math-and-meta

[primal flicker]

It is both a 2 way splitter and a buffer.

[spice egret]

Is there an easy name for that contraption that I can just google and nobody will have to explain it? 😂

Why is the top right only 450 and not 480?

[median heath]

No, ISCs do not function as two-way splitters.

[primal flicker]

Do the outputs not alternate?

[median heath]

No.

[spice egret]

Oh. How do they flow out then?

[median heath]

One of them always has slight priority.

[primal flicker]

Well that's mildly infuriating...

[median heath]

Always been that way 🤷‍♂️
Unsure if they will change it for 1.0 or not.

[spice egret]

Is there an actual consistent number to that?

[median heath]

A lot of people would use them as cheap 2:2 balancers if they worked the other way.

[primal flicker]

I don't see a problem with that.

[median heath]

Not.. really.. sorry.

[spice egret]

Whaaaa… then how can that photo be sure of the outputs on each line?

[median heath]

Neither do I, but they would have to be recoded to function that way.

[median heath]

What photo?

[spice egret]

This photo

[primal flicker]

If there is some pseudorandom behavior, you could turn a set of ISCs I into a random number generator...

[median heath]

Link to <blocked message>
🤷‍♂️

[woven wharf]

https://cdn.discordapp.com/attachments/558721941410807812/1179139961094672456/Screenshot20231128-14202800000.png?ex=6578b2aa&is=65663daa&hm=c6f34599ae1c7e486237dc3119d2299c28dcaad0b2533aa0e135edb657c1767f&

that

[spice egret]

Yeah that

[median heath]

Yeah that doesn't work.

[woven wharf]

ok so im not stupid xd. cause i didnt understand it fully

[median heath]

It may "kinda work" to the point it works "good enough"
But it isn't precise.

[primal flicker]

Sev says this won't work. But someone blocked someone 😆

[spice egret]

Oopsie

[median heath]

If they recode ISCs, this is worth revisiting.

What are you trying to accomplish?

[primal flicker]

We talking 95% precision? 99%?

[woven wharf]

I have 4 conveyers with 275 iron ore/min. I need these 4 lines to go into:

1 line with 360/min
1 line with 450/min
1 line with 290/min

Keep in my mind i dont have mk 5 belts yet.

https://cdn.discordapp.com/attachments/558721941410807812/1179134052238884974/image.png?ex=6578ad29&is=65663829&hm=91fe18997afc635f991d68375f6eccefaf78e79883b3b6c58999fe051205fd4f&

[median heath]

That's a question for someone else.
My care level stopped at "is not precise" and I accepted it.

[woven wharf]

this is what im trying

[primal flicker]

6sigma mentality

[spice egret]

Indeed

[median heath]

1. Use Tools 😭
2. Injection manifold.

[spice egret]

In my noob experience any type of manifold that’s possible seems to be the easiest solution 😅

[woven wharf]

im doing injection manifold right now (i think... im not an expert xd)

[median heath]

Screenshot your Manifold and I can confirm/deny 😁

[median heath]

Correct.

[prisma kraken]

injection manifold was my first suggestion, but sometimes people say things won't work without trying them out

[woven wharf]

well its kind of like between 1m foundations with like a 3m gap so its a little hard to show xd. looks like spaghetti to me

[median heath]

Do you have smart splitters?

[woven wharf]

yea im using smart splitters

this is like the best way i can show it. its not finished yet ofc.

[median heath]

That is not injection manifolding.

[spice egret]

Doesn’t look like it to my noob eyes

[woven wharf]

idk what it is but i think its going to work xd

[median heath]

If you have smarts there is a simple-ish way to get the numbers you want.

4 lines of 275 right?

[woven wharf]

yessir

[median heath]

And you have mk4 belts?

[woven wharf]

yessir

[spice egret]

Injected manifold is essentially normal a splitter line manifold except somewhere along the line there’s a merger input right?

[woven wharf]

yea im using mergers somewhere in there

[spice egret]

Just doesn’t look like a manifold to me lol

[median heath]

Take 3 lines.
Smart splitter with main output to a mk2 belt.
Other output to Overflow.

Combine the 3 mk2 for your 360. 👍
Combine the 3 overflows for a 465.
Smart splitter off the 465, main output onto a mk1.
Normal splitter the mk1 in half, then in half again.
Take one of the quarters, merge with last 275 line for 290 👍
Merge remaining quarters with what WAS the 465 for a now 450 👍

[median heath]

Not quite.

[spice egret]

Nvm then

[median heath]

Will explain when the above is confirmed/denied understanding.

[woven wharf]

this seems pretty easy. and i think i understand it :O
you are truly a conveyor magician. i will do this now, thank you!

[median heath]

👍I do what I can.

[woven wharf]

[median heath]

Injection Manifold is normal Manifold with a merger on the line, yes. But you need a smart splitter immediately prior to the merger to ensure everything that is supposed to be in the section before injection stays in that section.

[spice egret]

So you don’t let anything from the first section pass into the next section? Is that the purpose of the smart splitter?

[median heath]

You let stuff pass.

You just only let what is meant to pass, pass.

You don't let stuff that is supposed to stay pass.

[true junco]

idk if a smart splitter before the injection merge is strictly necessary. but it is definitely advisable as "best practice"

[spice egret]

Is there a simple example of how that would be configured on some simple line?

[median heath]

The only scenario where it isn't necessary isn't used that often in my experience.

[spice egret]

Is it just overflow forward or

[median heath]

You just replace the last normal splitter in the first section with a smart and set the forward output to Overflow.
Done.

[spice egret]

Oh. Okay, that makes sense then

[true junco]

well. if your line is made correctly almost everything prior to the merge could be consumed off the splits. so whatever is passing that smart splitter could be well below what will cause the merger to fail to take in the full injection. that and you can also put a smart splitter on the injection line prior to the injection merger and overflow that to another injection merger further down the line. but thats just a less effective way of handling the issue resolved by the smart splitter on the main manifold prior to the injection.

[median heath]

It isn't about the merger failing.

[spice egret]

What’s the purpose?

[median heath]

Making sure the last machine in the section gets what it needs to.

[spice egret]

Okay what I figured. That’s pretty simple then

[median heath]

Because if the amount needed is a 50/50 split of the of amount supplied-> you're fine.
But you also don't need an injection manifold then. Because you will perfectly feed that and the next machine and can just end the manifold and start a new one.

Most likely you get something where the amount needed is 30 and the amount supplied is like 45.
Which if you split in half -> failure.
So you use the smart to make sure 30 goes into the machine and the 15 gets passed on to the injection point.

[visual bane]

hey guys! i updated my program!
https://www.reddit.com/r/satisfactory/comments/1865j3y/the_satisfactory_factory_planning_utility_has/

[spice egret]

Yeah, for some reason I was just picturing the last machine in my constructor line as being fine and didn’t understand why adding onto the line with an injection would break it, but obviously it isn’t currently being split

[median heath]

👍

[woven wharf]

(i know there is clipping but whatever) i think i did it now. only thing i need is to connect the last 275 to the 15. idk if you can really see if it looks right but i hope it is

[median heath]

If I had a proud face, I would be wearing it right now.

[woven wharf]

:D Thank youuu

[median heath]

And then I would scold you harshly because we don't clip in this family.

[spice egret]

I will admit

[woven wharf]

😭 im sorry. but its going to be hidden underground so noone else will ever know :)

[median heath]

Acceptable loophole.

[spice egret]

When you first send the photo I thought that was supposed to be your manifold and I immediately had a 😳 … then I looked closer

[woven wharf]

xD

[woven wharf]

do you guys know if this works? or do i need to put them on seperate mergers

[median heath]

The horror

[woven wharf]

😂 what do you mean, its beautiful

[prisma kraken]

fyi, i'm about to start running a test on my idea with the ISC's

[woven wharf]

let me know how it goes

[prisma kraken]

more constructive to do the experiment than argue, i wish others would do the same

also, that merger looks like bad mojo, lol, i only ever swallow the lift one way, let me know if THAT works 😉

wish me luck

[woven wharf]

oh god...

[prisma kraken]

i actually had the testbed more or less primed, it was just a matter of dropping the correct numbers of constructors

it looks like it is actually going to work, the 290 manifold is almost completely backed up, once that happens, i believe the 450 and 390 will have the correct number distributed between them

sorry, it was 360, wasn't it?

i have an extra smelter in there, lol

[woven wharf]

360
450
290

the 360 are going to foundries

the 450 and 290 are going to smelters

[prisma kraken]

yeah, i'm just using smelters, easier to build

[woven wharf]

looks like this for me right now. one of the pink ones are overclocked

i think i got the beltwork correct. you just cant see it cause its under the floor

everything should get the correct amount

[prisma kraken]

i'm going to assert that i believe that if you have x amount of consumption on a line, it shouldn't matter what the machine types are

its worth tossing an asterisk next to that assumption for furter review, but i'm just letting this test run as-is

[woven wharf]

yessir

[prisma kraken]

i can say now that the design works

the 450 manifold is now nearly completely full

once that happens, all remaining stuff will go to the 360 manifold

that one is taking a bit longer, of course

nah,it doesn't work, just noticed that ore is starting to back up in a container

it needs that second set of isc's

[woven wharf]

Noted for later use. thank you :)

[spice egret]

Okay so it doesn’t technically work from a “math” standpoint but just relies on everything backing up and balancing itself out, correct?

Earlier I was just trying to figure out how those numbers worked

[prisma kraken]

i think i figured out something that does work

i realized that essentially each manifold wants a full 275 line plus a bit of the fourth, so why not just split one line into 3 and merge that into the other 3 lines?

in theory, it should overfeed the 290 manifold, and then the 360 one, and the splitter should then direct the right amount to the 450 line

[spice egret]

Makes sense

[prisma kraken]

if it works as i think it should, its pretty simple:

nah, doesn't work... it would need a priority merger to work correctly

[spice egret]

Why doesn’t it work?

Made sense to me that the 290 would get backed up and then it would be giving more to the other 2

[prisma kraken]

the straight through line for the 290 miner starts backing up

there's too much going to that line, and the merger for it prioritizes the lower volume belt

so that causes the miner to start jamming up

i'm thinking maybe something can be done to use a smart splitter to prioritize the 450 line, but pretty sure that doesn't work either, which brings us back to a full-on 4:3 balancer

i'm trying to think if that will even work properly

when i say 4:3, i mean split each of the 4 lines into 3 equal, and then merge 4 of the 12 lines for each of the 4 outputs

that's worth doing the experiment to see, but i have to run for a few hrs, i'll build that up later and let y'all know

[warm basalt]

Hay im looking for a 7 to 8 belt balancer, but haven't had any luck. Does anyone have any idea on how to make one, because ive failed like 10 times

[prisma kraken]

Following up, a full on balancer does work as thought

[snow dove]

split each belt into 8 then merge 7 of each?

[vapid gorge]

there's apparently guides on a reddit post? easiest way would be to group the producers into 8 groups in thefirst place though

clocking is your friend

[prisma kraken]

for that many inputs, it may be simpler if your volume is way under belt spead to do something like split one line into two and merge that into two other lines so you're instead doing a 6->8

always ask yourself as well if you can control the input and output speeds through other means

another thing i'll point out is an 8:8 balancer should function if you just don't connect one output

[warm basalt]

this is all great. thanks

[desert crater]

how much coal per min does a coal generator use

[vapid gorge]

press e on the coal gen

always be looking at the control panels of machines

[desert crater]

does not show it

just says insert fulel

[vapid gorge]

put a coal in it and it should
you can also use the wiki for things like that

!wikisearch coal_generator

[brisk shore]

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

[fair mirage]

hello, if I have a line pushing 240 iron ore and i split it between 1 mk3 belt and 1 mk1 belt, will that make it a 180 belt and 60 belt? i assume so but I'm doubting myself XD. I'm setting up a simple steel factory and have 2 pure nodes.

[vapid gorge]

it'll split between 2 outputs as evenly as it can so yes

if you have a manifold system where 1 end needs 60 and the other needs 180 though it'll automatically self balance if the consumption and production match up

so in that case you don't even need different belts

[fair mirage]

alright thank ya. i know its a bit of a dumb question but i didnt know if some logic to it would cause the mk1 to get like 55 per minute or something. nah im basically going ahead and pulling 15 ore out of the system as to not cause a back up

[vapid gorge]

perfectly reasonable question

are you just turning them all into ingots?

[ebon crater]

rates for my next factory(green are outputs I care about, red are inputs not being produced at this factory)
edit: creates 4340 fuel/min, or 54.25GW

[fair mirage]

as of current yea, based on the recipe asking for 45 per min im running 4 foundries and turning that extra 15 into iron ingots for a small funnel of screws

[vapid gorge]

ah well consider this:
if this is all happening at basically the one location you could just run that 1 belts with 240 past smelters and foundries to make steel and iron w/o making 2 distinct systems and it'll balance on it's own

(even if its not the same location it still works just looks weirder)

[fair mirage]

true. I was just gonna load the unsmelted ore that got kicked and throw it on a ttruck back to my main set up

[vapid gorge]

up to you - if you're shipping multiple item types from one truck you have smart splitters already right?

[fair mirage]

indeed but not the full ability to make them on the regular. thats gonna be my next step

[vapid gorge]

yeah look theres basically infinite ways to set up logistics and using clocking and selective merging (only grouping the outputs of selected machines) is super powerful. It all depends on how you like to build and what your goals are.

if it works then a set up isn't 'wrong' but there are definitely simpler or more complicated choices you can make

and some methods work better/worse depending on exactly what youre trying to do

Oh tip with screws - try to build the screw machine next to what uses them. Makes life easier. And the Steel Screw Alt is great for that

in fact you could just make all the steel for screws too there instead of iron if you find hte alt

[fair mirage]

unfortunately I don't have it yet but since I know now that that's a thing I probably will. I have like 6 wrecks around me atm that i cant open cuz I don't have power to this place yet. Thank you for the help though, much appreciated.

[vapid gorge]

No prob 🙂 the big benefit of Steel screw is that 1 steel beam turns into 52 screws so often you can just have 1 constructor next to the machine that uses screws to feed it. Think of Steel beams as a box of screws that just get unpacked 😄

it's not hte most resource efficient option but might be the easiest logistic choice

the most resource efficient is probably steel ingot > steel rod > base screw

[fair mirage]

ok thats insane, as of current i have i think 18 constructors making screws? thats to feed 2 assemblers making the alt reinforced iron plate that takes like 250 screws and 4 assemblers making rotors

im prolly wrong on those numbers but not close to it to count

[vapid gorge]

yeah steel screw is GREAT for space saving

you do get significantly more screws per resource with the Steel Rod Alt, but it's as lot more machines, space and belts

260 screws pm will satisfy the most screw intensive recipes

It's one of hte reasons people choose NON screw alt recipes tbh

You can pretty easily cut out Screws and Iron Rods from your world with the right alts

[fair mirage]

makes sense. ngl i just like seeing screws go down the belts, other than that ya, they're a pain

[vapid gorge]

hah yeah but it's when you need 100s pm for 1 machine that the number of belts you need gets to be a pain

[mystic moon]

Steel screw + copper rotor ftw

[vapid gorge]

Steel Rotor supremacy!

[fair mirage]

what in the world!

i couldnt imagine

[vapid gorge]

that recipe is pretty silly w/o Steel Screws

[fair mirage]

i could see that being a nightmare very quickly without it

[vapid gorge]

But there are alts in the game whichn means you never haveto make screws again if you want 🙂
In fact you can produce every non weapon item in the game w/o - coal, iron ingots, screws, iron rods

I think everything else is needed at some point

(all automated stuff, not buildables)

Anyway - there’s a million solutions to the game xD enjoy

[fair mirage]

thank ya thank ya, now i gotta go deal with someone tryna hack my steam lol. have a good one

[true junco]

Resource efficiency is pretty subjective. Some insist the steel rod to default screw is most efficient... some say neither steelrod nor steel screw alts are efficient because they both use a rarer resource, coal. Vs default and caste which use no coal at all.

Personally. Steel screw is my favorite. Its almost as resource efficient as the steelrod to default screw path. But its the most logistically efficient by a huge margin.

[vapid gorge]

well you get more Screw per iron/coal from steel > steel rod > screw than Steel screw so that's fairly objective in that sense. the choice then becomes steel source or iron source for the screws

[true junco]

Basically most ways of making screws mean for the more complex items you need multiple machines to feed a single machine that needs screws.

Steel screw can out produce (with overclocking) literally any machine that requires screws as an input (even with overclocking) so you end up with 1 to 1 or even a single machine feeding many machines needs for screws.

Steel screw is 💪

The only exception i see is that you cannot feed 1to1 the "flexible frame" alt if you overclock it to much... but you can OC that alt past the limit of a mk5 belt anyways, so thats not something to do anyways.

Steel screw constructor at 250% is 650/min.

Heavy Flexible Frame at 100% is 390/min... so OC250% is 970/min, so not possible. So 200% is 780/min. And would take more than a single steel screw constructor to feed.

Anyways. Overclocking heavy flexible frame production is definitely not common. Lol

[grand talon]

that feeling when you realize u messed up math for turbo fuel generaaation / pipe max

[vapid gorge]

just a bit of pipe rebuilding surely?

[grand talon]

got 80 blenders pumpin out 9000 turbofuel/min now to figure out how to evenly supply generators

[vapid gorge]

groups of 600 fuel in pipes

easy peasy

the bigger issue is the number of fuel gens that are going to sprawl

[grand talon]

figured it out 4 blenders feeding mk2 pipes at 450/min feeding 40 gens per line. 20 lines to get it even per pipe w/o valves

[vapid gorge]

never use valves

[true junco]

Yeah... im lothe to do another big fuel generatorum for that reason. Even OC'd to 250% it can be difficult to configure all those generators in a way that i personally find to be SatisFactory.

[median heath]

But they look pretty 😭

[true junco]

I use valves all the time... 😆

[median heath]

Wait, can you adjust a valve that isn't connected to anything?

[true junco]

... i actually dont know.

[median heath]

Because I know the adjustment visuals do show on the model itself.

[true junco]

Guess im going to find out... lol

[thorny cedar]

try clipping it to a wall 🤷‍♂️

[median heath]

Build pipe.
Snap to end of pipe.
Delete pipe.
Interact with floating Valve.

Plz.

[thorny cedar]

the wall thing is prob less steps

but idk about valves i only know pumps and crossings clip to them

so im sure valves do to and you can rotate them with mousewheel :3

[vapid gorge]

just clip through them

[median heath]

That's what the testing is for.

@vapid gorge because if you can adjust with no pipe, you can set them to make them look pretty and then clip through them.

[vapid gorge]

why would you need to test if it works? anything clipped through shouldn't affect it? or am I confused at what you're trying to do

[median heath]

Can you ADJUST them if they are not ATTACHED is what needs to be tested.
Because:

the adjustment visuals do show on the model itself.

Which could be used to make it PRETTIER

[true junco]

And because the adjustment is dependent on the attached pipe ... at least when there is an attached pipe. So. Once this update is finished ill check sev

[vapid gorge]

ah couldn't you have it connected, adjust it, then disconnect?

[median heath]

Idk, another thing to test 🤷‍♂️

[true junco]

@median heath you can set a free floating valve that has never known a pipe connection. also. the range is the full mk2pipe range in the dialogue box.

[median heath]

PRETTY

Time to "use" Valves

[prisma kraken]

cycling back to a conversation from many hours ago, this contraption does in fact work for making a 4:3 split... it isn't a great way of doing it because it requires one of the ISC's on the output side to completely fill with product before things start overflowing the correct ways, but it does in fact work

it isn't a great way of doing the split in and of itself, i will absolutely agree, but if the rear input containers were instead a train station, well, that IS something useful and space-saving

[trail obsidian]

What's the goal for this machine?

[primal flicker]

What advantage would this offer, vs one that only uses splitters/mergers? (3-to-4 shown, but could easily be reversed into a 4-to-3)

[delicate chasm]

That looks ugly as sin, so there is one advantage.

[primal flicker]

😂

But it was up on a spire where I would never look at it.

[delicate chasm]

That's what I thought about my ex-wife too, but I'm here to tell you that dismantle is Zod's real gift to humanity. (edit: it's "Zod")

[primal flicker]

And it was compact

[wind spade]

@cunning heart here's some help for the alt recipe choices: #math-and-meta message

[prisma kraken]

the advantage to my contraption is that it could be built around a receiving and transmitting train station which isn't a huge space savings, but having something smallish on both sides could be more convenient

also, for reference, i did a 4:3 balancer last night as well, it is just about as unwieldy as what you pictured above:

[primal flicker]

Nice
I didn't bother with lifts, even. Just clipped belts and said fuhgeddaboudit 😅

[prisma kraken]

personally i think the injection manifold to be the superior option of the three, but i think it is worth exploring the differences

another top-down

[primal flicker]

And sometimes solving a puzzle suboptimally can be more fun . Which is really more worth optimizing.

[prisma kraken]

took me a bit to figure out a 4-way vertical merge can be done like so

really figuring out that compact vertical merge was very worthwhile all the other time spent

that's something that's just been nagging at me for a year or two

and i don't think any solution to the problem is really sub-optimal, just they have different characteristics that make them more or less suited to specific situations

one thing i'm finding as i use trains more and more for infrastructure is that i'm pulling in several cars of ores or ingots (or plastic/rubber, etc), and what is the most desireable behavior is if all the train stations deplete at the same rate

that's a trivial problem if you're doing one miner per train car and just build lines to process each train car at miner's speed

[primal flicker]

I haven't touched trains at all.
I have one single tractor on a raw quartz route. That's my whole experience with vehicular logistics. 😆

[leaden pendant]

you'll need to have over-speed belts, though. cause it needs to catch up from the delay.

[prisma kraken]

but if you are mixing together some pure mining node's 780/min output with some normal nodes at 600/min, to even out the rate the cars fill (advantageous for not needing to run a second train on the line, since 780/min limits the RTT to around 4.5 minutes), it sometimes does end up being useful stuff to understand

tbh, you have to be building pretty big to even encounter these problems

[mossy flint]

That's fine. In early mid game you only need it for select resources, and even those resources can be covered with portable miners anyways

[prisma kraken]

really i think its more stuff like phase 4 elevator parts where this stuff comes into play

[primal flicker]

OK but that's not my personal context.
I'm very much of the "just belt everything" school of thought, and challenging myself to NOT, this save.

[prisma kraken]

and it really is a handful of problems... cheap silica is one of the nastiest ones

[mossy flint]

Where is your base?

You can set up a circular train route going around the general area with your bases that can visit all of your factories and take them to one central location

[primal flicker]

Dotted area for my next major project.

[median heath]

Imagine having a main base

[primal flicker]

[mossy flint]

You could group togehter a couple of factories and then make a longer route to reach the farther factories

[mossy flint]

Bro doesn't have the HAB

[median heath]

The game doesn't have a HAB.

The game has a HUB.
Also, I have the HUB and still don't have a main base.
Because imagine having a main base.

[primal flicker]

Definitely relocating the hub as soon as this fuel power plant / rubber/plastic/fabric/gas filter factory is launched.

Then again I may not dismantle/relocate it until I start aluminum production.

[median heath]

Mine is just off somewhere with a nice view.

[primal flicker]

I zipline or drive a truck (if I need to transport stuff)

[cunning heart]

THX

[mossy flint]

Get the explorer bro

[primal flicker]

Small trunk.

Do mergers give priority to any of their inputs?

[oblique hollow]

not really

in fact, they give equal priority to every side

#satisfactory-experimental message
theres this technicality resulting from that

[primal flicker]

Does this mean that in the case of two Mk2 belts and one Mk1 merging onto a Mk3, the Mk1 gets priority and the Mk2s can back up?

[oblique hollow]

merger indeed always tries to fairly accept inputs.

if you have 100 and 780 entering a merger with a mk 5 output, then the 100 will be entirely consumed, and the rest will be stocked up with 680/min from the incoming mk 5

[primal flicker]

In other words, it can be worthwhile to flow balance before merging, but not necessarily in all cases.

Or at least slow down the low-volume belt so that the belt ticking is more proportional to the inputs. 🤔

[median heath]

Not having to build multiple km's of belts.

[deft lichen]

I'd argue good use cases for trucks are pretty narrow, but they exist

[median heath]

I would angrily argue against this assertion.

[deft lichen]

I know you like them

[median heath]

Not even about preference.
They math supports uses for them.

[deft lichen]

the downsides don't outweigh the positives to me

[median heath]

People just preferentially dislike them, even though they know the math.

[primal flicker]

The math supports always using belts for everything.
It's the framerates that don't support that.

[wind spade]

depends what math

[primal flicker]

The "no upkeep cost" math

[wind spade]

so, one parameter out of many 🤷‍♂️

[primal flicker]

List your top three prioritized parameters then?

[wind spade]

depends on situation?

[primal flicker]

Most likely to be prioritized, then, based on your experience or common applications, I guess.

[wind spade]

any of

• simplicity
• terrain
• distance
• complexity of build
• verticality
• resource efficiency
• reusability
• future plans
• ...

[primal flicker]

I have found belts to be simple, do not consider terrain, distance, verticality, or build cost to be obstacles to their use, and I probably don't plan far enough ahead to envision repurposing logistics that I place down in the first place 😆
My approach is clearly imperfect, but at the end of the day it is a game, and its job is to entertain.

[true junco]

As a job, this game doesnt pay me enough to run belts everywhere for everything. Lol

[mystic moon]

Also the setup time doesn't favor belts

They're really tedious to run cross-map

[true junco]

They literally do not pay us enough for this. Lol

[mystic moon]

Wheras rail networks only need to be set up once and then you can just set routes

[true junco]

@ivory wedge

30 crude, 80 water, 90 plastic

Scale to suite.

Can sub diluted fuel blender for diluted packaged fuel loop with packaged water.

[prisma kraken]

i'll add that it naturally scales to the size of 150 oil input, at which point you start exceeding mk4 belt limits

i'd be interested to know how many people actually build recycling loops

every now & then i get asked about how you belt the thing up

[true junco]

I have. And will again.

But i have a ton of HOR+DF to Generators i plan to build, so ill have a decent amount of "residual" plastic and rubber before i really need to ramp into recycling.

[prisma kraken]

what i've been doing in my current save is actually something a little different

when i have a factory that has a demand for plastic or rubber, i've been tapping 300 oil and tuning the loop to produce the exact amount of product i need, and using the excess fuel for power

it's been kind of nice just having a few extra gw of supplemental power from it all

[true junco]

Makes sense.

[prisma kraken]

it also has made building it quite a reproduceable build

that has been a big win

[true junco]

Sounds good.

Im vaguely just making fuel power plants with residual plastic/rubber on the side. Then when I ramp up nuclear ill pivot the fuel into recycled loops and shutter the fuel gens.

[prisma kraken]

yeah, i'm shoring up the production lines i need for nuclear atm, but not shy about dropping recycled loops, i have 80 gw of tf power right now

i'm hoping that's enough to get nuclear rolling

what i'm doing now is a big motor build that will kick off all the stators i'll need for max nuclear

ECR's are just fricking annoying to make

[true junco]

Yeah. Still not sure which path is best for ECRs. Lol

[prisma kraken]

one of those components that can be made easily, but if you take the easy route they eat the world

when you're maxing, you have to start horse trading between copper/quartz/caterium & bauxite

i've not found a path through ECR making that doesn't require silicon HSC with cat or default CB

ECR's are going to eat a lot of caterium if you use ai limiters or default HSC

surprisingly, neither path is that much different on resource costs

but that's why ecr's are annoying 🙂

[vapid gorge]

@wind spade : idea for tools - a couple of buttons that can sort out a pick of recipes for a given production based on fewest types of resources and one for fewest unique items in a build - and after the sort be able to go in and do some tweaks by altering recipes as needed. No idea how feasible that might be 😄

[wind spade]

I don't think it's possible for current algorithm to do "fewest amount of [different things]", I'd have to find some other algorithm that can do that (which I'm not sure is feasible)

[prisma kraken]

the number of permutaions of base resources seems small enough, could you just loop through them smallest to largest?

[wind spade]

12 resources = 4096 calculations

"small enough"? 😄

[vapid gorge]

orly? it can't be look at different recipe groups and go 'this uses only 3 resource types , whlie this uses 5' ?

[prisma kraken]

amortized 1/2 of that, i guess

[vapid gorge]

liek caterium computers in the right chain only need oil and cat

[prisma kraken]

plus you'll have a lot of degenerative cases like 'limestone only'

[wind spade]

the problem that not many recipes use raw resources, and if it uses intermediate resource, you don't know from which raw resources it is made (or can be made)

[wind spade]

sure but even with 1k calculations, that's still 1k more than I'm doing now 🤷‍♂️

[vapid gorge]

interesting, didn't think it'd be that complicated

[wind spade]

and even though one calculation is super fast, 1000 of them would definitely be slow

[prisma kraken]

i know, but you're tool is pretty snappy as is

are the calcs done on server or client side?

[wind spade]

(not to mention that the new gen tools I'm designing will have mod support in mind, so raw resources can be more than just those we have)

calcs are server side (but would be doable on client side if I find a nice library for it)

[prisma kraken]

well, that's a deal breaker, lol

[vapid gorge]

ok what about having a data set where the program does a required output for arbitrary amount of a resource and it does the calculation once to figure out the steps for fewer resource types and it can just draw on that?

[wind spade]

uh, I'm not sure how that will work with the current algorithm

[vapid gorge]

fair enough, I bow to your wisdom

[wind spade]

currently I'm using linear optimisation - a set of inequalities and finding lowest/highest variable set that fullfills the constraints

so basically I just put all the data in a "black box", let it run and then draw graph from result

[vapid gorge]

and having a table to pull info from for already solved chains wouldn't mesh?

[wind spade]

the calculations are not done in any way like going one by one recipe or passing through recipe tree

[prisma kraken]

i'm surprised all the game problems are able to be solved with just that, honestly

[vapid gorge]

unfortunate but not the end of hte world 🙂

[wind spade]

I mean yeah I could pre-calculate things, but the amount of different possible requests user can make is just... insane

f.e. even with all the same recipes and raw resources, only changing amount that you want to make can lead to different production chains (hitting limit of one resource)

[vapid gorge]

based on volume sure but what if there was info just on someone wanting 'x computers' , the table knows the minimum types and recipes for it.
or 'x computers and x motors' , chain already found ect ect ?

[wind spade]

and then you start combining it with things like maximisation and multiple products, etc.

[vapid gorge]

hmm what if it was separate from the volume calculations and you could just ask for a chain of recipes w/o volume?

[wind spade]

multiple products at the same time are the biggest "feature killer"
the fact that you can add a new product which causes production line for the old product to change is just... bad for any of these features

[vapid gorge]

oh yeah, even I could gather that xD

[wind spade]

(basically the solver can change the production of old product to something slighly less efficient, but making some byproduct in the process which gets used in making of the new product)

[prisma kraken]

what you kind of are looking for is something that drills down from an end goal and allows you to choose recipes based on ingredients

[vapid gorge]

it wouldn't help if it was kept apart from actual numbers and just looking at the chain? Since types of resource to get to X Y Z item types would be much simpler

[wind spade]

I mean... it's all just linear problems in the end 🙂

minimise [ore usage]
with [list of recipes]
while [production of plates] >= 10

[prisma kraken]

(which is exactly how i use runesun)

yeah, i'm just surprised that all the game problems can be boiled down to that is all

[wind spade]

not really, see

(basically the solver can change the production of old product to something slighly less efficient, but making some byproduct in the process which gets used in making of the new product)
that isn't related to amount of stuff being made, just different combinations of stuff

[vapid gorge]

huh cool beans

[wind spade]

f.e. using residual fuel to make fuel but then changing it to normal fuel because the resin byproduct from it can make extra rubber for newly added item

[silk hornet]

4 250% reinforced plate assemblers, mow many iron ore does that require? all default recipes except the bolts, im using the cast recipe, also the r plate recipe is the bolted one

[wind spade]

that's pretty easy to calculate tho?

just start from the RIP assemblers and calculate backwards

[silk hornet]

i changed the assemblers to 200% bc the bolts wont be able to keep up

[wind spade]

*screws 🙂

and you can just build more screws if you don't have enough 😉

[silk hornet]

i dont want more 😭

imm jsut hook up 3 nodes to it

3 normal nodes

IT USES SO MUCH POWER

[wind spade]

build more power then

[silk hornet]

its so satisfying to watch all the iron ingots weave in and out of the system

[silk hornet]

i did 😎

cast screw is the best alternate recipe in the game

theres like 1 that fights it for #1 spot

[wind spade]

it's not 😄

first of all, alt recipes are all subjective, so there's no "best". Obviously you can have your favorite one, but that doesn't make it best

and second - the cast screw recipe gets outshadowed by pretty much any other screw/screwless recipe, basically as soon as you get steel (and partially even before)

it's pretty much only good for "pre-steel screws" and that's if you even want to make screws at that point (and not just use screwless recipes)

[topaz tree]

cast screw -> default reinforced plate fans when iron wire -> stitched plate enjoyers enter the room

[flat orbit]

Is it possible to do this? I am making 15 renforced plates per min (5 per machine) and its telling me to I need to split those 15 into 5 modular frame assemblers. But each modular frame take 3 per min? How could I possilby split that???

[true junco]

Steel screw is 👑 and recipes that require Screws are 💪

[median heath]

Manifold.

[flat orbit]

ah

I dont believe we have that

[median heath]

???????????????????????????

It isn't a thing you have. It's something you build...

[true junco]

All the RIPS fit on a mk1 belt. Manifold of mergers does that part.

Then a manifold of splitters distributes the RIPs to the MF assemblers.

[median heath]

!wikisearch Manifold

[brisk shore]

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

[flat orbit]

oh im stupid hahahaa

thanks lol me and my buddy have been sitting on our calculators and paint tryhing to figure out how to do this

[prisma kraken]

this looks pretty great as an early RIP build:

[primal flicker]

If you start near accessible caterium.
Early on, I don't think there's anything wrong with using copper for all the wire and cable. As soon as you hand raft a few RIPS you can make use of 120 ppm ore rates...

[random charm]

I prefer manifold too, but in this case, 1:5 balancer is very easy

you split output to 6 , 5 of them connect to machines and remaining one back to input.

[prisma kraken]

something else is that whenever you see a need for a 1:5 split, you can always turn it into a 4-way split with clocking the consumers at 125%, and a 2 way split at 250%

[primal flicker]

And a 6way at 83.3333% (and save a bit of power)

[primal flicker]

I tend to try to make everything at once, early on. With copper and iron close to each other, I would typically have something like this.

[prisma kraken]

eek!

i wouldn't actually bother with dedicating the screw output for storage... when you need them it is easy to drop a constructor and fill it with beams or rods and a few power shards

[cinder silo]

I added one underclocked constructor tossing screws to stores for the hell of it, not like it impacted my production at all.

[primal flicker]

That represents the overflow from unused iron elsewhere. It's easy to just let it build up in the machines and grab a stack if you need to build a shop.

I should post a screenshot of my "temporary starter factory" from this session. It's not quite what I show above but reasonably close.

This one was a similar approach except I wanted to see how well it works to keep iron and copper in completely separate production lines. I think I used bolted frames.
#screenshots message

[prisma kraken]

yeah, making screws except for rotors is an afterthought for me in the early game too

[prisma kraken]

so i just sat down and looked closer at what you are doing, and i have to ask why not push the copper use to 120 instead of 90?

[primal flicker]

You could, but there's not a lot more you can get out of it in terms of early game usefulness. It just translates into excess wire/cable/sheets.

[true junco]

Personallt I love: Pure ingots > Fused wire > Stitched plate.

Just wish there was another alt for iron plates that wasnt oil based. For earlier uses at least.

[prisma kraken]

yeah, i just saw it as a combo that is very low power, i know that fused is better all round, but i just did a hub0 run last evening (up to coal power), and did the caterium research a little early... fused & cat wire kept popping up for me, so i figured 'lets look at stitched + cat wire'

i want to poke at that a little more and see whether there's any synergy with it & using the copper for sheets->ai limiters

[true junco]

Yeah, cat wire has some advantage in simplicity and compactness. Being made in a constructor and fed by only one resource is worth considering.

[prisma kraken]

i'm not finding it really compelling, but if you're going to hang out on biomass for a while, it isn't a bad combo

[true junco]

Well. I gather that both fused and caterium wire become increasingly more important once you need copper powder. Lol

[prisma kraken]

heh, yeah

[true junco]

I do generally like the idea of using very dense producers for local construction materials at least.

[prisma kraken]

tbh, fused wire solves a lot of problems in the game

autowire is a heckuva lot less scary with it

but 90 wire for 16 gw vs 12 gw for 3 constructors - the extra 4gw is saved on the mining

[true junco]

wait... 12GW for 3 constructors?

[prisma kraken]

mw, sorry

total brain fart

[true junco]

Lol. Fucked me completely up.

[prisma kraken]

something i'm kind of thinking a bit about is what is the 'minimal factory' you need to build

so in phase 1, you can make 5 rip/min, for 60 iron/min, 4 rotor/min for 45/min, 15 rod for 15/min, there's 120/min

and that gets you an assembler on mf's too

still need a constructor for plate off of another node

but if you're tapping another node with a miner, you might as well toss 4 smelters and constructors onto plate, and mix stiched plate into it to

i'm kind of approaching it from a perspective of 'what recipes can i guarantee i'll have?' and trying to figure out a fixed build order for everything

[primal flicker]

The map-wide resource distribution still supports pure iron ingots to iron wire in that case. In my opinion, obviously.

[random charm]

sure, there is more ways with under/overclock. I just wanted to mention, that balancers are relatively easy, split to equal parts and send overflow back to input. But as i mentioned, i prefer manifold. Thru is, that it starts slowly, but comparing to huge balancers...still i can pre load internal storages and manifold in 1sec do same job as balancer.

[leaden pendant]

i typically build manifold, once done wtih build go explore and start something else. by the time i need what I did from that build, it's fully loaded and working. basically balances itself by the time i need what I need

if it's a chain of micro builds into a more advanced componenet, I'll power on the early sections. so all the smelting/iron plates/copper wire leading into RIP into MF, I'll wire up the smelting and base iron/copper stuff. then ill work on organizing and adding the plate assemblers and getting the belts in place. once that's done and i hook up the first phase, the first phase will have been loaded well enough

[true junco]

Totally valid. I mostly use Fused wire, but suppliment with iron wire.

[primal flicker]

Fused wire is a great alt, don't get me wrong! Especially for keeping a low building count.

[true junco]

I especially like fused wire for any factory where i am already going to use fused quick wire because i already plan to feed it Copper and Caterium anyways. Most of the time fused wire and fused quick wire easily make all the wire and quickwire i could possibly need. And i combine that with insulated cables. Wire/QW/Cable needs handled.

[prisma kraken]

i'm not sure if i rate fused wire as s-tier recipe, but it is very close

its like one step below solid steel & copper alloy

[primal flicker]

Fused quickwire is S, fused wire is A, for me.

[prisma kraken]

see, i'd flip their rating

[primal flicker]

It's all arbitrary anyway 😅

[prisma kraken]

yeah

if fused qw had the same 1:4 cat/copper ratio instead of being 1:5, i'd place it s-tier

(the difference there is between needing pure copper vs copper alloy to use a caterium node)

[true junco]

In the end. The only firm purpose of caterium is to make quickwire. It can be used to make wire, but isnt necessary to do so.

Using copper to boost quickwire production is only needed if you need more than the max default quickwire (local or global depending on what you are figuring on) and freeing up caterium to supplement or simplify the production of regular wire, or to free up copper for copper sheets, aluminum parts or copper powder.

[prisma kraken]

honestly, whatever copper sheet production you need ends up being very limited

it really doesn't have much of an impact in global limit allocation

the two things that use copper sheet are ai limiters (can't hide from using it for them) and rotors (and for rotors, fused wire + pipe is pretty comparable outlay)

you could toss circuit boards into it all, but they should probably be made with cat cb as the best choice

but when you start looking at allocations for world plans like a 60/60/30/30 world or something similarly large, you look a the number of rotors you need, and come up with a number of btw 40 and 100/min and for ai limiters, you'll use them in making oscillators, but that's it at a rate of 1:1 to oscillators, and the sheets required there are very small too

i think it is 12 sheets per oscilator

[true junco]

I feel like its a toss up between CatCB and SiCB. But I was messing around with different calcs to find good ratios a while ago and i think default CBs matched a lot better with some things... been a while and i hadnt reach a conclusion regardless.

[prisma kraken]

yeah, cat cb works with cat computer nicely

[true junco]

Oh. And i meant just overall need for copper, a sum of cu sheets, Alclad sheets and casings, copper powder etc.

[prisma kraken]

everything else is a little messy with it b/c its one of the recipes dominated by a 7

[true junco]

Which one is dominated by 7?

[prisma kraken]

cat cb

pretty sure it is

yeah 10 plastic+ 30qw = 7 cb

7's in the game are like a bad penny, when you find one, there's no getting rid of it 😛

if you are using cat cb for things other than cat computer, what i think makes sense to do is just set up a recycled plastic loop for 900/min, build a cb factory around that and yield 630 cb/min from it, dealing with the remainder as overflow

[true junco]

Ah yes. And caterium computer requires 7 CB per computer.

Ok. It is default CB and Crystal computers that are a 1:1 match.

DefCB is 7.5CB/min out.
Crystal Comp is 7.5CB/min in.

[deft lichen]

steel screws have a nice ratio with default computers iirc

like 1:1 or 1:2

[true junco]

Imo, Comparing CB alts is one of the first and biggest examples where people looking for the overall "best" recipe are really wasting time.

It truely is just picking which resource you want to spend.

A) Copper+Oil
B) Lots of just Oil
C) Copper and Silica
D) Caterium and Oil

The only real leway you have is making cheap silica for the SiCBs, or making fused quickwire for the CatCBs, and of course efficient Petrochemical products.

Other then that its just how efficiently you want to process the ores.

[true junco]

Yep.

1 Steel Screw Constructor to feed 2 Default Computer manufacturers

A perfect example of the arguement for steel screws i often make. 1 steel screw constructor can feed a manifold of machines. Vs 1 machine needing a manifold of other screw constructors to feed it.

Incidentally, with the "mini-game" ive created for my self of highly productive designing multi step BPs, with overclocking, a 125% steel screw constructor can feed a 250% def computer manufacturer... hmmm. I can fit 2 MNFs in a BP. I could probably make one that has 2 MNFs making defComps at 250% and fit a single steel screw constructor at 250% in there. Just make the manifold array bring in steel beams instead of screws

[prisma kraken]

btw, pretty much any recipe except for default rip & default rotor take screws in multiples of 13

[true junco]

Or even manifold steel ingots and stick a steel beam constructor infront of the steel screw constructor.

Steel beam output at 100% is 15/min.

Steel screws at 250% consumes only 12.5 beams.

[prisma kraken]

the oddball one is bolted frame, but there's something neat about that

it takes 140 screws

[true junco]

True. But if im screwing around ill use bolted plates and rips anyways.

[prisma kraken]

now, neat thing about 140

[true junco]

Bolted plate and bolted frame go very well together

[prisma kraken]

add 120 to 140 and you get 260

add 250 to 140 and you get 390

that's default rip and bolted rip

[true junco]

1 bolted plate assembler exactly feeds 2 bolted frame assemblers.

[prisma kraken]

mhmm

not the most efficient way of making mf's

but it is convenient

what's a nice middleground is to take the excess rips and feed them into default mf with steel rod

let me find the math on that

this is the cute one:

[true junco]

I just like how simple it is.

Besides. Steel is really easy to make in huge quantities, so im going to run out of other things way sooner. 😆

This is just simple as hell. And it can be scaled up to 250% OC and still easily fit inside a BP.

[prisma kraken]

can you select for that one to use coated iron plate instead?

i'm curious if that ends up in pretty numbers

i don't think it does, the screws probably throw it of

[true junco]

I dont think it does because coated iron plate is 75/min vs 45/min

Its also less convenient in terms of input. The one i have here only needs plastic and steel. That change would lead to plastic steel and iron ingots. Not a huge deal but still.

Fyi this is what i have put i to a BP using all the overclocking.

[prisma kraken]

coated plate is meant for this stuff, you often have a normal coal node and a pure iron node, its what you're supposed to do with the additional 180 iron 🙂

[true junco]

Ok. The coated iron plates do make nice plastic and iron ingot input numbers

[prisma kraken]

there's another build that is better than all of these, btw

[true junco]

I dont doubt it.

[prisma kraken]

that beats everything but is so many machines

[true junco]

Thats the issue. Everything i was doing was to meet a specific design criteria to maximize production with as few BPs as possible.

[prisma kraken]

imho, they should list that combo on the wiki

that one actually is pretty easy to blueprint (or else i'd still be building it)

[true junco]

To be honest. I wish ALL the production comparisons on the wiki assumed that every recipe used is always the default except for the last step which is comparing all the alts for the specific item whos page you are on.

Let the players find cost savings on previous steps themselves, especially since its all subjective. Just make the examples always follow "default" so i dont have to figure out which alt path was assumed by whoever wrote it. Lol

[prisma kraken]

you break it up into solid steel, which is a 26 smelters in manifold and 19.5 foundries in manifold, both very easy to bp

[wind spade]

no matter what you pick, you'll lose tons of paths

recipe comparisons are weird, recipe paths comparison is what we should do

[prisma kraken]

then the rest of it is 2 assemblers + one constructor modules for the frames & rod, and 3 assmblers for plate+rip

yeah, the part on the wiki that i'm a little iffy about is how it lists comparison recipes and shows steeled frame as the best

[true junco]

Understood. I just wish the comparisons on the wiki assumed default recipies in all previous steps.

[prisma kraken]

which it really, really isn't

[true junco]

I am just perpetually annoyed that iron ore shows up everywhere because the wiki assumes iron wire all the time. Lol

[prisma kraken]

the wiki doesn't even factor in the recipes that take oil

i blame greeny 😉

[wind spade]

I didn't have anything to do with wiki analysis

[prisma kraken]

i'm sure some of it was done with the aid of your tool, but i'm just ribbing you about it all 😛

and its correct, iron wire is the lower resource way of making wire

shoot me if i ever have to build pure iron + iron wire, but it is more resource efficient

iron wire is 5->9, right?

[primal flicker]

It is, strictly by WP.
I was discussing something relevant to that, earlier, with greeny.
I think, unless you have a COLOSSAL save, it sounds beneficial to approach very large builds with more consideration to your computer's resources than those available on Massage 2(A-B)B 😆

[prisma kraken]

true

[primal flicker]

In other words, fused wire (as much as possible) due to lower building count.

What's the theoretical max Sink factory? Around 310 ADS and 30 TPR? Which would use many thousands of buildings. And most people will never produce a tenth of that, so "most efficient node resources use" isn't as relevant of a goal.

[true junco]

This is exactly what recently brought me to focusing on high output per machine as a design paradigm.

[prisma kraken]

what would actually be nice is if there were a 'sucks to build' ranking on the wiki

[wind spade]

depends how you define resource efficiency 🤷‍♂️
if by the subjective WPs, then yeah

[primal flicker]

Ooooh, yes!

[prisma kraken]

no idea of what the criteria for that would be, but 9/10 times you want easy to build instead of sucks to build, lol

[primal flicker]

I feel that pure iron probably sucks to build, relative the abundance of iron on the map. It's so easy to plop down another miner.

Also, for compactness, building count, and I think power, coke steel > solid steel 👀

[prisma kraken]

more than that pure iron has the wacky 7:13 ratio

[delicate chasm]

💯 This this this.
The only way to draw any real comparison between things is to first understand what two things you are comparing.

Making assumptions about how things can be made is exactly the can of worms that the whole alternative recipe analysis thread was all about.

[prisma kraken]

that really sucks to build

like how do you find 420 iron and make it into 780 ingots? what do you do with the other 180 or 360?

[primal flicker]

Whatever you want

[prisma kraken]

it ends up being a belting headache with a pipe running down the middle

[delicate chasm]

And the point is in the beatiful simplicy of the system; you can easily figure out the "WP" or whatever personal metric you might be using, per part as you look at their wiki pages.

You can see clearly what the specific tradeoffs are between the recipes only when you are familiar with recipe chains anyway.

So it does seem like a good call to just base all of them on the default (milestone, if not available via milestone, MAM) recipe chains leading up TO the final product.

[true junco]

The fact that the wiki appears to go by WP most of the time is source of my problem with it.

[prisma kraken]

i can see a flaw: what's the default way to make silica?

[primal flicker]

You don't like coke steel? 🥺

[wind spade]

technically there's no "default" or "alt" recipe

for anything

[delicate chasm]

Directly, not as a byproduct.

Why do you ask?

[true junco]

Sure there is. The default is the one you get by going thru the milestones and dont involve any hard drives.

The only items you absolutely have to use hard drives for are compact coal and turbo fuel. And you never actually need either

[delicate chasm]

And in the case of silica, the MAM unlocked recipe is the default for silica, and it's impossible to determine if it's intended behavior that alumina solution can be potentially marked as the default silica recipe.

It is however definitely a byproduct of the recipe and not its main output, though that distinction is hardly worth making.

[wind spade]

fuel or residual fuel?

and my point is about "technically" = from game's perspective

in game there's no distinction between types of recipes
you just have list of unlocked recipes

[median heath]

I would posit that some recipes have defaults, but not all of them do, personally.

[delicate chasm]

That's fine, except that we do have a codex that parses that information and it makes an attempt to display default recipes and distinguish them as such.

The ACTUAL issue here is that we don't know if it's supposed to be ambiguous ever.

It may be patched later to have absolute 'set in stone' defaults.

[delicate chasm]

This is the case.

[median heath]

Set in [Lime]stone

[primal flicker]

Compact and simple

[wind spade]

codex simply displays "default" as first one you unlocked, anything else is labeled "alternate X"

that's why you can have default recipe for silica being "alumina solution"

or why "residual fuel/rubber/plastic" are labeled as alternate

[delicate chasm]

I think I've been pretty clear that I understand that? That's the behavior I'm describing right now. Do you have another related point though?

[wind spade]

I've been reading messages one by one and replying while being on call, sorry 🤷‍♂️

[delicate chasm]

No, you're good - I know that you have spent a lot of time on this topic though so that's a genuine question, if you have another point.

[wind spade]

my opinion is that we shouldn't categorise recipes into two boxes (defaults/alts)

because there's already recipes that are kinda gray area and it can get worse with future updates

[delicate chasm]

It can also get better with future updates with patches that don't display Alumina Solution as a silica recipe in the codex or offer unlock access to Cheap Silica.

[wind spade]

sure, but for now we can't really objectively categorise recipes

[delicate chasm]

I think we have enough information to do so. Fuel and Silica are the only questions, right?

You have a required first recipe in the case of turbofuel.

[wind spade]

only "possible" way I can see is "does the recipe have alternate in name", but that leads to some questionable cases with turbofuel

[delicate chasm]

So Fuel and Residual fuel are both default unlocks. They get unlocked at the same time too. Do they display in different order depending on language?

[wind spade]

what are the "residual X" recipes then? they are not alts (not from drive, not have "alternate" in name), but they are not default either (since that's already "fuel/rubber/plastic" recipe)

[delicate chasm]

If they are always marked the same way respectively, then there is already an objective answer in-game (for now).
But if language settings or something else can cause the residual recipes to display above the non-residual ones, that's a question.

[wind spade]

I think it just depends on order of recipes unlocked in the UE blueprints

so if they for some reason move residual fuel above fuel, it will display as default

[delicate chasm]

Since the same milestone unlocks fuel/residual fuel, they should always display in the same order then. The one marked as default would be the default, or you could just as easily distinguish through the intent:

Fuel is default. Residual is a pre-unlocked alternate recipe.

[wind spade]

SFTools call them "base" recipes, maybe we can use that? or "standard"?

[delicate chasm]

i.e. if your pod crashed, the next pioneer would probably not start with residual recipes and would have to use your hard drive unlocking it. =p (this one's a joke)

(ahhh but now it's my headcanon that our pod has the residual alts preloaded)

Any of those words are fine. I don't think it really matters as long as we can be at least confident that we're not misinforming anyone.

[wind spade]

because (imo) we have:

• default = the "default" recipe to craft thing, usually named the same
• alternate = recipe that has "alternate" in the name
• XX = all other recipes that don't fit in the two categories above
• YY = category for default + XX recipes (essentially non-alt)

[delicate chasm]

I feel like it's not a really big deal to begin with, but it would be more helpful to people who NEED the wiki and might not think of it in post-game framing like we do, to do @true junco 's suggestion.

IMO we have the expected first version of a recipe you will gain access to or realistically use.

Then you have alternatives to that recipe.

And which one that is would be premeditated based on CSS's intent, not mine, so it's not exactly objective but I feel like it's not a difficult deduction, either.

[wind spade]

and this discussion affects much more than just naming things here

• wiki has "alternate" tag
• SFTools have two different categories (both wiki and SFTools use the same parser, which just checks for "alternate" in the name and adds turbofuel and compacted coal)
• game has categories (since devs may rework it, maybe some feedback from our side can help them decide)
• etc.

and also each of the above may have different definitions, so it's kinda hard to unify it all 🤷‍♂️ also hard to agree on a definition

another problem is "alternative" vs "alternate recipe"

since the first word basically means "any other recipe that makes the thing", while "alternate recipe" is usually defined as recipe with "alternate" in name

[delicate chasm]

Why does the distinction exist for recipes that actually have the word "alternate" in their name?

[wind spade]

tools, wiki, ...

[delicate chasm]

If the devs change the name of the recipe to some other arbitrary thing, your category is meaningless though.

It's just a search term?

[wind spade]

(and the most problematic thing - how to automatically decide if a recipe falls into given category, since that is relevant for the parser I've written)

^ is the reason why we use it

because then I can just hardcode "turbofuel is alt", and everything other gets automatically marked as alternate

[delicate chasm]

hdd_recipe
mam_recipe
hddmam_recipe
milestone_recipe
pioneer_recipe

You need a total of 5 categories I guess to really distinguish every recipe source?

[wind spade]

eh, if you want to do it this way, it's even more

[delicate chasm]

hddmam_recipe is redundant but I guess...

[wind spade]

build gun recipe f.e. 😛

since building things from build gun is recipe as well

[delicate chasm]

pioneer_recipe is your always-have stuff.

I did miss shop_recipe however so replace hddmam_recipe with that.

[wind spade]

on the other hand, milestones are not recipes 🤔 (I'm looking at this from the POW of data files)

[true junco]

1st. I think The "residual X" recipies that alternates are just the plastic and rubber ones. Not sure tho.

2nd. my bad on terminology. Whenever i say "default" i always mean "base" recipe. But thats an issue anyways since there are still multiple recipes for some products that are all "base" in SFTools.

Basically
To be the "base" or "default" is never unlocked via hard drive research.

If its unlocked via tiers or the mam, then its "base" aka "default"

[wind spade]

shop isn't either

[delicate chasm]

Aren't you distinguishing from the player-perspective SOURCE of a recipe though? That's what I'm talking about here.

From a data perspective it doesn't matter what anything is called. You group things as is convenient for whatever actual calculation you are doing...as you know.

[wind spade]

well first of all the question should be "what do we consider a recipe", since (again, I'm going at it based on their data export, since that's what I work with in both tools and wiki) currently it's:

• all actual recipes (things you set in machines)
• recipes for equipment workshop
• recipes for build gun (building machines and buildables in general)
• customisation recipes (convert wall to steel wall, etc.)

anything else is not a recipe (not even burning fuel or shop or milestones)

[true junco]

I also suspect that the fact that compact coal and turbo fuel are unlocked on the mam using an HD each, may be a place holder from when we used to have to find those alts via HD scanning and those may end up using parts in the future.

[wind spade]

I think you may be confusing "residual" vs "recycled"?

[delicate chasm]

Parts recipes then, to streamline the discussion.

[true junco]

I thought you said the residual recipes were alts. Lol.

[wind spade]

no, I'm saying residual recipes are not alts and not default 😛

[wind spade]

alright, just asked to be clear 🙂

[true junco]

There are no "alt" recipes for buildables. Every buildable is exactly made by one recipe only. Since even things like ficsit foundation vs concrete foundation, the buildable is different.

[open jungle]

Not to add fuel to whatever dumpster fire this discussion is but alt recipes are technically also just different recipes

[wind spade]

well, not exactly true
because you can make a foundation in two ways:

• building it
• customising it

and I'm not exactly sure if that's two recipes or one 🤔

[true junco]

After the item is made in a machine you csn never distinguish the item from another of the same item made with another alt. Therefore alternative recipe to make the same item.

[wind spade]

(I guess I have a bias for looking on things from different POV than other people, but since the data I'm working with is what is used to generate wiki/tools data, I think that my POV is at least a bit relevant)

[delicate chasm]

Parts have 'first' parts recipes for making the part, and then there are <synonym for alternative> recipes that you can unlock only after acquiring that one.

The actual word that is used to describe this aside, residual recipes seem to be intended as an alternative method to producing, given that they require secondary production steps from oil, and therefor I feel like it's really not a big logical leap to call the non-residual recipes the base, standard, default, first, etc.

[true junco]

I reinforce my claim with this example. If you use the dismantle filter and set it to a 1m concrete foundation, you wont be able to highlight a 1m ficsit foundation.

[delicate chasm]

Dinner time for me.

[wind spade]

sure, but each of the buildables can have multiple recipes

[true junco]

But do any of them?

[wind spade]

if we count customisation recipes then yes

[true junco]

See i wouldnt
Because the game doesnt treat them as the same buildable.

[open jungle]

Functionally they’re the same, except for ingredients. Foundations serve the same purpose regardless of their ingredients.

[delicate chasm]

You can replace-in-place actual buildables with different recipes (Merger/Splitter with ctrl) too.

[wind spade]

that's something different than customisation

but yeah

[delicate chasm]

Wait, you don't need the 5 concrete when customizing, right? It just requires the secondary resource and gives you the other secondary resource (or 2 concrete) back?

I was gonna say I think it's probably the same function, together with upgrading (miners or belts), but I'm not sure now.

[deft lichen]

you only need the new material's ingredients when customizing iirc

[delicate chasm]

I'll check some in-place replacements as well as customizing foundations right quick. Curious about walls now too.

I think you are right.

[deft lichen]

we're still not showing any customizable buildings on the wiki because finding a balance between clarity and spam is hard

[delicate chasm]

You only need the new materials for customization, AND it shows as such in the build menu (rather than being the recipe for the foundation).

Replacements in place with a wall vs. a different wall require the full buildable cost, similar to upgrading a miner.

Merger/splitter same, full build costs followed by a refund of the old buildable's costs.

Yeah. This is so far removed from production recipes that the only commonality is the word 'recipe'. I for one would urge anyone to not include it in any list of "recipes" for the game and instead relegate it to its own category.

Customizations working differently from buildables is reason enough to separate the two.

[deft lichen]

for the purpose of the wiki, building a buildable (building/vehicle) is also a recipe

it's "produced in the build gun", so customizations are "produced in the customizer"

[delicate chasm]

That's fine for now, but in the final form could we not have a distinction along the lines of "parts/production recipe" and "buildable recipe"?

The internal workings of how to sort and parse data are one thing, but the displayed information is what I'm referring to.

[deft lichen]

the problem is that it would need two tables with two headings

[delicate chasm]

Why? They wouldn't be on the same page, surely?

[deft lichen]

and due to technical problems, we can't generate headings (separating the tables is easy and already kinda done)

so you need to
a) guess if said item is or isn't used for building and/or crafting -- prune to human error, especially with game updates
b) always include both, leading to empty tables

the table is already sorted, production recipes are first, alternate recipes second, buildings last

[delicate chasm]

dummy.wiki/Buildables
dummy.wiki/Recipes

or
dummy.wiki/Recipes_(Buildables)
dummy.wiki/Recipes_(Production)
dummy.wiki/Recipes <- disambiguation?

I'm saying I don't think these groups of things need to be displayed together for any end-user. Internally if it's all considered to be a recipe that's fine.

If I'm wanting to look at parts recipes, I don't want to also look at building material requirements for buildings.

Those are two different concepts in my mind and I'd posit it's the same for a majority of players, since they are vastly different activities.

[deft lichen]

I agree, so my question is, is sorting the table not enough?

base recipes, alt recipes, buildings

[wind spade]

tools have buildings even separate, but I guess you said that it's problematic on wiki

[deft lichen]

#math-and-meta message

[wind spade]

yeah

[delicate chasm]

I think it would help to fully separate them, in light of the conversation above.

Especially if Canadian_F_H's idea is implemented - it's less of a concern if there will always be assumed chains and things in the individual parts pages though.

[deft lichen]

could you please link the exact message?

[delicate chasm]

#math-and-meta message Here you go! There might have been more before this but it's what I replied to initially.

[deft lichen]

ah, recipe comparisons

[wind spade]

if we're talking about wiki's recipe analysis, then that's getting removed (eventually) in favor of text descriptions

[deft lichen]

well our idea is to get rid of the math altogether, because noone can agree how to do it

[wind spade]

(math is better done in online tools anyway due to sheer amount of variables)

[delicate chasm]

I agree with that.

[wind spade]

(unless we figure out how to inline some light build of SFTools into wiki)

[deft lichen]

one could say it's hypocritical to point out how every "objective recipe analysis" on reddit isn't objective and then have the exact same thing on the wiki

[wind spade]

well reddit has tier lists, that's even worse 😄

[delicate chasm]

If I had my druthers, the alternate recipe analysis would just be a few lines of commentary on each that describes its intended use case (if known), MAYBE points out some neat math (this recipe pairs with this other recipe 1:1), and leave it at that.
For some of the recipes that generate controversy, maybe an extra line or two explaining the niche they fall into at most. But the wiki is for information, not justification, so not too much of that either.

[deft lichen]

you pretty much described what we want to do

[wind spade]

intended use case
I'm not sure if you mean the same thing, but we were thinking about just pointing out advantages and disadvantages generated by the recipe (and some combos in case there are some)

[delicate chasm]

I wouldn't mind seeing a tip that charcoal or biocoal can make small runs of things that normally require coal easy to do anywhere and without coal. Initial steel before going and setting it up properly, that kind of thing.

Because long-term it's not useful by definition as it involves pioneer input, but it's good for a few niche purposes...
...IF you create those purposes!

[delicate chasm]

Basically, yes. Like with modular frames:
"Bolted Frame demands a larger volume of input than Modular Frame and trades resource efficiency for production rate. It pairs well with cast or steel screw to minimize machine counts on the production line, without which it requires a large number of screw constructors."
Inputs: 3x Reinforced Iron Plate (7.5/min) 56x Screw (140/min)

That's all the info I need/want for that alt, other than the fields for 'produced in, products, unlocked by' in the existing table.

How to put this... You're bound to hit subjectivity when analyzing this with someone else. If I suggest a way to use any recipe, it's probably not the only way to do it.

If we can remember what our questions were before we learned better questions, then we'll make a better wiki.

[prisma kraken]

tbh, i think what would be nice as a clue for those interested in keeping cycle times integral would be that it pairs with steel screw AND one of the screw taking rip recipes

[wind spade]

that's why what I wanted to do is not "suggest way to use it" but "say what the recipe does"

e.g. "saves space and changes ingredients needed"

[prisma kraken]

wasn't @oblique hollow working on something like that?

[wind spade]

we were, there's even a doc for it but people had no time so it's on hold

[prisma kraken]

what would actually be kind of cool is a recipe cookbook

there's stuff like 'clock an ai limiter assembler to 90% to direct belt fused qw and steamed sheet into it'

[delicate chasm]

Is that a "consult the wiki for the ANSWER " thing or "consult the wiki for the numbers I will then use to calculate" sort of thing?

I offer that as the qualifying distinction between something that does or doesn't get mentioned as a protip.

[prisma kraken]

they're things that aren't immediately apparent

[wind spade]

ideally "consult wiki for how to choose a recipe", not "consult wiki for which recipe to choose"

(if we're talking recipe picking from hdds)

[prisma kraken]

well, i'm not sure the wiki is the place for more than factual

[delicate chasm]

Do you expect it to be pointed out on the wiki though? Or would that be the kind of thing that is incidental to playing when you are playing with that kind of thing in mind?

[wind spade]

see https://satisfactory.wiki.gg/wiki/Tutorial:Picking_an_alternate_recipe

[prisma kraken]

more editorial and interesting things should probably be hosted elsewhere

but there's a lot of interesting recipe combos for oddball things

[deft lichen]

important to keep in mind the wiki should be useful for new AND advanced players

[prisma kraken]

like i was a few days ago wondering if insulated cable and coated cable dovetailed with the output of a default rubber refinery (they do!)

hope i never need that much cable, but that's a pretty sweet way of making lots of it, lol

[delicate chasm]

I expect a maximum theoretical production calculation for a given part on each page, but not really much else as far as math is concerned.

Maybe a table that has a quick reference to clocking each recipe to common amounts, and the last field of that table can be the explanation for when that number might come up.

Like the relationship of 300 crude to 800 diluted fuel on the fuel page.

[prisma kraken]

@Ondar, if i wished to get onto the wiki's train throughput page a table of round trip times per stack size for train loads, how would i go about doing so?

[delicate chasm]

Then you're able to insert the subjective/interesting stuff into the page without it taking over the data aspect...you're still looking at a spread of numbers That Are, regardless of whatever else you might be doing, and then you have listed next to a few of them the reason that you're being shown those numbers.

[deft lichen]

1. register a wiki account on wiki.gg
2. edit the page (you can use visual or source editor)
3. briefly fill out the edit summary and submit

[prisma kraken]

simple as creating an account? nice

[prisma kraken]

the flip side is that the wiki shouldn't end up being a complete set of actuarial tables only useful to someone that swims in the numbers

[delicate chasm]

Yeah, agreed 100%. That's the main worry I have - and why I think limiting ALL of the "you can ____ " to just one section/table is a good idea.

[deft lichen]

each item page would have descriptions of recipes that produce said item, and there would be a "master page" with all the descriptions in one place

either alphabetically or grouped by item, haven't decided yet

[true junco]

Hmm. Are buildables actually called recipes in the game? Idk why, but for some reason I feel like they were at least referred to as "schematics" at some point...

[delicate chasm]

More important I feel is that there definitely exists a distinction between them and a production cycle.

[deft lichen]

a schematic is anything you can unlock, so each milestone, MAM research (incl. individual alt recipes), shop unlocks and such

the to-do list refers to buildings added as recipes, but that might just be an oversight

specifically, when you add a building to the to-do list, it shows Recipe: {building name} above it, but the recipe name is just the building

[wind spade]

they are recipes in game data

and they have "made in: build gun" property

instead of "made in: manufacturer" or whatever

or rather... buildable is a separate entity
but there's also a recipe that has product equal to that buildable

[true junco]

Ah. .

Well. We are deep in the weeds now. This whole convo started about things made in machines. 😆

[warm wren]

Hrm, it's 900 meters with a fairly drastic elevation change from the bauxite to the sulfur, then another 600 meters to base I've build a train to carry bauxite here before but I think the distance and the elevation messed it up. I distinctly remember tapping into another bauxite source just to massively over supply myself so I wouldn't run out of bauxite between train runs (extra cars). drone needs the batteries which needs the bauxite so that's a bit of a "hrm" as I'd need to at least get the batteries started before I could run the drone(s). Considering what to do.

[prisma kraken]

you have a couple of options

all of them eventually lead to needing bauxite to copper or copper to bauxite

what i like to do due to its convenience of transit and proximity to sulfur and oil, is to build aluminum at crater lakes, pulling down the bauxite via train and pulling the copper up from rocky desert

if you don't do that, you can at least belt some of the crater lake sulfur up to the red forest and make some default batteries there

i personnaly feel its worth getting the train infrastructure going for moving the copper as the least sucky option, but getting trains up and down elevations isn't a quick thing to build

you can get a truck up and down the ramps to crater lakes instead, and possibly run some trucks up into pink forest to round up the bauxite, but you're probably going to want to build some roads for the terrain up there and need at least one truck for every 2 bauxite nodes you'd want to move

other thing possible is an idea of Sevrahn's of using a train at elevation to drop off stuff at a cliffside to belt vertically down to another station that delivers where you want it

[sudden geyser]

heres the limits for all the stuff

[prisma kraken]

you get 60 ADS/60 MFG/60 TPR and ~55 Pasta/min out of it all, to save you some time in twiddling the what-if's

[sudden geyser]

im early tier 8 what does any of that mean

[prisma kraken]

those are the 4 phase 4 elevator parts

assembly director systems, mag field generators, etc

you can't quite get 4 mk1 belts of the items out of the world 😛

[sudden geyser]

thats too bad

its so close too

[prisma kraken]

honestly, building that i think near impossible as a goal

there's too much moving of resources from different factories to make use of little fractional byproducts from other factories

more modest goals if you drop the tpr and pasta count down to 16 & 16 are quite buildable though

[sudden geyser]

I just want to stop fkin running out of base items to be honest

like ill be building and all of a sudden why dont I have rotors automated already

[prisma kraken]

rotors are like the first thing i automate after rod/plate/wire/cable

[primal flicker]

There's also the cumulative effect of all those calculations ganging up to melt your CPU.
Or so I've heard.

[mossy flint]

You are transporting things by hand

[viral ravine]

where does the water limit come from btw

since you can just stick extractors whereever as long as the game considerds it deep enough

[primal flicker]

I think it's just "arbitrarily high number"

[sudden geyser]

I mean why does it matter

If you can get a higher number then lmk but until then thats probably the right number

[viral ravine]

curious

[primal flicker]

If it was based on how many water extractors you could place, it would be a multiple of 300.

[primal flicker]

Does anyone have a SFTOOLS link for max plutonium?

IDK how much uranium waste to add as an input, it seems.

[prisma kraken]

i think i have the numbers for that somewhere, give me a moment

[primal flicker]

Would appreciate
I've been toying with this idea to store plut waste near the center of the map and just see how quickly the radiation ball grows ☢️

Wiki lists theoretical max number but idk how to get there (or actually how to scale it back to 6/7 because I won't use the spire except for nobs)

[prisma kraken]

i need to recreate the calc, give me a minute

[primal flicker]

TYSM

[prisma kraken]

the numbers are ugly AF for it:

[deft lichen]

note that max nuclear doesn't use max PFR

[prisma kraken]

max power you mean

that design isn't supposed to actually be built, i think, just every number is a repeating decimal

there's a slightly less evil build that trims out a few pfr's from the yield that is a lot nicer

oops, that's wrong, mb

this

i'd add in the power generated from burning the pfrs, but the tool i use calculates that wrongly

[primal flicker]

Probably what I'll eventually pursue. ❤️

[prisma kraken]

the difference btw the two is that one takes all the nuke waste and converts it straight to fertile uranium, the other only does that for half and uses the other half to make plut pellets

[primal flicker]

That's a very useful ratio!

[prisma kraken]

it takes a while to twiddle the numbers to find the ratios, yeah

[wind spade]

this is max power

2520 waste

[primal flicker]

I want max plutonium waste tho 👿

[prisma kraken]

yeah, that's the more productive build

[primal flicker]

Is that the beta build for SFTOOLS? Looks awesome 👍

[wind spade]

yeah it's the (outdated) beta

[primal flicker]

Oh no, what got outdated about it?

[prisma kraken]

that's the one i posted with the fugly numbers

[wind spade]

uses U5 dataset

(which should be fine for most production recipes, but still)

[prisma kraken]

does that mean the water consumption for the generators is wrong?

[primal flicker]

Once I get a better handle on the system balance, I'm going to try to get nicer numbers than the extreme, while staying reasonably close to 300

[wind spade]

may be (idk if it has changed since then), but water is like... never an issue 😄

[prisma kraken]

yeah, i know 😛

why that last build actually is kind of nice is that watch what happens when i tally the sulfur for making sulfuric acid

[wind spade]

@fair fiber @ashen stirrup see #math-and-meta message for alt recipe choices 🙂

[prisma kraken]

that build has a bunch of interesting things that ripple down into making the ECR's, oscillators & beacons that are attractive too

[ashen stirrup]

Thanks! I haven't been asking for advice, even though some have reacted that way. Just sharing frustration at the myriad irrelevant alts I'm getting 😂

[delicate chasm]

Only irrelevant alt is automated miner.

Impossible to place somewhere that saves you time vs. making them by hand on the spot.

[ashen stirrup]

Maybe for farming sink points?

[delicate chasm]

Nah. It takes a motor, which is itself worth more points than the miner.

[ashen stirrup]

At any rate, any alt that uses a refinery or manufacturer is currently irrelevant to me

[wind spade]

automating miners 🙂

[primal flicker]

Maybe 25.6 PFR and 21.6 UFR... There's a lot of messy decimals throughout the production web, but at least the power plants can be clocked for 100% uptime.

[delicate chasm]

Possible isn't the same thing as relevant.

Zenon said 'irrelevant' so I went ham with my opinion because when you are talking 'relevant' not 'possible' my opinion sticks.

[primal flicker]

(again, assuming only 1800 ppm uranium ore used)

[wind spade]

I'm talking about relevant tho

@misty lion see #math-and-meta message for alt recipe picking 🙂

[delicate chasm]

Relevant means it's appropriate for what it is intended to do.

Appropriate in turn means suitable or proper to the circumstance.

The circumstance is that you need n portable miners in your inventory.

I do not believe you if you assert that you can save yourself time or effort, and it's irrelevant if it's theoretically possible to do something but inappropriate because of other considerations like costing more time or effort.

This opinion does not value "the only possible way to achieve a specific narrow goal in controlled circumstances" because the controlled circumstances are exactly what make it irrelevant.

[wind spade]

given that "worth" or "relevant" is subjective, you can't just say "it's not relevant"

for me, if I can automate it, it's worth it 🤷‍♂️

[delicate chasm]

I said opinion enough times to be a little bit exasperated that you missed or ignored it.

I CAN say my opinion.

[wind spade]

Only irrelevant alt is automated miner.

Impossible to place somewhere that saves you time vs. making them by hand on the spot.
don't see any "opinion" there ^ 🤷‍♂️

[delicate chasm]

greeny — Today at 1:06 PM
automating miners 🙂

The Avelon — Today at 1:09 PM
Possible isn't the same thing as relevant.

Zenon said 'irrelevant' so I went ham with my opinion because when you are talking 'relevant' not 'possible' my opinion sticks.

First response to you. I didn't say "DISCLAIMER: This is my opinion!" at the beginning and I didn't need to.

It's carried by the fact that I am making an assertion, that I am sharing an opinion. Relevancy is itself as you said a subjective thing.

It's built right in.

If I were asserting something about the math, that would be me making an objectivity-based statement (and if false, should be debated).

[prisma kraken]

the purpose of the automated miner alt: to provide an off-meta max item build... those things are world-sucking to make in volume

[delicate chasm]

Yeah, since moonchild came up with that, my argument that there exists no use case (including ones you can invent) has no leg to stand on any longer. Sad.

And that really was an argument. =/

[prisma kraken]

haha!

what's sad is i actually contributed to it being left in the game with the idea 😦

[delicate chasm]

That's not just sad. That's one of the greatest injustices of our time.

We really should be living in a world where miners and drones don't use portable miners in the recipe to begin with.

[prisma kraken]

there's a mod that adjusts that

...if you really care about it that much, lol

[delicate chasm]

I don't. I just slap down an equipment workshop and don't have a dilemma. 🤷‍♂️

If it persists through 1.0 though, I might do some tasteful edits to stuff like that when I go to modding my game.

[prisma kraken]

more i feel the mk3 miner upgrade or placement is generally a pretty time-consuming task due to the low stack sizes of their build materials, the portable miners do add a bit of ugg to it, but really its more the marshalling of the other materials

dropping the miners for a new factory is usually a couple of hr prep task for me

[delicate chasm]

I can safely say there will be no unforeseen losses in challenge or fun factor if I switch portable miner out for something else. Probably equal amounts of rotor, which is actually harder and disrupts progression slightly since I wouldn't get the ability to make them until assemblers are unlocked.

[prisma kraken]

you know, i wouldn't mind that if you were 'gifted' a few rotors at the start for making the miners, new idea so i don't know how that would work in the tutorial

[delicate chasm]

It's really nice if you skip onboarding - you start with rotors so can do a fully hands free run.

[prisma kraken]

i kind of like the idea of being capped on the number of miners you drop in phase 1 though

[delicate chasm]

The pod could perhaps open a compartment on the side with 4 portable miners that you can grab, before the dismantle step?

[prisma kraken]

ehh, just make it part of the dismantle if you do it that way

BUT i like the idea of being gifted rotors to make the miners instead of the miners themselves

[delicate chasm]

What, like the FICSIT Freighter bringing down some building materials at tier 1?

Ooh, the same inventory you'd have if you skipped onboarding in fact.

[prisma kraken]

nah, just dismantling the pod gives you hub parts + rotors

[delicate chasm]

What about an explicit "you hand chiselin' and hand craftin'." phase wherein rotors are unlocked almost right away?

Seems strange to me to dismantle the pod for rotors specifically in addition to the HUB parts. Why just rotors? No other mats, really?

HUB Parts is just a collection of various bits so I get it, but then it does feel strange to have specifically a few rotors. I guess put it with an Ada line about standard issue with every pod...maybe a joke about how they will definitely not survive a crash however?

Explains why 0 rotors in world.

[prisma kraken]

no rotors in the world make part assembly a gate to completing phase 1 now

tbh, i think there should be a crash site somewhere in possibly the swamp or titan forest that has 4 rotors

somewhere very difficult to get to

[delicate chasm]

If that, I fully expect a secret achievement "Look Ma, No Hands!" for putting up the final phase with no hand crafting.

[prisma kraken]

yeah, i was just thinking that would be a cool achievement

changing topics...

[delicate chasm]

Definitely agree there should be a way to get there without skipping onboarding.

[prisma kraken]

what are all the items that can be produced with iron ore only?

i have rod, screw, plate, rip, rotor, mf, beacon, wire and cable

am i missing any?

[delicate chasm]

Ingot, rod, plate, screw, RIP, MF, Wire, cable, beacon

Ah! Rotor, hilarious I didn't think of that.

[prisma kraken]

are there any other (useful) ones we're overlooking?

[delicate chasm]

Portable Miner too, but only the hand crafted version.

[prisma kraken]

doesn't count

my reason for asking is i want to revisit my build yard bp

which is essentially a couple of bp's that snap together to produce the goods we listed on 300 iron

[delicate chasm]

So also not smart plate. Which I think is the last one. Just scanned over SFTools' recipe list.

OOH! Rebar!

[prisma kraken]

ehhh, maybe for style points, but not necessary

i'm revisiting this old design, and trying to figure out if i can do more with it or do it better

really the idea is just to be able to drop a bp for basic iron goods wherever you're building that slow-makes stuff just for building

the problem is it doesn't make things quick enough to really feed anything else and you'd be using it in areas where you probably don't have anything other than an iron node to tap

[delicate chasm]

In other words, during the scouting/hard drive hunting phase, with intent to come build here later. =)

It occurs to me that with blueprint dismantle, a set & forget box factory blueprint that does a bunch of stuff slowly would be tremendous for speeding up a run.

[prisma kraken]

really i find the utility more in avoiding resupply trips, traveling light on base iron goods & concrete allows you to fit more things in inventory like motors & hmf's and that is very useful, i've found

conceiveably, you could drop a bunch of boxes on a biome's iron nodes and run a truck route to gether them if you tuned it all to make smart plates/autowire/vers frames

obviously smart plates are the only one that would be iron only, but you could probably do it for VF's and autowire in grass fields or dune desert

just playing with the numbers there a little bit, vers frames don't work so well b/c they really want some plastic and rubber to reduce the cost of the steel beams

autowire otoh is disgustingly easy to bp, taking the same materials as stators

i could see doing a town style cube build connected by trucks for small factories like what we're discussing

[steady ivy]

I'm thinking it's one or 2

[wind spade]

as I sent in #satisfactory , read #math-and-meta message

[vapid gorge]

what are you building next?

[steady ivy]

about to scale up with miner 2s and build a larger more segmented facility

thinking it makes sense to start moving initial processing and secondary (to frame/assembly stage parts) to their own areas

I think I will take the rotors, those are slow with iron and I'm not heavy on copper consumption yet

[scarlet sparrow]

yeah, prolly the best one

[wind spade]

the rotor recipe is basically "pay more resources to have same ingredients as stator"

[wind spade]

best is subjective 🙂

[scarlet sparrow]

you're right

[mystic moon]

Copper rotor my beloved

[ashen stirrup]

Steel rotor sounds nice for a motor factory though tbh

[wind spade]

that depends how much you like to spend extra resources for slight convenience 🤷‍♂️

[steady ivy]

at the moment I have more spare copper than iron so offloading some of the load to that prod line seems like a good idea to ease scaling up

once I have trains, things will change

[wind spade]

I personally find it not worth (I'd rather save resources)

but that's just my opinion

[ashen stirrup]

Or maybe motors from steel + caterium with copper rotor, caterium wire, and/or quickwire stator

[scarlet sparrow]

lmao true

[steady ivy]

I'm not convinced either of the other 2 is great, caterium I assume I'll want for computers, and the frame is nice but more involved

[scarlet sparrow]

gold rotors

[wind spade]

I mean both are "good" in what they offer. Whether or not you like what they offer is up to you 🙂

[ashen stirrup]

Frame recipe is less than half as much iron as base recipe

[wind spade]

caterium wire is most wire per ore (but obviously uses rarer ore than copper)

[ashen stirrup]

Sulfur doesn't really become relevant for production until phase 4, right?

[steady ivy]

if I get that again I think it might be next then. Rotor and reinforced plate production is bottlenecking me a bit now and I'm planning a larger rework to go for a more distributed manufacturing layout

[ashen stirrup]

Well, just bear in mind you do need steel production for that

[prisma kraken]

kind of misleading if you look at the wiki for that... they aren't factoring in base recipe + steel rod+adhered plate+steel coated plate which is much less expensive on iron & coal

rips are far less iron intensive if you mix in rubber & plastic, so the savings of 2 vs 3 rips/mf ends up not being very good, and the pipe's steel requirement hurts the recipe's efficiency

i really can never find a place in a build for steeled frame because it always ends up using a bit too much steel that i'm trying to instead use for hmf's

[steady ivy]

Ok I think I am actually quite happy with this steel and copper turbine recipe, means I can produce it with the same pieces as stators as I spin those up

[prisma kraken]

yeah, it is nice for a motor factory, especially with fused wire

copper rotor is arguably more efficient if you use pure copper, steamed sheet and steel screw

[ashen stirrup]

Steel screw?

[prisma kraken]

(until you assemble voltron for that combo, steel rotor is better)

5 beam->260 screw

[ashen stirrup]

Gotcha

[wind spade]

steel rotor costs more steel than base rotor with steel screws

[prisma kraken]

in the end though, when you look at the grand scheme, rotors end up being a pretty low-volume part... even if you aren't using the most efficient recipe, the amount of resources you lose on the choice isn't going to change a max-world calculations

[prisma kraken]

why would you ever use defalut rotor with steel screw?

[wind spade]

to compare steel cost 🤷‍♂️

[ashen stirrup]

Base rotor with steel screws and steel rods?

[wind spade]

basically steel rotor is most resource expensive out of the three rotors

[prisma kraken]

base rotor with steel screw and steel rod is kind of a thing, but not great density-wise

[ashen stirrup]

Density meaning number of buildings?

[prisma kraken]

so, i'm just going to say this as a thing... the max number of rotors you're really going to ever need in a world is somewhere between 100 and 200/min

here's the numbers on the different ways to make them:

[ashen stirrup]

Including for motors and everything subsequent

[prisma kraken]

we're arguing over +/- 100 ore/min

[ashen stirrup]

Oh sorry, I'm not trying to argue or anything. I'm mostly trying to get my head around alts a bit more, because I know I don't really understand how to think about them

[wind spade]

essentially:

• look what it does
• think if you like what it does
• use it if you like what it does

[ashen stirrup]

Developing good taste in which alts I like and don't like is the goal 😂

[prisma kraken]

i generally prefer steel rotor over copper because it doesn't require water

[ashen stirrup]

or refineries, which I don't have

[prisma kraken]

but i'm using copper rotor in the build i'm doing now because the water is right there for me and the steel is more of a premium, and losing 75 copper ore is a good trade-off

[wind spade]

copper rotor doesn't require water

[prisma kraken]

you end up doubling the copper cost if you don't use steamed sheet

[wind spade]

well yeah but it's not a requirement 🤷‍♂️

and it's still good without water, just not as much

(good = resource efficient)

[prisma kraken]

nor is using steel for default 🙂

i personally do weight copper much more highly than steel

but that's a personal preference sort of thing

if you change the comparison of the recipes to avoid alts, it looks like this:

and, really, it isn't much difference in the recipes between them considering that those are totals on-par with what you'd need for a 60/60/16/16 factory world

[delicate chasm]

1 stack of rotors/minute is about all you need? Huh.

[prisma kraken]

there's not a lot of recipes that require them... smart plating, motors & the electric turbomotor alt

i mean, in my tallies for my world and what i'm building towards, i need to make 120 total

[wind spade]

Why does copper have 135

[prisma kraken]

natural recipe multiple

what you'd size things to if you wanted an integral number of machines at 100%

could trim 11.25 or 22.5 off of it if you wanted

[wind spade]

If you want to compare resource cost, you have to do all at same rate

Since you're not even showing machine count

[prisma kraken]

you can use power as proxy for machine count and see that any way you compare them, power is comparable; default always loses out over the other 2 on power b/c it has lower production density

revised numbers for copper dropping it to 9 machines instead of 12:

it just isn't a build you'd really do, instead opting to size it at 12 machines instead

or at least i would

my overall point is that the efficiency of the rotor recipes is pretty close together and the criteria for 'best' should really be more what is easier to build for your location and plan

even machine counts are similar; same view of steel rotor:

fyi, i chose to use cast screw instead of rod+screw for the rotor recipe intentionally even thought i said no alts b/c it was less clicking and same resource usage

[sacred orbit]

anyone have a blueprint able to output 12 Reinforced Plates per minute?

[snow dove]

can fairly easily design one

actually i can

gimme a couple minutes

iron only input i presume?

[mystic moon]

That's a fairly simple design

[prisma kraken]

its not bad to do with stitched plate

2 assemblers on stiched plate, however many constructors for iron wire & a few for iron plate with 3-4 smelters

i'm right now trying to figure out how to get 30/30/30/30 rod/plate/wire/cable+4/4/5 rip/rotor/mf into a 32x32x32

machines all fit but the belting is tricky

internally that should make about 12 rips

pretty much what i'm thinking:

...with the correction that i'll be 4 smelters at 250% instead of 10

there's enough space for all the machines, the belting is a pain though

[sacred orbit]

Need 12/min for modular frames

[prisma kraken]

numbers for that work out better with 5 mf assemblers + 2 rip assemblers

[sacred orbit]

rip?

[prisma kraken]

reinforced plate

here's 5 mf/min as a bp

not my best work, i'll admit

that's pretty much the base recipes for mf and rip for making 5/min off of 120 ore

used cast screw to fit it on a single level

[snow dove]

wait can we send BP files here?

[prisma kraken]

dunno, i'd do dm for it

tbw, on the subject of bp's, this is annoyingly close to not sucking as a solid steel bp

[true junco]

well i am about to load in and stuff all this into a single BP... note that the assemblers for plate,rips and MFs will be overclocked to 250%. The steel beam constructor will be ~169.9% and it will have 3 constructors each dialed in to direct feed each of the "bolted" rip and MF assemblers.

5 assemblers and 4 constructors. 25 MFs per minute out of one BP (besides making ingots and plastic elsewhere but I always do those seperate anyways.)

[prisma kraken]

so, if you are going that route, try running the numbers again with an equal number of bolted frame and bolted rip assemblers, they end up looking MUCH nicer

you'll end up making excess rips

but the steel requirements for the beam & screw end up working out to really nice numbers

[true junco]

Why tho? Oh...to make the BP also kick out rips...hmmm.

Might break the limit of what fits in the BP without clipping and still having decent structural elements to look legit...ill try tho.

[prisma kraken]

um, i've been doing a lot of packing of things like this lately

[true junco]

Wait. Did you mean reduce the MF assemblers? Or increase the RIP assemblers?

[prisma kraken]

what i would do is 2 rip & 2 frame asm's each at 250%

[true junco]

Ok. So increase the RIP assemblers to 2. Got it. Ill run the numbers before i go in.

[prisma kraken]

i figured out how to get 6 asm's into the bpm the other night cleanly on a single floor

its an idea for you

what you can then do is actually use the rips in steeled or default mf for a larger yield

i think i have the numbers on that drawn up somewhere

[true junco]

Wow. The ratio of buildings are infact perfect. The question is can i comfortably fit 8 assemblers and 4 constructors. 🤔

[prisma kraken]

that's kind of what i'm thinking

[true junco]

There are options. It becomes an easy module to then feed other things. Quite a few things can chew up rips if you want them to.

[prisma kraken]

i think that is about the limit of what you can fit in the bpm