math-and-meta

[vapid gorge]

what are you trying to understand?

[normal latch]

so basically, the system should have the right numbers with 2 water extractors combined with the output of the R2 (aluminum scrap refinery)

[knotty hornet]

Yes

[normal latch]

that totals 360 water, which is what the two R1 (alumina solution refineries) need

i just wanna make sure this system actually works since pipes can be tricky

[vapid gorge]

it's not just about 'right numbers'. Fluids flow both ways
and when merging fresh and waste it's really easy to have back flow that stutters and clogs the whole system

this may seem really strange to say but pipes are not belts

[normal latch]

so if i put a pump on the output of R2, it will be fine?

[vapid gorge]

no. not nescesarilly

back flow and clogs can still happen

is it possible to work? sure

but if thigns are a little off, or a little hiccup happens it' can clog and you'll have to manually restart it

[normal latch]

what if i have 3 water extractors instead, so 2 with proper clocking to supply the 180 to the first R1

[vapid gorge]

and that's if it works at the start

[normal latch]

and 1 with clocking to combine with the 120 output from R2 to supply the second R1

[vapid gorge]

is the waste still connected to the fresh in this scenario?

[normal latch]

yea it connects lol

[vapid gorge]

then you haven't changed it at all

[knotty hornet]

Bro, it's gonna work, just try it

[normal latch]

im trying to learn how to make it work, ultimately

[knotty hornet]

You have literally nothing to lose

[normal latch]

tru, but its late, i wont try till tomorrow

[vapid gorge]

the most reliable way to mix waste and fresh is using a VIP junction.

but those work off back magic and sometimes don't work when built

[knotty hornet]

And if it doesn't, you'll learn why

[normal latch]

im well aware, by the way, that the "easy" way would be to not combine waste + fresh water

im interested in learning how to make fresh + waste water combined

[vapid gorge]

it's not 'easy', it's reliable.

[vapid gorge]

there's no trick to it.

use a powered pump or VIP. Hope it works

if it doesn't work. Rebuild. Try again. Hope it works

rince and repeat

[normal latch]

guess we'll see how it goes manana

i'll try my initial setup first

[vapid gorge]

oh and don't have any repeating numbers in the clocking of the machines

[normal latch]

initial setup doesnt use clocking, but noted

[vapid gorge]

repeating decimals will eventually kill a direct feed system that isn't using a VIP

[knotty hornet]

🤔

[vapid gorge]

usually

[knotty hornet]

What's wrong with repeating decimals?

Also, would it make a difference if you typed as its fraction form? (31+1/9 as opposed to 31.1111)

[tired gazelle]

Is this doable in a world that I also want to have a maxed Nuclear Power Plant in?

mainly worried about the large sulfur draw

[brisk urchin]

i have also just recently come across this issue, here is my version of how i am gona build it

[vapid gorge]

subtract the sulfur used in this with the world total and find out

[tired gazelle]

this is good

[brisk urchin]

thank kyo29 for it

[tired gazelle]

would you happen to have a link to the calculator preset to this recipe/

[brisk urchin]

#1414244918376927383 in there i think

dont got much time rn sry

[floral mica]

There's only about 10,000 max sulfur in the world, so depending on recipes you used, you might not have enough. You could loop the compacted coal byproduct back into the input to eliminate some sulfur, or look at the nodes on the map you have available and use those

[oblique hollow]

Maxed out nuclear should at best use one third or half of all sulfur on the map

Besides Turbo / rocket fuel, instant scrap and batteries (and maybe the leached alts) theres not much you can actually burn sulfur on.

Turbo or rocket fuel are actually the main offenders when it comes to sulfur usage

Very sulfur hungry

[crimson moat]

turbo more than rocket, funnily

[oblique hollow]

Because rocket can skip turbo and partly because of that feedback loop

[unique cypress]

3108 is the number I remember, but I don't remember for which recipe pathway 😂

[oblique hollow]

Yeah that about checks out

For all purposes, nuclear really isnt that sulfur hungry

[crimson moat]

Even with neither of those, 60 (2GW) turbofuel makes 100 (6GW) rocket fuel

The byproduct compacted coal just expands that gap even further

[oblique hollow]

Ah, if you go with sulfur per MW / GW then yeah thats a different metric. Much more efficient when viewed like that

[crimson moat]

per unit of fuel rocket takes much less than turbo too, even without feedback.

[oblique hollow]

Rocket truly is too good for its own good

[crimson moat]

for turbofuel's good 😄

[oblique hollow]

For anyone's good.

Solution: buff Ficsonium xd

[unique cypress]

Just checked, 3108 is for 50.4/22.4

50.4/12.6 uses 2856

[oblique hollow]

i dont get the second number after the /

is that plutonium

[crimson moat]

plut rods yes

[oblique hollow]

alr

[crimson moat]

those are the numbers for rod and fuel unit w/ all uranium

[oblique hollow]

(not counting converter alchemy uranium i assume)

[unique cypress]

No, that's from 2100 uranium

Max wasteless uranium w/ conversion is like 200 rods

[crimson moat]

my 51.2 / 12.8 / 64 build (all uranium to minimum fics, overprovisioned 63/64) uses up to 5212 sulfur, but 2225 of that is sulfuric acid for caterium which could be pure.

Just putting it to use because my mapwide plan was leaving most of it untouched

[oblique hollow]

Leached Caterium? Eh, still neat and still only 50% of all sulfur

So yeah nuclear really doesnt waste sulfur

[crimson moat]

only really worth worrying about sulfur if you are trying to make the most power possible i think

[oblique hollow]

That would likely involve mixing nuclear and rocket fuel

[crimson moat]

yeah and power augs

KYO gave me this a bit ago 😄

15.32 TW waste free

Ionised is good enough to poke its head in a little. It's much stronger relatively when power augmentation is in play because instead of being somewhere around the break even point, the power augmentation makes it produce like 5x of the power that it costs

Ficsonium is strictly worse than other wasteless options in all circumstances

[unique cypress]

Actually, I need to check that plan lol

I don't remember whether I made it before or after I learned how to optimise for production

[crimson moat]

ye i deleted a bit cause i don't understand it fully 😄

[unique cypress]

So it might be optimized for net power instead

[lone jewel]

not if you actually enter the clock speed as a fraction 🤔 the machines may only display a few decimal places, but it's saved with 20 decimal places

[unique cypress]

Uhh source?

Last I checked the wiki, it said that 4 decimal places are saved

[lone jewel]

i don't remember where it was specifically, but someone took a save editor and the clock rate had 20 decimal places of precision when the rate was entered as a fraction

[vapid gorge]

as far as every tech geek and math geek on this server has poured through, even when inputing fractions, it'll only round to the 4th.

Happy for them to correct me but that's been the concensus for a long while

it's possible something has changed and I haven't heard about it? I don't stay rapt for info like that but it seems like something that would have at least been tossed around if it had

[lone jewel]

i've seen this several times recently

[unique cypress]

Hmm, in scim when you go to update clock speed, it does show more than 4 decimal places

[vapid gorge]

by whom?
and you can always input more numbers into a field, doesn't mean it recognises them

[lone jewel]

the screenshot i saw was a json export from a save editor, that confirmed the fractional input in the game was saved with 20 decimals of precision

[vapid gorge]

oh I totally believed you you could input numbers beyond 4 in a save editor. But that has no bearing how many digits the game acknowledges

[unique cypress]

In have not touched the clock speed in that window

But on the other hand, a machine set to 90 shows this

[vapid gorge]

I still believe you, but I'm just basing it off what code geeks have said in the past and they've been reliable on all the other things over the years. That no matter what the decimal input says, that the cycles will only acknowledge up to the 4th

I'm not going to pretend I understand it. Coding and programing is far from my knowledge set

Maybe ask Baldur next time you see him. He's probably the most techy person on the server

[unique cypress]

Yep, confirmed, that's just the floating point representation of 0.9

So it's probably SCIM displaying it as a floating point number, regardless of how it's actually stored in-game

So this confirms nothing

[vapid gorge]

Baldur is often about, just @ him next time, or even now, I don't think he has notifications on so don't think it'd bug him

[unique cypress]

Someone would need to interrogate the actual save file or set up a test to show the discrepancy

[lone jewel]

you'd have to enter in a clock value as a fraction that repeats, checking a 0.9 clockrate doesn't prove anything

[unique cypress]

It proves that SCIM displays a floating point value

Which means it's not a reliable source of what the clock speed actually is

[lone jewel]

disagree

[unique cypress]

The actual save file would have to be checked manually, not with a save editor

Or at least not this one

[lone jewel]

if it shows the repeating value when enterer as a fraction that causes a repeating value, it clearly is accurate

otherwise the value saved wouldn't be a repeating value

you can check it against that convertor

if precision is lost, the stored value won't match what it should

[unique cypress]

Tf you mean "disagree"??? SCIM shows the floating point representation of 0.9. That means that at some point between me typing in 0.9 into the box in the game and SCIM displaying it, 0.9 got converted into a floating point number. I have no clue when that happened and I have no proof that it's not SCIM itself doing that

Therefore SCIM is not a reliable way to check what the actual clock speed is

[lone jewel]

you're missing the point

0.9 IS that value above

[unique cypress]

It could be the game rounding it and SCIM displays it exactly but I have no way of knowing that

[unique cypress]

0.9 is 0.9. 0.899... is the floating point representation of 0.9. which one is actually saved in-game?

[lone jewel]

if you enter in a value as clock speed in game that has a repeating number, check that precise number in the IEE 754 convertor, if precision is getting lose in the game, the value displayed in SCIM won't match what the expected value is

[lone jewel]

there is no difference between those 2 numbers, that's my point

they are the same

0.9 exactly is not a possible number representable by a floating point number

[unique cypress]

Oh, I get it

Gimme a sec

I'll have to see if I have any machines set to 33.3333% or similar

Cause I'm on my phone in the doctor's office rn lmao

Either way, 0.333333 is 0.3333329856395721435546875

And adding extra 3s (0.3333333333333333333333333) becomes 0.3333333432674407958984375 in FP

[oblique hollow]

0.333333... in float is 0.333333343267440...

oop im a bit late lol

[vapid gorge]

nah, convo over time

[unique cypress]

Ok, got a miner likely set to 20/min (60/min at 100% clock speed) and it's:

[oblique hollow]

float moment

[unique cypress]

I have definitely not typed in 33.3333% into that one. I've either put in 100/3 or 20/min into the output field

[lone jewel]

what would be more interesting is something like 190 + 10/21

[unique cypress]

It'd suggest that clock speed gets fully transferred into SCIM

The question is: does the game use all that precision?

[lone jewel]

it would make 0 sense that it wouldn't

[oblique hollow]

I can only assume the game calculates a float and stores that - thats it

[knotty hornet]

Hey wow, my question finally got answered cuz someone other than Cobalt finally got involved.

[edgy leaf]

what was the question?

[knotty hornet]

Thanks guys, I asked about decimals and fractions and stuff last night, but Cobalt is such a -----, they have me blocked for some reason.

This was an insightful read

[oblique hollow]

Oooh hey Laura, good timing.

Can you do me a solid and send over that save with the cursed bottom feeding you showed me once?

[unique cypress]

That seems likely. But now the question is: what gets used to calculate the new cycle time

[oblique hollow]

Cycle time is most likely default cycle time for this recipe / said float

[unique cypress]

And even if it's all the decimal places in the clock speed, how many places does the new cycle time get?

[unique cypress]

Well, considering it is (used to be?) on the wiki, the belief that only 4 decimal places get saved has to have come from somewhere. So either that was true and has changed, or rounding happens somewhere

[vapid gorge]

I've stated a number of times why you're blocked. Not listening fits with everything so far.

[knotty hornet]

Sometimes, I hear his voice in the wind

[oblique hollow]

Regardless, if you were to assume the game actually only uses 4 decimals for the math, then a cycle time of 1 second with 33.3333% clock speed would become 3.000003129 (roughly)

which gives you 19.999979 cycles per minute

Assuming it is actually only as many decimals as the UI fits

[knotty hornet]

I imagine that's just what it shows though, right?

[oblique hollow]

I cannot say

I do not know

If it actually was that though, then you likely still wouldnt experience issues for a long time

[knotty hornet]

Right, that's what I just gonna ask, does this ever actually manifest, if we're talking disagreement after 7 or 8 decimals?

[oblique hollow]

My most common answer to this is "if an error happens, it usually takes a couple thousands of active playtime hours to manifest"

[knotty hornet]

So, practically speaking, no, it won't manifest.

[oblique hollow]

for practical purposes, no, it wont happen.

If anything ever happens, its likely one missing or excessive item every like 4000 hours, if we assume the current example

[knotty hornet]

So the timescale on which it happens is so large and the effect so negligible, that it's effectively a non issue

[oblique hollow]

Thats my take on it, yes

[knotty hornet]

Good to know there is nothing wrong with decimals again.

[oblique hollow]

My estimates always put it at a few months to a couple years of active playtime for an error to manifest itself in actual interruptions, albeit small ones

Most people barely clock in at the required time for a given save

do you have a 4000 hour old save file?

[knotty hornet]

Yeah, so this is a case of "theoretically sure, but in practice, no"

[knotty hornet]

I don't think i have recorded 4000 hours on any single game

[unique cypress]

The absolute maximum discrepancy is 1 item every 27.7 hours

[oblique hollow]

Assuming 19.999 vs 20 means you already are missing one item?

[unique cypress]

If the clock speed is indeed saved with 4 decimal places of accuracy, then the maximum rounding error is half the 5th decimal place, or 0.00005%

0.00005% of 1200/min is 6e-4 items/min extra/missing

Or 1 item every 27.7h

That's the absolute max

Most of the time, the rounding error is not gonna be near the max, and 1200/min is the maximum input/output

[oblique hollow]

99.99995% accuracy is a pretty high standard if it were anything else

i think that counts as 5 sigma

[agile junco]

Is there a mod which tracks a machine's duty rate?

[unique cypress]

That's 5.026 sigma lol

Spot on

[knotty hornet]

So the accuracy is so good, the variances are literally negligible.

[unique cypress]

And that's assuming the accuracy is 4 decimal places

[knotty hornet]

Which it's almost certainly higher, yes?

[unique cypress]

In reality, you'd have to calculate relative accuracy so 5e-5/x

But it's 5e-5 for 100% speed

[unique cypress]

Idk how the cycle time is calculated from clock speed and if there's any rounding happening there

It's also possible that the cycle time is quantized just like the physics are

[knotty hornet]

That would make sense

[agile junco]

Why wouldn't it just be 64 bit floats?

I don't know that much about game design.

Or even 32-bit...

They have weird non-communitive properties, which I assume might cause issues.

[unique cypress]

Idk how Unreal handles that, but generally making cycle time that's not an integer number of game ticks is harder than a cycle time that is

At least if I was coding from scratch, I'd base everything on a number of ticks, and then just make sure the ticks happen x times per second

[agile junco]

Yea, you want to decouple the logic from real-time to some degree.

It would be funny if overclocking your computer actually worked like overclocking your factory. But it's been a while since I saw a game which was tied to the system clock in some way.

[mossy ibex]

making rocket fuel has turned out to be much tricker than I thought. It looks easy on paper to just make nitric acid and combine it with turbofuel, but given the reality of my logistics it's been spicy

[unique cypress]

It's generally best to isolate your power plant's fuel supply from all of your other production

[crimson moat]

could concievably cause a problem, but yeah, extremely unlikely to cause a noticable problem

just for something really sensitive like a merged alumina setup or a single-input sushi belt it could drift in a very slow yet eventually problematic fashion due to floating point accuracy loss, which might break it after thousands of hours.

My longest save is i think around 840 hrs? 😄

even if it would be fine with perfect numbers

[mossy ibex]

yeah I basically did that. My nitric acid is overproducing but the portion of teh factory that goes to rocket fuel is separate machines with separate lines that get prioritized over the other section. I'm more talking about the literal pipework, etc, to get from point A to B

[knotty hornet]

Rocket fuel is a lot easier with the Nitro blender recipe.

[unique cypress]

Sushi shouldn't rely on machines to regulate it tbh. You should either make it dynamic with prio splitters and mergers or manage the i/o with belt limiters or something

[crimson moat]

Yeah, not really something that you should do in any case, but technically you could make it work

Whereas if the game uses float for clock speed and you don't account for that, it could cause a misalignment of inputs which trends towards a deadlock

good building practices will generally be able to absorb some variance without causing any noticable issue, especially fraction of a percent level variance.

[mossy ibex]

I bet you're right, but I already have a working turbofuel factory dumping 750pm right into my rocket fuel factory, so shrug

[knotty hornet]

Well yes, the easiest setup is the one you already have 😭

[crimson moat]

Nitro is the easy way, inputting turbofuel is the efficient way and it's a pretty solid way to do midgame power. Make enough fuel and later turbofuel in P3 that you can later redirect it into your rocket fuel production, and you increase your power by like 3x

[mossy ibex]

6x actually

in my case -- 750 turbo becomes 2500 rocket in my set up, and instead of 40 OCed fuel gens, I'll have 240

[crimson moat]

I'm getting 2.1x conversion rate of turbo into rocket which is 3.79x power, are you slooping or something?

key point though: no additional sulfur input

[knotty hornet]

Is doing the turbofuel chain actually more efficient? I'm seeing that that standard chain uses more sulfur and coal per unit of RF produced. huh, i guess it only saves on coal and a tiny bit on fuel.

[mossy ibex]

yes very slooping

with fully OCed and slooped blenders, I could eat 750 turbofuel with 5 blenders each producing 500 rocket

I'm actually doing 10 blenders at 125 but it's hell sloop expensive so I'm thinking of it as temporary

I'm basically temporarily "wasting" 20 sloops to save 1000MW of power

[crimson moat]

In order of Sulfur efficiency:

600 sulfur into nitro rocket fuel = 900 rocket fuel (54 GW)

---

600 sulfur into regular turbo fuel = 25GW turbofuel, available early in P3.

If you make that into rocket fuel, you get 1578 rocket fuel (95 GW)

---

600 sulfur into turbo blend fuel into rocket fuel with recycled compacted coal via the regular turbofuel recipe gets you 2526 rocket fuel (152 GW).

---

So you can directly turn your P3 turbofuel into rocket fuel and it's not that bad. It costs you 60% more sulfur to make the same energy if you make the turbofuel purely on the P3 recipe, but the sulfur cost isn't prohibitively high (1200 sulfur for 180GW worth of rocket fuel is fine)

[knotty hornet]

I skipped turbofuel entirely, so I guess that's where my problem is.

[crimson moat]

Skipping it is a good option

but if you do it, you can build in a way where you don't have to tear down everything and clean slate to make rocket fuel - either reasonably efficiently, or max efficiency.

There are still those 17 turbofuel refineries at the bottom.

In P3 you use sulfur to make compacted coal.

If you're converting for P4 max efficiency, you would redirect that sulfur into Turbo Blend Fuel, and run your old refineries from byproduct compacted coal instead.

---

but just straight up using the regular turbofuel recipe and then making it into rocket fuel is a fine option too, since sulfur efficiency doesn't really matter and it's not that far off. It's not like it uses 5x as much.

I think regular turbofuel uses more coal and less oil too, which can be easier or harder depending on location and scale. The Turbo Blend Fuel uses coke instead of coal.

For example at the blue crater you can get 1800 sulfur and around 2.5k coal / 2.5k oil using mk.2 miners

so the least efficient one will make >150gw with mk2 miners, and double that with mk3, limited by the amount of sulfur in one corner of the map (which is like 17% of the total map sulfur)

[knotty hornet]

Ngl, making power plants is my least favorite part of the game

And nuclear still seems not worth it to me yet.

[unique cypress]

The only advantage turbo has over regular fuel is the oil efficiency. And considering you can get 20 GW from 600 oil with just regular, I don't see a point in burning turbofuel, ever

[mossy ibex]

I also don't love power plants. It's funny though, there's no real difference, but my brain just loves there being a widget at the end and doesn't love it when there's no widget

[unique cypress]

Cause it's really not, unless you're aiming for 1+TW of power

[crimson moat]

It can be a decent stepping stone to rocket fuel

because if you make the diluted fuel for rocket fuel and run on that, you wont have enough power to comfortably reach p4

so it makes more sense to do 2 seperate powerplants, whereas 1 is actually doable via turbo

[mossy ibex]

yeah my overall plan involved diluted -> turbo for P3 for about 25GW of power, and getting a few basic P4 things set up so then I could expand to 125GW using rocket fuel. I'm doing the rocket fuel expansion and just turning on the blenders will put me more or less right at my current power limit before the new gens come online

as a bonus I had infinite packaged turbo fuel for my jetpack during P3

I'm using hover now mostly, but I'll use packaged rocket fuel once I can

[orchid brook]

only 210 points lol

[edgy leaf]

all in all this might actually be /totally/ negligible

I've noticed that when I start my game, my power grid fluctuates. that implies some machines stall, which they don't usually do.

if that's the case in any setup and not a weird quirk of mine, it would mean the error Cant even accumulate due to the bigger errors from startup

[oblique hollow]

i know that the game has trouble with keeping flow rates up after a game load, thats a bug i specifically tested for

it normalizes after, but in the time right after a game load, long pipes fall behind with their flow

[edgy leaf]

yea then the small rounding errors are probably totally irrelevant, not even negligible

[oblique hollow]

its other bugs that are more prevalent

[edgy leaf]

because the game restarts every 24 hours or something iirc.

[edgy leaf]

at least that's my cope, don't take it away lls

[oblique hollow]

speaking of cope, i dunno if you were keeping up over on pipe cord, but i believe we found an explaination and solution to pipe manifold issues, including that one infamous save of yours

[edgy leaf]

damn, I gotta take a look at that. I won't be at home for a while so it'll have to wait :/

thanks for informing me

[runic kraken]

this is going to look so cute when its built

[unique cypress]

it's gonna look 2x larger and more expensive than it needs to be because SCIM is a horrbile calculator 😂

[runic kraken]

which one should i use?

[unique cypress]

https://www.satisfactorytools.com/1.0/production

[runic kraken]

Unfortunately we couldn't calculate any result.
This can be due to many things: missing resource required for the production line, not enough resources for the requested amount, disabled recipes required for the product, etc.

i think im sticking to scim

[unique cypress]

you need to enable recipes

alts are off by default

and turbofuel is an alt

• alts will allow you to make everything much cheaper so you want them

[runic kraken]

what the fuck 💀

[runic kraken]

dont have hard drives for alts

[unique cypress]

the hell did you do to that calculator

I didn't even know that was possible

[unique cypress]

ah, maximize

[wind spade]

you know you can just press the blue share button and share a link? 🙂

[runic kraken]

is that a bad idea?

[runic kraken]

https://www.satisfactorytools.com/1.0/production?share=dVvsjjfgnj7L996T2pJ9

[unique cypress]

do not even look at the output in maximize mode. paste the calculated number back into items/min mode so it can calculate properly

[runic kraken]

3.14GW usage

not bad ig

positive 19.3GW cool ig

[unique cypress]

compare the resource cost of what you sent and this https://www.satisfactorytools.com/1.0/production?share=NJEmZmvxBn32j7JJYemh

[small cloud]

I'm not the best at math, if I have 120 iron ore/coal p/m and I want to split it into 3 foundries, how would I calculate the clock speed to make sure all 3 are running at 100% efficiency?

[runic kraken]

thats the exact same thing i ened up with after a few tweaks

[runic kraken]

each of them to 40 steel output

hope that helps

[small cloud]

Okay, but how do you get the calculation, I don't want to have to ask every time I need to figure that out 😅

[runic kraken]

1 steel

= 1 iron + 1 coal

so 120/3 = 40

[unique cypress]

[how much they'll get]/[how much they take at 100%] * 100%

[small cloud]

thank you <3

[unique cypress]

also, wtf is up with people typing p/m when they mean /min???

[small cloud]

I saw it on reddit, my apologies

[unique cypress]

literally nobody outside of this community does that and it's incorrect

[small cloud]

just thought that was how most people communicated per min

[unique cypress]

some people do. incorrectly. for some reason

[small cloud]

Gotcha, I'll keep that in mind as well <3

[runic kraken]

do ppl use load balancers with pipes?

or should i just use a manifold and prime the system?

[unique cypress]

theoretically you can, but I wouldn't recommend it

[unique cypress]

I manifold and not prime the system

[runic kraken]

is there any advantage to that?

[unique cypress]

lower chance of sloshing causing issues in my experience

[runic kraken]

sloshing?

[unique cypress]

fluid flowing back and forth along a pipe, not doing anything useful and reducing available pipe capacity

[runic kraken]

ah

i don’t understand pipes at all

idk how i managed to complete a EA playthrough

[brisk smelt]

nondirectional conveyor belt with a temper

[unique cypress]

@red tangle you do one of these things:
either a blueprint with 1 of each machine, a fully closed loop
or set of blueprints with 4 of 1 machine, designed to connect together with auto-connect 1:1:1

either way, you put in empty canisters, packaged fuel, or packaged water into one of the machines to make sure that gets pasted with the BP and you don't have to put them in manually or set up production to fill them. in my experience, 20 per loop is enough, but you may need more or less depending how long and fast your belts are

[red tangle]

Right I see how it works closed loop for each of the packagers and unpackager I’ll do this when I go on and finally finish thanks

[vagrant pilot]

Can someone help me with the math for my solo run?

I’m still very new and don’t know how math works

[unique cypress]

https://www.satisfactorytools.com/1.0/production should help

[agile junco]

Oh no, I'm hit the 1200/min limit on my plastic/rubber belt because I was rounding too many things up.

This is actually going to be a pain to fix.

Possible without adding external belts though, since like half of it is used to feed itself recycled materials.

[normal latch]

@knotty hornet my aluminum setup seems to work flawlessly (-:

[knotty hornet]

nice

[normal latch]

whatever concerns others had, i havent had any hiccups

just a clean 480 aluminum ingots from the scraps heh

[knotty hornet]

important to note, it doesn't scale up super well, you can start experiencing problems when you get near max speed on a pipe mk 2

[normal latch]

hmm why is that?

i was thinking to scale it up by one extra section (2 refineries alumina solution and 1 refinery aluminum scraps) since i unlocked faster belts to support it

[knotty hornet]

pipes do have weird things going on with them because of how they are coded, and water going in conflicting directions can cause backflow, stuttering, all sorts of werid things in larger pipe systems

so as you scale up, you will need to keep other things in consideration, or keep working with small blocks of machines like your current setup

btw, you should definitely tag Cobalt, he'd love to hear it

[normal latch]

each block is isolated so it should be fine

as in, the water is only between the three machines mentioned

and the second block has it owns three machines, with no connection other than the bauxite and coal being split on conveyors

and i gues the 2 water extractors per block

but those are isolated to their respective block

[knotty hornet]

yeah, keeping fluids in managable sections helps a lot

[normal latch]

@vapid gorge with no special setup, the aluminum system that mixes fresh and waste water seems to work flawlessly (-:

btw i didnt have to do anything special when setting it up, literally just connected it all and turned on power all at once (didnt need to prefill water)

[dusky bronze]

as long as you dont back anything up that will work fine, just remember to put awesome sinks at the end of the aluminum belts

a good safety measure for the future would be to do what they said and keep the fresh water refineries separate from the waste ones

[normal latch]

yep awesome sinks are there

and i merge the overflow of silica and scraps into said sink

then i use a priority merger to have that silica get used at my main factory (i train in the scraps and silica)

otherwise my main factory uses the quartz generated silica as secondary to meet the 400 extra silica demand

to convert it all to 480 iron ingots

does it make sense to aim for rocket fuel for my next big power boost or turbo fuel first?

i current run on 10 gw of power i think it is

but will need to scale that up eventually

[dusky bronze]

turbo fuel isnt worth going for if rocket fuel is an option

[normal latch]

ok i think i will get to rocket fuel fast enough then

[dusky bronze]

nowadays i think turbo fuel is better used as a cheap boost so you can get a bigger power plant online

if you've been watching kibitz's 1.0 playthrough basically that

[normal latch]

have not been watching it, though i like kibitz

i guess my thought was that turbo would make sense as an intermediary for me

since i think rocket fuel requires the turbo fuel anyways

[dusky bronze]

yeah but its better to use the alt that doesnt

although they use basically the same resources

iirc the best recipe is nitrogen + fuel + coal + sulphur although i dont remember the name'

[normal latch]

will keep that in mind

i think i need turbo still in some dosage for unlocks in mam at least

so maybe i'll just make a single refinery for that

[dusky bronze]

yeah

[unique cypress]

I'd say that's questionable. the alt is significantly more expensive in everything but oil

[normal latch]

how so?

keep in mind, i never made rocket fuel or turbo fuel yet heh still newbie

[dusky bronze]

might also want it for turbo ammo but thats if you like the rifle shooting really fast and completely covering whatever direction you're facing with bullets

[normal latch]

i wanna have a nasty rifle, yea

i explore a lot and die a lot as a result lol

i just got the hazmut suit so that should help

i have explosive rebar but it isnt as strong as i expected, still good at clearing trees lol

also cluster bombs werent as good at killing enemies as i hoped

[dusky bronze]

explosive rebar is baically just a nobelisk you can shoot iirc

i havent used it for a while tho (i ran out of it months ago and havent gotten around to making more)

[unique cypress]

it's faster to shoot than nobelisks

[dusky bronze]

yeah

dont have to charge

[normal latch]

i'll make a dedicated rifle ammo setup to upgrade the ammo

[knotty hornet]

i love making ammo factories

[dusky bronze]

i need to make a weapons factory

[normal latch]

i automated nobelisks which was nice

[knotty hornet]

i just shared some pictures of one i made recently

i made one that makes everything except nukes, and both filters too

[dusky bronze]

i've been using "magic" to get nobelisks because i havent had a factory for them for over a year

[normal latch]

btw for hazmut suit, does that mean with filters i can start ot mine uranium?

[dusky bronze]

yes

[normal latch]

to get it off the map lol

[dusky bronze]

iodine filters, and you'll want a lot of them

[normal latch]

woohoo theres a somersloop that i think is inside a uranium ore thing

[dusky bronze]

they burn REALLY fast when you're at max radiation

[unique cypress]

5/min is enough

[normal latch]

i'll automate those and rifle ammo before next exploration mission

gonna go in like a stacked up monster heh

[dusky bronze]

just talking about how much you want to have in your inventory, automating any amount would be enough once you get a couple stacks to start you off

[normal latch]

i still need the other filters too right?

one type for gas the other for radiation

[dusky bronze]

yes

hazmat + iodine for radiation, gas mask + the other filters for gas

[normal latch]

kk

[sour bison]

im going to go insane.

[fresh escarp]

not that bad honestly

finished this one the other day

[sour bison]

using screws

ew

[fresh escarp]

i don't mind screws

i used casted screws so it was less brainrot

[unique cypress]

automating project parts

ew

[sour bison]

...you want me to hand feed my machines for project parts??

[fresh escarp]

wait whha

[sour bison]

and make 2500 modular engines by hand??

[unique cypress]

that's what I do

a few hours and it's done

[sour bison]

yeah no

[fresh escarp]

based

did that my first half dozen or so starts

[unique cypress]

that's my phase 4 factory

[vapid gorge]

Jfc my dude. I did say they can work, I just explained the dangers and how common it is for them to fuck up. Also the need to manually fix things.

Unreliable doesn’t mean ‘impossible to work’. If that was the case I would have said ‘impossible’.

[unique cypress]

I see no point in doing anything else tbh 🤷‍♂️

[fresh escarp]

sink

[vapid gorge]

And I let people know they are unreliable because almost every day i help fix someone’s direct feed system that looks ‘fine’.

[unique cypress]

you can sink everything else and it's more than enough

[dusky bronze]

this is pretty basic tbh

dont worry about the future tho

[sour bison]

fear

[unique cypress]

you know you've reached peak Satifactory gaming when your plans look like this

[sour bison]

what the FUCK.

[fresh escarp]

yeah average

[dusky bronze]

average phase 5 factory

[fresh escarp]

there's just a lot of nodes that's why

[unique cypress]

it's the plan for a modded 10 TW Ficsonium plant

[dusky bronze]

what are you even using that much power for lmao

ig mods can do anything

[unique cypress]

well, the mods are just alt recipes

to make it possible in the first place

[dusky bronze]

ahh

that sounds fun for the future, which mods are you using?

[unique cypress]

my own lol

[dusky bronze]

ahh

[unique cypress]

they're on ficsit.app too tho

[dusky bronze]

what're they called?

[unique cypress]

links in my bio

[dusky bronze]

these look fun

[unique cypress]

there are fun things in there such us

[dusky bronze]

well done good sir

[unique cypress]

520 fully OCd refineries for copper

[dusky bronze]

some of these should be in the game ngl

gives much more options for phase 5 stuff

[unique cypress]

also, idk how this graphing software draws nodes because the matrices are on the far left and the ficsonium on the far right. that's gonna be fun to parse

[unique cypress]

thank you

cause that was exactly what I was trying to achieve

[dusky bronze]

in future saves if i can remember i'll definitely think about them

will probably do it for my ficsonium save

[knotty hornet]

i told you he'd love it

[normal latch]

whoa relax there lmao why are you taking offense to a non-offensive mention?

[normal latch]

wild lol

i thought he'd be happy or at least neutral

[knotty hornet]

naw, the trick is, he irrationally hates me, and so every conversation that reminds him of me irritates him

he could've been happy for you getting it to work, but ....

[normal latch]

eh, i dont wanna press him further, just gonna let it go and move on

[vagrant pilot]

How do I use this calculator I’m so confused

[unique cypress]

You're gonna have to be a bit more specific because there's like half a dozen to a dozen calculators for this game

[vagrant pilot]

The Satisfactory tools 1.0 one

[unique cypress]

What's confusing about it? You select the item you want to make, choose the amount, and it gives you the most resource efficient recipe combination for that output given the constraints (available recipes, resources, machines)

[dusky bronze]

in the production tab you can choose what things you want the factory to make, how many you want to make (can set a number or tell it to maximize from the available resources)

in the input tab you can change how much resources the factory uses, as well as add any inputs (stuff like aluminum or plastic or rubber or what have you)

in the recipes tab you can choose what recipes you want the factory to use/not use

in the machines tab you can choose what machines you want the factory to use/not use

and then it calculates everything for you

[vagrant pilot]

How do you change the value of how much ore?

[dusky bronze]

it optimizes for least amount of resources to get the most amount of items i think? im not very well versed in it yet

and then all the tabs below the 4 main ones show you basically everything you'd want to know about a factory

[unique cypress]

It calculates how much ore you need for the output you chose

[dusky bronze]

you can play with the numbers here if thats what you mean

[vagrant pilot]

Ahhhh okay thank you that part was unclear

[unique cypress]

It assumes all of the map's resources are available but tries to use as little of them as possible

[dusky bronze]

like if you dont want to use too much of one resource you can bring the available amount down to whatever you want

[agile junco]

Man watching a manifold of 30+ frames and heavy frames fill up is like watching paint dry.

Going to just lazily kick start it a bit

[normal latch]

i saw a manifold trick where you have a fast conveyor down the line and then the slowest conveyors (depending on how much speed is needed to meet the machine consumption rate) to feed the machines

it cuts the manifold wait time down by an insane amount

[vapid gorge]

iirc it takes longer doing it that way. Because more machines start consuming the products sooner. Most of the time feeding machines with slower belts takes longer #1138254823196725419 message

[knotty hornet]

Yes, if machines (or blocks of machines) are only taking in the amount they need, then more is available for other machines, so that system starts up faster but it takes longer to fill the machines inputs.

[vapid gorge]

longer fill time in this case of the graph means more machines stuttering longer

[knotty hornet]

No it doesn't

[agile junco]

Yea, I do that generally... but I'm limited by the amount of production.

[knotty hornet]

Inputs don't need to be full for a machine to operate, they just need to have enough to do the recipe again.

[vapid gorge]

that's what the graph simulator shows in the link. A longer total spin up time

[knotty hornet]

If the input rate is greater than the consumption rate for any given machine, it works at 100

[unique cypress]

Unless the side belt speed matches the machine consumption exactly, it's actually gonna take longer to reach 100% uptime

[agile junco]

wdym side belt?

[unique cypress]

The belt connecting the main spine of the manifold to the machine

[agile junco]

Interesting, that's not immediatly obvious to me.

[knotty hornet]

Hold up

Let's say you have a 780 belt full of screws, and a side belt of 60 going into a machine, which is using 60. Then 60 goes to the machine at all times, cuz it can't back up, to 720, continues through the manifold. Is that right?

[vapid gorge]

that's load balancing

[agile junco]

That's not what he's saying.

[old zephyr]

doing aluminum, the scrap byproduct water is being looped back in with 1000 water a min from extractors. my refinieries for the solution need 1500, which is the total amount from the byproduct + extractors, yet the soltuion refinieries are all getting full and backlogging the scrap refiniery, any thoughts on where im going wrong?

[agile junco]

Longer than what? using a balancer. In that case, it's not surprising.

[knotty hornet]

He said a side belt matching the speed of the machine's consumption, isn't that what I described?

[vapid gorge]

you're mixing fresh and waste water

[unique cypress]

Then that has instant startup, yes. But if the machines take let's say 45, and you use mk1 belts on the sides, it's gonna take logner to reach 100% than if you used all mk5s

[agile junco]

Side belt is the spine, not the direct feeders.

[old zephyr]

and this is bad why

[vapid gorge]

you often get back flow and clogs

[knotty hornet]

What???

That's wrong

[agile junco]

Oh I misunderstood.

[old zephyr]

so should I just add enough extractors for the refineries and package/sink the waste water?

[knotty hornet]

Why would that be the case?

[vapid gorge]

no just split your fresh and waste up

clock 1 group of refineries to just use fresh water

clock the 2nd group of refineries to onloy use waste water

[knotty hornet]

It would seem that would then be forced to fill up the machines in sequence

[vapid gorge]

it's just a little bit of basic math but makes it basically unbreakable

It's not impossible to direct feed waste water but it's often fiddly and can easily break

[unique cypress]

Because you get more machines consuming items, and therefore lower excess that can fill them up. So you get more output at the start, but then a slower rise to 100%

[unique cypress]

That's if you used smart splitters for every step. And that is the fastest

[agile junco]

The fastest should be balancing, and filling them all up at the same rate...

[knotty hornet]

I think i might see what you're saying

[vapid gorge]

practical example and ratio diagrams #1287987061361610793 message

[old zephyr]

ty, gonna have to add a machine or two but at least it will work

[knotty hornet]

But if each machine is still getting enough for itself to run and the total in to the manifold is > or = what that block uses, i still can't see the problem.

[unique cypress]

Imagine a more extreme scenario. The machines consume 59/min and the belt you used is mk1. They fill up at 1/min. With stack size 100, that's 100 minutes to fill. But because they're all pulling off 60 from the main belt, the last 2 machines are gonna get less than 59 until all the previous ones are full. Which takes 100 minutes.

[agile junco]

For a manifold though, I would think it's dependant on the number of splits in the manifold. If it's one long single manifold, then I guess it's a matter of the ratio between the inner and outer belt?

[vapid gorge]

unless you're working with a system that's completely over clocked you can just increase/decrease clockings

[agile junco]

Yea, but if you're only producing 61/min, it's basically unavoidable and the speed of the belt doesn't matter.

That's what I'm dealing with.

[knotty hornet]

This makes it make sense

That's what I needed, an extreme example.

Okay, so I get what you're saying, but then in that case, just always oversupply, I guess

[unique cypress]

And if you use a manifold calculator, and assume 10 of these machines (59/min each), it takes 9 minutes to reach 100% assuming infinite belt speed (so basically no belt limiting flow)

[knotty hornet]

Which would mean, like you said, using the fastest belts you can to fill it AFAP

[normal latch]

i mixed fresh and waste water in my setup with no issue, but mine is also small modules that are self contained

[vapid gorge]

yes, as mentioned, it's possible. but often breaks. In such an example that has been presented.
I don't recommend unreliable and shitty options to people.

[unique cypress]

Yeah, if you oversupply enough that you can saturate every single side belt, then it's instant again. But then you need an input of n*belt speed, instead of just n*machine consumption

[agile junco]

If you can supply 1000/min, but need 500/min, it's best to use a Mk6... that said, it may slow down other places if you also split that 1000/min.

[unique cypress]

So the fastest manifolds to reach 100% are ones using smart splitters to fill one by one

[agile junco]

So I generally make the belt as small as still fufills the manifold.

[knotty hornet]

But where do trees rank in this game?

Cuz i still imagine tree feeding is faster than manifolds

[normal latch]

what is tree feeding?

[crimson moat]

yes, they just scale more poorly

manifold is linear building count

you should directly split with 6 or fewer buildings, but for 60 it takes a lot of buildings and time to do that 😛

[agile junco]

Is this just because of the idle timeouts? I'm not sure I see why this is the case.

[knotty hornet]

Split one into 3, then 3 into 9, etc

[vapid gorge]

load balancing

[crimson moat]

Manifold fill time is machine capacity plus consumption on the way to capacity.

If you feed more machines a little bit, consumption increases but the amount of stuff needed to fill all of the buffers stays the same. Therefore fill time increases.

[agile junco]

I would have guessed load balancing is the fastest, but now I'm not sure.

[knotty hornet]

In my aluminum foundries, they are grouped into cells of three that split inputs and merge outputs, then 4 cells of 3 makes a block of 12, and I feed those through a 2-2-3 tree

[unique cypress]

The more machines are running, the more items they're consuming, the less items go to actually filling the manifold

[unique cypress]

That's a balancer, no?

[knotty hornet]

Is it?

I just call it a tree

[agile junco]

Oh, sure. Smart splitters reduce total fill time for that one manifold... but if the output of that manifold is powering another manifold, I don't think it changes the production, does it?

[knotty hornet]

Cuz it just branches

[normal latch]

all this stuff could be tested with the conveyor measurers

[crimson moat]

probably not, no. In that case having more production earlier in the warmup counterbalances the longer warmup.

Just talking strictly about time before reaching 100% production, feeding 1 at a time gets there sooner. Preferably zero, even (no power).

[agile junco]

Right. That's the part that I was confused with.

[knotty hornet]

You tell me, is that a balancer?

[vapid gorge]

or just with Math
which has been done.
esepcially since those belt counters aren't super reliable

the magic of math #1138254823196725419 message

[agile junco]

A 1-N balancer where N is even, is just a bunch of splits.

[unique cypress]

Draw a diagram I guess

From the description it sounds like it is

[agile junco]

I forgot you can split into 3, lolol

And I'm also wrong.

[unique cypress]

Not even. A number whose prime factors are only 2 and/or 3

[agile junco]

Ignore me, but yes, the idea is that you have the same amount of output on each line.

[agile junco]

Yea.

[knotty hornet]

Well, you can do 1:5 now too

[unique cypress]

That needs mergers tho

[agile junco]

But that involves mergers and a bit more thought.

[unique cypress]

So it's no longer "just a bunch of splits"

[knotty hornet]

Yeah, but taken as a block, you can do it, so factors of 5 are allowed too

Psh, alright, purist

[normal latch]

my strat is to manifold and just go do something else, assuming it'll eventually work lol

in the time the manifold fills up, you can explore or build other stuff hehe

[agile junco]

Yea, but I was being impatient tonight and just wanted to cram a bunch of extra product down the line from my depot.

[normal latch]

you could hand feed the machines to fill faster i suppose

[agile junco]

Anyway, this conversation has basically finished the fill.

[normal latch]

heh

[agile junco]

I did the lazy version of that and just put a merger in the main split and filled a container.

splitting up 31 parts of something is a pain.

This humble 12 heavy frame / min feels so good tbh.

[normal latch]

nice

[agile junco]

In the past I've had like 2/min and it consumed all my supplies while it was going and nothing was balanced correctly.

Now I can focus on cleaning up the design tomorrow.

Middle building is

Wish I could have it show the numbers in the node, not just the sum of each edge.

[normal latch]

you could make labels/signs in your building to display the numbers

kinda like a visual stamp saying "here is what i made"

[agile junco]

Yea, when this whole thing is done being tuned, I will do that.

[knotty hornet]

after reading the wiki page, the things i am describing as trees are, in fact, balancers

[agile junco]

I just know I'm going to need to adjust things because I've been rounding up a lot.

[knotty hornet]

in which case, i am also on Team Balancer over Team Manifold

[agile junco]

I built my first 3:4 balancer the other night after @unique cypress insipred me to learn about them.

[normal latch]

im team manifold

[agile junco]

They just take up a lot of space and don't generally help with thoughput once things are going.

[knotty hornet]

that's a pretty fun one, send a screenshot and tag me if you can

[normal latch]

i also saw a vid of a compact design for foundries (that has no clipping)

so im team compact too heh (with no clipping)

[unique cypress]

I am team both, because both have their uses

[agile junco]

@knotty hornet

[normal latch]

i havent found a use for balancers yet, besides like basic "it happens to split in half or in three perfectly"

but im also not far along yet, only pushing into tier 7-8 now

[unique cypress]

Balancers' main use for me is feeding manifolds
And loading trains

[agile junco]

It's not the smallest possible, but it's nicely (mostly) symmetric, and reasonably small.

[knotty hornet]

nice, i wonder if there is a clever way to compactify it using lifts and snapping to utilize levels

[normal latch]

im team not this lmao

[agile junco]

It does clip a little bit, which I generally try to avoid.

[normal latch]

how would a balancer help with train loading?

[unique cypress]

Loads all wagons equally

[normal latch]

seems like (at least currently) i just pump my full belt into one station

[agile junco]

Minimizes the "until full" condition.

[normal latch]

oh i havent needed multiple wagons yet for a single product

[dusky dust]

One common use for balancers is to use them on nuclear production lines to minimize the amount of ambient radiation along your production lines (no full machines due to manifolding, etc)

[normal latch]

i just put plastic in one wagon, sicila in another, scraps in another, and rubber in the fourth

[normal latch]

that sounds fun and interesting, excited to eventually build that

[dusky dust]

(Though to be fair, it's not like the radiation generated from machine manifolds is awful. A single stack of anything is only gonna push so far out)

[unique cypress]

You can't get the maximum theoretical throughput without having all your wagons load and unload perfectly equally, and a balancer is often necessary to achieve that

[normal latch]

can you give an example?

i might be too early into trains for my case to need it yet

[agile junco]

It would be nice to be able to set a number of stacks which is considered "full" for a train...

[unique cypress]

Unfortunately it's 31 stacks + 1 item

[normal latch]

gotta fill it up before it gets there then

but also that seems like something they could add into the game to check for full stacks

[agile junco]

wait... what. Where does that 1 item come from?

[normal latch]

i think it means like the 32nd slot, if it just has a single item in it (rather than a full stack of 200 or whatever)

that triggers the condition of "all 32 slots are occupied"

[agile junco]

Ah

[normal latch]

but like with trains, i dont really get why i would need balancing

[agile junco]

That makes sense, but is rather dumb.

[normal latch]

i can only put two max speed belts into a wagon, and two max speed belts out of the wagon

isnt that just the inherent limit?

[vapid gorge]

because you can then fill the cars at the same rate and not have tthe train wait for others to fill while others back up

[vapid gorge]

also you can't move 2 full belts per platform

[normal latch]

i thought you can pull on two belts

[agile junco]

You just need to make sure none of your belts back up once a single dock is full. This generally requires some form of balancing.

[vapid gorge]

you can attach 2 belts, but when trains dock, input/outputs stop for 27 seconds

meaning you can never move 2x belts of throughput

often because of hte system you're limited to maybe 1 belt's worth or less

[normal latch]

my setup is:

• station connects both belts to industrial storage
• industrial storage pumps two belts into network

[unique cypress]

Train throughput is complicated. With default settings, the theoretical max throughput is only achievable at a single point. If your train comes in earlier or later than that, you get lower than the maximum throughput.

The depart when empty/full setting will automatically delay the train to exactly that point if it arrives earlier than that, letting you achieve the max theoretical throughput at any length of route, as long as you have enough trains on it.

The issue appears when you have multiple wagons on your train and use that setting. If one of them fills slower than the others, it'll be filled last. And the train won't leave until it's full. The others fill earlier, and then sit idle, waiting for the last one. That loses throughput. The only way to avoid that is to make sure none of the wagons are slower than the other and all satisfy the full condition at the same time

[vapid gorge]

yes, buffers help the lock out time.

but if you pause a belt even for 1 second you lose throughput you can never catch up to

[normal latch]

i guess the thing is, i wasnt waiting for wagons to fill

i basically treat it like a magic belt line

[vapid gorge]

you often don't need to no

[normal latch]

so if i put 480 speed into the wagon

i just assume its magically 480 speed out on the other side

like, wherever the train goes to dropoff

[vapid gorge]

just direct feed what you need to onto each platform and as long as you have enough throughput it'll be fine

[agile junco]

That 27 seconds may or may not matter, but if the margins are tight you can optimize it a bit.

[unique cypress]

No, because the platforms stop moving items between themselves and the belts for 27 seconds while the train docks. That means it's impossible to get 2 full belts per platform unless a train never docks, and it's also a reason why a train arriving too early is bad - the pause happens more often and you lose more throughput

[normal latch]

my system is also, admittedly, super simple

and my first train system here lol its just one train

it picks up four wagons of items (rubber, plastic, silica, aluminum scraps)

im sure theres something im doing wrong, but so far it seems to work

[agile junco]

Why not aluminum ignots?

[normal latch]

it sounds like, though, i will eventually need to factor in the 27 second thing

[unique cypress]

Yeah, you only need to start caring about those things when you have like 20 trains on the same network carrying a collective 10k items/min

[normal latch]

not enough silica on the scrap site

although i did consider having one of the wagons unload silica there

[agile junco]

Pure recipe FTW.

[normal latch]

and then ship the ingots back

[vapid gorge]

you only need to care about the lockout time if you care about consistent processing

[normal latch]

sounds epic

well i mean, my ingots seem to always have scraps

it sounds like the idea though is to eventually want fully loaded wagons

[vapid gorge]

no? just put on whatever you need on each platform

[normal latch]

and enough time inbetween train stops so that it has 27 seconds + time needed to belt those items off the wagon

oh ok, so kinda like what im doing now then

[agile junco]

https://satisfactory.wiki.gg/wiki/Tutorial:Train_throughput This page has some great info.

[unique cypress]

If you want to minmax your trains, yes

[normal latch]

eventually i wanna minmax things

and learn the nuances, especially through trying ideas and failing and learning

[unique cypress]

You can also just use more trains/wagons than the absolute minimum and that works fine too

[normal latch]

makes sense

[agile junco]

I'm thinking that's going to be somewhat unavoidable for me.

[normal latch]

does the 27 second thing only affect wagons that are unloading?

like if the train has other wagons that arent unloading, i mean, do those portions of the station still belt out items as normal?

[unique cypress]

It's just that I abhor having trains that aren't fully utilized

[normal latch]

or like the whole station and all its loading docks get frozen even if its just for one wagon to unload?

[unique cypress]

The pause is per wagon/platform

[normal latch]

ok great

i can work with that, so basicaly each wagon needs to be managed indepently, to an extent

[unique cypress]

When that one's loading/unloading, that platform stops moving items in/out

[normal latch]

cool

im gonna start making drones soon btw

im making the radio control unit things for it and have the other ingredients automated

[agile junco]

Drones are cool.

[normal latch]

ur cool

[agile junco]

But I really wish they could do more complex routes.

[normal latch]

im wondering if i can use a drone to like, transport me places

[agile junco]

.<

[normal latch]

like drone having a dock at point a and another at point b

[unique cypress]

Not reliably unfortunately

[normal latch]

so i can just hop in and fly between bases

nooo

[unique cypress]

Unless something changed since I last tried that

[agile junco]

You can ride on top for a moment to boost you really high. And if you're good they may launch you in the right direction.

[normal latch]

that'd be so sweet if it were added

do devs take suggestions from community?

[unique cypress]

You used to be able to glitch inside one and it'd carry you to the destination like 90% of the time but you needed to dismantle it or respawn to get out

And sometimes it'd drop you in midair

[normal latch]

i wonder what the devs are cooking up for 1.2

maybe its drone transportation for the player

[unique cypress]

I'm hoping it's alts for some t9 stuff

All the alt budget in 1.0 went to diamonds it seems

[normal latch]

what about extra tier?

i kinda dont want to "finish" lol

[agile junco]

At the moment I mostly want better controls of trains and drones tbh.

[normal latch]

two questions:

• how do drones work? like you make a station and then just point it to another station similar to trains?
• when i explore now with my new hazmut suit and iodine filters, and i mine uranium to remove the ore from the map, what do i do with the uranium? delete it? put it in a storage container?

[vapid gorge]

point A to point B
do what you want. Main option is to not hand mine uranium.

[normal latch]

how else do i get rid of them from the map then lol

[unique cypress]

When I asked about a potential t10, they didn't say yes or no. I suspect that unless they just drop development at some not some distant point, it'll happen eventually

[vapid gorge]

why would you need to?

[normal latch]

they radiate me when i explore

[vapid gorge]

unless you're building there

[normal latch]

well i think a somersloop is inside one

[vapid gorge]

health pack and grab. Faster

[normal latch]

unless im wrong lol, but the radar thing kept going nuts

[unique cypress]

You can just spend a few hundred hours on a nuclear power plant if you want :)

[vapid gorge]

or just wear the haz mat suit and then go

[normal latch]

well i think the somersloop is literally inside the uranium

[vapid gorge]

I'm not sure what you mean about radar going nuts sorry. I don't use them.

[normal latch]

but maybe im wrong, that was just what i assumed

[vapid gorge]

eh, bin it I guess?

[unique cypress]

Yeah, there's like 2 sloops that are literally in/under a uranium rock

[vapid gorge]

just don't sort yoru inventory

[normal latch]

like "oh, to get this one, you gotta mine the uranium to open it up"

can i delete from inventory?

[normal latch]

heh figured

[unique cypress]

Yeah, you can just drag the uranium to the trash

[normal latch]

oh nice

[agile junco]

Took me way to long to learn about trashing radioactive stuff.

[unique cypress]

It's only waste and processed waste that can't just be trashed

[normal latch]

i figured since i kept hearing "uranium cant be sunk" or something about waste, that maybe you couldnt delete those ore

glad i asked lol

[vapid gorge]

you can also just leave it on the ground

[agile junco]

no no no

[normal latch]

feels like that will still radiate lol

[agile junco]

That's how you get ants.

[normal latch]

heh

nice archer reference

ok so drones are just station to station

[vapid gorge]

and? unless you want to build there doesn't really matter if it radiates

[agile junco]

how long does radiation take to clear up anyway?

[vapid gorge]

and while I'm sure some people will want to build in those spots, it's probably not most. They aren't very convenient to most resource nodes

[normal latch]

idk how to explain to you that i would rather not have radiation randomly on the ground lol

i also dont want the game to have to track more random items than needed

[unique cypress]

As soon as the items are removed

[agile junco]

Oh that's nice.

[normal latch]

so with my trusty hazmut, i can mine uranium ore and delete promptly

plus im gonna make rifle ammo

[unique cypress]

Make sure you got filters tho

[normal latch]

i will be an exploration machine

yea im gonna automate and dimensional depot filters etc

[unique cypress]

Having uranium in your inventory is instantly max radiation

[normal latch]

with filters, i live, ya?

[unique cypress]

You die in like 2 seconds without hazmat + filters

[normal latch]

how long with filters + hazmut?

[unique cypress]

With, it's no damage and 1 filter every few seconds

[normal latch]

im invincible

[unique cypress]

8s? 10s? 12s? I don't remember

It's def on the wiki somewhere

[normal latch]

trust me, with how many times i die while exploring

going in actually prepared is gonna feel epic heh

oh btw

question on fuels for jetpack, i noticed people seem to use packaged biofuel a lot in vids

and like, it actually looks better than my packed fuel (regular)

like they fly longer, is that accurate or am i crazy?

[unique cypress]

!wikisearch jetpack

[brisk shore]

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

[unique cypress]

^ fuel comparison

In short, liquid biofuel is slow af but also lasts very long

Perfect for building

Decent for exploration

Probably meh for combat

(I wouldn't know cause I play in passive)

[agile junco]

Does it beat just tapping on the thrust for rocket/ionized?

[unique cypress]

It gets you about as much hover time as ionized

But due to being slow, both rocket and ionized give more height

I've also noticed fuels that are faster vertically also give you more acceleration horizontally

[normal latch]

cool just read the wiki page

[unique cypress]

Which is good for moving around but terrible for precise movement

[agile junco]

I can't wait to make some ionized fuel. I love launching myself high.

[unique cypress]

Imma be honest, when I play without mods to buffer the Hoverpack or just straight up flight mode, I use the jetpack with LBF for building over the vanilla hoverpack

[unique cypress]

Wait till the lagspike boost launches you 100m into the air

[agile junco]

I love that.

The physics are clearly tied to frame rate in some way.

[normal latch]

apparently the hypertube is also tied to framerate

[runic kraken]

are they not supposed to?

it’s a single player game how else would u expect the math to be done if not from frame to frame

[agile junco]

If my game skips rendering a frame, or the framerate drops, I wouldn't expect my physics to change.

[vapid gorge]

@last matrix hey, back from thing, did you want help planning out some motivation?

[last matrix]

as in not getting overwhelmed or what

[vapid gorge]

yeah 🙂

part of it though sounds like it's because yo udon't have an end goal you're working towards. Like a bigger end goal

you're not a newbie that's just learning the ropes after all

So 2 things we could look at that might help:

1. breaking down your current plan into steps to help manage things. Because sometimes getting on the path is just the blocker for people

or

2. planning an end goal you'll work towards after you unlocked everything. As a motivation.

which do you think would helpyou more?

could also do both 🙂

[last matrix]

I do not see 2 as being possible

that is simply not how I operate

but there is likely basis to 1

it is very plausible that part of the reason I feel so overwhelmed when looking at a new factory plan is because it is a huge largely-indivisible task, which will have no real sense of progress until it is completely finished

like refer again to my upcoming oil factory https://i.imgur.com/4MACIPp.png

how might you subdivide that in a way that actually makes sense?

the numbers do not even allow that

[vapid gorge]

yeah it's not an uncommon problem being able to take the first steps

so are these all final outputs you want?

[last matrix]

those are the final machine/recipe counts (at right) yes

the rates being their share of the input/output

[vapid gorge]

ok are the containers for diluted packaged fuel or something/

[last matrix]

they are the containers used for the packaged fuel

I would make 360/min containers, all of which are filled, making 360 packaged fuel/min

[vapid gorge]

ok so not all of these are final ouputs then right? like there's a 120 fuel all on it's own

[last matrix]

the final outputs are in the top section

"Net Output"

water is yellow as that is surplus rather than actually desired output

[vapid gorge]

and it looks like you're making the plastic, fabric and rubber all from the waste resin?

[last matrix]

Correct, as that is way more efficient per oil than the direct recipes

also this is where the factory is planned to go

[vapid gorge]

well, if you wanted to go really efficient, you could do recycled plastic/rubber

[last matrix]

though the water extractors will likely have to go off the top

[vapid gorge]

but using resin is also fine 🙂

ok give me a minute to draw up something pls

[last matrix]

I have never used those recipes; are they markedly better?

in terms of plastic/rubber per unit oil

[vapid gorge]

depends how you're going about it, but it looks like you're wanting almost all the fuel to go to power plants so this is a very solid plan

[last matrix]

yes

[vapid gorge]

ok back in 5

[last matrix]

the plan is technically not quite accurate as i will be using modular power rather than vanilla

[vapid gorge]

oh and is it diluted packaged fuel for the fuel process?

[last matrix]

but that does not change the amounts, just the power yield

[last matrix]

there is no diluted packaged fuel, but the packaged fuel is for my jetpack and for trucks

with surplus going to the sink

[vapid gorge]

oh you're using blenders for hte diluted fuel. got it

[last matrix]

yes, though I do not actually have those unlocked yet so that part will have to come later

that is not really a hurdle

[vapid gorge]

oh intersting, you're using the alternate Resin recipe too.

then would you mind if I tweaked it a little? the recycled recipes would probably be simpler and more oil efficient then

[last matrix]

go for it

do you want to use the actual planner application I use

I can even share the .factory file so you can just import it

[vapid gorge]

I'm breaking it down in tools, I think it might be easier to do step by step for you. And muuuuch faster for me

just a quick thing though - on your image part of hte final output is 30 water pm, but I don't see any fuel output?

but it looked like you were making a TON of fuel at the lower section

[last matrix]

I am, or will be

[vapid gorge]

the 1600?

[last matrix]

mostly just to use up the residue into something useful

[last matrix]

because it is all being consumed

[vapid gorge]

got it 🙂

[last matrix]

either by the generators or the packagers

[vapid gorge]

ok! I have broken it down for you.

your main issue I think is that you were going for a lot of interlinked end products. Which can happen with oils

the easiest way around that is to simple split it into multiple independent factories. It may not be as efficient but it saves on brain

I've managed to break this down in such a way that it's still linked up though

the first plan - Just your fuel, for both packaged and generators https://www.satisfactorytools.com/1.0/production?share=7d70WbRffEBdFjye4fvX

part of the waste goes directly to fabric https://www.satisfactorytools.com/1.0/production?share=RKvgSbU0aIvL8iXfyAa7

[last matrix]

except none of these numbers work out at all

[vapid gorge]

then you have your total plastic/rubber including hte plastic you need for the packaged fuel https://www.satisfactorytools.com/1.0/production?share=JcNFLLSP9exLPJ2iZeRU

I would further break this one down into independent rubber/plastic factories though to make your life simpler

[last matrix]

the issue is not so much that there are too many kinds of recipes but the sheer scale of the project

100+ refineries for the two main products alone

[vapid gorge]

about 140 with the recycled plubber. But you use significantly less oil

do you have a bunch of power shards?

[last matrix]

like from a past playthrough this really burned me out https://i.imgur.com/moUiFPi.jpeg

this was just part of a pure copper ingot plant used to feed into what was going to be the quartz facility

I never got that far

[vapid gorge]

even using blueprints?

[last matrix]

blueprints help but not enough

chiefly because they save only a small amount of time

you can only place a few machines at a time with them, and you still need to do most of the belts manually

[vapid gorge]

you weren't connecting them to power sockets, junctions and splitters? that's the real time saver. Even if you only have 1 machine per BP , if you have all the connections done up already it's a huuuuge help

[last matrix]

power connections take a couple seconds per machine - that would not be a major time saver

and no I do not see how you would, since the power mount is "external"

[vapid gorge]

well depends how yo uuse hte poles or sockets. An easy way is ot run a beam along the top of hte machines with sockets and wire them up like that

if you have 3 or 4 machiens in a bp you just have to link up the BP sections

[last matrix]

I typically use wall mounts

[vapid gorge]

yup. Wall mounts can be put on beams

[last matrix]

on walls, ceiling - which never looks very good - or on beams running across the room

here is my oil facility in that last playthrough https://i.imgur.com/L7eNY5f.jpeg

that too burned me out badly

[vapid gorge]

could run them in front of hte machines? mostly an aesthetic choice, but it sounds like yo uneed to sacrifice some aesthetics for ease of building to get you going

[last matrix]

I took like two weeks break after that

and the later turbofuel expansion did not help https://i.imgur.com/WW29oDx.png

[vapid gorge]

the other option, on top of breaking it down into multiple plans, is to shrink the overall plans. You may like the idea of really big productions, but if they get in the way of you actually enjoying them is it really worth it?

[last matrix]

that thought has occurred to me

the reason I do that is because I know from experience that only building a little screws you over later

as you end up with far too little production for later tiers and then have to wait around for hours to produce enough

[vapid gorge]

make dedicated factories for dedicated processes 🙂 make new processes for later

[last matrix]

that is not unique to Satisfactory either

[last matrix]

that is what I do, mostly

a bunch of dedicated-purpose regional factories using the nodes in that region

[vapid gorge]

well, how about this. Halve this project plan, use a bunch of power shards, and you'll really shrink down the number of machines you need

[last matrix]

that also has occurred to me

[vapid gorge]

sometimes you just need to sound ideas out with folks

[last matrix]

historically shards were so limited that i have developed a habit of never spending them unless absolutely necessary

but i need to keep reminding myself that they are not limited anymore

that said the power cost from using them might be so high that i cannot actually manage that

[vapid gorge]

a fully over clocked factory only uses 33% more power than when all the machiens are set to 100%

[last matrix]

random example from past playthrough

[last matrix]

really? that little?

[vapid gorge]

yup!

[last matrix]

I always thought that the power penalty for overclocking, especially at 250%, is like a factor of four

I know they rebalanced it in U4 or U5

[vapid gorge]

per machine? yes. But you're also reducing the number of machines by a lot

[last matrix]

true

[vapid gorge]

so for example if you're making 1000 plastic per min with everything at 100% and X machines

vs

1000 plastic per min with all machines at 250% with X / 2.5 machines, the total overall increase is just +1/3 🙂

iirc they just changed OCing generators. But I could be mistaken, it was a long time ago

[last matrix]

that is the part I explicitly remember too

but yes i will try the power shards

i have about 350 right now

somerslooping the slugs really helps

[vapid gorge]

just as a thing though, I think you might be crippling yourself from the start , this makes ME want to give in and I've made massive systems before

[last matrix]

I actually finished this one

[vapid gorge]

that's so many outputs to manage

this has 500+ refineries and that last image made me go 'ugh' xD

that made 9000 plastic/rubber and almsot 10k alum ingots

[last matrix]

this was the main base in that playthrough https://i.imgur.com/VedxCLK.jpg

the long linear section was the actual project part assembly, plus a few core parts

the foreground area is mostly for a bunch of T1 stuff like frames and reinforced plates

[vapid gorge]

Fuel https://www.satisfactorytools.com/1.0/production?share=B3dyTfdyW9cywMl94Ifx

Plubber https://www.satisfactorytools.com/1.0/production?share=rZ01MccaZ7QvQDjiPwfK

Fabric https://www.satisfactorytools.com/1.0/production?share=rZ01MccaZ7QvQDjiPwfK

I'd split the plubber into rubber and plastic factories , make life easier. Use some of hte plastic for hte packaged fuel
(these are the halved plans)

[last matrix]

frmo the other angle https://i.imgur.com/fGO7Aqs.jpg https://i.imgur.com/qQmUQHw.jpg

[vapid gorge]

You like big builds, I can appreciate that 🙂

[last matrix]

yes, though that is probably a major contributor to the burnout

here is the motor plant when it was nearly finished https://i.imgur.com/a2x0JhD.jpeg

[vapid gorge]

I imagine having to focus on so many outputs at once doesn't help :\

[last matrix]

that is no much of an issue so much as the fact that it is just a ton of work

and as mentioned you do not feel like it is making progress until iti s completely done

[vapid gorge]

oh sure, but having to juggle everything and start the steps makes it worse

executive function I think it's called.
and managing that can be hard

But I gotta have a shower and make dinner for the family 🙂 feel free to ping me if you run into any hurdles. I check the server most days

[last matrix]

not helping matters is that when I built half of those I was dealing with a kidney stone

so that really did not help focus

the aluminum facility in particular suffered from that

[vapid gorge]

.... why would you be anywhere other than the hospital for that >.>

I've had kidney stones 7-8 times and they are not fun

[last matrix]

it took several months for a diagnosis and then another month until the surgery and then a month and a half of recovery

[vapid gorge]

oof. Thankfully mine were all one and dones :\

[last matrix]

having a history of them will help expedite the process if it is ever to occur again, fortunately

[vapid gorge]

one would hope!
But yeah, ping me if you have any problems. Refer to the links I think the smaller break downs and layouts will help you

hell, with tons of over clocking that won't be very many machines at all 😄

[last matrix]

overclocking will be the biggest difference i think

[vapid gorge]

only 5 refineries and 4 blenders for the fuel! amazing!

[fallen iris]

what website is that?

[velvet warren]

should I use default rotor recipe or the steel rotor alt given the default with the right alts are cheaper on everything but coal?

[vapid gorge]

completely dependent on what resources you have available and if you want to make anything else in that area

and how you like to lay things out

[knotty hornet]

That's a pretty significant percentage of savings with that first method, for just a tiny bit more coal, seems well worth it to me

A few extra machines ain't nothing

[velvet warren]

yeah currently I have a production line up and running for the first recipe

[knotty hornet]

Steel rotors is nice cuz it's the same ingredients as stators, so it makes making a motor factory much more simple.

[wind spade]

Copper rotor 🙂

[wind spade]

At the cost of more resources

[knotty hornet]

But using nearly half as much water, less than half as much iron ore, less limestone, that's all major.

[knotty hornet]

Yes, they have the diagram up there.

[deft lichen]

wowwww what is this?

[vapid gorge]

LOTS of outputs in one plan

[deft lichen]

is that graphic automatically generated?

looks like an analysis of an existing factory orr

[vapid gorge]

it's some planner? not sure which.

I don't think I've seen it before

[deft lichen]

interesting

[velvet warren]

why am I doing this to myself

[deft lichen]

if only...there was a planning tool that would plan the production line automatically

[knotty hornet]

You got this, bro

[deft lichen]

correct approach to do this from scratch though

it looks huge but you have like 1 of each machine

[velvet warren]

it's gonna be a chain of modular factories bringing their product into one

[deft lichen]

at that point you could 3x the output without increasing complexity at all

[vapid gorge]

basically every image of modeler someone shares makes me further baffled on why people use it

[deft lichen]

how can you tell what recipes to use???

even if you made it yourself, the chart is just unreadable

[vapid gorge]

forget that you would have had to take time to make this by hand, how much extra time do you have to spend to read it eveyr time you want to look at it

[velvet warren]

it's more about I want to see how much of each resource it's going to use

[deft lichen]

you can use Tools that does all that, it's readable and automatic

[vapid gorge]

I just made this in 2 seconds https://www.satisfactorytools.com/1.0/production?share=dK5pfexLWLdsVeoRElkU

although I would use some alts to simplify it

but I'm not going to make the effort for this

yeah that plan could really be simplified

[deft lichen]

https://www.satisfactorytools.com/1.0/production?share=F8fu1DiXRJ4QY7R4fwVn

this should more or less match

[frosty owl]

Someone-who-can't-read-icons-says-what?

[eternal hare]

https://tenor.com/view/hilarious-hey-gif-9959410

[sullen light]

update:
all my factorys are now in these containers

[tiny leaf]

does headlift reset method still work for byproduct systems?
(aluminum for example)

[wind spade]

not sure what "headlif reset method" is, but I wouldn't rely on any hacky solutions and instead do a proper one

[deft lichen]

copper factory looks fine
the factory near your hub uses too much space, you can just have the buildings...closer to each other
running a long pipeline means you have to deal with head lift, and a full pipe can support less generators than a full (mk2) belt

those pipeline pumps in front of each generator do nothing except waste power, pipeline pumps are needed to push fluid up, not forward

[tiny leaf]

saw a video from a while ago (Update8) and it was very effective

but idk if pipes have changed since then

[wind spade]

yeah, I wouldn't trust this, even if it's reliable now. We may get some pipe fixes in the future. I'd rather do the "separate" setup

[tiny leaf]

i did check some reddit posts

[tiny leaf]

isnt reliable post 1.0 unfortunately

seperate setups it is

time to use all the bauxite this world has to offer

[frosty owl]

Unless I recall incorrectly, there haven't been any changes to how fluid works in a few Updates, including U8, 1.0 and 1.1 ^^

[noble timber]

It definitely does work

[visual ocean]

How much nuclear pasta are you guys making?
I'm trying to understand if I'm going to need more than what I am preparing. I have a built factory expected to do 7 nuclear pasta but i feel that's gonna be little for future endeavours

[limpid vapor]

7 seems like a lot to me

[visual ocean]

Oh good, i thought it wasnt

it seems massive the builds needed

[vapid gorge]

its a lot. I would probalby only do 2 and make a bigger system after unlocking everything

[visual ocean]

I cant even run it efficiently due to belts yet tho

[limpid vapor]

i was gonna say, i am planning to make 2 currently

[visual ocean]

I already have 1400 copper powder ready to go (well the mk6 would make it perfect). Also the drones for the pressurized cubes are ready. All i need is mix and match

[limpid vapor]

if you have the capacity for the needed frames and RCUs, go nuts