math-and-meta

[unique cypress]

but if I gather the data from it from half an hour, it should be mostly accurate

but it's not

[knotty hornet]

I feel like that's expected, but i'm not sure how to quantify it.

[crimson moat]

It would go up when more stuff goes through, and down when less stuff does, but the MEAN of the display should equal the MEAN of the belt - especially over long time periods. It won't be 10% off without a major bug.

[crimson moat]

The variance is expected if the averaging period isn't what you want, but the incorrect mean is not.

[knotty hornet]

No, that's what im taking about

[crimson moat]

Well, you're wrong about that. The periods of above-mean and the periods of below-mean should cancel each other out to the correct average. They must have a magnitude, length and frequency which perfectly cancels each other out, because any deviation will create an equal and opposite shortfall/excess.

If you put 600 through a belt, the average of the belt indicator should be 600. Not 520.

[knotty hornet]

I think the incorrect mean is expected, because of rolling averages and long production cycles

What was the original context this came up in, so i have a better understanding.

[crimson moat]

If you put 600 through a belt, the average of the belt indicator should be 600. Not 520.

This.

600 belt through an indicator for 100 minutes. 60,000 stuff arrives at the exit. Belt counter reads 520/min on average, which predicts 52,000 stuff arriving at exit.

that cannot be a consequence of using a rolling average of any length, that's mathematically impossible. It's something much worse like a rounding error.

[knotty hornet]

How did you make this 600 belt? Machine producing, belts merging? Is it a smooth 600, or lumpy?

[crimson moat]

Issue is worst when lumpy

[knotty hornet]

Yeah, that's the part im saying is probably expected.

[crimson moat]

but basically, if you put 1200 through the belt half of the time and 0 through the other half, the MEAN of the counter should be 600. Not 520.

It's not expected. No matter what the rolling average interval is, it will have a mean of 600 over the long term.

Any deviation upwards in the rolling average will cause an equal and opposite deviation downwards, and vice versa, which will cancel out to a mean of 600. You should not get a mean of 680 or 520.

Will the counter show 680 or 520 sometimes? Yeah, cause it's lumpy and the rolling average period is too small - but it will average out to exactly 600 if it's doing the math correctly. The only way to deviate is to have e.g. rounding errors - like 6 missing items on the belt causing a larger and longer dip in the number than 6 extra items causes an increase.

[knotty hornet]

a part of this is reminding me of the Borwein integrals

[crimson moat]

It's much more basic than that, like early highschool math

[knotty hornet]

i would suggest that, since your "early highschool math" is giving errors, maybe it's not as basic as that.

im gonna run some integrals and see if i can figure out what might be going on

[crimson moat]

The math works. The game doesn't. It's been proven.

[knotty hornet]

who proved? i see no data/observations/evidence

[crimson moat]

KYO yesterday

It's impossible for the mean of a rolling average of any period to deviate significantly from the actual average rate over a long period of time (that is, say, 100x the rolling average period), but he demonstrated it doing so by over 10% IIRC. Well outside of error margins.

[knotty hornet]

i'd have to see your experimental setup, cuz there are too many variables to say clearly one way or the other, but again, i feel like, under certain circumstances (lumpy throughput, etc.), that may be explainable.

it's absolutely not impossible for a rolling average to end up less than the global average

[unique cypress]

the original setup was for testing drone throughput, the monitors' issue came up in the middle

I'm building another test setup now that should have perfectly consistent timing - i.e. a machine

[earnest trench]

how do i power at a lower level? i'm making a coal setup that would generate 1350 MW, but it would consume water at 810 m^3/min, which isnt good because the mk 1 pipes i have access to can only handle 300. What do i do?

[knotty hornet]

break it into 3 pipes

[earnest trench]

i see. Also is using pumps for a medium distance feasible energy wise? how much do mk1 pumps consume and how far horizontally do they move?

[knotty hornet]

pumps only affect vertical headlift, so you only need to use them when you are raising the fluid. mk1 can lift 20m, mk2 can lift 50m

[earnest trench]

oh, thats refreshing to hear

[knotty hornet]

also, the headlift boost doesn't stack. if you need to lift 40m, don't do two mk1's in the same place. one has to be at the bottom, and the other halfway up

[wind spade]

make smaller modules that are easier to build

and build near water, don't move fluids 🙂

[earnest trench]

my coal's really far from my water :(

[wind spade]

move coal to water

[earnest trench]

i see

[brisk urchin]

in early game i suggest not making your power grid to complicated, go easy on it and search for a spot where coal nodes are near water

[earnest trench]

spots like that would be REALLY far away from what i can see here

[brisk urchin]

and then build it on the water so that you dont even need pumps

[brisk urchin]

well then just as greeny said, move the coal to the water

because the other way around is quite painfull if u didnt master fluids

[knotty hornet]

But then you cover up the water if you ever want to expand in that area

[earnest trench]

alright then tysm

[wind spade]

water is like everywhere

[knotty hornet]

But then if it's covered, you'd have to move it from elsewhere

[wind spade]

*build elsewhere

[knotty hornet]

Which is what you were trying to avoid

[wind spade]

build near water

if water is not enough, build also near another water

[knotty hornet]

Or (hear me out)

Build above the water and use a few pumps

[wind spade]

or... build near water?

[knotty hornet]

So the whole thing is available should you need to expand

[wind spade]

and don't need any pumps

[knotty hornet]

You have that much flat space on the shore of your waters that you'll never need to raise your water level at all?

[wind spade]

extractor gives like 10m headlift on its own

and coal gens aren't that big to need gigantic flat space

[knotty hornet]

That seems awfully limiting

[wind spade]

if you can't fit it in the space around, by all means use pumps 🤷

but it's not really needed in majority of cases

and "saving water" is hardly a thing given how half of the island is surrounded by oceans

[knotty hornet]

That's awful advice, ngl.

[wind spade]

?

[knotty hornet]

You gotta elevate most of the time to get buildings large enough to fit most things that are gonna use water, unless you find an elevated source and build somewhere farther off below it.

In which case, you are moving the water to the factory.

Why do people hate pumps?

[wind spade]

that's... not what I mean by "moving water to factory"

and people don't hate pumps?

I mean "don't build FAR AWAY from water". Horizontal distance. Not vertical

and pumps are fine when needed (aka when you're moving liquids upwards)

[knotty hornet]

Okay, the part i had issue with (which you didn't even say) was building on [the surface of] the water

[wind spade]

which is also fine in most cases, as you won't run out of water places ever

[knotty hornet]

You might in the location you are building, and if you do, you will have to pipe it (horizontally) from elsewhere

[wind spade]

if you do, you build elsewhere

so you don't have to pipe it anywhere

[knotty hornet]

Again, if things are all happening in on spot, then you'd have to move materials.

The easier option is

[wind spade]

easier option is to build in place where materials are near water 😉

[knotty hornet]

Build higher up so you don't block your local water source.

[knotty hornet]

Not always feasible depending on the project/game stage

[wind spade]

and I plan my build so I know how much water I need. So after I build that much, I don't care if I could put a few extra extractors, that will be a different project in a different place

[knotty hornet]

Do you ever go back and upgrade/expand existing factories, or do you just build all new ones?

[mossy ibex]

depends on whether I built them for expansion initially

[wind spade]

always build new ones. No reason to upgrade/expand, the factory is making the amount it needs to make

[knotty hornet]

Then my point doesn't matter to you, glad to know we are talking past each other

I always build for expandability.

[wind spade]

eh, I don't really agree with that choice (but it's your choice to make) 🤷

[unique cypress]

@crimson moat Ok, I have no fucking clue where the discrepancy in drone throughput measurement came from.

measuring nuclear pasta at 250%, I measured 249.76/min copper powder input

the actual value is 250/min

integrating one peak gives a result of 199.7 items moved, close to the 200 it actually is

and it's not like the flow is stable lol

[knotty hornet]

But that is still at least one peak a minute, so every rolling minute has at least one peak.

On your drone port setup, the peaks would be much farther apart due to travel time, so there are plenty of spaces to fit a one-minute measuring period that have 0 throughput.

I think those extended periods of 0 are what is driving the rolling average down.

[unique cypress]

eh...

is there anything else that takes/produces more per cycle than pasta?

[knotty hornet]

Somerslooped/overclocked manufacturer on turbo ammo outproduces an mk6 belt lol

[unique cypress]

I'm asking per cycle

not per minute

[knotty hornet]

Not that I can think of

Singularity Cells are close

[unique cypress]

I could mod in a recipe I guess

eh, whatever, pasta will have to do

we'll see what happens without a 250% OC

but considering a single peak when integrated showed 200 total items moved during it, I'd say it won't change anything

I'll only do like 20 mins this time

I don't have time for more

[brazen arrow]

So is the priority pipe merger build just not working correctly? Or did I not build it correctly? Because it seems like no matter what the "priority input" backs up and causes issues...

[dusky bronze]

drones use fuel based off how far they travel right?

[vapid gorge]

There are more coal nodes next to water than not, go find some distant ones 🙂

[vapid gorge]

They aren’t very reliable

Just split your waste and fresh fluids to different machines

[brazen arrow]

I ended up putting a valve on my fresh input and limiting to the refresh rate

[wind spade]

valves rarely help

[vapid gorge]

That’s generally an even less reliable method.
A tiny bit of slosh can collapse the system

At least with a VIP if it works at the start it’ll keep working

[vapid gorge]

Also wouldn’t the ‘limiter’ be whatever you clocked your extractor to?

[brazen arrow]

True

[devout ingot]

you make it sound simple like "extractor to machine, dont overcomplicate the middle part"

[dusky bronze]

does anyone know how much one fully OC'd nuclear reactor makes in terms of waste?

[wind spade]

depends which waste

[dusky bronze]

uranium

[wind spade]

also, wiki has the answers 🙂

[dusky bronze]

ahh

25/min does not sound right but maybe it got lowered a little bit recently

[unborn dome]

Mfw I add a new train to the same route an existing one to increase throughput but forget to set the unloading filters, and end up with my entire bauxite manifold full of plastic...

[warm granite]

Hot take:
Rubber and Plastic are free resources much like water is considered a free resource.

Why? Most people just consider sinking Polymer Resin instead of mixing it with water.

Instead of dedicating fuel for the recycled plastic/rubber process, folks should just make use of all their polymer resin.

And if rubber is free as Polymer Resin is free, then using insulated cable is a great way to save on copper ingots

[vapid gorge]

it has not changed

[dusky bronze]

maybe i was just hallucinating. doesn't really matter, apparently i actually planned out my control room well

[vapid gorge]

I have no idea what would make you think people didn't use the resin as the starter seed for recycled plubber

[unique cypress]

the byproduct resin from power production doesn't make nearly enough plastic and rubber to cover their demand

[vapid gorge]

?? you can start the seed for the plubber loop with like 1 rubber pm from resin if you wanted to?

what's with these hot/bad takes this morning

[warm granite]

You can template refineries being pre-filled with material. I had my refineries feeding into one another and overflowing into a main output line

[knotty hornet]

This is a bad take.

[vapid gorge]

thankfully 'you' don't make up the majority of people. I know. a shock

[gloomy gyro]

Hot take McDonald’s sucks

[vapid gorge]

that's objectively true

the dingiest street food stall probably has higher quality food

[gloomy gyro]

Hungry jacks is better

[vapid gorge]

It's been a long time since I've had either, but at that time I think that's accurate

[gloomy gyro]

Also I go to dominoes to eat like everything except pizza

[unique cypress]

tested with 100% speed pasta, so 1 peak every 2 minutes. basically the exact same results. slightly better, actually. the mean over the entire period is 100.5/min, over the first peak 100/min exactly, and the integral of the first peak is 199.8

[gloomy gyro]

You’re the first person to agree with me

[unique cypress]

I think it's adequately mediocre

[gloomy gyro]

Really gotta add the adequately ina hot take about mcdonald’s

I mean it is the maths channel

[warm granite]

How much of a demand you find yourself with rubber/plastic?🤔

[knotty hornet]

More than resins alone could ever provide

One pure vein can get 900/min each of plastic and rubber with no byproduct with recycling

[unique cypress]

like 6k/min each by the end of phase 4? I'd give you more accurate numbers but SCIM seems to be down

[knotty hornet]

Leaving more veins available for power (or more plastic/rubber)

[vapid gorge]

it's extremely low grade 'food'. The only reason people go there because it doesn't require thinking and they know what they are going to get

[knotty hornet]

So, what's the matter then?

[unique cypress]

no clue TBH. I have no reason to suspect why item counting in the drone test would be wrong, and it agreed more with what the drone ports themselves said

but in these pasta tests, the monitor's inaccuracy seems to be within margin of error.

though it does seem to consistently measure a little low. but more like 1-3%, not nearly 10% like the drone tests would suggest

the main difference is that in the pasta test, I'm measuring a full belt stopping and starting (because I'm measuring the powder input), and in the drone test, I was measuring the output belt (which was completely empty between dropoffs)

tomorrow, I'll cook up a quick mod to have something output 200 something per cycle every minute and see if that changes anything. or I'll test alien protein biomass, which puts out 100 per cycle. should be enough. though I will have to downclock it from the 1500/min default output lol

maybe I'll also test a miner with a really weird output, like 5000/7 (714.285714286) to see if it handles that

[knotty hornet]

Fun fact, all sevenths share the same 6 repeating digits in the same order, just starting at a different point.

[unique cypress]

ps. for the drone tests, I have no actual "correct" reference value, so it's hard to say which of these methods is the most accurate, only compare them to each other. I thought 2 mostly agreeing with each other, and the other being different would mean that the 3rd is wrong but maybe not. drone throughput is really inconsistent anyway

[knotty hornet]

I would argue there is no actual issue and that the monitor counts accurately.

[unique cypress]

I mean it's inherently inaccurate if used to measure something variable because the averaging window is only 1 minute

but by stretching that to 30 minutes in these tests, so far it seems fine

[knotty hornet]

Well, that's just the nature of a rolling average. It smooths out variations, but it still has them. And if it's rounding error, then it's fine.

[unique cypress]

I still wish it had a configurable averaging time and more decimal places

well, any decimal places lol

[knotty hornet]

Yeah, I agree, that'd be nice.

[wind spade]

the thing is that in practice, 99% of time you know (or should know) the amount of items on the belt anyway, so I'd argue they shouldn't even have added that thing

[thin owl]

hi, is satisfactory-calculator down ? can't reach the website anymore

[wind spade]

there have been reports of it being down

if you want to calculate stuff, there's e.g. https://www.satisfactorytools.com/1.0/ (which even works better in some cases)

[knotty hornet]

There's another really good app

It's called

Pencil and Paper

[thin owl]

damn

@knotty hornet yeah i like it but i wanted to view my world actually

[knotty hornet]

Oh yeah.

[knotty hornet]

You use any mods? Cartograph is really good for that

[crimson moat]

eternalUnion has found quite a few pipe bugs recently, including one which elevates liquids without headlift (the code sees certain connections as being horizontal despite them ending at a higher elevation)

[silent vessel]

Hi, I'm Mijael. Does anyone know how I can upgrade all my conveyor routes more quickly? It's actually taking me a long time to upgrade each section from MK1 to MK3.

[crimson moat]

Usually people don't really build in a way where they have to go back and rework/upgrade old stuff. It can stand on its own, and you can build a new thing next to it yeah?

Prebuilding on a large scale will continue to be a pain in the butt to integrate new tech with unless you play from a gamestate that has everything unlocked before you start, as every technology unlock will just have more and more stuff that you have to go back and change which will become unreasonably tedious.

The easy/quick way is to upload the save to Satisfactory Calculator Interactive Map, where you can drag select all of the belts and just convert them to mk.3 and redownload the modified save.

[thin owl]

will take a look, thanks !

[wind spade]

why do you want to upgrade anyway? existing factories don't really need the upgrade and new factories will be built out of the new belt

[silent vessel]

@crimson moat , @wind spade Thank you very much for your answers. What I was trying to do was improve the mining by overclocking the mines I already had and then increasing the speed of the belts. But I see that it's a very mechanical and tedious job that ruins my gaming experience. I sincerely appreciate you taking the time to explain this to me. In short, I should stick to the traditional game of finding new mines and directly applying the new technology there, right? The Satisfactory Calculator also seems like a good idea. Does this option block my game achievements on Steam?

[wind spade]

you can also upgrade existing miner and just put a splitter after it. That way the existing factory doesn't need any upgrading and you can split the excess new material for new part of the factory

[crimson moat]

Nps! What i would suggest for mines is just adding a splitter in front of the miner if you're going to later upgrade or overclock it.

If it outputs 2x as much stuff, then 50% of the splitter can feed all of the original manufacturing while the other 50% goes to new stuff on a new belt.

It will also self balance; if you have a miner outputting 120 ore per minute and one of your belt lines is consuming 40 of that, then (after a delay to fill belt space and machine buffers) the overflowing ore which is unused will go out of the other side of the splitter. That means that if your first usage of the miner only consumed 40 ore, there's 80 available for you to use instead of just 60.

[silent vessel]

@wind spade @crimson moat I was drowning in a glass of water. I'm so grateful. The idea of the divider in front of the miner is revolutionary to me, haha. Thanks guys.

[winged rune]

Keep in mind you can upgrade things in place without re-placing them (idk if someone said this)

So it means the fewer lengths of belt you use the easier it is to upgrade

[rotund frigate]

Guys sorry for the ask. But how does liquids work in terms of math?

[plucky tusk]

Wym

[rotund frigate]

I’m going to start with oil generators

I’m asking if the math is almost the same as like modular frames or something like that

[plucky tusk]

Well math should be the same but new variables such as pressure and headlift

You also got waste you have to deal with

So either sink or more refineries lol

The math is always gonna be the same just the complexity goes up with each teir

[rotund frigate]

Okok thank you a lot. I think I understand. Isn’t simple but it I’ll eventually ahah

[inner gulch]

Attempt at a directional manifold design targeting 600 flowrate. Junctions must have their welds parallel to the ground (like the second picture) for it to work properly.

[vapid gorge]

try it with something other than fresh water or crude oil. Those regularly can be done w/o a loop and just a straight pipe

I would also really get rid of that buffer

[inner gulch]

what would be a good example for testing?

[vapid gorge]

HOR to fuel? any secondary step really

I've seen people try things like this before though, no idea how reliable they are vs loop

[inner gulch]

this one intentionally exploits an issue with junctions though

[vapid gorge]

I suppose an excellent test would be something like a turbo fuel line where blenders are feeding refineries , 25 of them, on a single side of the manifold

that's something a loop has a tough time with

[vapid gorge]

what issue with junctions?

[inner gulch]

it is a little bit detailed but in essence, when the welds are parallel to the ground like that, junction acts like it is horizontal. for example, it can output water from the top even when it is not fully filled, a junction turned 90 degrees cannot do that.

[vapid gorge]

I don't know why that's beneficial

[inner gulch]

it is beneficial because the junction can transmit its full pressure without gravitational loss to all 4 connections. pipes do lose pressure to gravity, so the whole system tends to flow to the right

[vapid gorge]

generally the issue is that the empty gaps in the pipe made by the machines at the junction is the issue

[inner gulch]

I did not experience that because I tend to break intermediate fluids into smaller systems, usually <600 flowrate required. but I can try to modify my existing turbofluid line to see if it works.

[inner gulch]

did something like this. 10 packaged fuels (=600 flowrate) feeding 26 turbofuel refineries (=585 flowrate), both using the design above. for now, it seems like a fixed 600 consumption, will observe for some time to see if there are any changes. (E: It seems to be pretty stable so I will just leave my laptop on for the night and see whether it worked or not tomorrow) (E2: Tested for 2h 10m, still stable)

[young agate]

If I want to use all these crude oil spots, what would the max amount of power I could get from them (normal fuel only) and should i use all of them or half.

[crimson moat]

75 GW with diluted fuel

18.75 GW with no alts

[young agate]

what would u recommend me aim for? Im in phase 3 rn

[crimson moat]

heavy oil residue and diluted packaged fuel alts

and maybe make 25GW to get into p4 and build more power there? Been a while since i played at that stage

[young agate]

What if i dont have a Alt

[crimson moat]

then your life would be far harder, i'd recommend going and unlocking them (either ingame as intended, or via the cheat menu / SCIM)

[young agate]

what about turbo fuel?

[crimson moat]

It's not great, rocket fuel is the real upgrade

(or nuclear)

[young agate]

how do u get the blender?

[frosty owl]

Visit your local Ficsit-sponsored supermarket

It's unlocked in the HUB

[young agate]

phase 3-?

[frosty owl]

If you don't see it in your current phase, then you're not there yet ^^

[spiral hornet]

Am I correct that Crystal Computer is a lot better than normal Computer since I can cut out oil entirely and make more per second per machine? With Silicon Circuit Board.

[vapid gorge]

70gw of diluted fuel will get you to the end of the game if yo uwant

[fallow siren]

it depends, if you don't want to bring oil, crystal computer + silicon cb is a really powerful recipe for it

while normal computer does use oil but far less complicated to make

[spiral hornet]

cutting out oil is always a big plus for me lol

[light fox]

good or bad planning for heavy modular frames?

[eternal wharf]

I needed some help with balancing fuel for early game Biomass Burners, I've basically made 2 Columns of 3 Biomass Burners in each facing inwards connected by a Manifold. The issue is that idk how to balance power load, I don't think I can. This leads to the issue where my power grid keeps powering off even though the first two burners have 100+ Biomass in them while the last two being empty is what causes the grid yo shut down. It keeps needing me to either manually balance out the Biomass or periodically keep going out for leaves/wood more than should be necessary. Is there a more efficient way to resolve this during like Phase 1 and 2?

[unique cypress]

Not great, not terrible

Heavy Encased Industrial
Encased Industrial Pipe
Steeled Frame (optional)
Steel Rod (optional)
Iron Pipe (optional)
Pure Iron Ingot (optional)
Coated Iron Plate (optional)

[light fox]

give us a screenshot

manifolds take a long time to spool up depending on the amount of machines and material coming in so im pretty sure your issue is that youre not giving it enough time to prepare

manually fill the first few burners and then the biomass should automatically move down the line and that speeds up the manifold starting time

[eternal wharf]

Does this give a better idea? It's a configuration I saw in a Let's Play but that person neither mentioned this issue nor did I see them experience it throughout the video. It's 2 rows of 3 Biomass Burners each facing inwards with 3 Splitters in the middle. The main lane is connected to storage which gets input from two constructors thaat create Biomass from Leaves and Wood each

[light fox]

also make sure your belts feeding the burners can carry enough materials at that rate

[unique cypress]

Biomass burners only burn as much fuel as they absolutely need to. No matter how you fill them, they'll consume fuel at the same rate. The only downside of a manifold is that you have to fill them first, which does require some extra biomass. But this is a one-time cost

[light fox]

belts have max throughput

that could be a problem for you

[unique cypress]

A mk1 belt full of biomass can supply exactly 6 burners

[light fox]

nevermind then

also id suggest adding in some more crafters to make solid biofuel or whatever its called its way more effective than biomass

wood/leaves -> biomass -> solid biofuel -> power

[unique cypress]

Yeah, solid biofuel is better than biomass but it's a tier 2 unlock

You'd be stuck with biomass in the meantime

[light fox]

ohh right

[eternal wharf]

I actually did plan on doing that initially but ran out of room and then got distracted with other stuff so didn't get to it. I do have it unlocked

[unique cypress]

Solid biofuel burns at 4/min so you can supply 30 burners with a mk2 belt

[eternal wharf]

I'm just gonna flip the constructors 90 Degrees, build another constructor behind each so all the Outputs are facing inwards

[young agate]

Can anyone help me with the calculations for diluted packaged fuel

[wind spade]

diluted packaged... biofuel?

[young agate]

Im building a new and bigger oil factory in the green but i need a way to optimise it

[young agate]

mb, fuel

[wind spade]

also, any specific reason you're making the packaged variant?

[young agate]

I dont got the blender

[wind spade]

no I mean

why not make the liquid fuel

then you won't have to make canisters for it

because they will loop in the system

[young agate]

Cause i need blender to make it without the package

[wind spade]

not what I mean

[young agate]

Am confused sorry

[wind spade]

I'm asking why your final product is "packaged fuel" instead of "fuel"

if you're planning to use it in fuel gens, you can just make fuel

[young agate]

oh-

I thought diluted fuel was better

[wind spade]

diluted fuel is equal to diluted packaged fuel

[young agate]

• the calculator did the final thing so idk

[wind spade]

I'm asking why are you not doing this

[fallow siren]

the packaged fuel can be unpackaged into fuel, the empty canister from unpackaging fuel can be looped back into water packager, resulting 0 canister production

[wind spade]

why are you making specifically packaged variant

[young agate]

I was trying to do smth like that, mb

[young agate]

Can you send me the link to this so i can do some changes of the amounts?

[wind spade]

https://www.satisfactorytools.com/1.0/production?share=IULGPwDEUJIQN1VOpBS8

[young agate]

thank you

[light fox]

if you dont have blenders the diluted packaged fuel gives you the same as the diluted fuel recipe

its the best you can get out of an oil node before you unlock blenders

it just includes the extra step of packaging and unpacking

[wind spade]

even after you unlock blenders, it's practically equal

[light fox]

equal to what

greeny whats equal to what

[wind spade]

oil to fuel ratio

DF and DPF is equal

[light fox]

dude

he is making packaged

because he cant do regular

???

am i missing something or are you messing with me right now

[wind spade]

I know? why are you telling me this, it's not relevant to my point

[light fox]

okay so your issue with that dude, was it literally just because he didnt put the last step of emptying the packages?

no its because he forgot the loops okay yeah

[wind spade]

I was asking why they didn't unpackage it, if they are specifically making the packaged variant of fuel because they want it for jetpack or something, or if they just want fuel for gens (in which case the plan they made is kinda bad due to making packages from plastic)

[light fox]

packages still do have to come from somewhere

he just didnt know he can loop ig

how does sf tools work

im just using modeler because it feels simpler

[deft lichen]

Modeler doesn't solve for you

You tell tools what you want to make and it shows you how

[analog crag]

Is nitro rocket fuel alt just better than the default recipe in every way? Looking at the calculator, it takes much fewer resources for the same result, it outputs faster, and only takes a single blender and 2 refineries. Am I missing something?

[fallow siren]

wrong, nitro alt use twice as much sulfur and more nitrogen than default recipe in cost of simplicity

[analog crag]

Huh, I guess the calculator is wrong then

[deft lichen]

Which one are you using

[wind spade]

what exactly do you have issue with?

[analog crag]

Never mind, it was me not using it right. Figured it out

[light fox]

only being able to like put in one of a certain item or something

idk it just hates me i think

[wind spade]

not sure what do you mean? you can press the green plus button to add more items?

[visual yarrow]

Is this a floating point issue? Terminology question. Actual response to image: hnnngh

[deft lichen]

UI rounding, most likely

If you save the game and load it into SCIM you can check the exact amount

[visual yarrow]

Good to know, but it's not important enough to me. I knew it was gonna happen anyways. It just still makes me feel sad.

[deft lichen]

Fluids actually aren't floats, internally they're stored as whole liters

Even if they were, only 3 digits of precision would be used, which the game logic can guarantee to be exact

[visual yarrow]

Let me rephrase my original statement: what exactly is a floating point/floating point error? I feel like I have only half of an understanding of it and I use it in wrong contexts too often as a result.

[deft lichen]

Base 2 can't exactly represent some numbers

Notably 0.1 becomes an infinitely repeating decimal

[visual yarrow]

So it's like a leap day?

[deft lichen]

0.1 + 0.1 adds up to 0.3 if you round it

[visual yarrow]

That's horrifying.

[deft lichen]

I said it wrong

It adds up to 0.20000...1

[visual yarrow]

And that's just a quirk of the way that binary works? Number theory, and all that.

[deft lichen]

Much like how you can't represent pi exactly

0.1 decimal is 0.00011001100[1100]...

[visual yarrow]

Is the issue with clock speeds at 1/3 being infinitely repeating also a type of floating point issue, or is that something else entirely?

I say "issue" but I mean "how that value works when converted to a percentage"

[wind spade]

no, that's the game being internally limited to 4 decimal places

[deft lichen]

It's related, you only store 0.3333, not "exactly" 1/3

[wind spade]

(for clock speed)

[deft lichen]

0.3333 isn't 0.33333333333333...

[visual yarrow]

The game would need to store infinite decimal places.

Unless there's a point where the threes stop i'm unaware of.

[wind spade]

well decimals in base2 can't represent all numbers (sometimes even non-repeating)

[deft lichen]

It could theoretically store "1/3" as a fraction but underlying CPU math would still be limited to a fixed amount of binary decimal spaces

[visual yarrow]

But long story short: floating point errors in a Satisfactory context are just hiccups where these little quirks of binary align in just the wrong way, is what i'm getting?

Or any computing, I guess.

[deft lichen]

Yep

The clock speed imprecision is specific to Satisfactory because they chose to arbitrarily limit precision to 4 decimals

[wind spade]

yes and no

partially because of hardware limitation
partially because of arbitrary limit of 4 decimals

[visual yarrow]

Is it really arbitrary? There are UI concerns to having extremely high max decimal places.

Or is there a... "better" number of decimals in this context?

Or, I suppose I should say, is it any more arbitrary than any other amount?

[wind spade]

anything is arbitrary

[visual yarrow]

Your face is arbitrary!

[wind spade]

UI rounds the saved decimals to 3 decimal points anyway, so that's not a concern

the problem is that the saved value is also rounded

[visual yarrow]

Maybe i'm overthinking this, but, like... clockspeed settings ultimately are changing the production speed of a building, right? So wouldn't there be issues with... tickrate, or something, for extremely precise production times as a result of extremely precise clockspeeds?

[wind spade]

not really

most of things like these are calculated using delta

[deft lichen]

For clock speed, 4 decimal spaces are shown and stored
For stuff like... how far into a production cycle a machine is, a full float is used

[wind spade]

(aka "number of [extremely small unit of time] passed since last update")

[visual yarrow]

Hmm. Maybe i'm just looking at it from the wrong perspective. I've had a save that stopped being able to handle fully overclocked pure nodes because of... something, when it got to a certain level of complexity.

[oblique hollow]

And the discrepancy between 33.3333% and 33.333333333...% would in most cases take years of playtime to cause issues

[wind spade]

could be anything from FPS issues to bugs (depending on version), etc.

[oblique hollow]

at some point you just don't need more decimals for a given precision

[visual yarrow]

Okay, I honestly forgot how I got onto this topic, but I think i've gotten about what I wanted to know.

[oblique hollow]

its a difference of 1 vs 99.9999

[visual yarrow]

OH WAIT, no I remember.

[deft lichen]

10 m3 fluid isn't 10 m3 fluid

[visual yarrow]

Is there any way to force this last bit of DMR that's presumably in the pipe into the particle accelerator?

[oblique hollow]

nope

[deft lichen]

Is it a good idea to edge it like this?

[oblique hollow]

Pipe fullness dictates gas flow

[deft lichen]

Wouldn't it be better to have a stable excess in the system

[visual yarrow]

Ah well. It's not a big deal- I was just hoping that something like, removing the pipes would shift the contents one inventory over or something.

[oblique hollow]

If you dont want a lot of DMR to be wasted on filling pipes, have short networks or prefill it using the RSAM to DMR recipe

[visual yarrow]

This is a temporary setup.

[deft lichen]

Is this a temporary setup?

[visual yarrow]

Like... it's quite literally just a single dark matter crystal i'm not getting. Not a big deal.

[visual yarrow]

Yeah, i'm making neural whatsits to unlock the blueprint designer mk3.

Or I guess, have made, at this point.

I still need to automate computers and HMFs. Sigh.

On a different note... it still bothers me so much that the quantum encoder doesn't have snap points for catwalks despite being like, the single machine in the game that has it's walkable part of the model grid aligned.

[deft lichen]

You have neither of those automated by tier 9?

[visual yarrow]

I don't really understand myself either.

[deft lichen]

You can still build up to it from elsewhere without having to attach it. Your point still stands though

[visual yarrow]

Yeah, I just had to, like... nudge it from the base of the platform it's on because the encoder isn't on the same grid as the base foundation.

But I tend to rush tiers early to get access to stuff like the hoverpack.

I have to play on a "what do I actually want to do right now" priority system, so this has led to me finding ways to procrastinate in extreme ways.

My first 1.0 save, I didn't actually build coal, fuel, or nuclear power until after I had finished the final phase.

[deft lichen]

No coal????

[visual yarrow]

Not even the first time i've skipped coal.

Though, it was the first time where APAs existed.

[deft lichen]

Ahh fair

[visual yarrow]

I like to use crash site loot to rush for geothermal, since it's a MAM research, and thus you can get it super early.

I had a point back when supercomputers were also in the caterium research tree, where I did this, and had the recipe for supercomputers but not the recipe for computers.

[mossy ibex]

this seems pretty legit to me, thoughts? https://www.reddit.com/r/SatisfactoryGame/comments/1muhps5/why_your_pipes_frustrate_you/

If it's right then that means the bleuprints with the packed in fuel gens are a bad idea (the ones that are so close you have to put a pipe right between them and then clip a junction between to make it work

[unique cypress]

Great, a third pipe post on Reddit within 24 hours

[mossy ibex]

it's almost as if pipes are annoying and janky

[brazen kettle]

How much fuel are yall making for your drones? And what fuel are you making?

[fallow siren]

300/min packaged rocket fuel, enough to fuel hundred of drones

[unique cypress]

I don't use drones, but if I did, I'd probably do that

Or plut rods, if I wasn't making ficsonium

[oblique hollow]

Not correct though

Pipes have hidden extra capacity

Its just that they dont want you to build these tiny segements, same with belts

And further, pipes dont move once every second

They move once every tick, which is about 1/30th of a second

(usually at least)

[mossy ibex]

are ticks framerate dependent?

surely not?

[oblique hollow]

No

[mossy ibex]

btw thanks for chiming in, I was hoping to hear from you about this 🙂

[oblique hollow]

Framerate is ticks dependant though. If the simulatiom slogs too much for some reason the game will slow down to ensure calculatiom correctness

Its more so an extra safety net

Usually they both dont interact much

[mossy ibex]

yeah I would've been surprised if it were tied together, that's a rookie mistake

[oblique hollow]

ah neat, eternalunion already replied to that post

What they say now counts as much of a source as what i say since they provided some insight on pipe code and i talked with them a lot recently

[unique cypress]

Is the post deleted for you too or does Reddit show posts from blocked users as deleted?

[mossy ibex]

I see it just fine

did you block grimgaw?

[unique cypress]

Nope

[oblique hollow]

odd

[plucky tusk]

Reddit 🤓

[unique cypress]

I signed in with my alt and can see the post just fine

So either Reddit is doing Reddit things or the author has me blocked and that means I'm not allowed to see their post

[eternal hare]

Says OP deleted the post for me too.

Probably some kind of server/database load balance thing.

Actually, says the account is deleted as well. So post probably shows as deleted because the account was deleted.

[visual yarrow]

So this is a thing for ionized fuel I set up a bit ago. I've noticed something very interesting about it.

This is the setup I have- what i've noticed is that the contents of the two fluid buffers have stabilized at the exact same ratio that DMR is produced in. The left one has 80m3, the right one has 120m3.

So basically it's another one of those Pipe Things that I will never understand. But at least this one is kinda neat instead of kinda causing me problems.

Well, at least for now. I do fully expect the possibility of leaving for five minutes and coming back to find it not working for seemingly no reason.

Oh, the valves are unlimited, by the way, before anyone yells at me. I'm just using them to separate the two sources of DMR and also aesthetics.

[lone igloo]

Always wanted to know this. Does the game round this up to 56 per min? On the underclock it says 56

[wind spade]

game doesn't care about per minute values at all

[steady walrus]

yeah it does

[wind spade]

untrue

game works in cycles, it outputs 5 every cycle

[lone igloo]

well if you make it 93.3333 it says 56 so

[steady walrus]

to the human eye its gonna produce 56 per minute

[wind spade]

that's also rounded

[unique cypress]

the only value that is saved and used for calculations is the clock speed

and it's saved exactly as you see it

[wind spade]

you're asking in a math channel, you're getting accurate answer

[steady walrus]

ur not gonna be able to tell its making a thousand of a second slower

[steady walrus]

fair

[wind spade]

and since the question was "will this round up", the answer is "no" 🙂

[lone igloo]

as long as 56 is the output I dont care about cycles or any other under the hood shenanigans. It dont bother me at all.

[unique cypress]

what's the recipe and what's the clock speed?

[steady walrus]

for all intent purposes it will be 56

[wind spade]

seems like base quickwire

[lone igloo]

quickwire for 56 per min

[unique cypress]

I'm asking for clock speed

[lone igloo]

93.3333%

[wind spade]

56 won't ever be exact output of the machine
whether you're bothered about it being close enough or not is up to you

[wind spade]

that makes this

[unique cypress]

then it's gonna make 55.99998/min exactly

[lone igloo]

close enough to 56 and I wont even worry about numbers, says 56 per min on machine will go with that

[wind spade]

if you're bothered by this, you can clock to 93.3334% and machine will idle once in a few hours or so

[unique cypress]

it'd make 56 exactly if you left it at 100% and let it idle whenever it filled up

[wind spade]

just fyi, practically any number (apart from the clock speed one) you see is rounded, and isn't necessarily the one game uses

[lone igloo]

Kinda miss leading then ngl.

[wind spade]

well, if you need more than 3 decimal places, you should do the math yourself anyway 🙂

[lone igloo]

it was more that I needed 168 quickwire for 3 high speed connectors for super comp and at 3 HSC its 168 quickwire and that is 56 for 3 machines. Typed in 56 on the overclock and I got 93.33% 1st and then 93.3333% for 56.

[wind spade]

yeah that goes through at least one step of rounding

[lone igloo]

it also makes wire easy as hell to work with at 30per mi instead of 37.5 and circuit boards at 3 instead 3.75. Thats why I went with 80% on HSC and its 1:1 for Super Comp set to 1per min

[wind spade]

eh, I don't care about decimals, they are same numbers as non-decimals

[lone igloo]

we will see if my quickwire will lack I will just set it to 93.3334 and GG whatever xD

[unique cypress]

@crimson moat you're gonna give me some specific scenarios in which you think the throughput monitor is inaccurate

because I've ran 2 more tests today, and it was 0.6% and 0.3% off

[knotty hornet]

You can type fractions too. 93.3333... = 93 ⅓ = 280/3

[crimson moat]

you said it was wrong from the lumpy droneport deliveries, or did you just mean the droneport itself?

because both are a bit wonky, but maybe just the droneport is really bad

[unique cypress]

the issue with drones is that I have no way of knowing what's the actual flow

the monitor didn't agree with counting items in containers but that's it

in all machine tests, it's been at most 3% off iirc

I have no reason to suspect item counting of being wrong, but at the same time I currently have no reason to suspect the monitor either

[crimson moat]

the container is correct though

[unique cypress]

and actually, I don't know if you read that message
#math-and-meta message
but I used item counting to confirm IGT is accurate

so how can 2 methods not agree when used on the same test, but show correct results when tested separately?

[nova vortex]

can you kill these?

[unique cypress]

explosives

[haughty barn]

It’s their fault really. If they didn’t try to kill me they would be cool decorations

[grizzled grove]

After about 100 hours and several months 1/5th of my nuclear is approaching mega printing stage

[noble pine]

what do i do with extra water?

like in my aluminium factory im making 1440 water per minute and i dont have anywhere to take it

[unique cypress]

put it back into alumina production

[noble pine]

so remove the water extractors and make a loop?

cuz the refineries are already full

[mossy ibex]

I figured out my fuel question. Basically, to make a fuel station work you MUST pause at the station. If there's a pause on the path it will refuel. If you remove the pause, even if it passes by the exact same spot, it'll no longer fuel

in related news, it would be really nice if the "press F to load/unload" message was accompanied by highlighting the station(s) that would be un/loaded from. For the case when they are near each other

[dusky dust]

Yeah; I recommend keeping the fresh + recycled water totally separate, but you can keep all the refineries in a line as if they're all in the same manifolds. Basically:

1. Pretend you're using all fresh water, know how much of that you need (X/min)
2. Later on, the alumina solution will produce a certain amount of recycled water (Y/min)
3. So you slice off a section of your water-processing refineries so that they consume X-Y/min fresh water, instead of the "full" amount, underclocking/overclocking as necessary
4. Then you have the Y/min recycled water going over to the remaining refineries (under/overclocking as necessary)

You may end up with an extra refinery or two if you underclock, since you may end up with a fractional bit of refinery that you wouldn't have if they were all combined

Keep in mind that it's very helpful to flood the "recycled" pipes with water beforehand so that the system can start working at 100% pretty much right away

For example, a real small vanilla-recipe setup, and a bigger sloppy+electrode setup

(I realize I wrote it out like steps up there, but there's definitely no need to first hook it all up to freshwater and then rejigger it to use recycled; that's more just the logical way to chop it up)

[azure warren]

what u all think about this build am gonna do its gonna be 4 different factorys one for smellting and 3 for the items
https://www.satisfactorytools.com/1.0/production?share=Pm3SOwx8a4k3WoVQEycf

[unique cypress]

Well, I certainly wouldn't build this, or anything close to this

1. you could cut the costs by a lot with alts
2. automatic project parts is a waste of time unless you're doing it for a challenge or for fun

[wind spade]

or for sink points

[lone igloo]

Finally producing these bad bois. Only 2 per min atm but will safice to get tier 8 unlocks to start making those perma factories

[wind spade]

why do people hate decimals so much 🙁

[lone igloo]

cuz whole numbers much more fun to work with aha

[wind spade]

eh, you clock either way

[silent shoal]

I'm planning my final power plant... It seems there aren't enough resources in the game to do it at its maximum... What would you do? Of course there's somersloops but before that I'd like to think of another way

https://www.satisfactorytools.com/1.0/production?share=GL9c8EguEfLXL9RJrZ7O

[lone igloo]

• normal recipe requires 7.5 for 1.875 and making 9 comps per min. So I use 8 for super comps and have 1 per min just there for my depot. I already got bunch back up comps but having that 1 extra constatly running is nice.

[unique cypress]

Without sloops, the most Plutonium Waste you can recycle is 191.25, or 19.125 rods' worth

If you do minimum Plutonium instead of maximum, it'll be enough

Because then you'll only have 12.6 rods to recycle

Like so: https://www.satisfactorytools.com/1.0/production?share=8vC5IkIWysvRvobfhIZ6

[lone igloo]

can someone explain why sometimes on save load my fluid vanish from my pipes?

[hoary oar]

byproduct of 452 power shards, while having no resources left to convert them in to something that can be sunk seems a bit problematic.

[frosty tangle]

Oh man, I'm gonna have a lot of fun with these pipes

[vapid gorge]

managing the waste water is easy if you keep it split. Basically a no brainer

[unborn dome]

Would it make sense for a longer train platform to have a signal half-way down, so if there's a train waiting outside the station to enter, it can start moving faster? Probably not needed except in the tightest of throughput issues.

[unique cypress]

Theoretically, yes, it means you'd be able to get another train into the station faster. But practically, what does that accomplish?

On default stop settings, getting another train in that fast is actively bad. You barely unloaded any items from the previous train, and you immediately get another loading pause and a top-up. It kills your throughput to less than half of what the belts can handle

And with depart when empty/full, unloading the previous load takes at least 40 seconds. As long as you get another train in within that time, it literally does matter if it arrives a minute or a second before the previous load gets unloaded

[unborn dome]

True, but in this case, the trains are unloading different things onto different platform segments. Just sharing a station due to size constraints.

[waxen meteor]

currently am using the crude oil well just north of the bamboo forest gonna convert it all to jet fuel to power generators, i got a hold of the alternate recipes for diluted fuel and turbo fuel blend, so obviously those, is it better to import coke from a separate node potentially much further away or should i limit my potential power output from this well and generate the coke on site?

[vapid gorge]

entirely up to you, though I'd probably just move the oil to you

do you want more power is the question you should be asking yourself

[waxen meteor]

this is turning into power plant management simulator so yes i would like more power

yeah, i answered my own question there. thanks

[vapid gorge]

you can pretty easily get through phase 5 on like .. 30gw mind you. You could always do that and plan out a big permanent station aftewards

[tame harbor]

What're you making Coke for?

[waxen meteor]

turbo fuel blend

[tame harbor]

Ehhhhhhhhh

You can get 144 GW of Rocket fuel (and a little over 1 GW from the compacted coal byproduct) out of 600 crude with Nitro Rocket fuel

For phase 3 you can just use HOR-Fuel or use Diluted Packaged fuel

[vapid gorge]

if you go to the centre of themap, by the lake, you can make enough diluted fuel and TF blend for about 40gw without moving anything

[tame harbor]

Or go all the way down to the bottom right at Blue Crater

[waxen meteor]

i'm in phase 4

[unborn dome]

I've done everything in Tier 8 aside from nuclear power and pasta, on this much power

[tame harbor]

Nitro rocket then

[waxen meteor]

i am leading up to that

[vapid gorge]

eh, just go nuclear. It's far more interesting than spamming 1000 fuel gens

[waxen meteor]

turbo fuel blend is an attempt to squeeze as much power out of this one oil node before i move to nuclear

[tame harbor]

If you have any Nitrogen nearby you will get way more power from the oil with regular or nitro rocket fuel

[waxen meteor]

i know i should but i am a bit intimidated by the super colliders and their variable power cost

[tame harbor]

Just build a big bank of power storage

[waxen meteor]

i am gonna import nitric acid to get to rocket fuel

[waxen meteor]

probably gonna do that too, at least one battery on each power tower

[vapid gorge]

eh, just treat the variable power as the max power.

it's really nothing especially if you use the base recipes

[waxen meteor]

i need to look at the production line of uranium to plutonium just so i can have a base idea what the power consumption is going to scale to

[tame harbor]

Ir you consume every single bit of Uranium and use the best recipies you can make 50.4 Uranium Fuel rods/minute. Enough for 252 reactors (or just over 100 if you fully Overclock them by throwing 756 power shards at them

This will also make 2,520 Uranium Waste per minute that you will need to convert to Plutonium Rods to get rid of

But that 50.4 uranium rods is also 630 GW

[waxen meteor]

thank you, i don't have that many power shards so that will be shelved for the time being

[tame harbor]

Bear in mind this is also going full Fucking SEND IT mode

[waxen meteor]

i get that but this gives me an upper bound to work with

[tame harbor]

A single Impure Uranium node can make 7.2 rods/minute

[haughty barn]

1000 reactors :steamhappy:

[waxen meteor]

the only impure uranium node i'm aware of is the one on top of the peak that gets you that one achievement, which while nice is impractical

[tame harbor]

There are 3 impure and 2 normal uranium nodes throughout the world

[waxen meteor]

alright, i'll have to find those cuz i've found 2 normal ones

at least i think the one is normal

[tame harbor]

Volume of transporting Uranium is so low using drones powered by packaged turbofuel to bring them to a central location is genuinely worthwhile

At least until you then make Plutonium Rods and can use those as drone fuel

[knotty hornet]

Does nobody bother with Ficsonium?

[dusky bronze]

for what you get its not very worth it

going all the way through with ficsonium is more costly on power and resources than just sinking it at plutonium rods

[vapid gorge]

its basically for people who can't stop themselves from burning plutonium rods, but complain about waste

[lean magnet]

Either my math, or my understanding is off. Possibly both. I have 8 refineries producing plastic, which should output 800 Heavy Oil a minute (10 each x 8). Refineries use 60/m to make fuel. 800/60 = 13.3333 refineries, so I built 14 at 95.237% clock speed....but it looks like Im not actually producing enough oil for that? Am I misunderstanding the numbers?

[vapid gorge]

check your oil extractor

is it backing up?

[lean magnet]

No, the problem seems to be between the 2 refinery sections, like there simply isnt enough heavy oil output. https://i.imgur.com/GNljevY.png - 10 per minute x 8 is 800 right? Or am I mixing up the 1m^3 and the 10 somehow?

[vapid gorge]

well if you're oil extractors aren't backing up, but your oil consumers are starving it sounds like a math problem

[lean magnet]

Everything is running continuously except the 2nd refinery set, because they dont have any heavy oil to use. Just cant figure out what I did. 800/60/m should be 13.3 refineries

[unborn dome]

Try adding a pump in between, even if you don't need headlift

[vapid gorge]

ah the machines producing HOR clogged?

[lean magnet]

Im a moron that cant do basic math

I have an extra 0 in there. Ive been looking at this for an hour.

[vapid gorge]

it happens. I wasn't going to go over your math cause my brain is mush atm.

[lean magnet]

I thought it was too much even when building it, but noooooo. Did it anyway.

Appreciate the help

[silent shoal]

Power shards can't be sunk?

[robust panther]

hey guys im having some fluids issues, the issue is being caused by the nitrogen getting stopped when a train is unloading. i initially thought about a buffer but this just pushes the problem back because when the buffer is filling back up itll do 600 input but only send out ~500.
what should i do to solve this?

[silent shoal]

Are there any other options? I see that even with somersloops it would be a bit complicated to do this...

[unborn dome]

The buffer needs to be large enough to last through the train's entire unloading cycle.

Or multiple buffers, or mk2 pipes and junctions on both sides of the buffer, so it can output on both outputs.

[robust panther]

yeah it can, the issue is that after the trains unloading and the buffer is filling back up itll be taking in the full 600 rate, but itll only send ~500 down the line, and i need the full 600

[vapid gorge]

there isn't enough ficsonium to do mac plutonium even with sloops afaik

[unborn dome]

May need to rework your pipes feeding the buffer, if it's only outputting 500?

[vapid gorge]

dont ship unpackaged nitrogen if you need full throughput

gas doesn't like buffers

[unborn dome]

TIL, good to know. I haven't tried buffering gas yet, since the packaged nitrogen I'm shipping is a pretty small quantity.

I'd assumed it applied the same as liquids

[robust panther]

😭

[vapid gorge]

nah, I think it's because gas tries to equalise fullness rather than flowing to most empty

[vapid gorge]

why cry? gas compresses 4x in containers

much easier to move packaged

[robust panther]

because that means redoing the entire set up which like i dont think it would be to bad cause it would just be deleting the blueprints for the blenders making fuel

but cry because ive spent awhile tryna get this to work

[unborn dome]

What if you just ship more than you need, and unload with more packagers?

[vapid gorge]

shouldn't be much work. Make a few containers and use fewer platforms.

[vapid gorge]

I don't know what you mean here

[robust panther]

im not using packagers atm, this is raw nitrogen

[unborn dome]

Like buffer the packaged nitrogen instead of the gaseous nitrogen, and then have more packagers in parallel

Oh right by train

[vapid gorge]

they aren't packaging it though

[unborn dome]

Yeah I missed that part

[robust panther]

i had asked before setting this up if i should, and maybe i misunderstood the answer but i had thought i had been told i didnt need to and packaging just made it easier

[unborn dome]

Can gas be pumped, such that adding a pump would prevent the attempt at equalization by preventing backflow?

[vapid gorge]

no, gas just doesn't do well in buffers

there's no easy solution to it other than packaging it

or shipping more fluid gas than you need and

I'd aim for at least 50% more

[robust panther]

ye but each pipe line is already using the full 600 so still would require changing

[vapid gorge]

RIP

better package it

[robust panther]

at least im not dependent on this power yet so i can turn it off safely

yeahhhh

[unique cypress]

Other options for what?

If you just want waste-free nuclear, you can sink the Plutonium rods. And if the 630GW the 2100 natural uranium gives isn't enough, you can make more via conversion. It's cheaper per MW than Ficsonium.

If you want to cut the SAM usage to the minimum, you can use Superposition Oscillators or Power Shards to generate DMR without using any SAM. That cuts the usage to like 6-7k. Alternatively, you can also use AI Expansion Servers to generate DMR, and SFTools prefers using them, but they do cost some SAM

[young agate]

If i make 640 fuel a minute, how many fuel burners will i need?

[wind spade]

build one, it will tell you how much it needs

[silent shoal]

The goal I want to reach is to use 2100 uranium for electricity and from there also use plutonium for electricity and finally for fissionium That was the original idea but it seems to have a few complexities

[azure warren]

I used a few of alts in it. the build was so much bigger than this but if there is something im missing I would be glad if you tell me. Making this smaller or cost less would is my goal but I can’t find anything

[modest shell]

I did this with no alternate recipes, you'll need to transfer quite a lot of resources over the map to wherever you want to setup which takes a while to do.

I think this screenshot shows the full setup best

Not the best for using space efficiently and I think I made some adjustments since I took this as some belts/resources didn't flow properly

[silent shoal]

Cool
How many rods of each thing do you make?

[modest shell]

I don't think it's worth doing all the way since the complexities for just making ficonium fuelrods are most of this setup

I think it's 21 uranium, 5.25 plutonium and 26.25 ficsonium per minute

Most of the power generation is done by uranium too, I can't find my notes/calculations as to how much though

So it's probably not worth going further unless your goal is just to use all the uranium to make a complete power setup, like this

[silent shoal]

It's a little smaller than I planned to make...😅

[modest shell]

If it is, I wish you luck and much sanity

I'm using 2100 uranium all the way, no sloops (yet)

And no alternatives

https://satisfactory-calculator.com/en/planners/production/index/json/{"Desc_FicsoniumFuelRod_C"%3A"26.25"}

[silent shoal]

Yes I know... I'm still having problems... Why don't you use ALTs?

[modest shell]

This is what I followed, I overclocked everything too, hence why it looks small

Didn't feel like using alternative recipes

One day I might go and redo this with alternatives and sloops to see how much I can get out of my 2100 uranium/m

[wind spade]

fyi this planner has some issues in mid-lategame productions, I'd recommend switching to something more reliable which doesn't have issues with loops and byproducts

[silent shoal]

And how will you do that? The way I see it, there aren't enough resources to process all the waste.

https://www.satisfactorytools.com/1.0/production?share=HI14Cei26fil6eT9ASbR

[modest shell]

Not sure how I'll do it other than tactical use of sloops

[modest shell]

I got around this by not using the loops and sinking some byproducts

[vapid gorge]

I'm almost certain someone came around the other day and was unable to sloop enough to do it

[modest shell]

So it's reliable enough

[vapid gorge]

they ranted at it at length in fact

[modest shell]

Lol, I've not properly thought it through, it's more just a consideration I had while I was losing my mind over some inefficiencies and other things I had in the setup

This shows everything better, along with where all my sloops have gone. It's a little disappointing this setup + the amps don't make 1 terrawatt alone

[unique cypress]

I mean with 63 Ficsonium it's absolutely possible.

Max Ficsonium from 2100 uranium and without using sloops on the nuclear stuff is 153 and that's barely possible in an otherwise empty world

It's not easy, but even pre-1.0 max nuclear wasn't easy

[unique cypress]

if your goal is to cut down on the resources, just allow SFTools to use every resource, recipe, etc, and then disable anything you don't like/don't want to use. SFTools optimizes for least resources, given the constraints you set

[unique cypress]

why do you have nearly 60 plutonium rods? making more than 22.4 requires fertile uranium, which requires uranium ore, which requires bauxite to uranium conversion
which means you're using nearly 6k uranium, not 2100

[gleaming fiber]

Hi, this may be better asked in the Q/A section but..

If I wanted to train in fuild from one location to a factory elsewhere, how can I math the fluid needed at the factory to figure out how many train cars I need including travel time? Is this even possible?

[vapid gorge]

there's equations on teh wiki, in the locomotive page

[unique cypress]

you need to know how long it takes the train to drive its entire route

though with fluids, the calculations can't be perfectly accurate, because fluids are impossible to balance properly

[gleaming fiber]

I was afriad of that. The unknown is how often the train will stop at a junction or get clogged up on the rail by another train. I'd probbaly need it to be on its own rail line to avoid that

[unique cypress]

or you can overbuild to make sure that it's enough

[silent shoal]

I see that imkibitz is doing something similar to what I'm trying to do but I haven't seen a video where he explains quantities... Basically I'm trying to get the most out of nuclear energy... For that you need max plutonium and for that you need max uranium. Then you have to deal with the plutonium waste and there aren't enough resources for that... Maybe make every possible sloop to SAM and do a material conversion 🤔 wow actually I might cook something here

[gleaming fiber]

I'm sort of/kind of doing that with Nitrogen now. But with Nitrogen, if I run low and staff production, its not going to turn the lights off 😉

[unique cypress]

max nuclear power involves making exactly 0 ficsonium tho

wasteless and with storing waste

[silent shoal]

Sorry my bad
https://www.satisfactorytools.com/1.0/production?share=r14iVns4MrmN9iNfA8pf

[azure warren]

I will try that and let you know how it goes

[unique cypress]

with only 34 sloops, you can double all RSAM, making this possible
https://www.satisfactorytools.com/1.0/production?share=L82v2dOLKhyj3A7PKSmm

[silent shoal]

Where to store it😅 🥲

[vapid gorge]

run the train back and forth, time it

[vapid gorge]

why would it get clogged? are you not running two lanes?

[unique cypress]

in containers. it's like 500-1k plut waste per min, so 1 container lasts 24 minutes
so 500 containers last 200 hours

just dump it all in a corner of the map

[languid hawk]

thats teh output of the 20 refiners all into that

and then taken out 4 lines

[modest shell]

20 refineries into 1 pipe network split down to your power?

[languid hawk]

yes

all mk2 pipes

i am using a mod for teleporting the fluids in so reverything is very level and straight and i didnt have to think of lift

[oblique hollow]

Dont use mk 2 on everything

[silent shoal]

And then what?😅 🤡 let future me to deal with this 🤣

But seriously
For 10200 SAM you need 34 constructors to create Reanimated SAM and in order not to waste all the SAM you can also make a Sloop for 17 converters (the machines are at 250%) This seems to solve the problem🤔

[oblique hollow]

Use mk 1 for everything thats slower or equal to 300

[languid hawk]

does it change anything using mk2?

[oblique hollow]

Using mk 2 for everything creates pressure waves and those generally are unwanted

[languid hawk]

huh

so all in puts and out puts would need to be mk1?

[oblique hollow]

Yes

Thats preferable

[silent shoal]

Won't the radiation scatter? Does it have a certain fixed radius?

[languid hawk]

cause 20 refiners makes 400 fuel so thats mk2 right?

[modest shell]

Could do 2 mk1 instead

[oblique hollow]

Yes, but the pipes directly out if the refineries can be mk 1 and so can every pipe up until the point where you go above 300/min flow

[unique cypress]

it decays exponentially, so it effectively does have a fixed range. it doesn't in theory, but it'd take an inordinate amount of time to have it expand beyond 1km

[languid hawk]

thats a bit more backtracking but back to the problem like is there any reason for this flucation or do my fuel need to run for a bit

or will replacing all these to mk1 fix it

[oblique hollow]

Which fluctuation exactly

Your fuel gens or something?

[languid hawk]

fuel gens yh they on and off

but I JUST made this

[oblique hollow]

Its likely they still need to fill for the next like... 30 minutes

After turning a system on it can take quite a while to fill everything

[languid hawk]

yup that was all i asked initially haha

[oblique hollow]

I didnt know it was only 400/min flow

[silent shoal]

Is there some kind of function? That sounds interesting.

[unique cypress]

yeah, it's on the radiation wiki page

[languid hawk]

actually i think my math is wrong cause a refinery does 40 fuel and theres 20 thats 800 m3 so do i need to split it in the middle? so i have 2 mk2 lines each for 10 refinery

[oblique hollow]

Yeah split this up

Its better to keep the 2 networks seperated

[wet scarab]

Might be working on a lil-something
All automated, just need to input your final output and (and for now, the items required for it), and it gives you the perfect power consumption, downclock and overclock values needed for 100% efficiency, and I'll be adding more features!

But now that means I have to manually input every single recipe in the game into my spreadsheet 🥲

(yes I know Quantum Encoder's power isn't linear but I plan on adding that in the future)

[still blade]

which alt recipe should i take?

[unique cypress]

encased pipe 100%

you should basically never use the default recipe for those

[gaunt marsh]

"um, akshually, all recipes have uses 🤓"

[unique cypress]

or you could just use an online calculator, take the number of machines the calc spat out and type in [result]/[rounded]*100 into the clock speed field

[wet scarab]

But where's the fun in that?
What if there's a global internet outage

• I mostly do it to incorporate modded recipes and stuff

[unique cypress]

SF Optimizer is an answer to both of those issues

[lone igloo]

Is OC Super Comp good recipe or its over complicated on the process? I mean it makes it so its only 2 items but those 2 items have decent chains them selves.

[unique cypress]

it's expensive af

super-state is better

or default, if you don't want to use any alu on SCs

[lone igloo]

thats what I thought

@unique cypress Pure or Default Aluminum Ingot?

[unique cypress]

depends what you want. default gives you more ingots per scrap, but costs a huge amount of silica

in most cases, I'd say pure

[mossy ibex]

yeah I've been debating that. My next build coming up is aluminum, I've been considering

which recipe

[lone igloo]

I'm about to start setting up LARGE aluminum production so I can make train chain to get cases and sheets for other things for T8 stuff as you already know. So just gathering some recipes.

[unique cypress]

default:
https://www.satisfactorytools.com/1.0/production?share=hOBOU8m9VQ5hHSX9PhUb
vs pure:
https://www.satisfactorytools.com/1.0/production?share=C5Ut2tRy2yLyJmBb6hrl

[lone igloo]

I like those tools to see end result but never use myself for actual game. I like looking for nodes making plan where I will place things and manual. Kinda feels more fun ngl.

[lime yew]

Is this enough computers?

My friend said this is light work.... he has like 20 hours...

[wind spade]

Whichever you like

[muted galleon]

nuh uh, I have 300hrs

[lime yew]

you play like you have 20

[unique cypress]

I mean those are certainly awful recipe choices

and an awful calculator choice

[lime yew]

what would you change?

[unique cypress]

and why 93.75/min?

[unique cypress]

https://www.satisfactorytools.com/1.0/production?share=2Ti01laWf7XdWy3hN0Bn

[lime yew]

its the max we can make pulling all availible resources within about 8k

[unique cypress]

oh, you can make way more

[lime yew]

do share?

[unique cypress]

I would have to know the resource limits

besides, you can figure that out yourself using tools

and also, making the maximum amout you can is usually a terrible idea

make a handful for personal use, and then make however many you need for more complex parts

[silent shoal]

@unique cypress
https://www.satisfactorytools.com/1.0-ficsmas/production?share=jTViysHhGacT0JTgWlUE

Looks promising...
And that will only give me 1470GW😅

[unique cypress]

I'd suggest turning off the ore conversion recipes

no reason to waste SAM on extra sulfur

[silent shoal]

Wow, you're right. I don't know when it happened.
https://www.satisfactorytools.com/1.0-ficsmas/production?share=UvsnBn5gZIMIRdvVkVde

Do you want to take a look to make sure I don't have any more retarded mistakes?

[unique cypress]

I would do this:
https://www.satisfactorytools.com/1.0-ficsmas/production?share=O08n7iGy6yzgk5kvYGWJ

but I'm not saying you should go with these exact recipes

I just optimized it for my own personal preferences

the only things I'd definitely suggest disabling are default alumina and default and bolted modular frame

though I guess the default frame is not terrible, steeled just saves space

tools often uses default alumina to make silica for silicon boards, but it often results in an absolute mess of an aluminium setup, if it decides to mix recipes

[silent shoal]

Wow, that's beautiful!
It doesn't destroy all the Bauxite😅

I need to build a production line now for the parts I will need in the future for this project and I was already thinking about doing it in the appropriate size...
For example, I need a 44.8 Fused Modular Frame, so I will build a line that can produce 50 and I will just have to transport them to where I will build the reactor...
And so on with additional components
Is this a good idea?

[unique cypress]

I built like 80% of my nuke plant in U8 in one spot, and I'll probably do the same in 1.1 when I finally get around to building it 😅
but it's not like distributed production is a bad idea

[silent shoal]

And how did you get the products?
I want to invest in trains but I'm afraid I'll have supply problems (in terms of quantity...)

[unique cypress]

the only things I processed offsite are oil -> plastic + rubber and bauxite -> alu ingots

those things and the raw resources were trained and/or belted into the main site (depending on distance)

well, except water, because I built the entire thing above an ocean

this is all to avoid transporting liquids. I only had to train in some nitrogen, but that was easy thanks to packaging being extremely efficient with nitrogen

[silent shoal]

How can I know when I'm crossing the line with moving items by train?
If for example I want to collect all the Bauxite on the map in one place, one train is enough to make a loop or two, etc...
Because I want to build the train system now...
In addition, if there are good blueprints I would be happy to receive a link🤓

[mossy ibex]

it basically boils down to how long teh train takes to make a loop and how many cars it has

your ultimate bottleneck is the speed of your belts into and out of the station

[silent shoal]

In other words, I'm fine/ If there's a problem, are there any "simple" solutions?

[mossy ibex]

solution is always "add more cars" or "add more trains" if you're not getting enough material in

[unique cypress]

yeah, if you're deep into the "train limited" region, the only thing that matters is the total number of wagons on the route. it doesn't matter into how many trains they're split up into

but you can have too many trains. at some point, you're gonna end up being station limited, and at that point more trains will at best do nothing, and at worst decrease your throughput

[silent shoal]

Do you happen to have any blueprints that look good?

[mossy ibex]

it depends what vibe you want really. There are lots of options, although I ended up making my own

[upbeat summit]

are these stats show in the train station relaible? also do they update automatially based on the amount of items are loaded / unloaded (say, that i have an X thing, so after the train container is full, it shows the rate of transfer to be 0, cuz im notusing it, but some other number if im using it)??

[hoary oar]

for the most part they are reliable.
its not gonna be the exact number you're getting on average, but its close enough

[upbeat summit]

okk

[visual ocean]

Hello, I'm trying to do a slight upgrade to my power structure and since I'm now confortable to do it, I opted to go with rocket fuel at the blue crater.

However I have a lot of compacted coal. I could sink it but I wanted to know what else could I do with it that would be productive

[unique cypress]

they take a while to average out, but after that they're about as accurate as your round trip time is consistent

[unique cypress]

with default rocket fuel, you could use it on more RF, saving coal and sulfur

like this

[visual ocean]

So, instead of Diluted packaged fuel, using RF

[unique cypress]

no, not instead

default RF uses turbofuel, which is made from fuel

you make the turbofuel partially with the compacted coal byproduct

and the rest either with turbo blend fuel, or with default turbofuel but made from compacted coal made by you

[visual ocean]

Okay thats already a little more power

[unique cypress]

or this

[visual ocean]

Damn, why is tools not giving those combos..

[unique cypress]

tools does this by default

you need iron for iron plates for the nitric acid

that might be why it doesn't work

[visual ocean]

Gave me that

only condition i added was crude to be at 300

[unique cypress]

the other options are generally less oil efficient, but more efficient in other resources

also, if you're using maximize mode, make sure to switch back into items/min mode before you build anything

maximize mode doesn't optimize for raw resources, unless it abolutely has to

[visual ocean]

i did not know that

what a game changer, ngl

i just switched to items/min and added manually the 1200

Much simpler build as well

[unique cypress]

this is what 1200 looks like with default RF

[visual ocean]

yeah, loads of crude there tho

Well you kinda saved my build, so thank you

[languid hawk]

so im trying to learn this game a bit more and pipes are still something that allude me, just wanna ask I have 450 oil in. One 300 line and one 150 line. 375 is going into refineries for plastics. 60 goes into rubber line and 15 into fuel line for not wasting.

I could just combine the two lines into one giant line and let it all trickle down but would there be any benefit of making the 150 start at one end and the 300 line at another to then make sure the tiny bit needed goes into the machines first

if that makes any sense

[vapid gorge]

no. In fact in general you want your fluid lines as simple as possible, fewest merges and splits that you can do to limit flow issues

[languid hawk]

with splits here i have 16 machines that those 450 oil needs divided on does that count for the splits? and thus it would be better to make some segments

3 lines that then each fuel a segment from source

[vapid gorge]

so technically , yes. The more splits you have along a manifold the more points where it's a little unstable

but it's far worse when you're splitting things before the manifold

so if you have a branched manifold fed off one pipe for example. You can do that, but you have to build around it

[languid hawk]

manifold is where it like trickles downward when a machine is full right?

[vapid gorge]

basically. Fluids are a bit different though , and it often helps to flood the system first by underclocking a few machines down the line

[languid hawk]

aah that is actually very fair

this isnt the power gen or anything this is just to produce resources so

[vapid gorge]

still, good habits to make 🙂

in this example you could have a 300 and a 150 line, completely seperate, and just clock the machines to make what you need overall

and in general you have more leeway with lower flow pipes. More tolerance for shenanigans

[pastel shell]

any feedback or better alternate suggestions to use before i start building this?

[vapid gorge]

would need to see what recipes you're actually using to comment on it
modeler is not a planner to share info from

it's more of a layout designer than anything

[heady vine]

But it shows recipes used, even if in a convoluted way.
You meant "available alts"?

[knotty hornet]

If you have limestone/concrete, do Molded Pipes to get better yield.

[unborn dome]

And heavy encased frame

Works great, boils almost everything down to steel pipe and wire.

[knotty hornet]

I think they are doing that.

[unborn dome]

Dunno, I didn't squint at the item icons to figure out which alts they were

[vapid gorge]

if you want someone to look at info to analyze it you don't hide it behind a bunch of icon you have to deciper each and every time you want your eyes to pass over it.

[pastel shell]

ye rn i got 10 belts of pipes rn, if i got 7000 concrete it would be 15 belts instead

[heady vine]

"a bunch of icon you have to decipher each and every time"
Yes, but....
Didn't you just describe the process of reading text? As in 'bunch of icons' being letters.
Funny

[fallow siren]

at least you should be able to tell which recipe by having model name

you cant figure if its normal reinforced plate recipe or bolted ones, they both use the same input type

thats my only gripe with modeler rn

[heady vine]

I tot agree that Tools make better layout for understanding recipes involved

[fallow siren]

modeler is great if youre using it for yourself, but to share it with others? not so much

[unborn dome]

(Molded pipe would've worked great here too, but I only had a normal limestone node to work with and I didn't have mk3 miners yet when I built this factory)

[vapid gorge]

it's not even great for yourself. The only benefit it has over some others is making nodes for layout. but it's so god damned slow

[heady vine]

It shows volume of input, so technically you can.
And i would even argue that direct numbers and ratios are more important then how recipe is named, but...
Yeh, we used to thinking in "recipe A os better then B"

[fallow siren]

also one thing modeler cant do is link share

you can only share them with pic, and it gets convoluted when its really big

unlike in tools where u can directly share links for said production line

so much to be desired

[knotty hornet]

Whereas I'm still here and have never used either website and never will

My abacus hand is strong and my notebook is filling up

[fallow siren]

thats fine tbh, i also still use my enote for my own sf+ planner

[knotty hornet]

I teach and tutor math, math is fun to me, so why would I outsource the fun?

[vapid gorge]

because with tools you can tweak 1 alt recipe in a large plan to see the knock on effects backwards and forwards and get an idea of the raw resource changes w/o spending an hour doing arithmatic you learn in year 4

spending several hours doing arithmatic a preteen can do does not sound anything like fun

[unborn dome]

I started out doing the math manually, but the lesson it taught me is I make a lot of minor mistakes that I don't notice until things start messing up after I build it

[knotty hornet]

Sounds like coward speak from both of you.

[vapid gorge]

I wouldn't force my students to do grade 4 maths for hours either. They've got better things to do with their lives

every interaction with you seems to be tedious. GL with things

[unborn dome]

It's "don't wanna rebuild this factory again" speak tbh

[frosty owl]

With the appearance of Modeler, I discovered that a surprising amount of people like to act like boomers that don't even recognize in-game icons... I don't know how else to word that 😅

[unborn dome]

A bunch of small icons, and the recipe in use isn't described.

[frosty owl]

99% of recipes can be recognized by their input

[unborn dome]

If you're already familiar with all the alt recipes, yes

[frosty owl]

If you're not, I'd argue you should be learning the game rather than using a planner. The comparison I like to make is: you're using a calculator to solve the simple arithmetic problems someone gave you for you to learn math (ie: the game's first "challenges")

[unborn dome]

Lol I'm almost at tier 9. I've learned the game. I just haven't memorized the 100+ alt recipes.

[frosty owl]

So if you see the inputs of a recipe you know, you don't recognize it...?

[unborn dome]

I'd have to go look it up, when I'd rather just read the text that says what alt it is.

I'm kinda confused why this seems so perplexing to have more info and not less.

[heady vine]

Why do you care about the name of recipe more then actual ratio of inputs and outputs?

[frosty owl]

That's not the perplexing part imo. Ofc more info can make things more "readable". The perplexing part is "I can't read this, because there is no text stating the name of a recipe"

[unborn dome]

Maybe it's a matter of being a heavy reader, but I can skim stuff really easily in text, and icons just don't parse as quickly for me.

[frosty owl]

Can't you see how easily one can make the opposite argument, based on simple preference? (Ie: "I read icons faster than text")

[unborn dome]

Anyway, I find the Modeller diagrams nigh-unusable.

Yes, that's the point - preference

The Satisfactory Calculator planner, at least, has both icons and text.

[frosty owl]

And is, objectively, the worst calculator too (it's the only one that leads to incorrect solutions)

[unborn dome]

Oh yeah, I don't use that one either lol

[frosty owl]

The thing is: if you prefer something, you don't normally go around telling everyone to do things your way. That's when you both dislike something (Modeler), like something else (other calculator) and feel like your isn't an opinion based on your preferences but a universal fact

[unborn dome]

I don't recall ever stating that tbh, just that I wasn't going to squint at the modeller diagram to try to decipher it

If you wanna use Modeller, all power to you, but I can't help with it

[frosty owl]

I'm not pointing fingers against you, just on anyone rudely stating how they won't help until the user changes production planner. If you fit the bill, I haven't noticed :P

[frosty owl]

BTW, I think it's clear that there's a difference between refusing to engage with something and actively engaging only to say "do things this way or I won't help".
I mean... It should be clear they are different things, right? 😅

[young agate]

if i make 640 fuel per min, dont i need 32 fuel burners?

[unborn dome]

You can do less fuel gens if you overclock them. No downside aside from using your power shards.

[young agate]

But i am correct

[unborn dome]

For regular fuel, yes, correct

[young agate]

Should I use 16 with 2 power shards in each?

[unborn dome]

That's what I've done in my game

200% OC on my fuel gens for a nice clean 2x in capability

[heady vine]

That's unusual
Most advanced minds develop interface for critical instruments and technologies aka powerplants, planes etc with design ideology that's a circle dial is much better to quickly read then write plain values as numbers, that color indicators are quicker to interpret then spitting out long error message.
And yet you claim it's better to have text. You could replace the whole game with just text. "Tree", "ore", "hog"... Welcome to text adventures.

[frosty owl]

People are different. It's not really unusual for individuals to deviate from any standard (ie: even if most do find icons better than text, it's not impossible for some to prefer the opposite)

[noble timber]

I'm somewhat the same - I find that when I have the choice between an icon or a word I would usually pick the text option. However, if there is colour involved i'd rather be able to see different colours rather than text or an icon

[frosty owl]

So, icon with color is peak?

[noble timber]

Perhaps 😅

Any spreadsheet I make for this game is just a colour bomb because I assign certain colours to specific values or items. It's cut the time it takes me to build things by so much 😂

[frosty owl]

Even with just text, adding colors already makes things much better

[wind spade]

even if I knew what the icon is (sometimes hard to see due to zoomed-out schema), it's still kinda bad, especially for new players
"hey this production line someone is recommending seems good, but idk what these recipes are, if only I knew the names so I could check if I have them"

[noble timber]

That's fair - an icon is only good if you actually know what it means

[wind spade]

and even if you know the icon, you may still not know the recipe
and since many times the planners are used to show new players a possible production line, I'd argue it should be more clear what recipes it uses

[frosty owl]

I don't see your point.
Picture bad because some npobs might not understand?

[wind spade]

my point is not "picture bad"
my point is "only picture bad"

[frosty owl]

"npobs"
Thanks autocorrect, very helpful

[wind spade]

add a label for the recipe

[frosty owl]

If your objective is sharing something with new players, sure. That's not what most of the sharing is about though, it's usually people asking opinions about something they have designed/planned.
This whole shabang started from experienced players complaining about how they don't want to waste time understanding a picture

[wind spade]

so many times people asking for opinions don't know the other recipes tho

[frosty owl]

They're the one making the production plan. If they don't know the recipes they have chosen, I'd say the issue is lack of game experience 🤷‍♂️

[wind spade]

not them
the person who replies with their own plan

[heady vine]

I am totally accepting that reading text is generally easier then some icons.
Only saying that in general, quick and critical communication is usually done with smaller visual detail and in tight context it works

[wind spade]

like, I don't see why it would be worse to have a recipe label somewhere. By all means, have a toggle so that you can turn it off if you're annoyed by it

[frosty owl]

Well, yeah, that's why if you want to show something to new players there are better and worse ways to do that. We just addressed that o.O

[wind spade]

so I have to only use the tool if I'm talking to experienced players? 🤔

[frosty owl]

I'd say to just expect some confusion if your trying to use it to tutor someone completely new to the game about recipes they've never seen before

[wind spade]

sure, but that's exactly what I'm talking about

I can't know if the receiving person knows about the recipe
I can't know who else will read what I posted

so why not use something that has the label in it? (or why not complain about lack of labels)?

I'm not saying to remove icons, people who know icons can easily use them
I'm saying that adding something extra (labels) would be a huge hep to tons of people

[frosty owl]

Because it's not always the most convenient (else everyone would just be using the most convenient).
Tbh, I've taken the "having a label is convenient" opinion at face value, but I'm rethinking that too... What do you gain from knowing the name of a recipe, over its inpit and output items?

[wind spade]

name is easier to communicate

[frosty owl]

Through text

[wind spade]

"why did you use stitched plates" vs "why did you use that recipe that takes wire and plates"

well discord is pretty much text-based, with a few images

[frosty owl]

I don't really see much value in that. You're already discussing an image, the terminology can be looked up if one wants (or they could use something that explicitly lists the names of recipes if they want that extra clarity)

[wind spade]

I know pretty much all recipe names, but I don't usually know exact resources they need

[frosty owl]

Even when just texting about recipes, it's very much not unusual to see a lack of names. Saying just "the alt" or "the one using X" is very much common

[wind spade]

and in 50% of cases that leads to "which alt do you mean" or "do you mean recipe X?"

and again, I'm not saying "replace images with text", I'm saying "add text to the images for more clarity"

[vapid gorge]

I cannot believe this one half of an argument which is as simple as 'names are easier to read and reference than multiple icons' is still going on. Or that anyone would try to argue against the obvious so long

You're wasting energy greeny.

[frosty owl]

(Which was not the point of the convo from before, BTW, that was "Modeler is unreadable because it lacks the name of recipes so only the creator knows what's going on")

[wind spade]

which is solved by adding labels

[frosty owl]

Yeah, I'm sure there's plenty of suggestions about that on Modeler's feedback forums or whatnot xD

[frosty owl]

The point wasn't wether it needed solving or not. Ofc being able to add labels would be nice over just not having a chance to do it at all; nobody said that.
The point was wether the output was readable just by the maker or not

[wind spade]

it's not tho

[frosty owl]

How so?

[wind spade]

it's not for me

I don't know recipes by icons, I know them by names

[frosty owl]

So if you see a machine making Quickwire with Copper and Caterium, you don't know what it means, but if it also says "fused quickwire" you understand?

[unique cypress]

Modeler has the option to add labels. They just have to be written manually

[frosty owl]

That's about as useful as not having them (exaggeration)

[wind spade]

not in this concrete example, but in many others yeah

also the icons are often ambiguous due to low resolution

[unique cypress]

Prolly because you haven't played the game in 5 years

I have 0 issues reading other people's Modeler screenshots.

The only remotely problematic recipe is bolted plate, because it's indistinguishable from default by just looking at the ingredients. You also have to look at the output rate or the plates/screws ratio

[frosty owl]

Can you give one such example? I struggle to understand the difference in information that you gain between seeing the icons for inputs and outputs VS seeing just the name of a recipe (or is it "name plus name of inputs"?)

Wether the info itself is readable is a whole different things (ie: just zoom enough)

[wind spade]

or because people have different capabilities

ie: just zoom enough
how do I zoom a low-quality screenshot?

[frosty owl]

The maker of a screenshot has to make it readable.
Just like SFTools' or any other screenshot is unreadable if not given enough zoom

[wind spade]

sftools is shared via link

[frosty owl]

We're comparing screenshots.
Ofc, SFTools can share the whole plan; that is extremely convenient; having the option to do something is better than not having the chance to do it at all... That's a whole other point

I mean, if the server allowed sharing files, Modeler would have a bit of an easier time with sharing, but that's, again, not the point

[unique cypress]

Modeler can share the entire plan too. With a file. Which is only slightly less convenient than a link if I'm on my PC

[wind spade]

I'd need to install modeller to see the plan

[frosty owl]

Yeah, that's fair. Still besides the point

[frosty owl]

From memory, I think Standard and Bolted plates are the only recipes sharing the same inputs and outputs... Maybe even the only ones sharing the same inputs at all

[wind spade]

point still is that the way modeller builds are shared is unreadable to me and many others here 🤷

[frosty owl]

Bumps

[wind spade]

I'm more of a text guy
I see icon, I have to convert to text
I see text, I'm happy

[crimson moat]

there is a rotor and stator recipe sharing both inputs, and they're used together for motors

[languid hawk]

I am trying to learn to make factories esp with sftools and this other site for splitter i found and this feels so illegal. Is this the skill of making a factory effecient or is this a necessary evil or am i just not used to how this game works. What feels illegal to be clear is the 15 into from the 75 line to make a 90 line

[vapid gorge]

load balancing is completely optional

manifolds work for any situation

[summer flare]

That distribution is based on a 5 way split. Splitters only do 2 or 3 way splits, so that arrangement does a 6 way split with 1 looped back to effectively give the 5.

[unique cypress]

1 single splitter would be able to do this in 99% of cases

[languid hawk]

I dont think i understand this

[wind spade]

If machine gets more than it needs, it fills up and overflows

--S--S--S--S--S
  |  |  |  |  |

[haughty barn]

It will take a bit to kickstart though

Especially if there are a lot of machines and/or high stack sizes

[wind spade]

Yeah, but it will mostly work while stabilising and you can do something else in meantime

[languid hawk]

yeah and is whats called a manifold right? What confused me is that is 5 splitters not 1 splitter haha im just a bit too literal there

[wind spade]

It can have any number of splitters

That was just an example

[unique cypress]

When you have 2 machines, or 2 groups of machines, and one consumes 30, and the other 45, if you split 75 between them, at first, each side will get 37.5

But the side that only uses 30 cannot use 37.5 and will fill up at a rate of 7.5 items/min. After all the machines' buffers are full, 37.5 cannot keep flowing and because machines consume 30, that's how much will flow

The other side, which was previously receiving 37.5, will now get the extra 7.5, now getting 45, which is exactly what it needs.

[storm geode]

if you've played Factorio, compared to this game, would you say balancers are mostly useless then ? especially with slugs and tuning of constructors ?

[unique cypress]

They're exactly as useful here as they are in Factorio imo

I use them in both games in pretty much the exact same way

[storm geode]

yes but where specifically ? have gone all the way to tier 8 without ever building one

[vapid gorge]

do you mean load balanced feeding? or evening out belts ?

if load balancing, it's purely aesthetic

as for evening out belts? pointless really. It's for people who don't like planning steps in advance and clocking machines appropriately.

[unique cypress]

I use them whenever I have more than one belt total of one item

By balancing belts, I can get away with way less tedious math

[wind spade]

They are mostly useless in Factorio as well

[unique cypress]

There's a difference between "not necessary" and "useless"

[wind spade]

And yes, here we just do clock speed changes to match ratios and such

[frosty owl]

Yeah, that's another one!
Though they differ by output

[unique cypress]

Yeah, I wasn't counting that one because you can tell the recipes apart by output

Copper and iron alloy are more easily confused because they both output ingots with the same shape and only differ in color

[frosty owl]

That's a preference (totally fine). It doesn't affect wether icons can be understood or not 🤷‍♂️

[crimson moat]

You can blueprint and clock em together so that 2 inputs go in, and motors come out

[frosty owl]

Yeah, I know, that's just a different topic xD

[storm geode]

another question : as far as moving stuff from A to B goes, what do you prefer to use between :

• belts belts belts
• trucks
• trains

i feel like the trucks just suck

[unique cypress]

Belts or trains, depending on distance and amount. I use nothing else

[storm geode]

yeah but when do you decide like : hey this is better served by putting a train line ? most of the time an mk5 belt (or several) end up easier to set up, faster and more throughput

(asking as an semi noob here, just my impression)

[unique cypress]

Whenever I think I will be able to reuse that train line for something else later

[vapid gorge]

entirely depends on how much, how far, terrain and personal preferences.

and when going up the tiers? I'll just use long belts for speed and lazyness

[hoary oar]

trains for large quantities, drones for smaller stuff, like <50 items/min
belts for the really close stuff, personally am not a fan of big belt highways, so i tend to build in locations where things i want are close together or really far away for trains.

[vapid gorge]

all the logistic choices have their place

[unique cypress]

Before auto-connecting belt blueprints, running a single pair of rails was also often easier than 12 mk5 belts when the distance is multiple kilometres

But with AC, it's difficult to justify trains if you never plan on expanding them

[storm geode]

I just wish trains were more powerful in terms of throughput, because when you have to cross half the map, it becomes cumbersome compared to belts

[unique cypress]

A single pair of rails can handle enough trains to transfer 100s of 1000s of items/min

[wind spade]

And since many people have similar "preferences", the tool is unreadable for them 🤷

[hoary oar]

an important thing for trains is also to look out which item you transport
say for simplicity sake, you wanna make reinforced iron plates, but you need to import 3000 iron
instead of importing the iron ore, you could import iron plates. they have a higher stack size compared to the ore, which means you can transport more with a single train

[frosty owl]

So, "prefer text" = "can't read icons"?

[vapid gorge]

you're failing to see the power of trains. Once you have the infrastructure you can have many trains going along the path

plus 1 platform can often do 1 belt of throughput. 1 platform between each point is a lot less work than draging a belt all that way

[storm geode]

but you still need to take into account, load time, travel time, unload time and travel time back

[unique cypress]

I usually move ores, plastic, rubber and aluminium ingots by train. Rarely anything else

[storm geode]

But I failed to recognize the reusability of train tracks, thx for pointing that out

[vapid gorge]

load and travel time aren't an issue - only throughput.

[unique cypress]

I was talking about the capacity of rails themselves. Half a mil of items per minute requires the tracks to be basically filled with trains all the way around, as densely as they can be while still driving at max speed.

[vapid gorge]

even if it took an hour for a tran to do the path, if it gave you the throughput you needed it'd do the job

[wind spade]

Indeed. I have hard time understanding them, even after focusing for long time.

[unique cypress]

The point is that you basically never need more than a pair of rails between any 2 points. As long as you keep adding wagons, trains, stations, the rails themselves are very unlikely to be a bottleneck

[storm geode]

how big do you make your trains on average ? how many wagons long ? and how much space between block signal sections ?

[vapid gorge]

however long as you need really. and you could do a couple max train lengths between block signals?

point to point trains, rather than trains that run loops is generally easier to set up

[hoary oar]

i still have ptsd about push/pull trains. not sure if its fixed, but in my U8 save they kept turning themselves around and breaking every factory imaginable by unloading/loading in the wrong order

[unique cypress]

Entirely depends how many items I need to transport, and how easy it'll be to build the appropriate balancer

So far, I don't think I've gone beyond 6 wagons, but that's simply because I've never needed to move more than 6000 items/min to 1 station.

If I need let's say 12 wagons on the route to maintain throughput because it's long, I'll either do 2 6 wagon trains or 3 4 wagon ones. Or 3x5 if that's more convenient for whatever reason

I add another locomotive for every 4th wagon after the first

Block signals go basically at every connection between rails, so every 100m. The only exception is rails before path signals, they get 200-250m of space

[unique cypress]

That's user error I'm afraid

Whenever a train pulls out of a station in reverse, it rotates 180°

You need an even number of stations with reversing, otherwise the train is backwards after completing its loop