math-and-meta

[frosty owl]

The amount/min doesn't matter much, what matter is: how many are the sources (machines making Rods)? Do you want to merge them?
How many are the destinations? (12 reactors)

[molten light]

12

12 to tvelve

oh shit, I realised smth

[wind spade]

then hook them 1:1?

[frosty owl]

Let them cook~ xD

[molten light]

Wait, there is a pluto waste 💀

[wanton sky]

How does this plan look ?

I might have to do trains I don’t think the trucks will be able too keep up the steel beam delivery

[dense cave]

looks like the iberic peninsula lol

[worldly oyster]

yeah nah it's water. i'm reaching the max throughput of mk 1 pipes so there isn't a lot of water in them or the internal storage of the generators.

built a few more and now i'm up to 2 gw of power

[sly fjord]

Opinions about Steel Cast Plate in a factory that already produce steel?

[prisma kraken]

It is an okish new recipe... I have to say that i'm using it in my hmf factory and i really don't like having to move the iron ingots past the solid steel foundries on a line because of the logistical complexity of it and that it makes manifolds a lot longer. Much prefered the steel coated plate recipe that cast plate replaces

[magic island]

before, you could completely replace all Iron Ingot use with Steel Ingot use

now, the Steel-based plate recipe still calls for keeping Iron Ingots around, so I can see that being an annoyance

[sly fjord]

Hm.
I started making HMF factory from the input up instead of other way around, I know realize that with that recipe its enough for a 30 HMF factory.

That may be a little overkill

[prisma kraken]

yeah, that's the thing i think i really object to is that you can't go all-in on steel anymore and supplant basic iron

i like the addition of the cast recipe as a solution that doesn't require oil tech, and adding a 3rd alt plate recipe feels sort of crowded

[prisma kraken]

i'm making 90/min and think i'll probably have to add another 10-20/min

[sly fjord]

Why so much?

[prisma kraken]

need a lot of pasta for singularity cells

i'm probably overshooting a bit, but i will end up doing a big nuclear build

[sly fjord]

btw, for input with the alts I use its only around 600 Iron, but 2,2k limestone.

[prisma kraken]

yeah, well here's the vitals on my hmf build

[sly fjord]

I havent even built computer factory yet.

[prisma kraken]

(should also give you an idea of how good iron alloy is as a recipe... 2400 iron in and 1740 iron ingots PLUS 90 hmf out??? WHAT???)

but yeah, it is consuming almost 4 whole pure limestone nodes worth of rubber concrete

HEF as an alt trades steel for concrete and that's a rather good trade to make. concrete is a lot less complicated than steel to make

when making steel, you are limited by both iron and some sort of black stuff and you are capped by both of them, concrete, you just find another limestone node and some water

[onyx breach]

Question: is Ficsonium really worth the effort if you plan to do a big nuclear power plant?

[amber jacinth]

The amount of resources it gobbles to create makes it not worth it- mostly looking at the SAM usage.

[onyx breach]

Alr, better off stock plutonium waster then

[amber jacinth]

It's a novelty more than an efficient power source, unfortunately.

[prisma kraken]

it is very resource intensive to make ficsonium. if you sloop the qencoders on the output step, you double the power yield of the plut rods you make, but for watts/resource comparisons, you're probably going to make more power by instead making more uranium rods and sinking the plut rods

[amber jacinth]

Using the SAM on converting bauxite to uranium is better than using it for trigons/DMR in ficsonium production

[prisma kraken]

given that at this point with rocket fuel kind of being so good, buiilding nuclear is pretty much a vanity project at this point, so how you do it for what amts of power isn't really a question you should be asking 🙂

[amber jacinth]

Waste-free 200 uranium rods

[vernal patio]

about trains. is it a good practice to cut blocks on an ascending incline? it's a 1m ramp, so not a big one, but still. train will struggle more to get to speed, if something stops it.

[prisma kraken]

if you want to build it for a challenge 'because its there', go for it, but realize that other means of generating power are as good if not better

[onyx breach]

Making a nuclear power plant that makes 30 uranium fuel rods per minutes, i checked and calculated for the ficsonium and 💀

[prisma kraken]

i have a 4 freight car standard train length and haven't had issue with them stopping on 1 or 2 m inclines at all

i.e. i don't believe your concern is really warranted based on a lot of experience with it

if you go with 5+ frieght cars on trains, yeah, you probably want to start thinking about that being a problem, but with 4 carraiges, it isn't an issue

[vernal patio]

ok, fair enough. in practice i don't think it's an issue, but trains are a topic on which some people hold very strong opinions, so i'm always wary of commiting a heresy.

[prisma kraken]

with such things, i'm a believer in giving it a try, but having a good Plan B in your back pocket and just empirically testing if it works

[vernal patio]

makes sense. it's not like putting or removing signals is a lot of work.

[prisma kraken]

a lot of people parrot other people's experience and it gets whispered down the line minus details

again, speaking from first hand experience, i can tell you that i've not had any problems with that

[onyx breach]

Okay, but why sinking plut rods when you can just put it in a power generator, i don't really mind the waste

[amber jacinth]

Closed-loop of production

[vernal patio]

good to know, thanks.

[prisma kraken]

i don't even have to travel far you show you an instance where i'm doing exactly what you're asking about

[amber jacinth]

Some folks dislike running open-ended production chains, even if that nuclear waste storage would take literal years to fully fill

[onyx breach]

Oh okay, not my case then

[amber jacinth]

If you're fine with burning plutonium and storing the waste, go for it 🙂

[prisma kraken]

i mean, it is your game to play how you want, you can set aside a corner of the map and just turn it into chernobyl if you are fine with the plut waste

it gets to be a chore adding and augmenting that storage if you have to come back to it every 40 hrs in game though

i'm a player that intends to continue building in a save for a long while, so that isn't a good option for me, but your needs and wants are probably different than mine 🙂

[amber jacinth]

And when your factory shuts down, 50 years after humanity has fully conquered the solar system... you will regret not making a closed-loop system 😛

[prisma kraken]

something like that, yeah

honestly, waste does stack up pretty quick if you afk at all

i often afk quite a bit because i have an hour or two before RL calls me away and don't want to have to lod the game up again

[amber jacinth]

Uranium, definitely. Plutonium.. less so, but yeah, any substantial nuclear build does make it pile up

[prisma kraken]

of i'm burning in a production line and testing

my goal in this current playthrough of mine is to get some amount of ficsonium running because it looks nifty to build

[amber jacinth]

I just started planning out my 60/15/75 rod nuclear yesterday. Gonna be fun

[prisma kraken]

i'm already at 250 gw and flush with power w/o any nuclear, so it is just because i enjoy building and the game 🙂

yeah, i have a pretty good idea of what i'm going to build and i think it is going to be a 30/22.5/225 build

i may revise that downward, but i'm working on all the stuff to supply it, and it is worth keeping the numbers big

[amber jacinth]

225 ficsonium? Slooping?

[unborn dome]

Side question, you put signals at every track connection?

I've only been putting them every three support pillars or so.

[unborn dome]

What is this ratio describing? Is it the number of uranium fuel rods/plutonium fuel rods/ficsonium fuel rods?

[amber jacinth]

Yep, that's the one

[vernal patio]

the point of blocks is to divide the track into bits where the train can wait if the block before it is busy. therefore, as i understand it, a block should be a bit over a typical train long.

although there's apparently nothing bad with a single train occupying several blocks at once.

[deep jacinth]

blocking ensures that no two trains encroach the same length of track. That's all

[unborn dome]

Yeah, that's what I'm using them for. I'm just spreading them out further.

I guess the closer together they are, the higher the theoretical max throughput?

[unborn dome]

So you're going to have 150 nuclear plants?? 😮

[deep jacinth]

Your ultimately still limited by the output of the production runs being transported and the maximum amount carried by a train. the distance just shortens the time needed to transfer the components. Faster transport in some cases just means less product per run.

Ideally you want to maximize the transfer amount per run of the train.

[amber jacinth]

262.5 reactors at 200% clock is the plan

[vernal patio]

yep, more trains can fit onto the same length of way. but it's a highly theoretical situation. it's not openttd, or even factorio.

[deep jacinth]

Ideally you would want enough material transport to ensure that your downstream production is never idle.

Whatever that ratio is, that is the target.

[unborn dome]

There, fixed. How does this look for my current computer factory plan? I realized earlier that nothing really uses plastic that isn't being made here, and so I can safely use most of the 480/min I'm producing of it.

EDIT: If I use Quickwire Cable, I can cut copper out too, hmmm

[past reef]

Caterium wire into cable is more convenient considering that you havent used the entire csterium node

[unborn dome]

You're thinking maybe I should bump up the output a bit to get the caterium usage to 480?

[prisma kraken]

it looks good, not anything near resembling what i'm planning, but pretty good

[past reef]

Its not really about rounding the number, copper can be cut from this setup, if you want to go through with it caterium wire into cable is more convenient than quickwire cable since you have to get oil instead

[prisma kraken]

one think you may wish to consider since you're using cable, is take a look at iron wire + coated & insulated cable and save the copper for fused quickwire

[past reef]

I dont know if your oil setup can spare a bit of oil/HOR, but as you dont need many pm I assume convenience is better

[prisma kraken]

i honestly unintentionally end up cutting cable out of everything

[unborn dome]

I don't really want to ship in copper, so I'm leaning towards cutting it out of this factory entirely.

Tweaked a bit, uses almost all the caterium node, and 390/480 plastic I'm producing.

[past reef]

You dont need many and its a small setup I would avoid shipping in extra things

Still, looks good to me

[prisma kraken]

yeah, usually i bite the bullet and make quickwire with the fused recipe, but idk, i think you can cut it out - builds i've done in the past have always ended up with a surplus of caterium. Also cat wire is pretty good™️ and quickwire cable 🤔

very off-meta stuff

[unborn dome]

Hmm caterium wire -> cable vs quickwire cable?

Guess it's an extra production step using caterium wire.

[prisma kraken]

it has been a while since i've played with the numbers & don't know if any recipes have been tweaked, but i think for mass cable, cat wire with the insulated + coated cable combo is still the most efficient way to make cable

[delicate drift]

im hunting these mercy balls

zzzzz

[prisma kraken]

this is pretty bonkers:

[delicate drift]

is there warp tech in game?

[unborn dome]

Wow that's a lotta cable

[prisma kraken]

ikr?

[unborn dome]

There's a point-to-point portal very late in the game.

[prisma kraken]

yeah, portal tech is t9

[delicate drift]

if i ever make it

lol

[plush gulch]

Bit off topic here, but it's math.
230 - 220 x 0,5 is actually 5!

Let that sink in

[unborn dome]

What are we sinking about?

[plush gulch]

My math being right 100%

[prisma kraken]

10 x 0.5 = 5. Whoa.

so i'm building a large quartz purification build. I'm sorry to say that my suspiscion that the recipe's wastewater product is as hard to handle as default battery.

I have a bake-off going of a bunch of different pipe configurations, and all of them seem a little unstable

i'm still routing out other issues as i find them, but i have to say that this hasn't been the smoothest thing to build

[unborn ermine]

Working on mine atm too, just tossing 600/min nitrogen at a chunk of quartz and going from there

I still love how clean this is.

Im just debating atm if I want the excess to be quartz crystal or double down on silica

[prisma kraken]

just to trade notes, this is essentially what i'm doing

each 4 pink refineries process 1200/min raw

i'm doing the 1/4 split between the blenders on wastewater with clocking them

[unborn ermine]

Mine is a little messier/spread through floors, due to having nitric acid on the top floor near my drones.

[prisma kraken]

yeah, this is my first time doing this build, so there's a lot of trial and error

it feels to me like my pipes are too short

[unborn ermine]

no colour coordination here yet

The large row is the caterium thats planned with the waste water

[prisma kraken]

i've seen similar in past releases with the classic battery recipe and was never quite able to figure out what's going on

errr default battery i mean

they're honestly very similar recipes

[unborn ermine]

Yeah... I stopped around the time I wanted to muck with battery production last I played

[unborn dome]

Age-old question - do valves actually prevent backflow, or do they only restrict flow?

[prisma kraken]

nowadays if i'm forced to use default battery, i just burn the water in coal gens

[prisma kraken]

i'm of the opinion that they do neither 😄

[unborn ermine]

https://tenor.com/view/well-yes-but-actually-no-meme-aardman-the-pirates-pirate-gif-26563702

The only thing ive seen prevent backflow is a pump, and thats if you build it right.

[prisma kraken]

i think valves don't really ever quite do what you want and it isn't documented anywhere what they actually do, so...

[unborn dome]

The description says it limits flow rates, so I guess that's the bare-minimum it should do.

[river plume]

pshaw, rookie numbers

[unborn dome]

150 MW is actually an insane amount of power for a single building irl

Your house likely only pulls 2000 watts max, and we're talking 150 million watts for that place.

[prisma kraken]

well, maybe soon Musk will achieve some level of sentience and realize his child is a human being 😄

[unborn ermine]

I think the thing is, people think valves work like a stopper, they are more like electrical resistors... but fluids also flow both ways.

[unborn dome]

Yeah, I guess the input and output are what throws everyone, like why does it have an in/out?

[prisma kraken]

people have done a few experiments and observed them leaking backward

i'm unsure of the specifics of the tests, but i can point you to some people who do know

personally, i just avoid designs where metering flow is a necessity

[unborn ermine]

Oh hey I can power Grok and have 58,000MW to spare

[river plume]

definetly, ...in the context of satisfactory tho lol
them blackwell gpu racks are insanely power dense

[prisma kraken]

its amazing how much dumb money is going into souped up versions of Eliza & calling it 'AI'

[unborn dome]

It's also amazing how many companies are call a string of if-blocks "AI"

[prisma kraken]

its cool stuff and useful and all, but if it can't figure out how to find logical consistency in data, it really is no better than a parrot

[unborn dome]

Yep, it's honestly just not trustworthy. "Confidently incorrect".

[prisma kraken]

the cynic in me thinks that very few people understand what intelligence actually is 🙂

[unborn ermine]

inb4 we find out a large chunk of that Grok is just mining

[prisma kraken]

yeah, garbage in, garbage out, as the saying goes

[river plume]

primagen viewer spotted? :p
I dont really see it as "if" statements per se, as its kinda just predicting the most likely token (glob of letters) that would come next based off of a giant pool of data

[unborn dome]

Like how the Google AI pulled a Reddit joke from 12 years ago about using glue to help cheese stick to pizza lol

[river plume]

that was epic lol

[unborn dome]

Yeah, like it's all probabalistic and whatnot, but possesses zero context on what it's presenting. So glue on pizza? Sounds great!

[river plume]

in fairness though, the google overview thing is kinda nice sometimes. a lot better than most of the other llm uses since I can still see the sites that I pulled the data from

[prisma kraken]

yeah, i'm no expert with it, but essentially most of what it is is baysean filters which we've been using for 20 years to counter spam on steriods and generate, not just detect stuff

[prisma kraken]

hehe, yeah, you can get it to do some really dumb things, lol

in any event, it actually does a very nice job of replacing my high school english teacher 😄

what i'd really like is to be able to give it basic tasks like 'can you look at the trend of immigration here in the US' and do a chi-squared analysis against crime rates'

that type of thing should be more than possible to achieve, but we're wowed by the pretty pictures and stuff

[native crater]

What's the short version of how I'd go about setting up a centralized location for drones to get a fuel source from? Had a giant megastorage in my pre-1.0 save that I plan on replicating but never used drones and want to mess with them this time around

currently have a fuel source getting routed to my storage but not sure how to get non-storage drones to have fuel - Essentially seems like it would just be setting up ports at the storage, flying to wherever to pick stuff up (and dropping fuel off simultaneously) and flying back, but the "wherever" port not having a destination is what's throwing me off a little

[prisma kraken]

pretty much you just set up a central port w/o a drone and point other drone ports at it to fetch fuel and then belt it out to where you need it

[native crater]

Mmk that was my next guess was just having an empty port for them to land on - thanks

[hearty frost]

Do you have some efficiency loss using mergers/splitters?

[unborn dome]

Nope

[prisma kraken]

yeah, you can do some interestingly complicated set ups with drones, but the one i like using for its simplicity is to have a many-to-one relationship with a droneless distribution port

[hearty frost]

Math is correct, but I am getting machines backing up pretty often.

[prisma kraken]

there shouldn't be any loss with splitters & mergers, but we are starting to realize that mk6 belts sometimes 'miss' items - I believe it to be caused by fps drops

[unborn ermine]

Are you using any 1m floorholes?

[hearty frost]

I'm merging a bunch of Silica to merge them all in to mk6 belts at 1.2k/min

[unborn ermine]

that can cause issues if you are expecting MAX throughput on a belt

[hearty frost]

No, just 2m holes and 4m

[unborn ermine]

Ok, I have ran into that myself with trying to do max mk5s

[hearty frost]

I'm noticing the mk 3 miners "skipping" some. * as well

mk 3 miner on a pure node with mk 6 belts

[prisma kraken]

if you can stay under the max belt limit, i think a lot of things can auto-correct and really the pronounced problems only occur when you're in render distance

[hearty frost]

Gotcha.

So stay around 1100/belt?

[prisma kraken]

and yeah, i've observed the miners doing that

i guess? we're still figuring it all out, lol

[hearty frost]

For sure lol

It's gonna be a long debugging process but honestly... best game of the last 5 years for sure.

[prisma kraken]

right now, my current strategy is to sort of overproduce and not care if things are at 95%

i'll eventually end up going down the rabbit hole of sciencing it all out, but haven't really gotten to that point yet

[hearty frost]

yeah. Just finished a 12000/min Quickwire factory to feed my nuclear setup lol.

Trying to make sure it's all as good as possible before I get to booting up the nuclear.

[prisma kraken]

i'm doing some goofy stuff with items like nitric acid to package it and unpackage it to meter out how much goes where

[hearty frost]

With that I have honestly just been overproducing because I can not have my recycling line back up.

Emergency backup system is on standby just in case as well.

[prisma kraken]

as i mentioned a little while ago, atm, i'm working on bringing up a big quartz purification build and kinda seeing some funny things

[hearty frost]

I ended up going the Cheap Silica route.

[prisma kraken]

i.e.

[hearty frost]

Just because I am using up sooooo much nitrogen for the recycling.

[unborn dome]

I wish dissolved silica was used in other recipes. Would be a nice variation.

Rather than just converting to/from regular silica.

[hearty frost]

Oh of course.

[prisma kraken]

i wish i could package it

[prisma kraken]

maybe a refinery alum ingot recipe from scrap?

[unborn ermine]

Oh yeah, I want that to be packagable.
would lend to some more drone/train use probably.
Imagining having that delivered to you aluminum plant, and metering out the silica for max ingots, working in that water for something even more gross

[unborn dome]

That'd be cool

[prisma kraken]

i mean, i really would find it useful to package the stuff

being able to ship nitric acid in via drone and ship out disolved silica? f yes!

would be cool if it could be done in fluid tanks too

[unborn ermine]

Yeah right now with the way it is, you are roped into making/shipping iron plates on site for nitric acid after sending in packaged nitrogen.
(or at least I am )
I think that would be really nice way to go about it.

[worthy talon]

is there a way to make blueprints externally similar to SCIM's megaprints but placeable in game?

[prisma kraken]

i'm doing a hodgepodge atm

[prisma kraken]

not aware of anything, sorry

[worthy talon]

damn. alr

[unborn dome]

I think it was probably a mod, but I saw someone with an MASSIVE blueprint designer.

[faint epoch]

can someone help me with splitter math, im trying to dividea belt of 270 coal a minute into 5 assemblers that 4 take 60 a minute and 1 takes 30 a minute, if i have to i will upgrade the belt but and the output but if its avoidable then ill do it, can someone help me?

[unborn dome]

Manifold it, no need to evenly split.

[faint epoch]

what does manifold it mean?

just overflow?

[unborn dome]

It's self-correcting. The only downside (which isn't really a downside) is that it takes a while to fill up and even out.

Since the furthest away machine gets a trickle of items until the closer ones fill up completely.

[faint epoch]

this is for my feul plant

[unborn dome]

That's fine. It's the same thing. Once all the inputs have a full stack of the item, it'll self-balance.

[faint epoch]

ill try it, let it overflow, then do it, thx

so like this?

[unborn dome]

Yep exactly, although you don't need that final splitter on the end.

And you can also manifold your outputs into a single belt too.

[faint epoch]

they're going to seperate refineries

[unborn dome]

That's fine. Manifold the refinery inputs too. 😄

[faint epoch]

the one on the left is going to just one refinery, all the others are going to 2 refineries respectively

[prisma kraken]

yeah, technically the english definition of manifold is any shape that provides connection for moving stuff around, but in the context of the game, we mean a system of belts or pipes where stuff overflows into each machine to meter the distribution

[faint epoch]

time to do that again

[unborn dome]

It's rare to actually need to equally-balance the belts leading to inputs. Nuclear power fuel rods are an example.

[faint epoch]

since it takes so long to make one the splitters reset

[vapid gorge]

Even then you don’t need to

[unborn dome]

Yeah it'll just take several days for the manifold to fill up 😅

[prisma kraken]

@unborn ermine think i found my quartz issue, the mk1 pipes i was using for 300/min dissolved needed to be upgraded to mk2

[unborn ermine]

Was it a flow issue somehow? Weird if that was the case.

[prisma kraken]

each 300/min is being processed by 2 blenders - one at 200% and one at 50% to partition the waste & fresh water

i'm guessing the uneven draw from the pipe between the 2 blenders was causing what's akin to a bad harmonic?

[unborn ermine]

Oh the mk2 to keep the flow "overflowing" to the new pipe instead of just being limited, that is funky.

[brisk smelt]

hey guys do you feed 2 rows of machines like 1) or 2)

[prisma kraken]

i've seen similar before, no idea really what's going on in the code, but i keep a few tricks in my back pocket for such situations 🙂

[brisk smelt]

orange means input green is output

[prisma kraken]

whichever is easier to build given the other constraints on what i'm building

[vapid gorge]

Just pre fill it with stacks of rods

[unborn dome]

Yeah true, if you build the rod production part first and let it stack up while you build the power plant portion.

[vapid gorge]

Depends on the product and the rest of the factory

[prisma kraken]

like if you're going to be immediately splitting the output into halves, you might as well use (1) and remove the merger, since that is easier to build

if it is something like cat ore->ingot, (2) may make more sense since 2 lines of ore will make 1 line of ingots due to how the yields work

[faint epoch]

alrighty, my second manifold is complete for the same assemblers, it took so long cause i had to drag the other belt over

[pulsar notch]

that injection manifold picture is wild. Never needed something like that, but it's wild to see.

[unborn dome]

Why's that?

[pulsar notch]

oh, the merger in the middle of the stream after you split off a couple flows with splitters. Is that due to belt limits or something?

[vapid gorge]

Injection manifolds are largely pointless.

It’s essentially just connecting multiple manifolds to each other. At BEST you might shenanigan 1-2 fewer machines, but then that really limits your clocking options for the next step

[vapid gorge]

Yes

[unborn dome]

I've had to do that in one place to avoid crossing belts lol

Ended up drawing myself a diagram on real paper with arrows and numbers to make sure it was balanced right at every point.

[pulsar notch]

generally, the most I do for load balancing is using belts as a really coarse throttle.

Also, typically, if I'm within 20 or so of a belt limit I just upgrade a tier if I can.

[unborn dome]

In my case it wasn't an issue with belt speed. It was because I'm feeding steel pipes from three places, and consuming them in four places on the floor above, so I had to get clever with the logistics floor in between.

Since none of the three feeders was sufficient to fully-feed half the consumers.

And one of them is actually two lines combined, so it was 90, 180, 90 coming up through the floor.

[prisma kraken]

when you're building very large factories where you're pulling stuff from multiple train cars and none of the rates really line up, there's a place for them or balancers there, but i generally agree. The thing is that it is important to have such tools in your toolbox for the one-offs

...speaking from experience, injected manifolds are really often very painful things to build and get functioning correctly. I don't recommend using them as a first option

[unborn dome]

YES

(Also speaking from experience)

The pain is real

[prisma kraken]

whenever i've built one, i'm always wondering if i mathed it all right

and there's no easy way of verifying that

[vapid gorge]

Process what each car delivers? ^_-

[prisma kraken]

yeah, definitely preferable to do that way

[unborn dome]

Yeah I would absolutely not want to do that frequently, but it was great in a pinch once I recalculated everything like a dozen times to work out all the kinks in the routing.

[prisma kraken]

last time i found myself using one was in u7 or u8 to process all the pink forest bauxite and was bringing all of the baux along with coke and copper together via train, and there just wasn't any good way of splitting it between cars and resolving the 780/min from the pure nodes

so far in 1.0, i haven't found a case where 1200/min hasn't been enough for a line i've made, but i'm still building new things and the book on it isn't written yet 🙂

also, just not needing to build to 780/min makes a lot of stuff simpler

[urban moss]

Can converters be slooped?

[amber jacinth]

Yep

[prisma kraken]

slooping ficsite ingots is a pretty good thing to be doing

[unborn dome]

Why's that? Expensive to make?

[prisma kraken]

yeah, they take a lot of sam & they take a lot of iron/cat/alum ingots

all of the t9 stuff just takes biome-spanning amounts of raw resources

[urban moss]

yeah

im just trying to get the neural things automated and oh my god its so many machiens

machines* and not even like lots of volume of things, just so many little parts needed

[frosty owl]

I've been using them for all my water recycling in aluminum.
Would those work without valves too?... I dunno 🤷‍♂️

[prisma kraken]

i'd put my money on 'yes'

i've been doing a bake-off on different pipe designs for purification

i actually have them all running stably at this point

the distilled silica pipes supplying it shouldn't be pushed to max capacity or it won't work

the blenders gobble too much fluid each cycle and it leads to a lot of 'backflow' - took me a while to figure that out

[frosty owl]

I've seen that picture already, but I don't understand what you're trying to say with it now

[frosty owl]

I'm willing to pass the save, but not to mess with it myself.
Ngl, the idea of testing fluids doesn't quite excite me right now 😅

[unborn ermine]

240 computers/min seems like a decent number to start with for a starter computer factory right?
60/min High Speed connectors as well.

Working around my planned silica + crystal outputs and its fun getting numbers

[past reef]

you barely need supercomputer tho

[prisma kraken]

only thing that 'needs' computers are ACU's. look at what you need for RCU's and supercomps with which recipes

[past reef]

I'd understand RCU in the hundreds but not computer lol

[prisma kraken]

my plan is to use the heatsink+hsc rcu recipe, so i've cut out a lot of them

btw, the radio connection unit alt is now a viable way to make them that saves aluminum

[unborn ermine]

Hmm so at that point, still worth it to make the oscillators enmasse then do some crystal computers, sending oscillators somewhere else.

[prisma kraken]

i'm not sure what was rebalanced to do so, but it no longer takes an unreasonable amt of aluminum

[unborn ermine]

Just have to see how much I can fiddle with in the end for my aluminum production.

Wouldnt mind sinking a ton of resources in the mean time anyways
me and my over 1 million points per minute with barely anything done.

[prisma kraken]

my accounting for aluminum atm is that i need 3000 ingots for slooped ficsite, 3000 for fmf's 3600 for heatsinks and the rest is sort of floating between fluid tanks and sheets for batteries and i need to nail that down

[past reef]

Heat sinks/cooling device?

[prisma kraken]

hey, i'm pushing 3m points/min and i feel like i'm not getting anywhere

[past reef]

Are you making 50s of rockets

[prisma kraken]

some heatsinks are going to cooling device, the rest rcu's

[unborn ermine]

Im basically tech wise in phase 3

im in phase 4 for power gen

(I say that with 2940/min aluminum ingots being made atm)

[prisma kraken]

i'm not even really thinking along those lines, and instead going biome by biome & using up the resources on what i need to build, lol

[past reef]

I havent made mass RCU for cubes yet, did 24 RCU with heat sinks though, seems a bit space saving

[prisma kraken]

i think what changed to make the connection unit alt more compelling is the default now uses more computers

what really did change in a good way is that the cooling device alt is considerably cheaper

both cooling system recipes now use the same number of heatsinks and the alt saves on nitrogen in exchange for motors

that really needed the rebalance

cooling systems just always took all of your baux AND nitrogen budget

that in turn made the OC supercomp alt unusable

[unborn ermine]

Thats my planned cooling system production yeah, just need to plan my numbers out

[past reef]

Interesting, I barely finished aluminum last time, good to know cooling device isnt as painful

[unborn ermine]

you save 30/min nitrogen using the alt recipe now every 10 Cooling Unit

[prisma kraken]

i mean, they're all endgame parts and a certain amount of pain is in order, but the rebalance made most recipes at least viable if you're mixing and matching

it doesn't seem like a lot, but that is considerable if you're looking for a meaty prod rate of the elevator parts and some form of max nuclear

[unborn ermine]

its late

staring at recipes they are blending

[vast jungle]

looking at heatsinks/cooling-systems yes, cooling-device looks quite nice...

cooling device and heat exchanger, I think I have the next project (just finished some copper powder)

[prisma kraken]

yeah cooling device really got buffed

both that and iron alloy just never made sense to use, like someone accidentally entered the wrong numbers for them somehow

[vast jungle]

and I built my "temporary" (only) Aluminium production directly across (the trainline) of my Plastic/Rubber production 😄

I still wonder that I got that far with only 270 rubber/plastic ^^

[prisma kraken]

well yeah, you make the rubber and coke and ship em to the bauxite at the same time, lol

[vast jungle]

close enough that I will just belt it over 😉

[earnest rapids]

I started a bus running from my train stations to the production side of the factory, but I just realized I have no idea how I'm going to split off a belt from the main bus when I need it without clipping through everything. Anyone dealt with this or have suggestions?

[prisma kraken]

last time i built out the pink forest, i actually had a train as a bus to move rubber, coke and fuel across crater lakes

[past reef]

nah you have to keep each resource type in one horizontal level to not clip

[vast jungle]

SF is 3D, so you can always raise a belt temporary to add a splitter

[prisma kraken]

you drop down or go up, split, then come back up, the building pattern is often called a belt weave

[past reef]

other than that can just use floating splitter on the edges

[earnest rapids]

so raise everything above the material I need up for a few foundations, split it out, then collapse the stack back down. Only problem is I'm scraping the ceiling already, I'll have to do that outside this logistics floor and make a higher ceiling.

[prisma kraken]

i personally tend to like stacking multitudes of belts vertically and use all the vertical space available before adding another line of them

[past reef]

well it's better to keep resources on the same horizontal level instead so you don't have to think about splitting like that

[prisma kraken]

(going vertically means you use a lot less space when turning)

[past reef]

if all copper belts are 4m off the ground you can split from the outermost ones and get more from the one inside

[earnest rapids]

I stacked it all so high because I didn't have enough room horizontally due to a cliff

no clue if I'll ever get anywhere near the capacity of this bus, my pre 1.0 builds were like a few per minute of each thing and I didn't get to nuclear

tore down all my production to start this base so all i have going is concrete

[prisma kraken]

most of my beltwork is mounted on ceilings

[vast jungle]

for a "main belt" its a good idea to make the logistic flood at least 8m high

[prisma kraken]

and space belt stacks at least a splitter width's apart

[past reef]

man's logistics floor must be like 18m high already

[prisma kraken]

often you want space for a splitter and a lift connected to it

[past reef]

you do need to design gaps though, a belt should a belt's width (2m) apart horizontally and a belt's height apart vertically to not clip all over

especially if the bus is not just raw resources

[prisma kraken]

i've found in this game, trying to shoehorn things into a tight space doesn't often turn out well

[past reef]

more like hard to make it turn out well cause half of the time you redo a certain section

[vast jungle]

I once had a logistic floor 3m high... that was painful

[past reef]

while with 2x more space you don't have to plan such connection

[earnest rapids]

yeah my floor is 19m

i guess it would be worth spreading them out so I can split something out between belts and on the floor, just annoying spending so much time on this repetitive stuff

[prisma kraken]

i've done a number of large builds in north forest off of speedrunner's bluff, and there's a spot in there between the cliff and the concrete node that is an absolute nightmare

always have a butt-ton of belts to run through there and there's never enough space

[vast jungle]

yes, that spot can be a bit tight...

too many pure nodes in one place

[prisma kraken]

@vast jungle i was sharing earlier, i've been working on a quartz purification build, and i have to say, the silica part of it is exceptionally finicky

[vast jungle]

I have looked at the recipes and decided "not worth my time at the moment"

[prisma kraken]

good recipes, but it is really tricky to get running at 100%

[vast jungle]

I AM using quite a bit "Pure Quartz Crystal" in some BPs

[prisma kraken]

yeah, that has a better crystal yield, but if you need both silica and crystal...

[vast jungle]

I don't like building a production for two products that depend on each other ^^

[prisma kraken]

altogether, it is one of the more challenging builds in the game - i'd rate it somewhere around same level as default battery to get working right

exactly the problem, engineering the ways for them not to grind to a halt takes some effort

[unborn ermine]

I wanna say I like purification, BUT, I need to figure out where I want to shove all that silica

[prisma kraken]

easy. make 4/3rd's the aluminum

[vast jungle]

you can always add 2 sinks... but then most of the complexity/efficiency is wasted

I would not consider using it unless I have a factory needing both lots of silica and quartz crystal

[unborn ermine]

But if Im nowhere near aluminum on the opposite side of the map as the free bauxite nodes... 👀

[prisma kraken]

how i'm generally handling everything is chucking it on a train whose ultimate destination is a sink

[unborn ermine]

I do have some nodes near my aluminum I could feed into it, but thats not where my other factory is right now

[vast jungle]

I tend to have one train for each "receiving" station... and sink at the production with an overflow splitter

[prisma kraken]

yeah, i'm pretty much 'as neeeded, wherever makes sense' my sinking locations for any factory get farther and farther away from production as the factories mature

[unborn ermine]

So if I am making 120/min oscillators, how much of that do you think would be a reasonable amount to dump into computers for ACUs?

[prisma kraken]

oft times i hang temporary sinks off of prod lines because i don't want to deal with something yet and circle back later to start refining it or stick it on a train to use elsewhere

[vast jungle]

120 OSC/min? Wow, thats a lot ^^

[unborn ermine]

Still processing that info myself

[prisma kraken]

i've built that before with the default recipe!

[vast jungle]

if the ISC attached to the outgoing train station overflows I don't care if the overflow goes into the sink 😉

[prisma kraken]

man, that was a build that burned me the f*** out

[vast jungle]

some factories are even just allowed to block when the trainstation is full

I have 16 OSC/min (with standard recipe)... and thats a 5x5 tower 320m high ^^

[prisma kraken]

it depends on the prod line, i'm flexible in my design principals with it all

i do want to keep my factories always on and at 100%

[unborn ermine]

its only 64 machines clocked at 100%
it could be worse.

[vast jungle]

for some time I was at 50% energy production... double max consumtion 🙂
blocking factories was a feature

[vast jungle]

I think your math is off...

or the number of sloops you have 😉

[vast jungle]

hmm... okay, my math was off ^^

[prisma kraken]

yeah, they changed the recipe speed i think

[vast jungle]

no, they did not... according to wiki

[prisma kraken]

its what, 1.875/min?

i'm thinking crystal computer

[vast jungle]

interesting that 64 machines gives you this nice and round 120/min number

[prisma kraken]

well, 8 gives you 15/min 🙂

[unborn ermine]

What gets me here, im only using around 900/min of the 2025 silica here and thats before computers' silicon circuit boards.

Its like, what do

[prisma kraken]

1000 silica will turn 600 bauxite (with sloppy+electro) into 800 ingots

[unborn ermine]

I wish I could package the liquid silica and ship it out

[prisma kraken]

it ain't cheap, but it's 1000 silica per 600 bauxite to gain an extra 200 ingots

[unborn ermine]

Yeah if I "upgrade" one of my modules for ingots

goes from 735 to 980

[prisma kraken]

i don't get why they made the numbers for the default ingot recipe the x.33333 machines

its like they really wanted it to be extremely bad to use

[unborn ermine]

I dont mind it, makes the silica from here become your storage trickle for building windows

[prisma kraken]

yeah, well, yeah, lol

anyway, grabbing some zzz's later all

[unborn ermine]

Same

Gn

[violet halo]

@lilac horizon If every other pipe led out, that would be much closer to working. But I still don't think it would be 100% output for this. #screenshots message

[brisk smelt]

very dead chat hmm

[outer vale]

momma always taught me to be quiet when I had nothing of value to say

[white bloom]

icemoonmagic's splitter calculator suggests this setup:

you said you don't have mk6 yet so instead of the 845 make a 3-to-2 balancer instead of the last merger

[wind spade]

@sturdy island see #math-and-meta message for alt recipe choices

[nimble yacht]

Wow, thanks man 🤯❤️

[shrewd oar]

can't just use a simple injection manifold?

the only numbers that matter in a simple injection manifold are "is there enough here to feed the next machine" if NO -> merge belt to refill + log quantity remaining on belt

[charred saffron]

Manifolds are usually the better way to go, but load balancing is fun and looks fancy

[shrewd oar]

i'll never consider waste-of-space-waste-of-time-giant-piles-of-mergers/splitters-for-no-reason either fun or fancy, but whatever floats your boat i guess

they're kinda like a teenager's first car with those overly loud exhausts... all the noise with no performance

[hushed kettle]

you just can't comprehend the beauty of having a belt full of items split and merge again until everything is coming out in unison according to the math you calculated

[charred saffron]

Achieving peak efficiency for minimal cost is also a form of fun to some, but the beauty of every belt never clogging up and ending up in perfect unison brings joy to others

Opinions differ and thank god they do otherwise the world would be so boring

[shrewd oar]

Very true

[deft lichen]

for future reference (because you won't be able to solve this without external help, probably)
change the input lines to make it possible to directly merge them without balancing

just clock the machines and merge them into groups accordingly, but keep the sum of their clock speeds the same

that gives you a "ratio splitter"

e.g. suppose you have lines of 60+60 and want to split them into 6+114
instead of a headache-inducing balancer, you just clock the machines to produce 6, 54 and 60, then merge the 54+60

unlike a balancer, this works for ANY number, and is trivial

[proud totem]

Would be interesting if there was a "trivial machine" so to speak. One that effectively does nothing (i.e. it just passes the item through it, always at a 1:1 ratio), but you can set the output rate like any other machine. Then you could just manifold into a set of these machines to get the desired balanced output

[wind spade]

That's the same as manifolding to actual machines, they will only eat what they need

[proud totem]

Right, that's what the idea is based on 😁 Except in a "trivial machine" no work is actually done to the items, so for example you can balance iron ore without smelting it into ingots in the process

[proper girder]

Does anyone have a breakdown or source describing the discrete values Valves are capable of?

[outer vale]

it's either 1/254 or 1/255 of the pipe throughput IIRC

[proud totem]

Yeah, I've heard its 8-bits, so increments of 1/255 would make sense. Though I am curious where that comes from, is it just purely the UI? Because in the headers there is the mUserFlowLimit for valves, which is a float, and therefore a 32-bit precision floating-point decimal, i.e. many fractional values, so I am curious to the discrepency if anyone knows

[proper girder]

1/254 would make sense if excluding the zero value. It might be an artificial limitation to reduce CPU overhead despite the 32-bit precision. Weather that's UI or engine based is the real question.

[proud totem]

Considering the flow limit is stored as a float, I'm assuming the fluid system still uses it as a float, and is just converted from the UI. I also don't know if using an 8-bit float is really even possible, or even well-supported by CPU hardware. From what I've seen, most hardware-based floating point operations work on 32 or 64 bit representations

But I'd have to dive back into x86 ASM I guess to verify if that is true

Yeah, looks like x86 only actually supports 32, 64, and 80 bit floating point numbers anyways, so the flow rate value is definitely treated as 32-bits for calculations

[prisma kraken]

8 bit floats are possible and they do exist, but the valve probably just uses an 8 bit integer

and um, more recent x86 processors do more than 80 bits but it is kind of really dependent upon the exact processor. Intel has always been a bit weird about specificying the exact precision of their fp implementation

[proud totem]

They're possible yes, along with 16bit ones, but I could not find any documentation in x86 that natively supports them. Same for higher bits above 80. Some processors may have hardware and instructions to support them (I think RISC-V supports quads), but the x86 standard itself does not.

All that to say, I do not think satisfactory is/would be gaining any performance boost if they are/were to be using 8-bit floating point values for valves specifically. Especially since it is already stored as 32-bits anyways

So my guess is it is a limit from the UI, and potentially for the save file. It may convert between 8 and 32 bits once when saving/loading data on disk

[prisma kraken]

i believe some of the mmx/sse/avx registers can deal with 8 bit floats for doing stuff like very imprecise fast alpha blending

it's been a while since i've done a deep dive into that all, so my memory may be off

the lack of meaningful precision with only 8 bits is pretty limiting

[proud totem]

Possibly I guess? But also, I would be very surprised if they tried to implement that when everything is stored as single floats anyways

IMO they should've done fixed point precision, but oh well

[prisma kraken]

yeah, when you are doing stuff like dithering, sometimes chucking precision out the window is a win because the result is going to look awful regardless and you just want it to be quick 🙂

idk what the state of the art is with that all, i just know i've come across it in my travels and that there's some practical applications. getting back to the valve, it is most likely it is just a value 0-255 that represents the fraction of 300 or 600 that corresponds to the flow rate

[proud totem]

Yeah, I'm guessing it is just an uint8 they use and multiply by the max flow rate (which is also a float) to obtain the value used in memory. Still decently frusturating though

[prisma kraken]

in any event, if you are relying on valves to meter flow, you're going to have problems with pipes. they don't really work intuitively to begin with

[proud totem]

But yeah, I forgot about SSE and AMD64 stuff that extended floating point

[proud totem]

Oh yeah, just the logic itself is not fully intuitive

[prisma kraken]

even when you can set them to the exact rate you'd wish, their flow is still dependent upon how full the pipe they are constricting actually is

[proud totem]

And since it is changing the max flow, your average flow is likely to be below that, since any dip (i.e. from a machine's regular cycle) will reduce the flow in that moment, and therefore reduce the average to below the max

[prisma kraken]

yeah, tl;dr don't use valves, lol

[proud totem]

Except for preventing actual backwards movement of liquid, in the case of forcing a full fluid buffer to only drain and not re-fill (useful for saturating a system).
Or in EA when you could scuff headlift through a 0-flow valve

[prisma kraken]

i don't think i've come across any specific build in the game after 1000's of hours that requires a valve

[proud totem]

Those are the only 2 I can think of. The first is a short use-case and definitely not required at all. The 2nd doesn't exist anymore

[prisma kraken]

i thought the headlift thing still works

it has been a few game releases since i've tried though

[proud totem]

Oh maybe? I guess I haven't tried but I heard it was patched in 1.0

[prisma kraken]

perhaps. i've learned to avoid things that happen to work without making sense

...especially with pipes, lol

[prisma kraken]

@oblique hollow

in each pair of blenders, one is clocked at 50%, the other at 200%

[oblique hollow]

seems a bit too complex

[prisma kraken]

yeah, well it is a big build... 3900/min raw in and partitioning outputs how i need them to be for other projects

its just taking the output of 5 logical blenders and feeding it into 4

(each unit is built differently, if you hadn't noticed - faster to parallelize the experimentation)

it took a while to realize the 300/min distilled silica feed needed an mk2 pipe

the rest of it, yeah, i have some cleanup to do

i was anticipating it being a bit tricky to get the piping right so i tried a bunch of different things at once

tl;dr, we're going to end up with a lot of questions about it

[plain fossil]

how do i signal this

one train is going the straight path and the other is making both turns

[prisma kraken]

just like you'd signal each fork separately

[plain fossil]

do i even use path signals

[prisma kraken]

personal choice

keep in mind that anything can be signaled with block signals

[charred saffron]

Path signals are for more extreme multi-route tracks but block signals will do most of the time. It's your choice and preference

[prisma kraken]

i often start with block signals only and if there is a bottleneck i'm seeing, then switch to path signals

with that rail intersection, i don't believe 2 trains could ever exist anywhere in it concurrently, so path signals probably wouldn't even do anything useful

i'll hedge that statement... if you have a bi-directional rail and trains are passing both directions on that track, then you do need path signals, but um... that path leads to pain

[plain fossil]

Is this correct?

[gloomy wagon]

path in, block out

[plain fossil]

should be

[gloomy wagon]

are the 2 branches going to a station? or just a intersection

[plain fossil]

station

[prisma kraken]

it looks correct, but without the rest of the rails around it, who knows?

[gloomy wagon]

yeah ^^^

depends on that mostly

[prisma kraken]

also it looks like you are trying to do this with a bi-directional track, and they are very very difficult to reason about

as i already said, that path leads to pain

[frosty owl]

What's your favorite diamond recipe?

[plain fossil]

its just this man, 4 stations 2 trains, seems very simple, but nothing works

[frosty owl]

Oh boy... Still wrestling with trains since yesterday?! 😆

[plain fossil]

yeah

[prisma kraken]

i'm favoring default and turbo, but for diamonds with how much you need, its kind of like whatever you have handy to depress the coal cost

[plain fossil]

signaling problems across the board, would be nice if they told what they were

[frosty owl]

Stay strong 💪

[frosty owl]

My first ones were with Coal, but then... Oil oil oil <3

[gloomy wagon]

quick tip!

https://docs.ficsit.app/satisfactory-modding/latest/CommunityResources/TrainSignalGuide.html

Take a look onto here, help alot in the beginning

[prisma kraken]

yeah, if you have a few spare oil nodes, toss them at it, though you ought to consider blended tf as well

[frosty owl]

Tbh, anything involving trains going back-and-forth rather than in a loop is more than just "simple" imo 😅

[frosty owl]

I kept hesitating between turbocoal and Oil, but in the end the appeal of using unprocessed resources won ^^

[prisma kraken]

we're not trying to be jerks when we suggest not to build bi-di rails

yeah, really diamonds is kind of a function of how many you need. past a certain amt, turbodiamonds starts looking very good

[frosty owl]

Maybe if you add one signal at the end of each station too...?

I really dunno how you could fix the "rail ends abruptly" issue or whatever it's called (I'm assuming this is what the signals are complaining about) in this kind of situation

[prisma kraken]

yeah, signals get unhappy when there isn't an 'other end' to a block

[dusty jetty]

IMO

let the train on the main path through and only let the green path train through if the blue block is empty

(straight path block signals a bit futher away)

anyone correct me if I'm wrong, I just like the main / side path separation

[frosty owl]

(This one is gonna make a lot of pioneers scratch their helmet xD)

[plain fossil]

yeah but 4 block signals is not gonna cut it

signal loops into itself

ive tried at least 50 combinations at this point is there really no straight answer

maybe found it RAAAAHHH

one train works the other one doesnt, even tho they have the same exact signaling

slightly going insane

oh wait the other train was just lazy and started working 5 minutes later

[charred saffron]

Wouldn't block signals on every path work, keeping the junctions and piece between them one block? That way the entire path for the train would remain open or closed so the trains don't block each others' paths-
nevermind, it's fixed now?

[plain fossil]

no it was actually path signals

path signals on block entry

block signals on block exit

so just path signals on the right of each train and block on the left

cant believe it took me 5 google searches to get the answer i needed that wasnt a 10 minute long video

[magic dock]

100 HMF/min no alt

@hushed kettle I created a thread so you can keep up with my monstrosity

[brisk smelt]

no pure is hell

[magic dock]

This was probably the biggest hurdle though

Because now I can just design blueprints for the constructors etc. and then connect the belts

[fringe seal]

the branches also need to have bidirectional signals

also
I'd remove signals in the shared track, to reduce risk of deadlock

ah

apparently it was already solved mb

[neat crest]

Looking at various rankings of alternates it seems like iron pipe basically makes steel unnecessary since you want the coal for late game stuff and iron pipe is so much easier to build modular factories with it

[wind spade]

pretty much no alt recipe ranking agrees with others on anything

[neat crest]

IMO css should add steel-only alternates for many iron parts to make using steel more worthwhile.

[deft lichen]

every "objective" alt recipe ranking is in fact very subjective based on how the author made it

[proud totem]

Clearly the objectively best alt is charcoal. Run all your power off of wood

[neat crest]

I'm aware of the subjectivity, thanks tho.

[proud totem]

(Except this actually still has use cases, for creating random things like filters it can be nice to dump excess wood)

[oblique hollow]

pardon the mod buildings for spawning in the resources, but heres how i would set up the "basic" 3 sets

disconnected water feed, shared disolved silica, shared nitric acid pipes, shared water output

unironically, a valve on the pipe going to the nitric acid blenders set to something close to what they need works here

just because that valve then forces a majority of the water to go to the silica blenders

[neat crest]

For quartz purification there's no need to connect recycle water to input water. You just need one extra blender and then slightly underclocking them is easy peasy

[oblique hollow]

thats what i did here

[true pier]

where is this node located at ?

of uranium

[fringe seal]

you need to enter from near the grassfields void iirc

the depression surrounding the grassfield void, there's a cave entrance

[true pier]

thanks

[unborn dome]

How can a 100%-filled fluid buffer I'm using for storage, with only a single pipe connected to it, simultaneously be filling and draining at the same time?

[prisma kraken]

really is more or less same as what i have

the point i was making earlier is that it didn't work right when i was using an mk1 pipe for the distilled silica

also, the small pipe network for the waste water is rather prone to lockup just due to its small total volume

[prisma kraken]

don't question. just accept and move on.

in actuality, pipes are bidirectional. you can think of the pipe and the buffer as discrete containers that can accept and transfer stuff during each discrete simulation timeslice up to their max rated speed. those numbers are how much has moved back and forth as average over a sliding window of time

[unborn dome]

Ah ok, so it gained 21 and the immediately lost 21 in the same slice.

[prisma kraken]

probably more like each 30th of a second a bit moves in one direction or another and that's the sum of total movement

i think really a lot of those numbers are there just to make the UI look cool than to really be of any use

sometimes you just want a blinking light on something to show activity and motion, ya know?

[unborn dome]

That's fair tbh. Same with a network rack.

[prisma kraken]

zactly

i think really the pipe simulation amounts to something like a loop every timeslice that is nothing more than 'for each pipe segment equalize volume with neighbors, starting high to low'

and then the equalize operation caps what can be transfered in or out at pipe speed

probably some horribly complicated if/else stuff in that that causes all the weird sh** that we see because of its size and special cases, but the simulation per timeslice is most likely very simple because it has to be

[unborn dome]

Yeah simple just for performance reasons I imagine

[prisma kraken]

it has to do it for every pipe segment in your world ever xth of a second

[worldly agate]

Making a rocket fuel plant in the southern crater. 1320 generators. Their footprint takes up the whole lake (granted, part of that is my blueprint having 4 gens with a buffer in a 6x6 space, the buffer is above the gap between them).

[unborn dome]

I wonder if the simulation gets simplified when you're not nearby? Like it stops calculating parts on belts and fluids in pipes, and just makes note of how much each machine is producing and comsuming, and where the output goes?

[prisma kraken]

in addition to all the belt logic it also is doing that with and the vehicle and train and machine simulation AND rendering stuff

[worldly agate]

It totally does.

Otherwise getting further from a factory wouldn't reduce lag/latency.

[prisma kraken]

i've seen several artifacts of moving within render distance that suggest that there's a near and far mode to the simulation

in past playthroughs, i've tried to keep flatline power graphs and have noticed that moving close/far from a factory actually changed power consumption

drove me bonkers for a while before i figured out that thats what it was and learned to accept it

so yeah, when stuff doesn't need to get rendered, the game does simplify the simulation

[worldly agate]

T9 spoiler building definitely has a different power curve when I'm further from it.

[prisma kraken]

i can't really tell you how, but i think just about everyone with some hours in the game has observed stuff that backs that hypothesis

[prisma kraken]

i haven't really set up continuous prod for anything t9 yet

[worldly agate]

It's not a hypothesis, pretty sure it's confirmed by devs. It's certainly mentioned in the wiki.

[prisma kraken]

good to know

[worldly agate]

Me neither, I just have a couple pets that I have sharded & slooped to make small amounts of stuff.

1 took up half my power, thus my need for rocket fuel.

[prisma kraken]

yeah, something slooped at 250% will definitely make that pop out for you, lol

on the subject of BC rocket fuel... you can make some power there 😉

horribly incomplete but functional 🙂

i'm hoping to eventually convert it to ion fuel - that's a project i'm not really even thinking about yet though

[worldly agate]

I can't even imagine how much floor space that'd need...

[prisma kraken]

in the shadows somewhere under the far side is a 90 hmf/min factory, lol

i'm already swimming in power between the 120gw from rocket fuel and a bunch of APA's

in terms of space, the conversion to IF should actually only be the addition of 12 generators per side of the octagon if i'm remembering the preliminary calculations i've made, not sure exactly where they'll go, but i have plenty of space

i'm still building stuff under it all, so it really is still very incomplete as a structure until i know how much space i need for everything else

atm, i'm up working up in east dune forest on a bunch of phase 3 & 4 stuff just to start getting that stuff automated in volume

[frosty owl]

If anyone feels like checking some math... https://www.satisfactorytools.com/1.0/production?share=Puzc6YqYq8dxoGlFsczI
I made a plan that I think should represent the biggest possible Nuclear factory with endless sloops, corresponding to the usual meta "all alts for Uranium, all Standard for Plutonium". Interestingly, this is the second best result for PFR/min. The best one is using the Plutonium Fuel Unit alt, but the worse ends up being Instant, as skipping one production steps also skips one Slooplier- I mean, Sloop multiplier. I haven't run the numbers for the Fertile Uranium route though!
Uranium Fuel Rods: 201.6/min = 2.52TW
Plutonium Fuel Rods: 645.12/min = 16.128TW

[prisma kraken]

you'll find that fertile uranium most likely doubles the plut rod yield

[fringe pawn]

If CSS regenerates sloops, foliage, and so forth again, then at some point it's fair to say the limit isn't 104. It's unlimited on an unknown respawn timer, with some indeterminate practical limit. How many sloops does it use?

[prisma kraken]

also, looking at your numbers, i've fit the same number of fics rods into a 15/15/15 elevator prod goal using the non-hack 103 sloops available on map with 2000 uranium used

i think my tally for total used is 92

which leaves you with 11 to use for an APA and some constructor somewhere

[frosty owl]

I find it hard to judge as I don't recall the numbers well and it's unclear to me how badly that recipe hurts Waste production
Anyway, I crunched these numbers before going to bed, so I'll be curios to see developements but won't be doing anymore math for the day 😅

[frosty owl]

"Same number" as what?

[prisma kraken]

fertile is pretty good, the trap with it is that you want to use instant cell with it when you should use plut pellet & encased cell instead

give me a sec to put together a listing

[frosty owl]

That'd be a bit long to calculate. There's 12'386 machines, assuming them all at 250% and Slooped at 2x, that's 2'477.2 (2'478) machines needing from 1 to 4 Sloops each

[prisma kraken]

this makes 225 fics rods for 2000 uranium - all totalled reactor yield is 1.5 tw

and, um, you CAN do bigger than this

the encoders for the ffr's need 4 sloops each (72 total), and you end up needing 10 converters for fingots that should be slooped (additional 20 sloops)

slooping the FFR's makes the production dm-neutral, but you still need some dm crystal from somewhere to make the singularity cells needed

if you were instead to use non-fiss uranium, you'd end up with your numbers and have 60 ufr/min that turn into 22.5 pfr's

if memory serves me correctly, that ups the yield to 1.875 tw

what i've come to realize is that max nuclear wishes to be built in chunks that process 500 uranium, and it is possible to go up the next full quanta or two to processing 2500 or 3000 uranium, but at that point you're starting to run into map limits for everything but the kitchen sink

where my planning is kind of directed atm is in making the stuff needed for essentially the build i just shared and probably end up building something a little bit smaller, but allowing myself to use the leftover 100 uranium/min on the map as drone fuel

(and to make some spider food 😄 )

[sullen knot]

Hey is the ratio for standard turbo fuel recipe 4 fuel to 1 compact coal makes 6 turbo fuel? That's what chat gpt said and I don't trust the fandom wiki and I don't have access to my game rn to check

[prisma kraken]

ai can't do math to save its life 😄

[sullen knot]

Thank you! Just discovered chat gpt, trying to use it to do the math for me

[sullen knot]

Ope well crap

[unborn dome]

Use the gg wiki, it's the official one

[prisma kraken]

like, don't try, it has no ability to calculate things, it just probabalistically responds with strings of symbols based on data it has 'learned'

[sullen knot]

Interesting, I didn't know that, I'm not tech savvy at all so I thought I discovered the coolest thing since sliced bread lol

Especially for this game

[prisma kraken]

it would be nice, but no, it isn't very intelligent

[unborn dome]

It can be cool, just be aware that it can be "confidently incorrect", so you need know enough to fact-check the result.

[sullen knot]

Thank y'all so much! I'm glad I came here!

[prisma kraken]

i'm dreading the apolcalypse that comes when people start basing decisions on chatgpt's 'calculations'

[unborn dome]

Glue will become a staple pizza ingredient

[prisma kraken]

it can replace your english teacher, but we still need the math teachers around 🙂

[vapid gorge]

ehhh.. I wouldn't say that about english teachers either

[unborn dome]

Slooping the constructor that turns alien protein into biomass is basically the "infinite biofuel" button.

Turns a 1:100 ratio into 1:200, it's nuts

And doubling it again by slooping the biomass->solid biofuel constructor too.

[unborn ermine]

I love when you look at raw numbers and see one thing is WAY better than something else in terms of just clean numbers and output

Looking between Pure Iron and Basic Iron atm

[unborn dome]

That's a really nice ratio

[unborn ermine]

Hmm I wonder which one will give me more trouble

Had a funky amount of limestone kicking around and was debating sending in more just to do something like this.

555/min leftover from doing my distilled silica (from a pure node)
and maybe planning to ship in some other materials for a factory.

As you can see, my planned factory dosent really support an easy source of water

I am already abusing that small pond beside the caterium here for a handful of extractors

[prisma kraken]

for basic iron, i'm not using it anywhere (yet), but in noodling with the numbers, i've found it more useful to ask 'how much can i make with 1200 limestone'

i say that because it will always be the limiting factor and you can use remaining iron ore in some other manner

in all honesty though, i haven't found that recipe very compelling to use anywhere

[unborn ermine]

I can only really see it being used in the rocky desert corner.

basic iron -> solid steel + the rest of the limestone -> HMF

like look at this mess

and thats only 1 impure and normal copper node in this whole area.

well 2 normal, but thats besides

7 nodes surrounding those 4 iron

You will run out of iron ore before limestone.

[brisk shore]

Official Satisfactory Wiki

This is the totally awesome community driven wiki for Satisfactory!! Come on in and check it out!
<3 @unkempt lintel

[unborn ermine]

I think I saw rocket fuel?
Burn rate if you are doing the numbers for it, 25/6 if you want zero rounding errors.

[prisma kraken]

yeah, a lot of that limestone is earmarked for silica 🙂

btw, never dismiss iron alloy... the recipe is stupid-good

[unborn ermine]

Area related tho

no copper

[prisma kraken]

there's an impure node in there!

[unborn ermine]

One thing I was thinking though, if you decided to use this specific quartz for diamonds later as well, these could all work together

but yeah, lots of hypotheticals.

[prisma kraken]

let me put it this way. i built a 90 hmf/min factory in blue crater. its inputs are 2400 coal, 2400 iron and 600 copper (plus gobs of concrete). it outputs 90 hmf/min AND 1740 iron ingots/min

imho, that's pretty nuts

[unborn ermine]

Oh look what I had still open

thats just with mk5 belts too

[prisma kraken]

haha

i gotta say i still really don't like basic iron, nothing about it ever seems to work out to nice numbers for me

[unborn ermine]

Lower number operations seems to be nice, like having a belt of 600 iron ore seems ok.

Thats 2x Mk4 belts of limestone and one normal miner of iron at max capacity.

very clean cut, just has to have its use.

[near hatch]

will this kind of pipe manifold work as intended even if some of the pipes do not have the same length ?

input is exactly 600 and each machine below this needs 50

[unborn ermine]

Shoot to fill the pipes to 100% plus the producer machines starting to fill, then start your testing.
This will make sure that if anything is at fault, its just the piping.

[prisma kraken]

it should even out if the pipes are completely full, but when you push pipes to their capacity, things sometimes just don't work as intended

[pastel obsidian]

Your pipes are beautiful what colour are you using

[near hatch]

#A964A4 with the default finish

[unborn ermine]

Crazy pipes always have issues, always good to test, learn, then make it simpler later.

[near hatch]

i color coded every fluid if you need other hex code x)

[pastel obsidian]

Yes please

[near hatch]

i am not in the endgame tho i might have some missing colors

[prisma kraken]

my gut feeling is that if you are pushing 600/min through those pipes, you'll have problems, but try it and see

[near hatch]

i've put too many hours in this build 🙏

[prisma kraken]

i also am worried that splitting mk1's off of an mk2 will cause issues, but that's easy to fix

[pastel obsidian]

The problem you have would be on the production side it seems. You should have no problem with the machines getting 50 * 12 (600) fluids after you fill up the pipes with fluids

[near hatch]

we'll see, first i need to bring the 2720 water lol

[vapid gorge]

what are you doing?is that just feeding from the top?

youv'e got so many splits and merges and hidden bits I couldn't begin to tell you if it'd work

[prisma kraken]

rule one is keep pipe networks simple

[vapid gorge]

if all that loop is doing is feeding some machines from above? probably fine.

I'm more worried about the pipes to the right of it xD

this bit is way scarier

[near hatch]

yes its from the top to avoid pumping issue

let me show you the full picture to better understand with what im dealing with x)

[vapid gorge]

well the red bit should be fine, but you already brought it up to that height, there shouldn't be any 'pumping issues' if you had them behind the machines

[near hatch]

the manifold remains above every other output pipes so thats why there is something a bit high

[vapid gorge]

I suppose? I can't see any obvious red flags but it's also hard to tell exactly what's going on on the roof.

[near hatch]

only god knows

[near hatch]

basically the rest of the pipes are just copies of the red bit manifold above its just not put together the exact same way

[vapid gorge]

I am worried about these ones though because it looks like the pipes are splitting from the manifold and then going up, which may end up having the same effect as bottom feeding

[near hatch]

thats the case actually

[vapid gorge]

I mean if it works it works? bottom feeding isn't impossible

[near hatch]

if it remains below the master manifold pipe its no problem right ?

[vapid gorge]

not nesceserally

a gap forms in the manifold when a machine sucks fluid out of it but the pipes here move upwards out of it

the left X point doesn't care about the right X point

but bottom feeding can work, just flood the system and go from there

[near hatch]

i will thanks

[earnest rapids]

The best ratio for pure iron I've found is 280 ore + 180 water into 8 refineries makes 520 ingots

[unborn ermine]

I was more or less just comparing what I had on hand that I wanted to blast out eventually.
all in one shot, least mess.

(aka no massively long water pipes and nasty extras to deal with/splitting productions)

Good to know though.

[astral hornet]

I built a factory outputting 270 rubber using HOR, diluted fuel and recycled recipes. It's filling a train station but I'm slight overproducing, I won't need all of the 270 right now. Is my assumption correct that if the recycled rubber machines start backing up, there'll be a cascading effect down the line and the system will get blocked? Should I add a sink right before the train station that puts the overflow rubber into it?

[outer vale]

Yes, though you should also have a buffer crate in front of the station if you don't already

[astral hornet]

yeah, I have that as well

[outer vale]

nice. Then yeah, smart splitter and sink before it should be all you need

[vapid gorge]

if you have a sink for the polymer resin I think you can let it back up and it'll restart as needed, but a sink for overflow is always nice for smooth production

[astral hornet]

oh that's a good point 🤔

I'm not aiming for smooth production though, I've given that up

[weak rose]

Always have a sink at the end if there is a byproduct, if needed use a smart splitter with the overflow filter

[astral hornet]

I'm building scattered factories across the map and have "base stations" where basic items like ingots, rubber... get produced. Those get distributed with trains

[vapid gorge]

depends on the by product and how you're using it

for example in the recycle loop you can have the resin used as the initial feeder stock

[astral hornet]

but I'd rather over produce the basics than constantly having to go back and forth to increase production a little bit

[vapid gorge]

but it's good to have a sink for that to make sure all the oil is getting processed

[weak rose]

Best is to make them completely independent

[vapid gorge]

for Aluminium you probably want to overflow scrap to the sink depending on your waste water method

[astral hornet]

Doing this

[astral hornet]

for now it seems to work, I'm pretty deep into phase 4

[vapid gorge]

yeah depending on how you're doing it just a sink for hte resin. But like I said, sinking over flow rubber is the all purpose solution

[astral hornet]

I'm seeing my train stations basically as huge manifolds, as long as I overproduce, nothing is running out for now

[weak rose]

Yeah, its possible but gets pretty shit later on

[astral hornet]

I'll see how hard it bites me, I've never gotten this far in the game without feeling like I built myself into a corner

[weak rose]

Imo its best to make everything as locally as possible and tranport the end product if needed

[vapid gorge]

they are, really. just gotta make sure you have the througput

[astral hornet]

That's basically what I'm doing

[weak rose]

Ur doing the opposite, ur basicly shipping the raw materials no? (Bars and such)

[astral hornet]

to satellite factories, that produce items that get then shipped to a main base

This is my world right now (the southern rail goes off to the blue crate where my rocket fuel plant is)

I produce steel products in the north, computer products in the north east, aluminum, rubber and plastic and cooling devices on the western coast (and there's an ammo factory there as well)

I have a caterium ingot depot and a quartz depot. Any satellite factory that needs quartz, aluminum or caterium goes to pick them up with a train there

[weak rose]

Oh like that, yeah thats solid. I thought you were shipping iron ingots by train lmao

[astral hornet]

Iron ingots often not, I am shipping alu ingots though

[weak rose]

Though steel products are pretty irrelevant cuz alternates exist

[astral hornet]

yeah, I hardly use any steel 😅

the central base has some mall production and then 3 factories producing elevator parts. But all the "larger" parts e.g. computers etc get shipped in through tractors/trains and put on a sushi belt. They go to dimensional uploaders first, then down the line to the factories and any leftovers are sinked

it's working quite well. I've just set up logistics for drones, so that's now the next part of the equation

all factories are completely modular with blueprints, so I can quickly increase output by plopping some extra modules on top

[weak rose]

I have alot of smaller factories and currently only one big electronics (super computers, radio control units, ai limiters, computers, hsc)

[turbid sapphire]

How far upwards do drone travel before heading out to their destination?

[astral hornet]

depends on terrain I notice

although I only have two flying about right now

[turbid sapphire]

well they have a decent amount of clearance, especially compared to the ground

[astral hornet]

oh absolutely

[astral hornet]

beautiful

[weak rose]

Doesnt matter that much, they can clip through terrain

[astral hornet]

yeah, aesthetics get left behind for now 😅 I've tried to make a couple factories a little more appealing, but most stuff is very much still floating platforms

[turbid sapphire]

Ye im just kind of trying to design around them

[astral hornet]

I usually do design while building up elevator parts

[turbid sapphire]

Im thinking one more layer of the middle tube should probably be enough vertical space

[astral hornet]

dang

[frosty owl]

Ohh, I just didn't include Ficsonium in my plans at all ^^
I (without much thought) judged it would be unnecessary with so much power aviable... But to be fair, most of that power would be coming from Plutonium in my plan 😅

Thanks for crunching and sharing the numbers and details though, it always makes for an interesting read 😁

[patent blaze]

damn i thought there was a recipe for resin -> coke

[wind spade]

why so?

[patent blaze]

thought i heard kibitz say it in a video and it stuck in my mind

would be kinda useful ngl

[rapid trail]

Hey, I just published Assistory: https://github.com/MaxWinklh2908/Assistory
It is a Python tool for Satisfactory where you can calculate optimal production plans and get hints about what problems are there in your factory. It still is an early state but I'd be happy to get your feedback.

[outer vale]

the obvious first two questions

• "optimal" in what respect?
• how does this compare to existing tools, what does it do better at?

I'm curious about the "get hints about what problems there are" bit, how does that work?

[vapid gorge]

yeah, not gonna install a random thing when SFtools can create optimal plans

[prisma kraken]

this looks somewhat useful

thanks for taking the time, and it looks like you've got the kernel of something useful and good cooking!

[rapid trail]

There are different objectives while sticking to certain constraints, e.g. Maximize the output rate of multiple items while enforcing a ratio between the output rates given some available resources. For further details see docs/optimal_production.md#Use cases.
The tool is a library/command line program in Python with so required integration into the game as a classical mod would be. This is less user friendly for normal users, but it is very easy to extend and adjust if you are familiar. Also for the game state monitor, I'm not aware of any other solution available so far.

[north mauve]

Fill those holes and you can rename it Blue <opposite of Crater>

[deft lichen]

if you'd like, it can be added to this list here https://satisfactory.wiki.gg/wiki/Online_tools (although probably under "other" given it's terminal-only)

[rapid trail]

Would be great...Do I have to do that myself?

[deft lichen]

I can do it for you, just tell me a brief description akin to what's already on the page and what category you think it fits best

[rapid trail]

Perfekt I will send it to you soon

[past reef]

rapid production is interesting I'll try it out

Was thinking of optimizing build location given a production chain, but it's probably not possible to do it in a general tool given that I need to quantify the effort of making a train line vs drone port vs belt

[plain fossil]

2nd try, im getting pretty good at this

[deft lichen]

you could compute the "optimal" location by weighing resource node locations against resources used in the production line and producing a heatmap, without having to factor transport modes

I think I already saw something like that once

[bronze barn]

There was definitely something along those lines around in U7 or there abouts

I recall a heatmap of Caterium CBs being shown obviously heavuly preferring the swamp southeast location, but it appears to be kinda gone

Ah here we go; #math-and-meta message

[deft lichen]

ah, excellent 😄

[bronze barn]

For obvious reasons this is outdated data, but it's for sure doable perhaps they still have the original

[deft lichen]

the script logic is unchanged, it would only need 1.0 map data

[bronze barn]

Yea - And this location in particular is unchanged so definitely still good for this usecase, just others have changed for better (or in the case of the goldcoast, worse)

[chrome dragon]

Not sure whre this is mostly relevant: Can somebody explain to me how the signals should look like for this layout?

I have: Path on every entrance, block on every exit. The loop in the middle confuses me though. I have a path when going into the loop, but nothing exiting, should there be something there as well? Been banging against it for 3 hours now and can't get it to work

[deft lichen]

you have right-hand traffic?

ah, yep

don't place path signals in the middle of the junction

also, your tracks are very close to each other, which may cause unexplainable signal errors

I strongly recommend 1 gap foundation between tracks

[rigid hemlock]

whats happening to my power grid? i get heartattack sometimes when this happens. im already nuclear

[deft lichen]

particle accelerators?

ignore the max. consumption line, it's bugged

[rigid hemlock]

oh that explains

[deft lichen]

all you need is this @chrome dragon

if the signal errors won't go away, it's due to the tracks being too close

[chrome dragon]

I fear you are correct. I have converted it to a roundabout instead. No signal error for now ^^

[deft lichen]

ROUNDABOUT?? 😭

if there's 1000 roundabout haters I'm one of them, if there's 0 roundabout haters I'm dead

[versed violet]

wont this loop thing result in a long train ramming itself when turning around?

[olive jacinth]

Depends. I have a couple of turnaround loops depending on the length of the train. Most of my trains have six or less carts though, so my tightest loop is generally okay.

(not the OC you responded to, just have a similar system)

[analog bluff]

it will clip through itself but it will not derail or anything

[wispy latch]

anyone tried leached ingot? since sulfur dosent have much use appart from power now.

[prisma kraken]

the copper and iron recipes for it aren't very good unless you are doing some converter jank. leached cat is interesting for its higher cat yield, but the numbers for it are pretty icky

22.2222 refineries to process 1200 cat with 666.66666 sulfur

[plucky tusk]

first time making the residual fuel power farm

[prisma kraken]

what is interesting to note though is that using 600 sulfur with leached to process 1080 ore and using the remaining 120 ore with pure cat gives you 780 ingots

[wispy latch]

I made a simulation with the calculator for phases 4 deliveries and it needed 0 sulfur, though maybe...

[prisma kraken]

that's honestly about the best use case i've found, it seems like maybe a situational phase 4 thing if your speedrunning

[unborn ermine]

Im phase 4 fast walking

[prisma kraken]

yeah, sulfur isn't as needed anymore (somehow - nothing really stopped using it)

[wispy latch]

only use I see is power now

alt recipes dosent seems worth it

[unborn ermine]

Yeah its mostly for Rocket Fuel, and even then you can STILL save it if you dont mind wasting more oil.

[prisma kraken]

being that i'm using 600 of it for 120 gw of power, i think i'm good

[wispy latch]

exactly, thats a lor of sulfur left

[unborn ermine]

yeah I dedicated a 780/min node towards 2600/min RF production

[prisma kraken]

also the map limits increased from 6700 to around 10k and like nuclear if you try barely will use 3000/min

in phase 5 turbodiamonds become a thing

[fierce ruin]

how u gunna load em?

[wispy latch]

right forgot that one

[prisma kraken]

but yeah, plenty of sulfur around still. even if you are making max nuclear, you'll still have 1000 or two free sulfur after turbodiamonds and possibly making some batteries

some of that you might want to reserve for doing some conversions to make logistics easier

[plain fossil]

Freights have just enough room for an elevator

placed next to each other

[prisma kraken]

you can fit an entire isc in that gap sideways

[fierce ruin]

ah okay hard to tell in the pic. dont forget your ISC buffers

[prisma kraken]

i've been taking to placing ISC's on an adjoining floor instead of next to platforms, but they fit fine next to platforms

[plain fossil]

yup

[north mauve]

How does radiation range work? I'm 109m from a drone port full of waste and taking damage. As far as I know, that's the closest source of radiation.

[fierce ruin]

I think there is a bug where sorting your inventory if you had waste in it, you will keep taking damage. Could that be the issue? 109m seems pretty insane
You can check the range using SCIM if you upload your save IIRC

[north mauve]

It's definitely a radius and not my pocket. I am hoverpacking around my (very radioactive) factory and I seem to have placed my BP designer so that the right side is fine, but the left half triggers the smallest amount of radiation damage.

[fierce ruin]

Ah I found this page on wiki that explains the math
https://satisfactory.wiki.gg/wiki/Radiation

[north mauve]

Oh no, the distance increases the more stuff there is

I didn't expect that

[unborn ermine]

Yeah, stack size and stack amounts.

[fierce ruin]

Most people build their nuke plants on edges of the map where they dont really need to be around

[north mauve]

Unless they quickly set up the factory to run to empty the ~48k U waste as fast as possible, before finishing the aesthetics or logistics cleanup @_@

[fierce ruin]

hahah have fun with it.

[north mauve]

Okay, NOW I have the bug where radiation doesn't go away. Earlier it was just a tiny little notch of radiation, and now it's full even though I'm far away. Great.

Copy and pasting settings between machines SHOULD NOT EMPTY THE CONTENTS INTO MY POCKETS

[wind spade]

Radiation is global and not a radius.

[queen pagoda]

Trying to use satisfactory tools calculator to set up a production plan for all basic items up to stators motors etc. Is it best to set it up using all items set to maximise?

[wind spade]

Just there's a minimal required amount for you to take damage

[wind spade]

Best is to not use maximise

[queen pagoda]

Ok i thought so. But how would i know of how much item to produce for each?

And should i be utilising all alternate recipes where possible?

[wind spade]

Small amount for storage is enough, usually 5-15/min

[queen pagoda]

when i set all to maximise it sets everything to make 25 of each to storage

[fierce ruin]

it really depends on your end goal and how you wanna play.
But I do agree with greeny, for most items 5-15 is good for storage purposes

[fierce ruin]

i believe it equalizes if you have multiple set to max

[fierce ruin]

this is a preference, some alts are better some are worse, some are the same but use different resources

[north mauve]

Are we talking real life or in game? I'd be surprised if the game engine is constantly calculating player damaged against every single source of radiation every tick

[fierce ruin]

sorry just noticed i tagged you 3 times coulda done it once lol

[queen pagoda]

the calculator would pick the best alts or no?

[fierce ruin]

greeny can explain that one, im not sure how he set it up (he made it)
but i would say no considering there is no "best"
some are better in different ways
such as # of buildings, resource usage, time, etc

[wind spade]

Ingame

[unborn ermine]

"Best"
Weighted on the total resource rarity/values, basically, in an isolated area there is no best, just what you want to move.

[wind spade]

Will pick cheapest considering weighted resources, but only if items/min mode is used (no maximised item)

[prisma kraken]

keep in mind that the rating that sftools assigns to resources is pretty good, but doesn't take into account proximity and other factors that might shift a decision in some direction or another

[queen pagoda]

I set it up to use all the nodes i have in the current area

[prisma kraken]

(it has to pick something, that choice won't always be the right one for YOU)

[queen pagoda]

so 25 per items is a good number?

[fierce ruin]

we cant really say, it depends on your specific goal
for me if you said is 25 concrete a minute good, id say no

[queen pagoda]

for these items

[prisma kraken]

@wind spade when you hand enter resource amts, does that influence the rating decision?

[queen pagoda]

yep i would change concrete to higher

[fierce ruin]

if i say yes would that satisfy you lol

[wind spade]

Don't use maximise 🙂

[wind spade]

No 😦

[queen pagoda]

can you help me greeny to set it up lol

[prisma kraken]

😦

[fierce ruin]

he is telling you to turn off maximise and to use items/minute

if you want 25 of each of those, enter 25 for each

[queen pagoda]

well maximise is making it it will be 25 of each

[prisma kraken]

yeah, when you set everything to maximize, it tries to make equal amounts of everything, which is almost always not what you want

[fierce ruin]

^^^^^^^

as mentioned, we dont know your goal so we cant tell you how much to make
for example, are u gunna just store this stuff?
are you gunna move to another factory?
are you gunna only move some of it?
are yoou gunna sink it all?

[queen pagoda]

its just a starter factory

basically just a starter factory and sink all excess

[fierce ruin]

okay and it also depends on where u are

for me when i use the calculator i like to check nearby nodes and plan it out, limit the calc to those values and decide what i want of each item

[prisma kraken]

for intro iron and copper, just figure out what your budget is and see how many rips/rotors/frames you can make reasonably with that budget. with copper do the same with wire, cable & sheet

[fierce ruin]

if there is an item i want alot of, ill use maximize on that

[queen pagoda]

i am limiting it to 6 pure iron 1 pure copper and 1 pure limestone 3 normal limestone

[prisma kraken]

for motors/stators/and rotors again do it as a separate thing

[queen pagoda]

all of this is in the 1 area

[fierce ruin]

here is an example of something i did

actually here this might be more relevant to you.
this isnt really a "starter" factory, but just gives you an idea of how you can use it

the reason i maxed modular frames is i want anything extra to be used for them because i know ill need alot of heavy modular frames

but lets say u knew how many heavy modular frames u wanted, you could just use items/minute for the modular frames and know you arent building more than you need to

[wind spade]

But doesn't optimise it

[prisma kraken]

well, that's a north forest grocery list if i've ever heard one. might as well count the coal nodes next door as well

[wind spade]

As long as there's one maximised item, it doesn't do any optimisation

[pallid knoll]

Should I put my storage in the middle of the factory or is it better to put it on the side of the factory? Or maybe build an extra level above production? 3th photo is the storage room i want to build

[queen pagoda]

yes north forest 😄

[fierce ruin]

this is really just a preference thing

i personally think about where my train station will go in the future and put the storages near there

[pallid knoll]

How big is the trainstation?

[fierce ruin]

idr off top of my head but they are beefy

lemme load my game and ill send screenshot example

[pallid knoll]

💀

[fierce ruin]

something im working on atm

just the stations and containers takes about 15x14 foundations

[pallid knoll]

Nice, I want to build a extra level but I don't have lights which makes it dark haha

[fierce ruin]

you can always use ur pioneers light 😛

B is the hotkey if you didnt know

[pallid knoll]

Haha i will wait until I have factory lights 😂

[fierce ruin]

yeah as u get further you kinda get a feel for your style.
I have a mixed style for my buildings. some factorys are just a single plane and some are towers.

[pallid knoll]

Yeah I just try to get it not messy as I build my production

watched 25+ hours of vids and still a lot to learn

[tough fiber]

Do you guys like aluminum rod?
There's no info on the wiki

[wind spade]

What info?

[tough fiber]

weighted resource points, space calculation etc

[bronze barn]

Those are temporarily removed for a new version

[wind spade]

Permanently

[wind spade]

Because that info was misleading

[prisma kraken]

yeah, the analysis stuff was pretty bad because it was so limited in the permutations listed

you'd either have to list everything from product to ore for every combo or to just go to ingredient level and have apples & oranges

[magic island]

these all depend on how you make the aluminum

and comparisons to the other rod recipes depend on how you make the steel or iron

the permutations do not fit in a convenient table

[tough fiber]

of course. but I still find it helpful

Otherwise it's just guessing for me

[bronze barn]

Not according to the comment I found saying temporarily

[magic island]

the tables on the old wiki were often deceptive, and therefore not as helpful as you might think

[prisma kraken]

best example of that was the mod frame recipe analysis

[magic island]

the assumptions those tables made about precursor recipes meant that certain stats could be wildly inflated or understated, and you would never know it if you didn't do your own analysis

[bronze barn]

Oh and Ondar didn't deny one being planned when I asked, just said they were not having the time right now

[prisma kraken]

it totally ignored adhered plate as a means of making rips, and as such skewed its conclusions to show that steeled frame was much better than it was

[wind spade]

There's plan to make something else, but it won't be the tables that were there pre-1.0

[bronze barn]

Well I never claimed for them to return as tables. I said "for a new version"

[wind spade]

Basically there won't be any analysis

[bronze barn]

Well something is planned, and this comment implies it replaces the analysis, but I ain't gonna lie I'm not involved so I could be completely wrong - https://satisfactory.wiki.gg/wiki/Aluminum_Ingot#cs-comment-6780

[magic island]

to illustrate how weird evaluating WP can get: any resource on the map can be involved in the iron rod recipe chain

(even technically uranium, if you use the byproduct water from plut processing)

[prisma kraken]

what would be nice instead is some suggested builds for things and have that kind of be a recipe book

[brisk epoch]

I’ve become convinced an unpowered pump does not reset head lift if that “head lift” was created by gravity

[bronze barn]

These things do exists for a lot of stuff on the wiki actually

[prisma kraken]

kinda sorta, yeah

[wind spade]

I'm part of the team that works on it 🙂

[prisma kraken]

some of those builds date back to the days of update 3

pretty sure you can still find a few pictures somewhere of the old refinery model

[bronze barn]

Well a lot of cleanup was done so they should still be relevant and correct - And if not you should ping someone involved with the wiki or create an account to create help out 😉

But stuff like; https://satisfactory.wiki.gg/wiki/Alumina_Solution#Gallery exists in a few places

[prisma kraken]

yeah, its community driven, its just that it takes time

i'm by no means criticizing, just saying that some stuff isn't very current

[bronze barn]

I figured as much 😉 Hence why I added I'm not involved and don't know specifically what is planned. I must say I wish there will be some analysis done. I really like the tables pre 1.0 to help me make a decision

[wind spade]

The problem is that 99% of the info was wrong-ish

[bronze barn]

What stops you from creating an account and updating stuff 😉

[magic island]

it's not that anyone objects to there being basic useful information

but the tables very much lied. a conveniently-packaged lie is still not good

[prisma kraken]

no one has invited me to do so or encouraged me to do so, i stay out of crowded kitchens.

[bronze barn]

Consider yourself encouraged 😉

[prisma kraken]

🙂 also, at one point there were a few personalities around that i didn't want to interact with

no sense in getting into that all, but it does make you feel less inclined to help when someone poo poos everything you say 🙂

[bronze barn]

Admittedly I really like this conceptually; https://satisfactory.wiki.gg/images/9/93/Plutonium_Fuel_Rod_stackable_layout.jpg

I'm just not sure if I will like it in game as a tower, so I'd personally need to do some work to make it more horizontally scaled, but that could likely be done with relative ease

And now that the designer mk3 exists, this is actually a possibility to blueprint

[prisma kraken]

has shaving foam been updating his stuff for 1.0?

some of those designs are pretty nifty

[bronze barn]

This is the same as pre 1.0 at least