not "fail" 🤷‍♂️

[proven prawn]

no its called out to accurate for what plants would logically produce

[proven prawn]

and yes it is fail, because item duplicate is noway documented or accounted for behavior in normal setups, its mentioned absolutely nowhere this should happen

and its also fail, because that means my entire grid shut downs, no more power

[wind spade]

in case you accidentally disconnect some belt or something and the waste processing will stop for a while, it will never be able to catch up (if you have it exact). Hence why I said "you should always have slightly more waste processing"

[proven prawn]

so yeah failure in this case is quite critical, as my entire world literally has no power when it fails

[wind spade]

same as you have buffers with water off-grid for restarting coal or power storages for preventing blackouts

[proven prawn]

So yes belts are quite unreliable, far more than i predicted

[wind spade]

your problem is that you don't have extra processing. duplication is just one of the ways stuff can go wrong

[proven prawn]

and all because splitters duplicate items when they shouldn't...so thats cool in a oh looking my entire waste processing plant is going to fail now kind of way

[ornate shoal]

one way to make it catch up is sink some uranium rods 😉 i've done it a lot because my plutonium processing is failure as well

[wind spade]

so it's kinda weird to blame belt duplication for the fact that your setup isn't issue-proof 🤷‍♂️

[proven prawn]

no because item duplicate isnt documented

nowhere on the splitter does it say it'll make extra items

[wind spade]

my point is that there's more ways stuff can go the bad way

so you should make the processing have more capacity than the amount you produce, to prevent all of these issues

[proven prawn]

yes but all of those are obvious, item duplication with something that has no indication that can happen is not one of them

[wind spade]

it's not documented as it's not a feature 🤷‍♂️

[proven prawn]

correct its a bug, but one that has now doomed my waste processing plant

[frosty owl]

Belts full of unsinkable items aren't much of a "good" thing

[wind spade]

or you just add one more waste processing machine and you're fine 🤷‍♂️

[proven prawn]

and im supposed to do that with something that isn't documented and i have noway of knowing how much i should add🤦‍♂️

[frosty owl]

Literally the only thing one can't precisely account for (in regards to belts)...

[wind spade]

any amount will be as olution

as long as you have more capacity than you need

also I just documented it 😛

[proven prawn]

sure its good, its good for no power, because nuclear waste has now backed up all the way into the plants and shut them down

[wind spade]

you also can't precisely account for how long will the factory be down if you accidentally disconnect it 🤷‍♂️

[proven prawn]

because if i didn't have enough waste already, look out its duplicated belt waste now, fun for all 😄

[frosty owl]

I can precisely NOT disconnect it
(Also, there are ways to do that, though annoying)

[ornate shoal]

is there any more information about the duplication glitch? like can mergers also duplicate items? does the item need to be in the splitter buffer to be duplicated?

[wind spade]

I mean yeah but if you have a setup that your entire factory relies on, adding a few extra machines just to be sure can't be a bad thing

[frosty owl]

Answering in orde
Mergers don't duplicate iirc
Items do need to be a specific state: game unclear on wether the item is in the splitter or on a belt, resulting in both results being true (quantistic splitting?!)

[wind spade]

you can then hook your lizard doggo waste container to your setup without fear of overflowing waste

[proven prawn]

its superposition splitters, both values are true now