math-and-meta

[wind spade]

but will be today soon

just need to fix a few things and I'm ready to go

[bleak coral]

Are you gonna just stay current with experimental, or will there be an early access version too? cause it's gonna be like a month before it hits early access

[wind spade]

there's gonna be separate url for U4

[west gale]

oh man, I'm gonna have to update my "easy aluminum" factory module

[upbeat tide]

Im waiting for some info before I go full U4

Need to know how to fix uo my nuke rod plant first

It uses all the alts for the rods and encased cells, soo should be interesting what actually needs done

[naive ingot]

Oh wow, control rods are getting used in several alts, I like this...

[upbeat tide]

Oo nice indeed

[strange jasper]

is there a list of all the alts and new recipes? or is it still being figured out as people are going along?

[wind spade]

I'm gonna have mine soon (10-20 mins estimate)

[naive ingot]

I'm looking on SCIM.

[strange jasper]

sweet. i don't even own the game and it takes up way too much of my life just calculating factories and other nonsense.

[naive ingot]

I don't know if it's a complete list though.

[bleak coral]

Just looking at the aluminum stuff. You still need to do a water loop back, they didn't get rid of waste, and it doesn't use any new machines

but damn they're faster, and the water input/output is all in multiples of 30 so it should be much easier to do nice ratios without decimals points

alumina solution needs 180m^3/min water, and scrap recipes makes either 120m^3/min water or 60m^3/min water

they also seem more efficient

[upbeat tide]

Also requires coke or coal, and produces silica as a byproduct

I am happy they did not remove the loop-back for water too

[wicked tinsel]

do you know if old diluted fuel is still working maybe? or was it removed completely?

[torpid robin]

We were told it wasn’t going to be removed

I havnt jumped on and looked yet I’m at work . But it’s what mark said in one of the streams that it will be retained

[wicked tinsel]

ye, but, clearly that recipe is now in the blender so i wonder if old remains or its gone

[torpid robin]

Probs be like the old packaging recipes . They would still work for old stuff . But anything new you won’t be able to put it in

[wind spade]

enjoy guys 🙂 https://u4.satisfactorytools.com/
https://www.reddit.com/r/SatisfactoryGame/comments/m6ito4/update_4_calculator_is_ready/

[shy mason]

Good thing I've shifted my stator factory to have ai limiters built on site, don't see as heavy use for stators now that they've changed the turbo motor alt. Still alot of caterium use though

[upbeat tide]

Confirm on diluted fuel still working in refineries?

I maybe diluting almost 2100m3 of oil...

[upbeat tide]

Actually, alot more than that

[wind spade]

I assume recipes set in buildings will stay there and work until moved tho

[upbeat tide]

Will it still work though

[wind spade]

but idk, you'd have to test

[upbeat tide]

Fun 🙂

[wind spade]

also rip that nice 1:1:1 diluted fuel setup

[upbeat tide]

3300m3 oil total diluted tho

[wind spade]

new turbofuel alt

[upbeat tide]

Never used the alt for TF

[sullen cloud]

what has changed regarding the diluted fuel setup?

[upbeat tide]

Never liked it

[wind spade]

it's a different one

[upbeat tide]

No canisters it seems

[wind spade]

no more loops, just one building making HOR -> fuel #math-and-meta message

[shy mason]

Drats, there goes my power and cpu factory

[sullen cloud]

yeah, for the new alt that's true. but the existing loop still works tho and hasn't changed?

[sly dawn]

Hi. I'm on my second play of Satisfactory. I want to do it properly this time and not hack through the Tiers.. Are there any good guides on the generic approach to designing and connecting your factories/resources? There's lots of specific guides on YoutTube (e.g. "How to make a bus", "tips and tricks") but none around overall strategy. I'm gravitating towards using a bus, but then puzzled about how I have a centralised storage rather than localised ones whilst having a uni-directional bus. So yeah, Wanted: a meta guide overaching design principles.....

[wind spade]

that depends, but I'd recommend not doing bus

I usually suggest this approach:

• figure out what item you need next
• figure out how much do you need of that item (e.g. using some of those online tools in pinned post)
• find good set of nodes that are close to each other and will provide you with enough resources
• build factory somewhere near those nodes
• ship the product to storage
• repeat

[upbeat tide]

Im actually not thrilled with simplifying diluted fuel

IMO

[wind spade]

I liked that 1:1:1 setup. Didn't get to play the game with it tho 😄

[sullen cloud]

has diluted fuel ever been really thrilling ;)?

it's more like grind

[shy mason]

Pushes it to tier 7 though with needing aluminum casing to build blender

[sly dawn]

thanks @wind spade , but that leads to so many more questions! Hence the question about any good meta-design reading/watching anyone knows about? I want to have my full approach planned out before embarking.

[sullen cloud]

there are quite a bunch of meta approaches out there. you could search in reddit, e.g. for 'cazanator', who wrote about two different concepts (mall and decentralized small factories), @sly dawn

[shy mason]

Spend time to get a nice convenient spot outside the factory so that you can just feed off it to newer additions. Especially with how much room truck or train stations take to set up.

[wind spade]

also if you don't want to force factories to be that separate, you can always build just a single factory. However with bus you'd have hard time making everything work correctly.

Some general good tips in my old reddit post, most of them still apply: https://www.reddit.com/r/SatisfactoryGame/comments/dn15y0/useful_tips/

[shy mason]

Blender diluted fuel will save my fps in that region at least

[sly dawn]

thanks guys - I'll do some reading

and dammit - a full day of designing an awesome bus out of the window

[wind spade]

however in the end it's best to stick to what works for you

[sly dawn]

why do all those youtube fellas reccomend them??

[wind spade]

most of them come from factorio, where bus is much more useful 🤷‍♂️

same as with balancers, they come from factorio, so they automatically apply what they learned there even though it may not be the best

don't get me wrong, anything can work and since it's a sandbox game, nothing prevents you from trying it.

it's just my opinion (tho I think a lot of folks here share it) that while a bus can work, it's pretty hard to get it right, will eat a lot of FPS (due to a lot of extra belts) and there are easier approaches to building a factory

[sly dawn]

yeah - but don't want to get 100 hours in and have to tear it all down for a more efficient system

[wind spade]

as long as you don't bottleneck items anywhere, any system can work and be 100% efficient, including a bus 🤷‍♂️

[sly dawn]

@sullen cloud - is this the artcile you were referring to, Marc? https://www.reddit.com/r/SatisfactoryGame/comments/bhq8d8/microfactories_an_alternative_way_to_build/

well yeah - on the fps thing - my main buss is enclosed - so shouldn't slow it down too much?

[wind spade]

belt FPS issues are mostly CPU related iirc, so enclosing usually doesn't help much

[sly dawn]

oh - i thought the drop would be due to drawing all those moving belts

[wind spade]

it depends from player to player 🤷‍♂️

but the game is mostly single-core, so if you have a lot of belts, it may get tricky to calculate all those positions

[sly dawn]

ok - but very confusing

[bleak coral]

@upbeat tide I haven't checked if the old diluted fuel recipe is still available, I remember them saying they might leave them both in, but since my power is still stable I'm assuming the old one still works at least

[frosty owl]

On that point, I'd say: a succesful bus is one that has the same or less belts then you'd need otherwise (without bringing them together, so when you can't conveniently mix them to save on belts)

[strange jasper]

nice! the calculator is already updated.
just a question though; is anyone else having trouble calculating plutonium fuel rods?

[wind spade]

try adding nuclear waste input, as you can't "produce" waste in any normal way

[strange jasper]

oh. i thought it was "produced" by power plants, but if it's a limitation of the calculator, i can understand.

[wind spade]

indeed, but that's probably harder to plan than the "normal" way

[frosty owl]

Probably so, unless you design your base revolving around it from the beginning

[wind spade]

it doesn't (yet) support power plants and power production, however it's something that's being actively worked on and will hopefully be released in following few weeks

[copper imp]

where can I find the new recipies outside of the game

[wind spade]

e.g. here https://u4.satisfactorytools.com/codex/items

[frosty owl]

Even then, it would only start being convenient over a certain scale, naturally

[strange jasper]

you work fast.

[copper imp]

Thanks a lot

[wind spade]

you can also check Recipes section here, they are all there 🙂 https://u4.satisfactorytools.com/production

[frosty owl]

He's an incognito member of CS studios

[wind spade]

I wish 🤔 my CV is somewhere in their inbox, so fingers crossed 😛

[sullen cloud]

yes

[iron prairie]

So, continuing from #satisfactory-experimental : I still think turbo blend fuel isn't as good if you have access to diluted fuel, though your point about sulfur is a definite maybe.

For this, I treat sulfur as having an energy of 261 MJ: after subtracting for water extraction & compacted coal assembler losses, regular coal has an energy content of 270 MJ, versus compacted coal at 531 MJ.

To make my life simpler, I normalized everything to "how much energy to create 30 turbofuel" or "how much energy, after losses, do the components of 30 turbofuel/30 turbo blend fuel, contain?"

30 turbofuel costs 6534 MJ to create, and the components can burn for 36954 MJ. That is a gain of 16512 MJ per turbofuel, versus components.

30 turbo blend fuel costs 5125 MJ to create, with the components burning for 44265 MJ (assuming you sent everything through the HOR -> diluted fuel path): a gain of 10610 MJ per turbofuel.

[icy lance]

1 + 1 = 2 (I did math)

[iron prairie]

Now, if what you care about is energy gain per sulfur spent, turbo blend fuel is very marginally better, at 707 MJ gained per sulfur spent on the process, versus turbofuel at 688 MJ gained per sulfur spent.

Given the much reduced crude oil cost of regular turbofuel, I'm not sure that marginal improvement in sulfur economy is worth it. "Much more turbofuel per sulfur" stops being as worthwhile when you instead look at "just a tiny fraction more energy gained by converting stuff into turbofuel per sulfur".

[wind spade]

I look at it from weighted resources point of view 🤷‍♂️ power was always secondary to map limits for me (as there's way more power than resources, or at least was in U3)

[iron prairie]

Granted. I don't think your tool is necessarily bad for caring more about output resources than power, but this is an edge case where I do think "you get more output turbofuel per weighted input resource" isn't telling the full story.

[wind spade]

sure 🙂 power optimisations are planned and almost finished (tho it will take a while till they get released)

[iron prairie]

The end product of this all isn't the turbofuel, it's the energy, and that's great news. I do really like your tool.

[oblique hollow]

12 m³/min fuel consumption is a nice change

[wind spade]

that's the plan in next big tool update, you'll be able to select "power" as one possible thing in the calculator. Oh, and also "sink points"

[frosty owl]

I've been long anticipating that one last statement <3

[wind spade]

both will come with next big update, that's been in the works for last 4+ months

the update isn't that big from user's perspective, most things are just rewritten

but should work a bit faster and unlock more possibilities

[torpid robin]

Needa crack the whip on greeny

[wind spade]

you can check out my paypal/patreon for much better effect than a whip

jk, it just takes time, sorry 😦

[strange jasper]

you don't know how long i've just been sat here thinking to myself "is it rude to assume they haven't thought of adding sink values to the calculator?" i work up the courage to come in hare and ask, and i'm half an hour late to the party.

[wind spade]

tbh, I'm always fine with people asking for features. Sometimes they are indeed useless features or impossible ones, but I always try to explain first why that won't be a thing. I'm just one guy tho and my time is pretty limited, so adding all those features is pretty hard for me. But the more people ask for something , the better, at least I know what people want and can change my priorities based on that 🙂

however now the #1 priority is finishing this big engine update and after that I'll start adding new features

[strange jasper]

now i'm getting esoteric with my thought process of what is and isn't possible to calculate.

[dark thicket]

oh hey i was just using your tool. i stumbled upon it on my last playthrough and its truly a godsend. i did the same thing by hand on a white board before this

[wind spade]

there's A LOT of things that's possible to calculate. The algorithm is pretty cool and handles everything nicely. There are a few things that it probably can't calculate, but you can always ask 🙂

[median thunder]

off topic but what's the most sink points you can get from a single, say iron, resource node

[wind spade]

hard to tell, without sink points calculator 🤔

[strange jasper]

ok, esoteric calculation question #1:
can you disallow certain recipes/alternates based on the sink values of component objects? to check for the most efficient production of an object instead of the maximum?

[wind spade]

well when sink points will come, they will come with an option to maximise for sink points

[strange jasper]

hmm. not quite what i was trying to explain, but it was a very vague and wordy question.

[wind spade]

how do you define "most efficient" production of sink points?

(or whatever object we talk about)

[strange jasper]

well let me give an example of the problem i am seeing.

i tried calculating the maximum of aluminium casing and it gave me 2 separate recipes for aluminium ingots. i'm not sure which is mathematically superior, but my mind is telling me that logically only one recipe should be the most efficient.

so there must be something going on in between.

unless of course i hit the maximum of some other resource other than aluminium... in which case it would be using the second option to calculate.

opp. yup. looks like quartz is at the maximum for that particular calculation. sorry for doubting you.

[wind spade]

that's most likely due to map limits. Simpler example would be e.g. wire: you can make it from iron or copper, once you run out of one, you can still use the other to make more wire

nvm, you figured it out 😄

I just read the first message and answered 😄

[strange jasper]

no prob. we both figured it out on our own in real time.

ok, if just saying the problem out loud to other people solved it, maybe it will work a second time.

[wind spade]

xD

[strange jasper]

ok, thinking out loud here. so if a factory was supposedly "100% efficient", by whatever definition that means, every resource would be used to produce the most things from it...

esoteric question #2:
can you use the calculator backwards? determining the most efficient use of a resource?

[wind spade]

well, pretty hard to define "efficient use of a resource" in a context of single resource 🙂

[iron prairie]

"Most things" would require further definition: otherwise, you just end up with octabajillion screws, wire, quickwire, silica, concrete, petrocoke, etc.

[strange jasper]

well we have a few contexts. if we use sink points or energy value or something as a base for comparison, we can determine...
that's just what you already have planed, isn't it?

[wind spade]

not all items have energy value, so we can't use that

[strange jasper]

but i mean the comparison is already there. you are already planning on calculating sink values for things, so just use only the recipes that produce the least amount of loss of coupons for a particular item.

did... someone reply with a coupon instantly?

is... is this normal?

[wind spade]

for sink points, if we ignore those few items that don't have sink point value, we can define if a recipe is good sink point-related or not. Base recipes are always [sum of sink points of ingredients] * 2 = [sum of sink points of products], alternate recipes are different, some increase it more than two times, some don't even increase it one time 🤷‍♂️

[wind spade]

it's a bot, ignore it 😉

[wind spade]

but I'm not sure we can define a path for a single item that's best for sink points

[bleak coral]

Before update 4 I believe max points per second was theoretically solved, if not mathematically solved

[wind spade]

since you most likely need to add extra resources and you may end up with byproducts

[bleak coral]

you just start from most points and work you way down making the most things, and make sure there's enough power

[wind spade]

yeah, they can be for U4 as well when I add sink points to my tool, but I don't think we're talking about that

[oblique hollow]

underclocking the accelerator is.... uh.... interesting

[iron prairie]

It definitely wouldn't be impossible to max sink points given resource limits (which would likely be 90% "how many turbomotors/min can you make", now replaced by "how many thermal propulsion rockets/min can you make"), but backwards...

[fallow lily]

Greedy algorithms have their limits.

[iron prairie]

It probably wouldn't be that bad given that each processing step is multiplicative: "find the most expensive item you can make, max that, find the next most expensive item you can make with the remainder, max that" probably wouldn't be very far from optimal.

[wind spade]

with my tool, I'll be able to solve it in a matter of seconds. Including power, including everything. Just need to finish it 😄

[bleak coral]

the trick is minimizing power vs making better stuff, though before U4 nuclear fuel rods were some of the better items to sink so it kinda worked out

[strange jasper]

ok, we have determined a way to find the most "efficient" recipes that could produce coup-
...uh, sink points. if we use that to compare recipes in order needed to make another item we can replace the "maximize quantity" with "maximize quantity efficiently". so you don't end up with recipes that overall would negatively impact all component items.

or the rest of your factory.

[wind spade]

the problem is that a recipe may be good, but the path that contains that recipe may be bad

[iron prairie]

... Let's say you have an alt which takes in 3 sink points of input, and produces something worth 8 sink points, versus its default going 4:8.

Problem is, the raw resources behind that alt's input could be sent down more productive pathways, as greeny mentioned.

[bleak coral]

good example of that is the heavy modular frame recipe that uses rubber, it undoes its advantages (less power and space by being faster), because the stuff needed to make that take up more power and space

[strange jasper]

then those are the two things we are thinking of:
the most efficient use of a resource
the most efficient way to maximize the item

[wind spade]

yeah, SF production tree is super complicated and you should never look at only part of it, but on the whole tree instead. That's how the tool calculates all the data as well. It optimises the given tree, it doesn't try to find best path through the tree.

[strange jasper]

they seem interlinked, so there must be some connection we are missing.

[iron prairie]

Another example of counterintuitive behavior is with steel screws.

Technically, you consume less coal via steel rods -> normal screws. However, that requires a huge amount of power due to the vastly reduced output/building, and the coal "saved" by going steel rods -> screws contains less energy than the energy saved by just using steel screws.

[wind spade]

for the first one, if we define resource weights, it gets trivial.

for the second one, it's not as easy. If it's only "how can I best use X without using anything else", then we can define weights on products, e.g. sink point values, but when you start thinking about "how can I best use X while having possibility to also add Y and Z", then it gets super complicated

real example:
"how to get most sink points from 10 iron plates" is simple
"how to get most sink points from 10 iron plates and 10 copper plates" is simple
but
"what's the best way of using iron plates (with possibility of adding other resources as well)" is pretty much impossible

and for those "simple" cases, it just turns to "pick all items that can be made out of these resources and maximise each one of them, then pick the best outcome"

[bleak coral]

btw @upbeat tide from earlier: diluted packaged fuel isn't gone there's just now an extra recipe that's called diluted fuel that's used in the blender

[strange jasper]

what if you reversed the inputs and outputs?

[wind spade]

which is pretty much the same as "maximise sink points with 10 iron plates input", which turns it into the reversed query

[bleak coral]

reverse the polarity!

[strange jasper]

hmm.

basically it would be turning sink points into iron ore in the most efficient way possible. then that would logically be the best use of iron ore, regardless of other inputs.

[wind spade]

"how can I best use 10 iron plates to produce as much sink points as possible"
is the same as
"how can I produce as many sink points as possible while using max 10 iron plates"

even though they seem to be each one from different "side"

[strange jasper]

at this point i think i've forgotten what i was actually trying to accomplish here.

[wind spade]

I may have drifted a bit as well, but for me it seems that your "other way" is the same as the "normal way"

(I may be wrong tho)

[hazy fossil]

For this max points thing (ive actually been thinking about it for a while)
first thing to figure out is the point cost of 1 MWH, presumably generated from the most uranium/plutonium efficient nuclear power w/ reprocessing

of course this is kinda hard since hadron colliders use random amounts of power im rlly angry about that decision tbh

[wind spade]

actually there's another way, involves complex math (the same as I'm using in my tools) and you can just set the problem as an optimisation matrix and it'll just give you the result

you'd just have to define if you want to calculate with collider's average or max power consumption

[hazy fossil]

probs just average /w some added energy storage

need power usage to be exact since wasting power = wasting points

[wind spade]

well since the calculation would be hypothetical anyway, we can assume user has built 10000000s of power storages and therefore power fluctuations doesn't matter to him

and count only with average power consumption indeed 🙂

[hazy fossil]

with all the power storages, will it be possible to use geothermal power with 100% efficiency?

will it be used as priority over nuclear?

[wind spade]

well there's no priority now

since all overproduced power just dissapears

[hazy fossil]

I suppose it's not a problem if the average geothermal power is calculated & it's added with a set amount of nuclear power for the total constant (average) demand

[wind spade]

well demand is now not constant either with those colliders 😉

[oblique hollow]

accelerators*

underclocking them is pretty powerful, but it becomes expensive to build mutliple

[wind spade]

called colliders in game files 🤷‍♂️

[hazy fossil]

yeah, so the average use will just have to be calculated (along with every other machine that wont run constantly even when underclocked) and then added to the total constand energy demand

thing is all unneccesary energy use cannot be allowed since that would be using points
that means no trucks, no trains, no drones, no hypertubes, no lights, no pumps, no hoverpacks, no jump pads and probably some other things im forgetting

only belts and pumpless pipes can be used to move stuff since they dont need energy

another thing, is it more energy efficient to extract water from resource nodes or with water extractors?

only miners can be overclocked, since maximum energy conservation is required

[wind spade]

you can also underclock everything to 1% and build 100 times that amount of buildings

or actually now it may be possible to underclock to 0.01% (idk how many decimal places it has)

[hazy fossil]

ill be honest thats the one factor im willing to ignore since even at 100% the number of buildings would be ludicrous

[wind spade]

it also depends on whether do you want to consider entity limit as a limitation for given problem

[hazy fossil]

shouldnt be an issue with everything at 100% right? (except for the buildings that are underclocked for constant demand)

[wind spade]

you'd have to really limit the amount of foundations, putting everything on the ground and making belts as short as possible

and even then I'm not really sure

there are people who hit the entity limit and they didn't even use half of the map's resources

[hazy fossil]

should be able to bring the count down by overclocking power plants
they're still perfectly efficient when overclocked right?

[wind spade]

don't know, but it probably didn't change

[hazy fossil]

cutting the amount of needing nuclear power plants by more than half would really help with the entity limit right?

since they're so big

[wind spade]

I don't think building size affects entity limit

some buildings count as multiple entities, but that depends on how many entities do they consist of internally

the bigger issue will probably be items on belts

[hazy fossil]

all the design stuff could probs be worked out later, limit might be increased in the future
first thing to do is figure out the point values of power and each individual ite, figuring out which items are the most point efficient, and figuring out the maximum of said item that can be made

[wind spade]

neither of that is needed if we use linear programming optimisations 😉

[hybrid frost]

I read in the patch notes for 4 that 17 new alternative recipes are added. If I collect all hard-drives will I be able to get all alternative recipes?

[cedar mica]

There where 15 unused hard drives and we got a few new once

[stark lichen]

The calculator is seriously bugged now, isn't it?

[cedar mica]

Its been updated for U4, just need to switch branch in top right corner

[stark lichen]

It decided to make computers on its on own with my spare circuit boards when I was calculating HSCs

t-thanks calc

[hazy fossil]

which calc u guys using?

[hazy fossil]

have all the new alts been discovered?

could I get a link pls?

[latent fern]

So I am running 2 pure iron and coal nodes for a steel factory. all 4 belts are maxed out at 270 have the miners overclocked. I am then running the ore into 12 foundries 6 Each. I am trying to maximize my Steel beam production while also making some steel pipes. I I am producing 270 steel ingots per min. I can then run one full line into 4.5 Constructors to make beams. But since each foundry makes 45min ingot base i am having some trouble with finding a way to get that .5 from my other one making pipes. could I just merge 2 belts of 45 for 90 and split it and take the third to the beam production would that even out correctly ?

[stark lichen]

https://u4.satisfactorytools.com/production

[cedar mica]

Works fine for me. Sure you are on the Update 4 version?

[stark lichen]

i guess i need to adjust he number of input circuit boards until there are no spare ones

yes

[cedar mica]

Just matched your test and no computer shown here

Oh, found the issue. If you put input tab and say higher number, it default add a production to use it up

[stark lichen]

yeah

[cedar mica]

Just skip it for now and manually keep track of what you have

Its probably auto set to use all input products

[hazy fossil]

So
Have people come to a decision on what new alts are good?
Looks like heat encased framed and cooling devices are really good alts

[cedar mica]

Its too early to tell. It takes time to calculate how good they are

[hazy fossil]

okay
trying to figure out which alt is better
radio connection unit or radio control system

[iron prairie]

Can anyone double-check my math here: https://www.reddit.com/r/SatisfactoryGame/comments/m6gi6d/new_turbofuel_recipe/gr6scqd/?utm_source=reddit&utm_medium=web2x&context=3 ?

I know for a fact that it slightly disagrees with math I did earlier here on #math-and-meta , though the general conclusion is the same (way more oil spent for a very, very marginal improvement in sulfur efficiency)

[hazy fossil]

The satisfactorytools calculator for update 4 doesnt seem to work
no matter what I put in it wont calculate anything yet

[strange jasper]

calculator works for me. have it open right now.

[willow igloo]

I started a new factory (it's what I do every new update) and I noticed that when I moved into coal that my bio burners stopped doing anything, and my coal burners are running at 100% production. Is this why we have power storage now?

[strange jasper]

all power except biomass burners are now not self regulating to simplify production lines. the coal generators being at 100% is normal. the biomass burners are a result of that, as they are the only remaining self regulating power source, and automatically shut off in the presence of the excess power generated by the coal generators.

and yes, the power storage was there to help store up power in case you overstep your infrastructure, give you time to fix your grid, and to additionally deal with fluctuating power sources and consumers (geothermal generators and the particle accelerator).

[willow igloo]

yeah, I don't suspect particle accelerator will operate 100% of the time, probably only as needed. Still I kind of liked being able to build a power generator buffer. I sometimes would build a coal plant as if I had mk 5 belts when I only have mk 3, knowing I'll upgrade the belts later and building up a buffer

Not that it matters much, buffers aren't long term solutions

[upbeat tide]

This makes U3 aluminum look tiny

my limiting factor now is silica. I have a supply of 2800/min limit

and this is only half my bauxite I have already mined

[cedar mica]

Speaking of, 812/m turbo motors, seems to be the new max...

Copper, Quartz, Oil and Bauxite, is map limit, with that setup

With only 5k coal left, not sure you can power that setup

[bleak coral]

So the new aluminum recipes are so bauxite efficient that they're now limited by quartz?

Is it even possible to use all the bauxite now?

[cedar mica]

I managed with 812 turbo motors, but dont think you can power that with about 5k coal and 4 uranium rods...

So unless fuel refinerys, accept Nitrogen, thats not a sustainable project.

[bleak coral]

Plus we need aluminum for even more stuff now, not just turbo motors and mk5 belts

[cedar mica]

And with power running at full capacity, storing excess, requires making excess compaired to before

[shy mason]

it's a toss up between assembly director systems, nuclear pasta, and Thermal Propulsion Rocket for most points per part.

[latent fern]

So i need some help im making steel ingots on 12 different foundries and combining them into 2 lines of 270 and trying to balance them all out between 11 constructors 8 of them will be making steel beams 3 of them Steel Pipe. Anyway to properly balance them between 11 constructors on one line. I am currently restricted by the mk3 belt witch holds 270

[shy mason]

could use an industrial storage container that has 2 inputs and feed the 2nd output to half the steel beams

[bleak coral]

Storages aren't balancers, they don't work that way.

[west gale]

get the smelter recipe for aluminum and you eliminate the silica input

[bleak coral]

storages have a primary output that starts as whichever was placed first, but then is randomly selected when you load a save

[ashen thicket]

You lose a lot of efficiency though, your ingots per scrap is significantly lower

[west gale]

You can use the smelter recipe in a production line, and you'll get 50 silica out of each alumina solution refinery you use in the production line. You can use that excess silica to feed into the regular recipe line

[bleak coral]

There's two solutions I can think off the top of my head:}

1. Put both the steel beams and pipes constructors in one manifold, and then merge the second 270 line onto the first one once enough has been taken off it
2. Don't merge them into two 270 lines in the first place. Create 3 lines that only carry exactly what they need; 2 for steel beams and 1 for steel pipes.

You could do balancing with the second option

Also you're underproducing steel anyway, 8x steel beams & 3x steel pipes needs 570/m steel

[west gale]

Currently I'm running this:

• 1 Petroleum coke refinery + 3 Alumina Solution refineries feeding into...
• 4 Aluminum Scrap refineries (electrode aluminum scrap recipe) feeding into...
• 18 smelters to convert the 960 scrap to 480 ingots per minute, feeding into...
• 16 assemblers making the 480 Aluminum ingots into 480 sheets per minute (with copper ingots fed into the system from elsewhere)

I get 150 silica excess from that system

[latent fern]

Been fucking with this for the last 2 hours and i must of miss counted somewhere along the way multiple times lol. Thanks for saying something lmao.

If i were to do the second option how would i Work the balancers havent really messed with them before and seems complicated slightly

[upbeat tide]

Yes but that alt heavily teduces your alumnium ingot making

Also, is the encased uranium cell alt bugged? In u4? Still lists uranium cells

[torpid robin]

Fuck be dead that’s unreal

[bleak coral]

I mean I wouldn't. Manifolding is easier. I just wanted to give you another option.

In case you don't like manifolding for some reason

[upbeat tide]

Yea alu ingots

[west gale]

It goes from 3 to 4 scrap per 2 ingots. I'm using the recipe because I need silica for other recipes

[upbeat tide]

You lose quite alot of ingots by going pure

25% less

[latent fern]

I havent made a manifold before there somewhere i can look at this at?

[bleak coral]

You might've already and not just called it that. Here's a breakdown: https://satisfactory.gamepedia.com/Manifold

Shows an injection manifold in one of the pictures too, which was the first suggestion

on a side note: anyone else wishing even more that the buffers in machines weren't the size of stacks? cause now that they've increased the stack size of some stuff they've made manifold warmup times worse

[latent fern]

Ohhhhh. I thought that setup was worse than balancers didnt think the manifolds were efficient for something like that. and yea didnt know they were called that lol

[west gale]

After the machine inputs are saturated, a manifold is exactly as efficient as a balanced input

[bleak coral]

yup manifolds just have a warmup time

cause splitters don't split if a belt is full

balancers are good at really low speeds cause warmup times can be.. problematic

especially on mk1 belts or just really slow stuff

[latent fern]

So how much does a splitter take off a line each time ? So i would want to wait like 2 splitters before merging the other line?

[bleak coral]

but at the rate most stuff are going, and if you hook up machines and let them run while you build other stuff, the warmup time is negligable

[bleak coral]

however much the machine is taking is what it will end up taking off

so if the splitter is going to just an assembler making steel beams it'll take off 60/m

[latent fern]

like the per/min number? Ohhh ok So splitters arent really worth it later game when you have upgraded belts and stuff?

[bleak coral]

what no, like how much it splits is dependent on where those belts are going

it's not fixed

that's the whole idea behind the manifold

[latent fern]

Ohh i see

[bleak coral]

when an item comes in it goes into the internal buffer of the splitter

it then tries to go out of it, and each output takes a turn

if an output is not available, because there's no belt or the belt that is connected to it is full, it will go to the next available output

so if you have 270/m coming into a splitter with two outputs, it'll average out to 135 and 135, but since the machine only takes 60 it backs up the belt

at that point it will only take 60 from it, and the 210 will go out the other output

[latent fern]

Oh That makes much more sense thank you. So If i do an injected manifold i want to put the merger after there is enough off the belt. So its either 240 or 300. if i do 300 i wont have enough from the first belt to continue it but if i do it at the 300 mark I have 20 left over but if im inputting 270 from the 2nd belt and continuing with 20 from the first wouldnt it bottle neck since they belt cant hold anymore

[bleak coral]

Right, I overlooked something, sorry. You can't inject a full belt because it would go over the max throughput no matter what you're merging it into. Sorry, that's my bad.

[latent fern]

Ill just run two different manifolds :) Thank you for all your help and explaining all that

[bleak coral]

No problem! I'm glad you figured it out

Oh and don't forget about clockspeeds, you may not be able to run 8 steel beam assemblers at 100% off of 270, but you can run 7 at 100% and 1 at 50%

[latent fern]

Couldnt i just make the last one of the first belt a merger and merge the end of the 2nd belt with that one and push it into the machine?

[bleak coral]

we don't have something that has 2 inputs and 2 outputs, mergers only go into 1 output

[latent fern]

So what i ended up doing and I think this will work out. isran a type of manifold for both belts but its two set up on the same side and they meet up in the middle which is the end for both belts and they merge into the constructor which should give me 60 for a full craft of a beam. since ill have 30 left over from each belt

[bleak coral]

Oh yeah that should work, I like it

[wind spade]

yeah, that's a new bug, it has been reported recently, I'm working on fixing it

[dark depot]

does anyone know if the coal ratios have changed?

coal power to be specific

[undone lotus]

If you're running math for experimental satisfactory fuel gens, they're borked. I made a vid on it on #streams-and-videos .

[dark depot]

i don't have fuel yet since i'm starting over so...

has the water/coal/mk.2 belt line changed?

[oblique hollow]

Doesnt look like it

[dark depot]

so this is still accurate?

[oblique hollow]

Should be

[wind spade]

yeah

[dark depot]

i'm having trouble understanding this new system....if i cut off the bio power everything shuts down even though i have plenty of coal power now?

does it automatically use coal first and bio as a backup?

[oblique hollow]

Biomass still self-regulates

If it shuts down you made some wrong connections

[dark depot]

in the past i always cut off bio when it was no longer needed

hmm

[oblique hollow]

Like, you should see that the demand goes higher than your capacity

Otherwise it wouldnt shut down

[dark depot]

but when i cut off bio everything shuts down

[oblique hollow]

Like, the fuse blows?

[dark depot]

i must be missing a connection somehow

[latent fern]

so is it 2 Coal gen to 1 water pump? and the mk1 tubes can hold 300m water so you can only run 6 coal gen off of one full water pipe? is that correct

[bleak coral]

the only change was the coal gens always running at 100%
so unless you've been habitually underfeeding the coal gens water or coal, you probably just missed a wire or something

[dark depot]

it's 8 to 3

it's not even like that....

[latent fern]

8 gen to 3 water pumps?

[wind spade]

I heard that overclocking a gen no longer is 250% -> 200%, can somebody confirm this?

[bleak coral]

sure, I can go check

or the game can crash

[dark depot]

8 gen 3 water pumps 2 water lines split 3rd water extractor as shown in my screenshot just a little bit ago

[latent fern]

How much coal would be needed for that? 120 right?

[dark depot]

supposed to be

kind of hard to find out for sure without peaking all 8

[bleak coral]

@wind spade nope, it appears to be the same, 250% -> ~202%
I checked coal gens and fuel gens with coal, compacted coal, fuel, and turbofuel

I don't have nuke stuff lying around to check that, but I doubt that'd change cause of pipe limits

[wind spade]

cool. It seems that the user got confused by the still-not-fixed buggy UI

[bleak coral]

yeah I'm surprised the bug is still there, it's annoying

they redid the whole UI, but I guess the overclock part wasn't actually touched

that's where the bug is, it's correct literally everywhere else

[wind spade]

yeah, I got a screenshot from them and it was there correctly 🙂

(in the top left corner, not in the OC menu)

[oblique hollow]

I tried with Fuel gens and it was still not correct...

[wind spade]

yeah, OC menu still displays wrong value, but it still works the way it should (250% gen produces 202% power)

[torpid robin]

😂 you can make 3307 super computers now. jesus christ

[dark depot]

probably need them for drones? 😄

[wind spade]

you can make 201.64/min of the most sink-point-value item (thermal propulsion rocket), netting 128.5m sink points per minute

(didn't check if power is enough tho)

[dark depot]

so happy just to finally get off bio fuel ulgh....

that upkeep start...

[oblique hollow]

Yeah, as for your power issue: you must be missing a connection somewhere

[dark depot]

i fixed it finally

[bleak coral]

what was the problem?

[dark depot]

somehow i did have 2 lines, no idea how it split

[latent fern]

Could i use the same middle line to add to another 2 water pumps or do they need their own 3rd pump/pipe

[dark depot]

you split the 3rd one into the other 2 lines

[latent fern]

I need to move my water pumps my current location limits how much and how i can place

May just move my water pumps and route it back to my base

[molten vault]

@fading flower ? this

[fading flower]

ya

[molten vault]

Nice

🙂 thx

[fading flower]

have fun

[molten vault]

i hope the new picture comes

i must loook nice

[oblique hollow]

Double pump. Remove one of the pumps from eaxh line

They dont stack

2 directly after each other is 100% useless

[dark depot]

yes they do...the head was 22 and i needed 2

now everything is 100%

[fierce ruin]

it resets the head lift at that pump

it doesn't add together

[dark depot]

it has split like it should

[vast jungle]

so 2 Pumps behind each other are 50% effective 😉

[dark depot]

i don't get what you are saying they are both now under 22

[fierce ruin]

Combined lift cant exceed pump limit

[dark depot]

the pipe is 300 all the way down

[fierce ruin]

If you add 20 lift to 20 it stays as 20

[dark depot]

it didn't work, i added another pump, it works, that's all that matters to me lol

[fierce ruin]

not counting the wasted headlift of the extractor or the over headlift

[upbeat tide]

Is the encased uranium cell alt broken in u4? It looks like the Ue version still with uranium pellets

[ancient shuttle]

Restarted with experimental build of update 4, is it just me, or is the Casted Screw alternate recipe insanely good? It cuts out an entire constructor and you have the same Iron Ingot ratio (counting Iron rods as a 1:1 ratio for ingots to rods) per screw made (1:4).

It costs you less materials for the constructor, reduces footprint and complexity (by removing a constructor), costs less power to operate fewer constructors (which I find is a big deal at the start of the game). It doesn't feel like an alternate recipe, it feels like a strictly better option to an otherwise inferior recipe.

[vast jungle]

yes, its nice...

[river night]

many alternates are just flat out better

[upbeat tide]

Remwmber alternates in this game are more like upgrades in others in some sense

[cedar mica]

Or they let you use different materials

[vast jungle]

Some Alternates are a bit weird...

[upbeat tide]

Some are forgetable...looks at you biocoal

Others seemed useless in U3 like pure aluminum ingot but in U4, seem maybe worth it

[ancient shuttle]

I'm aware of this, but it also makes it frustrating if your dicerolls are poor and you don't get this until significantly later.

[cedar mica]

Get drives and keep going till list is empty, no missing recipes...

[ancient shuttle]

I know, but that doesn't change the tedium.

It makes it worse when it makes more sense to delay your progression through the tiers (or the MAM) just to get a higher chance of getting a useful alternate recipe.

[cedar mica]

Setup machines to make space elevator parts, go hunt drives and unlock everything while you wait...

Plus you now have crash site parts, that saves you time later on

[frosty owl]

Remember you can just pop up a MAM anywhere to check the progress ;)

[upbeat tide]

Yea getting a bit of super computers or other parts early for research is nice

[feral summit]

Realizing that for the first time was a game-changer for me, especially during hard drive hunts

[lunar moss]

Good times

[upbeat tide]

Soo, this is my aluminum plan in U4 so far. Big limiter is silica. I only have 4 700a min belts available currently.

I do want to see what would happen with the pure alu alt tho. Since bauxite is not used as much

This is actually half the bauxite I have 🙂

[glacial hemlock]

Alternative recipes are mostly good. The hard part is you need at least some tier 6 gears to navigate around safely. Maybe zipline can replace jetpacks to some extent

Guns is a must

[upbeat tide]

Well, pure alu reduces total alu ingot by 25%, but may actually be needed now

Because if I want to max out my 4 normal and 1 pure bauxite nodes, I will need an obsene amount of silica.

[plucky dagger]

Let's turn nuclear waste into Nuclear Plutonium Waste 😄

[upbeat tide]

Im waiting for them to fix the encased uranium cell alt before fully going into U4

Its kinda mandatory for me

[median thunder]

plus massive amounts of power 🙃

[upbeat tide]

My U3 nuclear plant makes 31.5 rods a min using 600 uranium ore. In U4, to get this working, I will need all 4 uranium nodes, 2100 a min

Because the uranium cell alt is not updated correctly

[frosty owl]

Btw, how much waste does 1 rod make?
Was it 4 to 1?

Uranium rods

[plucky dagger]

@upbeat tide I now see what you mean, that's an unfortunate bug. Let's hope they fix it soon 🙂

[frosty owl]

Pffft
Gimme a few snowballs and I can conquer the world!
Much cheaper :P
:sf_nobelisk: work too

Though I'm more of a basher guy myself

[sullen cloud]

Only thing I don‘t really like is that alu is now even needed for nuke power generation as an input. Feels like alu products are over emphasized

[upbeat tide]

You can flood with alu products now tho

Not that big of a deal

[fierce ruin]

Any opinions on new alt recepies? (especially best process of making aluminium)

[vast jungle]

the new turbofuel recipe looks interesting

more oil, but less sulfur

and we just got even more oil

[fierce ruin]

but what else do we need sulfur for?

[vast jungle]

the point is we could get more total turbofuel ^^

[fierce ruin]

but I don't have pressureizerers

[vast jungle]

even without pressurizers you could use the recipe to get more turbofuel... in theory at the cost of plastic/rubber... if you max out oil

[fierce ruin]

Would have to check the math and get lucky enough

[upbeat tide]

I am playing with some alu setups atm. This is a silica restricted plan tho. Want to see what pure alu ingots does tho

[fierce ruin]

Dont like using oil here tho

gross coke

[upbeat tide]

Electrode scrap got switched

[fierce ruin]

I used coal cuz it's more abundant than oil

Not sure if the math adds up tho

[vast jungle]

130 oil for that much aluminium doesn't look bad

[wind spade]

did you hit any issues with new tool? I got a few bug reports, but I'm surprised that it works well so far 😄

[upbeat tide]

Not really no

I had a user error that caused the electrode scrap alt to break the chart, but was because I disabled the rubber/plastic recipies and forgot to turn on HOR alt

[wind spade]

cool

lmk if you find some

[forest hornet]

Any math yet on drone items/minute based on distance?

[magic shadow]

capacity divided by travel time of round trip

[forest hornet]

lol, I more meant like, what the round trip time was over a given distance

[magic shadow]

if u have the thing then you tell us 😆

[sullen cloud]

3/4 of the map is approx a round trip of 5min according to the UI

[neat stratus]

ok so with the new coal plants, how do I get 100% efficiency? I feel like if I do a manifold, the last plant barely gets resources. I have 8 plants, which means 15 coal each, so 120 coal with mk5 belts

[fierce ruin]

or mk 2 belts with mk1 belts to each gen

the gen will fill and the extra will carry over

[neat stratus]

I guess ill let it run longer

but the last gen in the sequence seems to barely get enough coal

[sullen cloud]

Let it run a while. You can try with 125/min as well

[fierce ruin]

it shouldn't since that's how probability works, but it'll balance in time

[bleak coral]

That doesn't change the warmup time. Restrictive belts only change the warmup time if they're exactly the speed the machine is taking at.

[neat stratus]

would over filling (240 coal) for a few minutes to fill everything up and then cut back down to 120 work then? maybe it just has a hard time filling up with the exact

[sullen cloud]

Absolutely

[fierce ruin]

your choice

[bleak coral]

It just takes time

You can also prefill them

[neat stratus]

interesting. Aight. I booted up the new build yesterday and my entire power grid blew up for some reason lol, so I'm going to each power plant and trying to get it all going again

[fierce ruin]

don't forget about water

[neat stratus]

built a huge power reservoir too

yeah been switching out to mk2 pipes

so much nicer

[sullen cloud]

Do you run any belts or pipes at max capacity for that?

[bleak coral]

If your power went out you've probably been underfeeding at least some plants and only now noticed

[neat stratus]

yeah most likely

that would make sense since beforehand I could get away with it

[bleak coral]

Yup, part of why they changed it: immediate feedback to mistakes instead of it lurking there to surprise you later

[fierce ruin]

that's a good add

[neat stratus]

haha yeah for sure. Overall I think it's a good change, just means I gotta spend a bunch of time fixing it 😛 but in the end that'll be good cause I can move on to mk2 pipes etc

[naive ginkgo]

ok i need some help figuring out my aluminum production, I've setup a factory that should in theory be able to use up an entire full 780 bauxite input, but i'm having issues having it backup on itself so the water it outputs can be properly re-ingested.

This is the layout of the section at issue

[fierce ruin]

you can take it else where to be used or you can recycle it and underclock your water extractors

[neat stratus]

oof I had issues with liquid backup in my aluminum production i'll have to revisit that as well

[fierce ruin]

Amelie of the sea made a good U3 video on it

[bleak coral]

Oh yeah all the aluminum recipes are completely changed, and the alt and vanilla have been flipped

[fierce ruin]

Whats the best approach for making fuel for generators AND plastic/rubber?

[bleak coral]

So everyone's aluminum factories need to be fixed for U4

[naive ginkgo]

yep, i'm trying to get something working in u4 but fluid backup is a constant enemy of mine

it took me forever to have my rubber production be constant

and honestly, i'm not sure how i did that

[sullen cloud]

There is a known bug that maxed out pipes and belts don’t support their max values

Like 10% is actually lost, especially when you have many junctions

This could contribute to your problem

[naive ginkgo]

my 600 pipe is showing a constant 600 water/m

actually, 602 going buy the buffer

these buffers are weird i swear

[upbeat tide]

Yes buffers are weird

I only use them in two spots

1. Fluid train buffer
2. Storage for my TF plant

[neat stratus]

my buffer fill rate went to 1000 at one point

[upbeat tide]

You also need headlift as high as the buffer tank if its going uphill

Or even downhill sometimes as flowrate can bug

[neat stratus]

is this a new bug?

[wind spade]

nah, been there since like forever

[naive ginkgo]

i always put them higher than what ever they fill

[oblique hollow]

Calculation errors. They arent THAT weird

[naive ginkgo]

i mean if i get more pipes per pipe i ain't gonna complain

[oblique hollow]

You dont actually get 602 out with only 600 going in. Its a display inacuracy, not a magic increase in flow rate

[naive ginkgo]

i know, that's why i struck it out

[oblique hollow]

Its kinda hard to tell at times what which formatting is supposed to mean. The internet is lethally allergic to written sarcasm

[naive ginkgo]

it's pretty well known that stuff that is struck out is usually a joke/unrealistic, or at least in my places of the world

i'd use sponge case, but /spongecase isn't a thing here

[oblique hollow]

Spongecases? Oh
ThIs KInD of WrItInG?

[naive ginkgo]

yes

[oblique hollow]

Ah 😆

[naive ginkgo]

my brain can't write this thing without spending 20 hours writing it

[naive ginkgo]

welp looks like adding a fluid buffer to every pipe-system fixed the issue... yay?

[oblique hollow]

That sounds like 100% hacky garbage. What was the issue?

[naive ginkgo]

#math-and-meta message

^ see original message

all the refineries are always active now muhahah

[oblique hollow]

I dont know why its so hard for people to understand....
Have you underclocked the water extractors yet?

[naive ginkgo]

nope they are all at 300%

well except the one at 50/m

[oblique hollow]

Reduce their speed

[naive ginkgo]

i mean i need them to be a 600

[oblique hollow]

How much recycled water do you produce

Byproduct water

[naive ginkgo]

see orignal post? 520/m

[oblique hollow]

Alright. Wanna hear an interesting theory?

[naive ginkgo]

shoot

[oblique hollow]

You only need 1 extractor at 80 m3/min

[naive ginkgo]

how, why and how

[oblique hollow]

The rest is recycled water

[naive ginkgo]

you do realise i need a total of 1170 water/m?

like i'm not sure i can have that much water out of an extractor

[oblique hollow]

Its not exactly easy having just a diagram to deal with

[naive ginkgo]

i mean it's an exact representation of my factory

it's custom made

[oblique hollow]

Yeah but its unhandly

Having written down the totals somewhere would have made it easier

Since then i dont have to look through the whole thing to figure that stuff out

[naive ginkgo]

i mean i can add them

[oblique hollow]

That would be great.

[naive ginkgo]

I mean the original graph has a quanity/m on every arrow tho

[oblique hollow]

Because your 500 fluid buffers didnt fix the issue, they only delay it

[naive ginkgo]

or at least have them
inputs:
water: 1170/m
bauxite: 780/m

outs:
Silica: 325/m
Water: 520/m
Scrap: 2040

[oblique hollow]

Alright, then you in theory only need 650 fresh water

Now it adds up

Though the current piping you have is not fit for this

[naive ginkgo]

what would be more fit?

[deft lichen]

@naive ginkgo is this for EX or EA

[naive ginkgo]

EX, I'm trying to get my aluminum production working again after the changes

[deft lichen]

so counting with 780 bauxite

https://u4.satisfactorytools.com/production?share=LZqxpslhDUCkkpHkWVYx

[naive ginkgo]

and water recycling is getting in the way

i'm using coke instead of coal, mostly because i want to see how it scales up

and because i can already have perfect value inand out

[deft lichen]

alr I'l toggle the alt

[naive ginkgo]

also we can limit to scrap since this the rest can do what it wants

i just have it plugged to the awesome sink right now, for maximum usage potential

[deft lichen]

the numbers become uglier

but still manageable

https://u4.satisfactorytools.com/production?share=rMsMDrXxEmOEAkXBCdg7

650 fresh 520 recycled water

5:4 ratio

[naive ginkgo]

i mean that's exctly what i'm doing

[deft lichen]

so split the 6.5 refineries in a 5:4 ratio

and that's seemingly all

[deft lichen]

what I sent says there should be 650 in not 600

o wait

I see now

[oblique hollow]

im currently revisioning a bit of your design in paint

[deft lichen]

I'd keep the recycled and fresh water setups completely separate

to avoid gridlocks

other than that the setup seems alright

[naive ginkgo]

i mean yeah, but i don't have an awesome sink for fluids

god i need that in my life

as i said adding some fluid buffers seems to have fixed the issues, but like why did it behave like that before

welp time to watch the bauxite belt to see if it stops again

[oblique hollow]

new suggestion that fixes some piping issues

[hazy fossil]

Btw did the satisfactory tools upd4 version of the calculator get fixed?

[naive ginkgo]

i'll give that a shot to see if it has the same issues

[oblique hollow]

i hope all the changes are clear and visible

i seperated the recycled water loop on top into 2 seperate ones

and changed the entry point for the fresh water on the left side

[naive ginkgo]

i'm guessing this is to prevent he pipes to reaching full capacity?

[oblique hollow]

aye

also it avoids the manifold issue better

large manifolds have some odd flowrate issues

especially at full capacity

[naive ginkgo]

i mean i personally suspect it may be due to some floating point imprecision shenanigans, but eh honestly i'm just happy to see it all green now

NOW TO FIND AN OTHER PURE BAUXITE NODE, FOR SCIENCE!

[oblique hollow]

it much rather seems to be some calculatory losses and inaccuracies on part of the junctions

[naive ginkgo]

calculation losses
that's what floating point inaccuracy can lead to

though usually that starts popping up at much more... decimalized number

[oblique hollow]

its real strange, i already made a QA post on it before U4

[naive ginkgo]

i mean that's just a result of how floating points work in computers

[oblique hollow]

pipes work real great as long as you dont try to squeeze 300 / 600 through more than like 3 junctions

[naive ginkgo]

we can't really store floating points with much precisions so the guys at IEEE became wizards and invented IEEE 754

[oblique hollow]

after that, you either need non-full pipes or to use a lot of inbetween splitting and some buffers

[naive ingot]

So, from the various changes to the recepies being generally positive, I'm assuming the old maximum of 156 turbomotors/minute has been increased by a bit, but has anyone done the math yet to come up with the new maximum?

[fierce ruin]

Ask Greeny

[iron prairie]

If the calculator is to be believed: 812.89 turbo motors/min.

Of course, the AWESOME sink points value of turbo motors got nerfed... though you can now use turbo motors in a space elevator part that's worth more than old turbo motors.

[glacial hemlock]

812? Wow that's crazy

[iron prairie]

Looks like the limiting factor is still bauxite. Several other resources are maxed, with some parts (like circuit boards) being done as a mix of an alt recipe such as silicone circuit boards, and their default recipes.

Thermal propulsion rockets, you can do 201.64/min: bauxite is extra-limiting there, as other parts of the TPR also require aluminum.

[strange jasper]

actually, limiting factor may be silica related. with the new recipes and alts, you can use up all the bauxite and alumina so efficiently that you have to use less optimal recipes to make ingots and such.

or at least silica reduces the efficiency overall of the line by just a smidge.

the normal aluminium ingot recipe is a 3:2 (or 6:4) ratio of scrap to ingot, while the pure recipe is 2:1 (or 6:3). the normal recipe is more efficient, but there is physically not enough silica, even with alternate recipes, to only use the normal one. you have to use the alt at some point if you are going to use up all your bauxite, unless you are using the alumina solution for batteries.

so while bauxite might be the bottleneck for turbo motor production, silica is a bottleneck for the bauxite production line.

hmm... the calculator is using the pure recipe a lot for a lot of recipes that are bottlenecked by things other than bauxite.

i suppose at some point, the most efficient production path of aluminium products is blocked by requirements elsewhere. making only aluminium items is somewhat of an unlikely factory choice, so i'd assume that the pure recipe is useful regardless.

[iron prairie]

You are correct: while the bulk of aluminum scrap uses the default, some uses pure aluminum ingot because of quartz limitations.

[wind spade]

well... "just a bit more"... yeah... around 813 per minute

[cedar mica]

The issue now is finding enough power, to make that many turbo motors

[oblique hollow]

Nuclear? xd

[iron prairie]

Nuclear would compete for turbomotor resources... though so would fuel and coal. Max turbomotors now really, really maxes out the map.

And yes, it does appear now that quartz is more strongly limiting than bauxite. With U4, there's more bauxite, the most bauxite-efficient alumina solution recipe no longer produces silica, and the aluminum ingot recipe consumes more silica. Also, other bits of turbomotors use a lot of quartz too.

[strange jasper]

i personally think this is a much more engaging progression to your factory. very unclear what items limit production, power generation competes with production, and more than 1 endgame item to produce. there are 4 new space elevator parts i think, compared to just turbomotors before.

[iron prairie]

I'd say now that probably the best bauxite path is sloppy alumina -> pure aluminum ingot, possibly with some regular aluminum ingots thrown in if there's quartz to spare for the silica.

The increased alumina solution yield from sloppy alumina outweighs the advantage of making silica for aluminum ingots. Specifically, 1 bauxite sent through sloppy alumina solution & pure ingots yields 1.6 aluminum ingots per bauxite, while regular solution recycling its silica yields 1.42 aluminum ingots per bauxite.

If you've got quartz to spare: sloppy alumina, plus a mix of pure aluminum ingots and regular aluminum ingots.

If you're quartz-neutral, sloppy alumina and pure aluminum ingots.

If you're quartz-deficient, regular alumina and pure aluminum ingots, diverting the silica to whatever else you need it for.

[strange jasper]

does anyone know if plutonium fuel rods can be used in vehicles? it could be a good way to still get some power out of the production line, and leave you with no waste.

[bleak coral]

It's crazy how bauxite efficient aluminum has gotten

I guess since they put it in everything T7/T8 they needed to make sure we had plenty

[bleak coral]

yes they can be

[iron prairie]

From 7800 bauxite (the old map limit), U3 maxes out at 4952.4 aluminum ingots (requiring 2971 coal and 6438 silica).

U4, meanwhile, does 12480 aluminum ingots with sloppy alumina, pure aluminum ingot, and just 780 crude (as 3120 petrocoke) as input.

[bleak coral]

no way would you use them at a decent rate though, they still have tons of power

[iron prairie]

(more if you count the additional bauxite nodes)

[bleak coral]

(the plutonium rods in vehicles I mean)

[torpid robin]

The way the game is heading through . I can’t really see any 1 item getting maxed out . Plus they are going to be bringing more stuff out. So if not now but soon you will be trying to spread the resources out over all sorts of things rather than just smashing 1 item out

Plus I think your pc would die . If you maxed out a production line now lol

[bleak coral]

turbo motors/aluminum now seem to be in the same place HMFs were in U3: technically possible to max out but eats most of the map and practically unfeasible

which if aluminum is the new iron for T7+ that's a good thing

[torpid robin]

Yea like I wondered what they were going to do . Cause it seemed wrong that the max of turbo motors was so low when it seemed there was so much stuff still to come . I thought they would increase things maybe with mk2 machines but doing it through alts is a good move I feel

[bleak coral]

they did say their vision of T8 changed, I'm guessing part of what changed was how much they wanted to use aluminum

and even the vanilla recipes are way more efficient

[torpid robin]

Or how little they actually had in the map of bauxite

[bleak coral]

I mean they could've added more bauxite as a blunt solution, but this is better

[iron prairie]

Fewer changes to the map accomplished by making recipes more efficient.

[torpid robin]

The recipe that made me laugh were oc super computers lol

Freaking 3k of the things lol

Havnt had a good look at the others yet though . Was more concerned getting my coal power running

[bleak coral]

between mk2 pipes, the new wells, and blender diluted fuel oil is also now crazy cheap

all those oil alts seem pretty tempting now

[torpid robin]

This is what I was thinking before the update came out . As you say with the mk2 pipes and wells . Those oil alts . Really step up. A bit more now

Pity they didn’t bring a new machine to refine ores though

[strange jasper]

right now the theoretical "end goal" of the game is to max out all your milestones and M.A.M. research, unlock every recipe, exploit every resource node to 100% efficiency, and then sink all your products. i'm really hoping that in future updates, when there is more emphasis on the story that we could have a reason to keep producing all these space elevator parts.

[bleak coral]

especially with the space between mk2 and mk3 miners being wider than before, there's a place in progression where the idea of saving mined resources sounds appealing

[torpid robin]

Oh man and with the gap comes different strats . It used to be you could rush t7 then start using mk3 miners and mk5 belts everywhere ASAP . Doesn’t work like that now

Gonna be stuck with those dam mk4 belts

[bleak coral]

hmm I think the belt strategy still works, aluminum numbers are different but not their logistics

[torpid robin]

Yea I was just about to ask where does it come in now ?

[bleak coral]

but mk3 miners are definitely out of reach for much longer

[iron prairie]

On a sidenote: useless calculation (U3): ignoring power constraints, one can make 265740 wire and 1817828.6 screws each minute. This maxes out coal, sulfur, copper, iron, and crude*... but that still leaves quartz, uranium, limestone and bauxite to even further destroy one's PC.

[bleak coral]

hahaha, try HMFs

for U3

I have this saved: https://www.satisfactorytools.com/production?share=43GOjkQHrd8Rfc41mnTd

This is for a sustainable max HMF plant, at least to the best of my ability

could be off a few decimals

[torpid robin]

That uranium for power ?!

[iron prairie]

There is a problem with that factory.
You don't have at least 2672 T5 belts, for the outputs alone, to destroy your frame rate.

[torpid robin]

That’s limited on the copper there ain’t it

Dam and the Sulfer

[bleak coral]

limited by oil too

and iron

and yeah uranium is for power

oh and coal is a limiter as well

[bleak coral]

are you sure? there's stuff like this all over the place

it may not be one mass of screws, but there's 100s of thousands of materials per minute on belts

[torpid robin]

Does late game still use a fuck load of quick wire and copper sheets ?

[jade minnow]

The new late-game seems to use a fuck load of eveything lol

[torpid robin]

Which imo is awesome

Gives end game variety

[warm island]

right here. on mine. this says a big number. but not how many nodes I need. does tell me oc or how many smelters anywhere either

[wind spade]

for smelters it says how many 🤔

for miners you have to do the math yourselves, since it doesn't know how many nodes and which purity do you have

[bleak coral]

or what miner you have, which also changes it

[warm island]

does that mean i need 1957 smelters

[wind spade]

1957 smelters at 100% + 1 smelter at 56/57%

[warm island]

wow

ok then

[wind spade]

well that setup maximises the whole map pretty much, it's not something you'd build normally

[bleak coral]

what are you trying to do?

[upbeat tide]

I am playing with aluminum setups, this alt instant scrap really sucks. I thought it would have been better for the asking price

[warm island]

I was just playing around with it to see how I should setup an optimized factory

[upbeat tide]

compared to electrode scrap

but the sloppy alumnia alt is soo nice

this build uses 2500 of my 2580 supply of bauxite too

[bleak coral]

if you're going to use the maximize function use the items, input tab to change the amount of resources it'll use

also note that maximize solves for most stuff, not best resource efficiency. so take the maximize number, switch back to items/min and use it there to get the most efficient

it's not always different, but it sometimes is

[warm island]

Ah I didnt know how it worked

[bleak coral]

yeah that tab sets the resource limits and defaults to max available on the map

[warm island]

Just trying to figure it out. Haven't played in a long time

trying to get back into things

[bleak coral]

you can also add other resources if you're already making them

[warm island]

so that would be me pulling every node in the whole map

[bleak coral]

yup, with mk3 miners

[warm island]

wow

[fierce ruin]

and 250

[bleak coral]

earlier sillySMS did the math and sloppy aluminum + pure makes more ingots per bauxite than the regular recipe + foundries/pure

[upbeat tide]

Yea I noticed myself

Too bad the instant scrap alt is trash tho

[wind spade]

sneak peek, this will work automatically on next big release

[bleak coral]

cool! I know it's been on the to-do list for a while, I look forward to it!

[fierce ruin]

greeny the REAL clutch

[warm island]

So how do I calculate how much I am bringing in

will the miner tell me

[bleak coral]

the miner will tell you, it's based on miner level and node purity

you can see the node purity by looking at the node

here's a chart for the speeds: https://satisfactory.gamepedia.com/Miner#Mining_speed

[warm island]

yeah i remember puritys

[upbeat tide]

For mk3 at 250%

Pure - 1200 a min
Normal - 600
Impure - 300

[torpid robin]

You might wanna get the right amount of water extractors to coal gens right before you start looking at using 70k of iron 😂

[bleak coral]

just looked at it and

sulfur to make scrap, no thanks

[upbeat tide]

Yup

IF it was insanely efficient sure and not alot of sulfur

But alas its worse on bauxite and everything else

[torpid robin]

Wonder why they add alts like that I.

You would think they would make all alts good . In one way or another

[bleak coral]

I blame the randomness of gathering alts

part of the design sensibilities behind random systems is you have to have both bad and good pulls, to make the good pulls feel good

[upbeat tide]

Instant scrap is instantly in the scrap heap along with biocoal 🙂

[bleak coral]

they do this in magic all the time, make bad cards as "filler"

or at least not very good cards

[upbeat tide]

Issue is even “bad”alts like sisemic nobelisks have some use

[bleak coral]

not always, some alts are just worse

compacted steel for example

or heavy flexible frame, though that one only falls after analysis

that one was probably a mistake

[torpid robin]

Or maybe it comes In with future plans .oil alts used to be bad for a lot of things . Now . Not so ouch

Much

[bleak coral]

oh no it's still worse, use the oil alts earlier in the process for heavy encased frame

the problem is it's supposed to be the fast one, but it's so inefficient that the factory needed to supply all the parts undoes all the space and power efficiency it has

a heavy encased frame factory using oil alts for maximum resource efficiency is about the same space and power usage as a heavy flexible frame factory that ignores resource efficiency in favor of power and space efficiency

[final spoke]

Question:

I gotta rebalance 3x belts of 780 into 4x belts of 480 (ignoring the leftover)

What would be a pretty repeatable way to do that? I've thought about mk4 belts and smart splitters, but merging the leftovers back and then using a smart splitter again is a mess

Is there a cleaner way?

[wind spade]

why not use the 3 780 belts?

[fierce ruin]

manifold?

[final spoke]

my output works out to 720/min so there's not much wiggle room out there

(for reference, making steel)
I'm feeding 12 foundries perfectly with 480 coal/iron ingots

I'm getting mk5 belts from my trains

[wind spade]

just manifold three rows of foundries with those three belts, overflow of each manifold would be merged to make a 4th manifold 🤷‍♂️

[torpid robin]

#screenshots message

I use systems like that for large scale changing of belts

[final spoke]

Alright, thanks

[torpid robin]

Just plug em all in . And pull anything out the other side at will

[final spoke]

it would seem my biggest problem is how I arranged the 4 "modules"

should've made them in line

[quaint abyss]

anybody know how to make a 60 belt into 26 and 34 belts with default conveyor splitters?

[fierce ruin]

just let it run

don't waste the space for single digits it scales ugly that way

[final spoke]

screw this, I'll just deal with merging the outputs instead

too much spaghet for my liking

[iron smelt]

The prime factorization of 60 is 2,2,3,5. Since you can only split 2 and 3 ways, you're not going to be able to evenly split belts into integer values that dont have a factor of 5

[quaint abyss]

ok, i'm gonna try and find another solution, probably another iron node.

[bleak coral]

there are ways to do splits other than 2 & 3, but for your specific example there's better solutions than creating a complex balancer

[sand epoch]

if one side only uses 26, a simple spliter will work..?

[bleak coral]

this^

splitters don't just split evenly, they can only do that if there's somewhere to go

if a belt is backed up, the items will just go into the available outputs

they don't disappear

[quaint abyss]

yeah ik that

[bleak coral]

right, so you can just use one splitter, and then once the 26 side backs up the other side will be able to take the full 34

[sand epoch]

don't think i'll ever understand the need to 'force' splits at specific numbers to prevent backups.. o0

[bleak coral]

I mean if that's what you want is to not have any backups, I'd not have a 60 line in the first place. I'd mess with machine clockspeeds and number of machines to have separate lines with the correct outputs

[iron smelt]

I think its the old load balance vs overflow argument

[sand epoch]

that's an actual debate? o0

[bleak coral]

there's people who really like load balancing

[iron smelt]

ye

I don't really understand why, but I believe its just a self imposed optimization challenge, that they can get all their machines running at full efficiency with 0 backed up

[sand epoch]

if i ever try to visit that world.. hit me.. a lot. i'll stick with always ensuring i have more than i need on the belts. 😛

[bleak coral]

I mean whatever, I try to not push people in either direction; I'm not going to tell them how to have fun

[iron smelt]

I thought that was the purpose of the internet though

[bleak coral]

I just usually offer the overflow/manifold solution first cause that's what I know best

also warmup times on like 60/min lines can be annoyingly long, so personally I don't switch to manifolds until I have mk3 belts

[torpid robin]

Yea like the devil @frosty owl

[bleak coral]

As a thought experiment I did think of how to do the balancer if you really want to split a 60 line into 34 & 26: Split into two lines of 30, split one line of 30 into 5 lines of 6 (using the 1:5 loopback as seen here: https://satisfactory.gamepedia.com/Balancer#Load_balancer ), then split one of the 6 lines into 3 lines of 2. Finally merge 2 of the 2 lines with the 30 line, and then merge the rest of the lines together.

[quaint abyss]

Usually i do Load balancing because it is just more Satisfactory to me

[bleak coral]

I can understand the fun of figuring out an interesting split, that was kinda fun to come up with.

[torpid robin]

Balancing is fine in smaller scale . You start trying to balance like 500 machines lol stuff that

[bleak coral]

But I prefer in practice avoiding actually making a complex balancer. Avoiding it either through clockspeed manipulation or manifolds.

[final spoke]

Say, if a machine uses for example 60 materials / minute. If I use mk2+ belts it doesn't really make a difference, materials just get to it faster (assuming you have exactly 60/min input on belt)
Is there any benefit for using mk1 belts in this case? In case of chaining multiple machines?

[bleak coral]

Not in the long run, no

Some specific, complex balancer setups use slow belts as limiters, and for manifolds if the machine takes exactly the belt speed it wouldn't have to backup

but outside of those cases there's no advantage to mixing belts

[final spoke]

makes sense, thanks

[torpid robin]

You don’t even have to have the exact amount . You could have 100 . Do a split to a manifold that needs 60 then send 40 off to another manifold

[digital halo]

hey could someone help me figure out how to split reinforced iron plate production - it produces 5 per minute and I need 3 of it to go for the frame and 2 into smart plating - since splitter splits half, I am guessing it will not do?

[strange jasper]

the wiki has a page on 5 way splitters. let me see if i can dig it up.

https://satisfactory.gamepedia.com/Balancer

[digital halo]

I only need to feed 2 assemblers not 5

[strange jasper]

well, you need one to have 3 parts and another to have 2 parts. that adds up to 5.

split it 5 ways evenly, then merge the lines into 2.

3 on one, 2 on the other.

[digital halo]

oh i get it now

[strange jasper]

or you can cut the load balancer and just deal with a manifold.

[digital halo]

ok

manifold?

[strange jasper]

let me explain this, it takes some thinking to get used to.

you know how fast your machine produces things? 5 parts. the first machine consumes only 2 parts, and will eventually fill up. the next machine will eventually take the overflow from that machine's input and receive the remaining 3 parts.

basically, just 2 splitters in a line.

it will start at low efficiency, but over time they will both receive the exact amount of parts necessary.

[digital halo]

ahh so i don't really need to do all that fancy thing I guess - just use a single splitter and it should do the trick?

[strange jasper]

yeah. here, look.

https://satisfactory.gamepedia.com/Manifold

[digital halo]

no i get it what you mean

ok thank you

[strange jasper]

no prob.

[copper imp]

so after the update, which is more valuable: 1 ingot of copper or 1 ingot of iron

because of copper now has many uses I do think replacing iron with copper in low tier items migh not be logical as before, any thought?

[iron prairie]

If you're talking about alts like stitched iron plate, I have two words for you.

Iron Wire.

[sullen cloud]

As long as you don’t aim for maxing out a production like 800/min turbo motors, copper shouldn’t be a problem

Esp if you produce copper ingots in refineries

[copper imp]

what do you think about copper rotor

is it still the best recipe for rotor?

well meant efficent about resources when I said best

[sullen cloud]

I don’t use it because my motor production isn’t close to copper resources

[copper imp]

would you use it if it was not a concern?

[iron prairie]

Copper rotor I have written down in my notes as a "maybe use" alt.

With steel rods + steel screws, you can do the default rotor recipe with pretty OK steel consumption: it's a bit slow relative to copper rotors, but sometimes copper just isn't hugely available.

[copper imp]

well before U4 all the copper on the map was more than enough for everything that uses it, now I did not yet calculated on my new save

before I calculated everything for max turbomotor now I don't even know to max which:D

[iron prairie]

Maxing turbomotors in U4 is now more of an issue with quartz than bauxite. There's simply not enough quartz on the map to max aluminum production: Satisfactory Tools winds up actually sending most of the aluminum scrap through pure ingots if you try to max turbomotors.

Otherwise: maxing quickwire takes up 27.6k copper ingots, which is a decent fraction of the map's copper supply. Beyond that, you can use copper to stretch aluminum (alclad casings, alclad sheets).

If you're at all worried about running out of copper, I'd try to avoid recipes which use copper to substitute for even more common iron.

[copper imp]

Thanks for your time. I got my answer:D

[torpid robin]

Copper

[mystic crag]

Question, has the maths for nuclear fuel rods changed?
E.g. has the alternative fuel rod items/pm changed?

[torpid robin]

All nuclear was reworked

[mystic crag]

Okay, just thought I'd see if the fuel rod revipies had as well.
Thank you!

[fierce ruin]

nukes got nuked

[torpid robin]

Not really lol

They got a buff imo

[marble coral]

is this a good place to ask what alts are the best for making [x]?

[iron prairie]

If you want a quick, automated answer, try this: https://www.satisfactorytools.com/production

[marble coral]

well yeah, im on that but Im trying to wrap my head around which is good and which is bad

[torpid robin]

I found the best way to see is just swap between the alts on that to se e

That’s how I learnt what’s good and what isn’t

[marble coral]

I was looking at pure recipes but it seemed like it just complicated it more than it was when there were no recipes

[torpid robin]

Well pures are always good to use

There is an argument for copper alloy though

[swift sage]

I'm looking for a math wizard

Transposition of formula

Y = X^n

How do I get n=?

I can go the other way to get X = Y to nth root, but I don't know how to get n=

[oblique hollow]

N = Log (Y)

In theory

But its the X-Log

Sooo, theres some transforming needed

Gimme a moment

[marble coral]

I managed to dumb it down to this, which doesnt look awful, but will probably make my save look a lot worse, if you know what I mean

[oblique hollow]

N = Ln(Y) /Ln(X)

[swift sage]

it's been about 15 years since I've used this maths you see

thank you

[oblique hollow]

Np

[swift sage]

Ln is Natural Log, right?

[oblique hollow]

Yeh

[swift sage]

which IIRC is different than just Log?

[torpid robin]

Oh . I mean it’s a prettt small set up . You could get away with normal ingots

[swift sage]

how do I do that on my calculator? ¦3

[oblique hollow]

It should work with the normal log too

[swift sage]

okay, thanks

[marble coral]

¯_(ツ)_/¯

[stable rapids]

ln is log base e i think

[oblique hollow]

Yeh, aka natural logarithm

y = e ^ something
Something = ln (y)

[fierce ruin]

it makes calculus "easier"

[oblique hollow]

General formula for Y = B ^ X
Log Base B of (Y) = Log(Y) / Log(B)

Thats true for any base

[swift sage]

it would probably be easier to just ask the final question directly...

[oblique hollow]

Yes

[oblique hollow]

What even are you trying to do

[swift sage]

Given a value for probability for an attempt, how many Attempts would be required for a specific percentage chance of success?
Formula is:
Percentage = Chance^Attempts.
rearange the formula so that Percentage and Chance are variables to work out Attempts.

Chance and Percentage are both numbers between 0 and 1.

Attempts = Log(chance)/Log(Percent)?

[oblique hollow]

Should be right

No wait

[fierce ruin]

hmm probability shouldn't that be a distribution

[oblique hollow]

Attempts = Log(percentage) / Log(Chance)

Exactly the other way around

[swift sage]

My friend said Thank you.

[trim current]

Site?

[fierce ruin]

man I'm out of shape mathematically.

[marble coral]

https://www.satisfactorytools.com/

[trim current]

Thanks

[marble coral]

no problem

[swift sage]

using my numbers.
Attempts = Log(chance)/Log(Percent) = 24
Attempts = Log(percentage) / Log(Chance) = 0.04
So... I think it is the one I said.

since I already know the answer SHOULD be in the 20s

by doing the formula the other way around with guessing numbers

[trim current]

I always do this on A4 paper I have a lot of paper here

[swift sage]

unless I'm doing it wrong

[manic oak]

@swift sage The answer that was given may be correct, but it sounds like you're talking about the Binomial Distribution: https://en.wikipedia.org/wiki/Binomial_distribution

[fierce ruin]

^

[fading valve]

Lol....... that moment when you realize you can replace 80 packagers and 40 refinery's with 24 blenders in favor of its diluted fuel alt....

[swift sage]

my numbers are wrong, that's why x3

[manic oak]

It describes the probability of getting k successes after N attempts. If you have a desired probability of success after N attempts (call it P), you would have P = k/N => k = NP

[iron prairie]

Well, the big problem with diluted fuel, you see, is...

Nothing, really. Simpler, less power, fewer buildings, probably less floor space, and doesn't alter the input:output ratios.

[fading valve]

yeap hahaha

To bad I just finished the packaging and everything 3 days ago

[swift sage]

anyway, thank you for the help, I'll figure out my maths.

[oblique hollow]

Makes sense, actually. If this is a non-exclusive event, then base chance times number of attempts should deliver the probability...

[iron prairie]

That said, when U4 lands, I'm keeping the old DPF setup at my turbofuel plant, for the simple reason of "CBA to change it all over for a modest reduction in power consumption".

[fading valve]

yea, the power difference is small. like 200mw

Would be good if you suffer from lag or something like that though

[iron prairie]

I mean sure, I could save a whopping 400 MW at my turbofuel plant when it happens (5 MJ saved per fuel, 4800 fuel/min = 80 fuel/sec), or I could keep what took me days to build.

[fading valve]

The blended alt is nice for getting a smaller scale TF plant up and running though

output is less but much faster to set up and use.

[manic oak]

I was not expecting an element of my research to pop up in the Satisfactory Math and Meta chat, so after seeing that question after working with the Trinomial Distribution all day, I as like "Oh baby, let's go"

[fierce ruin]

I've already forgotten everything from ap stats rip

[swift sage]

Got it.
the formula is actually
1-Percentage = (1-Chance)^Attempts
so
Attempts = Log(1-Percentage) / log(1-Chance)

[oblique hollow]

Oh, the counter-chance

[swift sage]

Given desired percentage is 90, and chance is 7.2...
Log(1-0.9) / Log(1-0.072) = Log(0.1) / Log (0.928) = 30.8~

Given a 7.2% chance of success, 31 attempts brings you up past 90% chance of success

[oblique hollow]

Heck why not make this a normal distrubution already xd

[fierce ruin]

normalCDF()

[swift sage]

considering the person wanting it actually wants a value of something like... 5000 of the item, simple 5000/0.072 exceeds 10 thousand, meaning Pseudo-RNG would probably just give the desired 5000 items anyway.

[manic oak]

Are you looking for the number of attempts it would take to have P percent chance of getting 1 success?

[swift sage]

it's what he's asking for, even though he actually just wants to know how many attempts to get the desired number of items.

[fierce ruin]

random sample time

[iron prairie]

For large numbers of successes, it appraoches (# successes / success chance)

Otherwise, use the binomial distribution and figure out the 95% confidence bound. You can't ever guarantee you'll get N successes, but you can figure out how many trials will give a 95% chance of at least N successes.

[swift sage]

it's for a game that allows you to literally just say "Attempt this X number of times", then the game repeats the rand function however many times you set that to... so... computer Pseudo-random number generators will in theory not be random when you run it in excess of 10k times

[fierce ruin]

rigged

[iron prairie]

Any PRNG which repeats itself after a mere 10,000 cycles is RAND levels of bad.

[swift sage]

Yeah, I tried to tell him what he's trying to do isn't necesary, because it's computer Pseudo random, and if he's rolling 5000 attempts, even on a 7.2% chance, he'll probably end up with around 400 items.

[manic oak]

What's the chance of success, and what's the desired percentage of success after N attempts?

[swift sage]

his numbers.
7.2% chance of success, and he wants a 90% chance of success

[fierce ruin]

how?

[swift sage]

those numbers using the formula worked out earlier tell me 30.8 attempts

(even though he wants over 5000 items) which means he's telling himself he needs to do... what, over 150 thousand attempts

in theory, just doing 5000/0.072 (which is only around 70 thousand) will probably give him around 5000 items, due to Pseudo RNG

[oblique hollow]

The heck

[swift sage]

Yeah, considering what he's trying to do, working out that 90% success rate thing is... probably overkill

[manic oak]

I'm actually seeing no solution xD

[swift sage]

I ran 1000 attempts:

[manic oak]

Just to be clear: a 90% chance that he gets 1 success after N attempts?

[swift sage]

Game is pretty... not-really random

[fierce ruin]

nvm

[swift sage]

yes

[oblique hollow]

Is this for some gacha game

[swift sage]

SpaceTravelIdle, made in Unity.

[fierce ruin]

if it's one success that makes the distribution geometric right?

[swift sage]

yes

[icy pumice]

so which of these is better? https://imgur.com/a/qgUqBb0

[fierce ruin]

the other alt

[icy pumice]

u should check the link for both images

[fierce ruin]

not shown

[icy pumice]

but it uses rubber

[torpid robin]

There is another

Uses steel

I think ?

[icy pumice]

so I wanna get the other one?

[torpid robin]

Heavy bolted frame maybe ? Shit can’t recall tbh

[icy pumice]

ill look up in docs

[fierce ruin]

it replaces screw with concrete

Heavy Encased Frame

[manic oak]

Ah, I see what I was doing wrong. I'm finding the probability that I have ONLY one success, which decays to zero as the number of attempts goes to infinity

[torpid robin]

Arrh yea that’s the one

[swift sage]

calculate the chance to get one:
7.2%, I would have done 100/7.2 = ~14 attempts.
he's trying to run it through that formula which for 90% confidence yields ~31 attempts.
from there, he wants to then multiple that value (31 attempts) by how many cards he needs (around 5000)
which means my method would yield around 70k attempts, but his around around 154k attempts.

[torpid robin]

Ok isn’t all that shit off topic? It has nothing to do with satisfactory ?

[swift sage]

Well, it's maths... this is the maths channel...

[torpid robin]

This is maths channel for satisfactory

Not random crap . People try and ask stuff and it’s flooded with shit

Read description

[swift sage]

I didn't know where else to look when I started trying to transpose the formula, I needed to ask a person, rather than google.

[fierce ruin]

wouldn't it be better to use a larger sample to reduce standard deviation

[swift sage]

and... people then asked why I needed to know

[icy pumice]

so i wanna get the Heavy Encased Frame thing?

[swift sage]

anyway, I'm done now, I'll leave.

[torpid robin]

Yea I get that . I fully understand . But you could always move it off topic now though

[torpid robin]

Yea it’s the much better one

[icy pumice]

okay

[fierce ruin]

you can use default rn if you want

then go hunting

[icy pumice]

thanks 😄

[icy pumice]

yea that makes sense

although then I need to redo the complete factory

[fierce ruin]

stockpile for trains

[torpid robin]

Yea as rso said . I build my 1st couple with standard just to get something going . Then go get the better one for mass production

[icy pumice]

well yea thats why I need them

okay

i also need to build a factory for computers, uh

[torpid robin]

Cat computer