math-and-meta

[upbeat tide]

Number 3 is what I use

[iron prairie]

I can tell you already that sloppy silica -> pure ingots is substantially more efficient than regular solution -> mix of pure + regular ingots if you're looking for 0 net silica consumption.

[torpid robin]

Cause if you keep going the most efficient ones you sometimes run out of resources . So you need to swap them around a little bit .

[upbeat tide]

I agree, but I break things down. If goal is x amount of y I like to break it doen into modular segments when possible

[torpid robin]

I do break them down when I come to make that line . It’s more just for the planning stage so I can see overall what’s using what . Then maybe change my alts to free up resources in other places . Then break it down later

[upbeat tide]

I am only losing 250 bauxite b using sloppy + slectrode + pure

honestly that is marginal loss

[keen flame]

So sloppy/electrode/pure is considered the best right now?

[nimble ridge]

hey, made this fluid output priority/overload for my aluminum, and figured i would put it here!

no leakage until the first pipe backs up

[idle knot]

nice

[nimble ridge]

-important- no pumps can be in line on the pipe, the one after the crossroads must be the only one

it has to be -pulling- from refineries without any other pipes in front of it -pushing- if that makes sense

otherwise i noticed it can spike and leak a little

[upbeat tide]

it would seem so yea. You do use more bauxite, but the difference is negligable

[keen flame]

When I was calculating it out, it looked like it was a difference of about 3:2, which is pretty significant in my opinion

[upbeat tide]

https://discordapp.com/channels/370472939054956546/558721941410807812/823645621255798816

vs

https://discordapp.com/channels/370472939054956546/558721941410807812/823640250471284766

first is normal, without pure, second is with pure

same end goal

1200 alclad and 2800 casings

[keen flame]

oh, wow, ok - only losing about 10% bauxite, that is actually VERY good

I wonder where my math messed up

[upbeat tide]

These are U4 numbers obv. In U3 pure alu is a F tier alt 🙂

[keen flame]

oh of course

[oblique hollow]

funny enough this looks like an exact inverse design of the usual overflow i use

[upbeat tide]

Its the sloppy alt. That is the reason this works with the pure alu alt

Imo

[oblique hollow]

this is the usual overflow i use

[keen flame]

I figured out what feels wrong. I'm comparing pure to regular both using sloppy, and if you compare the two directly, it is a bit of a drop - Pure ingots:regular is a 3:4 ratio, so it's about a 25% drop if you switch to pure

That's pretty substantial to me, though not horrible

[upbeat tide]

Yea thats the same loss in U3 too, 25%

[oblique hollow]

now that ive looked at it more: the reason it works is that the pump only acts as a one way valve here, a normal valve could do the same thing here

otherwise it does actually seem to be a normal overflow design

[nimble ridge]

@oblique hollow yes but the fact that its pulling is what makes it so that -none- of it ends up in the overflow otherwise, with a height difference thats smaller sometimes it still prioritizes the lower one but a little always gets pushed over the top

[oblique hollow]

thats why i usually go for 8 m. the only times when it goes over is when the stupid thing happens that fluid temporarily backs up for some reasons

[nimble ridge]

the pump ahead of the crossroads is what is causing the spikes

[oblique hollow]

to avoid any backfeed from the normal line, putting an upowered pump also helps

[nimble ridge]

its smoother if you let the refineries push

.. or a valve?

[oblique hollow]

same thing, but pumps are safer because they block head lift in the backward direction

[nimble ridge]

not that its a huge difference anyways to be clear what you're doing is completely fine

[oblique hollow]

the first pump is just to give it enough power to make it over that bump

you could make it 10 m tall or whatever

[nimble ridge]

yea see i made it a little shorter than the refineries to avoid that

[oblique hollow]

so the refineries have enough head lift to make it, i see

[nimble ridge]

yikes nice impostors

you can hardly tell the difference

[oblique hollow]

i prefer the pumps since i do not trust the default head lift from machines because its pathetically weak and doesnt even show limit ring projections

also the pumps ensure that nice and steady flow

and prevent backflow ( and in turn sloshing)

[oblique hollow]

amogus sus confirmeded

[magic shadow]

oh dear

i thought the LODs i got were bad lol

[ornate sinew]

Does anyone know how to find out the measures of Battery? I want to build a huge battery with tons of power storages.

[oblique hollow]

they are about one foundation wide, a bit less actually. Checked the wiki yet?

[nimble ridge]

just finished 430 casing & 430 alclad sheets per minute

time for a nap

[tiny sentinel]

not sure witch one to pick

[nimble ridge]

i use that heavy frame recipe

[tiny sentinel]

i went with the resin

[grave parrot]

I don't understand how liquids operate in a manifold design. I have 300 heavy oil in a Mk.1 pipe going to 5 refineries to make 200 fuel. It's an exact ratio, so there should be no leftovers. However as a test I had a single refinery making petro coke at the end of the heavy oil line. It's operating at about 70% efficiency so it's consuming a fair amount of heavy oil when I thought it shouldn't be consuming any (no leftovers). Does anyone have an explanation for this?

The fuel will eventually get backed up, but it wasn't backed up when I was running tests. And no pressure issues afaik

[iron prairie]

Unless you deliberately create a chokepoint, it won't be an absolute priority.

What you can do is set up a sudden rise in the pipeline, an upside-down U, such that the section after the U won't start filling up until the top of the U starts filling.

[grave parrot]

@iron prairie thanks! I really like that idea

[iron prairie]

Do note: I've heard of the trick, but never used it.

[magic sparrow]

Honestly, I tend to pick pocket dimension pretty much always because I use my pockets 100% of the time

[upbeat tide]

You know you have been playing too much ESIV: Oblivion when you read that as “I tried to pick pocket pretty much always”

[torpid robin]

When I was playing with pipe manifolds . I had a pipe at each end of the manifold connecting to each end of the feed manifold . And it solved my issues . Cause I was having problems too . What you also want to check is that you fuel refineries are all at 100 %

[upbeat tide]

Loops are the best way to avoid manufold end to end issues

[torpid robin]

Yea exactly what I was finding

[upbeat tide]

Pipes seem to behave when looped and many junctions

[torpid robin]

It was such a small thing . But was screwing up the whole line lol

Added a pipe at other end and boom all fixed . I was like dafaq

[upbeat tide]

I have something to blow your mind even McGalleon was surprised.

I give you this first build. This supplies a large diluted packaged array. It hates me

I give you a simple petro coke setup that loves me

Build A works but took several loops to get working

Build B has zero loops and works flawlessly

[torpid robin]

I can hardly see that on my tiny phone screen 😂

[upbeat tide]

Ill give you a clue

I colour my pipes 🙂

[topaz hedge]

pipes should be white :p

unless for fire supression, then they should be red or black with red markings.

[torpid robin]

No

[upbeat tide]

👀 USN pipework I see 🤔

[topaz hedge]

I've been around lots of pipes. never seen a purple one lol

[upbeat tide]

I use it to signify its a HOR pipe

[torpid robin]

Yea I paint my pipes too

So much easier

[topaz hedge]

Just giving you a hardtime. :p not usn, I don't even work with pipes, or the navy, just in buildings.. with lots of pipes lol

[upbeat tide]

All good

But navies and most maratime do color code pipes based on whats in it

Not military, sadly disqualified due to physical issues, but know several who are

[torpid robin]

Well . We don’t even have a military 😂

[topaz hedge]

That's kinda cool. they probably all follow the same code. I'm not military either, but I've worked in quite a few refineries, and buildings. and most of them are white, unless it's fire supression

[upbeat tide]

Pretty much or if its a gas pipe, liquid pipe, freshwater, seawater, waste water, etc

[topaz hedge]

back to factory, in ex I'm about to the point where I need to think about setting up yet.. another powerplant.. if my only choice is nuclear, would you wait until they take it to EA?

[upbeat tide]

Id wait to see what they do with the encased uranium cell alt as its borked atm

[topaz hedge]

yeah, that's what I was thinking. kind of a shame, but it happens i suppose

[upbeat tide]

already makes 31.5 nuclear rods in U3 from one uranium node 🤔

[versed violet]

"Check my math" question:
Is it more efficient to overclock miners, than use refinery for pure iron ingots? (both space and energy wise)?
If I got my math right,
MK3 miner (240 iron) + 7 refineries => 446 Iron ingots using 236MW
VS
MK3 miner at 200% oc (480 iron) + 16 smelters => 480 ingots using 156MW.

[Not nearly the point where I'm running short at iron nodes. Only using 3 norm nearby grass starting spot + considering an iron factory at nearby 3x pure]

[cedar mica]

Overclock miners, as there is a limit. If you have power issues, just underclock

[topaz hedge]

I suppose I have some preparation before I can think about going nooclear.

[cedar mica]

Nuclear?

[calm flax]

@versed violet More efficient in what way? I mean like you can both overclock them and use the pure.... so kinda hard to compete with that

[topaz hedge]

yus. my next factory might very well use almost 100GW

[versed violet]

OC + Pure ingot will only make sense when I max iron nodes nearby. Right now it would be more efficient to just tap the 5 impure ones than use refineries methinks

[topaz hedge]

OC everything and profit!

[versed violet]

not enough powah

[topaz hedge]

miners, 250% refineries using pure alts 250%!

[cedar mica]

50% clock speed, can run 3 machines on same amount of power...

[topaz hedge]

get moar powah! X3

[versed violet]

[well, just started my nuclear, but its flaky, and everything will collapse in amonth when u4 hits]

[topaz hedge]

unclocking machines to save on power waste the most precious limited resource on satisfactory - CPU usage :p

unclocking machines to make numbers work is completely acceptable though.

[versed violet]

trying to limit machine space too. So no underclock, unless its last one in chain

[cedar mica]

And building more power, can also stress the CPU...

Anything you build, will lag the game, to some degree. You just go to choose what you want

[versed violet]

I'll be trying for Distributed Lag ™️ next

[topaz hedge]

Sure, but if you want to try to squeeze as much out as you can, OC that power production stuff to.

Okay, that might be a waste of shards but. ohwell. what's waste anyway, we all know fiscit doesn't waste.

[cedar mica]

Na, just got to hire more dogs

[versed violet]

Got 400 shards, so gues I could put them in some coal plants

[topaz hedge]

we all have our own playstyle so there's no right or wrong, but OCing the machines at the end of a row, where you'll need for example 4.2 machines, build 4 and 120% OC on the last one definitely saves the amount of work you have to do, and saves one machine on your cpu usage, and doesn't have that much of a power cost.

[cedar mica]

If we are thinking about lag, foundations are really the worse thing you can do. But think thats a step too far

[torpid robin]

I hope they just delete it

[topaz hedge]

well, I've noticed that lots of walls and foundations cause lag in that area, however lots of machines cause lag anywhere in the world.

[cedar mica]

Build normally, then delete foundations and belt poles. Should buy the most amount of time

[torpid robin]

Depends how many you have . Lots of foundations start to lag the world

[topaz hedge]

True. Either way, I used to underclock machines at the end of each row, and then someone here was preaching about how it's better to OC them instead.. and honestly. it's changed my playstyle. I might've taken it a little too far when I OC'd a row of 28 manufactures to 200%..

[upbeat tide]

Herecy!

[cedar mica]

Then again, the fate of any building game is to die a slow death, as players add more and more stuff to the world

[topaz hedge]

This is also true

[torpid robin]

How . I feel how they have i atm is balanced . The amount of power you get from full nuclear and recycling is pretty good. You only get 223 waste now . Isn’t too resource heavy . I don’t think they should give people an easy out of plutonium by just going crazy on uranium . Like they added a new node for it all

Not to say that’s how it’s gonna go. That’s just my opinion on it

[upbeat tide]

You have to use that 4 th node tho to get up to 31.5 rods. Previous max was 94.5 soo 🤷‍♂️

[versed violet]

Is using up 31.5 rods/min realistic?

[topaz hedge]

I'm gonna try to... okay maybe I'll just try to burn 9 a min, with enough overclocks, that definitely seems possible.

[upbeat tide]

Yes...just need to build 158 nuclear reactors, a few thausand foundations, and countless pipes

[topaz hedge]

So with the power changes, does that mean max nuclear setups are dead, or will Kepler and others still do it. this is my question. lol

[cedar mica]

Max nuclear is not dead, just need to balance it out so Plutonium can run along side

[zenith hamlet]

I’m scared

[keen flame]

Just finished running numbers on nuclear with new info. Even with reprocessing, your maximum power potential from nuclear is still almost 20% less than it was in update 3, and that's before counting the ~175GW of power consumption to make the nuclear fuel in the first place. That's not with fully optimized recipes, mind you, but it's still a huge chunk since you have to use particle accelerators to maximize it. The absolute maximum is now somewhere around 900 GW at best, and that's if you use every ounce of nitrogen for plutonium reprocessing, leaving none left on the map for anything else...

https://www.reddit.com/r/SatisfactoryGame/comments/mak66i/ran_some_numbers_and_nuclear_power_seems_to_be/ Big discussion thread I've been active in on the topic

Even smaller bases will see a pretty heavy drop in their power potential

[zenith hamlet]

I’m more scared

[keen flame]

Put it this way...Uranium by itself now produces less power than fuel can. Only 393GW max from uranium alone, whereas with the extra oil you can produce well over 400GW with fuel power, and I was able to produce 151 turbo motors/min even with that much investment into fuel in update 3

I'd expect nuclear to be a muuuuch more substantial jump from that, as it was before, but I'm of the mind that max nuclear setups are dead now

[upbeat tide]

Well, this bypass has definately made my diluted fuel plant better.

It overflows into a petro coke factory on the side that sinks it

[obtuse scroll]

Can anyone help me optimize my Oil production?

[upbeat tide]

What u got? What you making? Alts?

https://reddit.com/r/SatisfactoryGame/comments/jmqw9d/pipeline_infographs_fluid_update_version/

Your one stop shop for fluid dynamics

Trains?

time the round trip (without loading/unload) first, then add 50 seconds. If the total time less than 4m06s, then 1 car per belt. If more than 4m6s, use 2 cars per belt or 2 equal trains. If more than 8min12s, use 3 car per belt or 3 similar trains.

Trains are very efficient as long as you time them right

You also need a buffer before and after each train station. On the load and unload side.

[obtuse scroll]

Do I still need a buffer if I am transporting Packaged Oil?

[iron prairie]

1. Don't. Packaging/unpackaging takes a lot of power.
2. Sorta yes, in the sense that you'd still want industrial storage containers at both ends to buffer in/out.

[obtuse scroll]

So, would you recommend I transport oil rather than packaged oil?

[iron prairie]

Yes.
Also, you'd need a separate train to transport the empty canisters back, so the "increased capacity" of solid freight trains is kind of a lie.

In terms of energy needed: pipes are cheapest, especially when little to no positive elevation change is required, followed by fluid freight trains, with packaged fluid freight trains being at least an order of magnitude worse.

[upbeat tide]

Oil trains work well. I use two of them.

I tend to split oil nodes into multiple cars. For example, if its a normal node that produces 300m3 total, that will split into two freight cars

If you do any liquid train, remember the golden rule

Pipe > pump > buffer > pump > station

And opposite at unload side

[obtuse scroll]

Can I send you a screenshot or screenshare, and you can tell me what I should get rid of. 😅

[upbeat tide]

Im half asleep, gonna shut down soon

[iron prairie]

I doubt you need a pump between the buffer and station so long as they're at the same elevation.
And do note that a single fluid freight car can handle 300 m^3/min so long as the round trip time is less than 5 min 20 sec.

[upbeat tide]

I know but these are 6:15 routes soo gotta follow my own train advice

I dont think the second set of pumps is needed either, its more a peace of mind thing since I know buffers can get weird with flor rates

[iron prairie]

Ah. Long train ride, it seems.

[upbeat tide]

Also use the second level of pipe outputs on the station. Might as well take advantage of some extra free meters of headlift

[obtuse scroll]

In this scenario, who's advice should I take on the train setup.

[iron prairie]

I know in my prior save, taking sulfur literally from the NE corner of the map to the SW corner of the map, I had to have triple freight platforms.

I... think we're largely in agreement?

[obtuse scroll]

Can I send you a few screenshots?

[iron prairie]

Sure? Won't have super-much time to look at it, though: it's getting late.

[torpid robin]

You actually can’t build the max if you want to recycle . Max is like 28.something . As you need uranium to make plutonium

[torpid robin]

Pure ? Or normal ? I found going from grasslands to desert I needed 3 for a nornal and 4 for pure

[full raptor]

You guys are nuts. I spent hours today just making oil production, getting distracted and ending up at the last recycling bit without finishing. Just 6 blenders worth of distilled fuel, not much either...

[iron prairie]

Two pure nodes on that train route, plus a normal. I might have been able to get away with just two cars for the normal node, but I figured it wasn't much difference to have 2 locomotives/8 cars vs. 2 locomotives/9 cars.

[torpid robin]

Well that was my thought too and there would be nothing worse than an oppsies and it goes another way. At least going over will allow you to have wiggle room is how I look at it

[ornate sinew]

I meant Batterys, not Power Storages.

[ornate sinew]

I now measured a Battery with the Silver thing on the Boarder of a foundation. It's 13 segments long and 7 Segments wide. I scaled that up to 39 and 21.

[oblique hollow]

this is the reason why i dont like when other people use battery when they mean "power storage". I read battery and instantly thought of power storage too

[ornate sinew]

When i say Battery i mean Battery, when i say Power storage i mean Power Storage. I want to build a huge building in the shape of a battery with Power Storages in it.

[graceful storm]

not advisable

[oblique hollow]

that many power storages could, at the moment, crash the game

aaaaaand are absolutely obsolete

[gilded maple]

Hills make trains faster? I notice down a hill auto pilot will go 200 and up it will go like 110, flat it’s 120, so are trains faster when used on a slope?

[gilded maple]

I made big box and I’m filling it to the brim with power storages

[topaz hedge]

I noticed that about the powerstorages, something around 400 and my game started to become quite unstable. 300 is fine though

[static rock]

Hi guys!!!
This here is my go-to belt compressor

I have a factory that im looking to compress down 20 lines.
(Yes i know this is a lot and im weird :-P)

Using this setup I would need a massive setup chaining these together.

I was wondering if peeps have a better design for a large scale belt compressor?

[oblique hollow]

the crisscross confuses me, is there something on the backside we are not seeing

[static rock]

Sorry, looking from the back obscured it too much, so ill break it down

Top Overflow gate

• Center - Any
• Right - Overflow

Bottom gate

• Center - Overflow
• Right - Any

[oblique hollow]

soooo.... in short: bottom one gets fed first..... top one gets overflow

how would this look for 4 belts though?

or 3

do you just stack them / add them ?

This remind me a lot of my variable flow divider from my pipe setups, just with multi input and output....

[static rock]

Essentially next to it is this same getup, but 1 merger higher.
So there would be another converyor belt running over the top of this that feeds the top of the next compressor, and the 'overflow' from this would be the mains of the next belt

Im mainly looking for a 20belt compressor 😕 And this doesnt seem viable

[oblique hollow]

now you got my gears runnin'

this is technically yours, just from the view of one input SS

[vast jungle]

main issue with multi-belt compressors is overflowing a belt...
easiest solution I know for n belts is a n x n matrix with a smart-splitter and a merger on each point

which tends to be a bit larger ^^

[oblique hollow]

gimme a moment

im thinking a multi-layer version of this

[oblique hollow]

which is probably just that

[vast jungle]

I once built a grid design for a 4x4 compacter... needed 16 merger, 16 smart-splitters, 4x4 foundations space and on belt-lift height

would scale to any size... but not sure its worth the effort

[torpid robin]

this is how i do mine. but i also do a loop back from the end to the start. i found it keeps things flowing

[static rock]

hey sorry, was finishing dinner

[static rock]

This is kinda what i was thinking.
doing this 5 wide, 4 high and then using my compressor infront of it

Still wouldnt be perfect but it would be compact

[torpid robin]

#screenshots message

and the 2 below. they are how i do mine. and i can just keep making them longer as i need to add more input or output belts

[oblique hollow]

this is what i was thinking, might be wrong, would have to test

its hard to think about all of this just in your head

[vast jungle]

yes, you are just missing the final "merge" line...

[torpid robin]

yea it took me a bit of stuffing around

[oblique hollow]

oh the one for all the overflow lines, ye?

[static rock]

Also this is the design for my compressor, but utilising crossing lines to compact it to my size
https://static.wikia.nocookie.net/satisfactory_gamepedia_en/images/d/df/Belt_compressor_schematic.png/revision/latest?cb=20200602073827

[oblique hollow]

Heres me inventing the Matrix Compressor again, out of thin air 😂

[static rock]

but doing this 20times makes me wanna vomit 😛

[vast jungle]

yeah, a 20x20 grid... you should ask yourself why you need it

[oblique hollow]

there might be something Programmable splitters could do, but im not sure

[static rock]

Because some dumbass built 240 pure iron ingot refineries and doesnt wanna rebuild it all to solve this issue 😛

and not a 20*20, just 20 belts compressed

[oblique hollow]

whats so bad about that, just put it in a train

[vast jungle]

20 input belts in one direction, 20 output belts in the other direction (90° turned)... space necessary: 20x20 foundations 😉

[oblique hollow]

what if its just 4 input belts

[torpid robin]

i do mine after all myh big stations. it makes it way easier to sort out a heap of 300/600/780 lines. just plug it into my systems and then you can take any amount out that you need

it is an easy way of balancing tbh lol

[oblique hollow]

wait, 240 times pure, whats the output of just one refinery

[static rock]

12 refineries = 780
240 = 780 * 20

[oblique hollow]

so compressing 240 refineries down to 20 belts, i see

do you have a top down view of your factory?

because if they all are in one straight line, this is easy

actually, if this is a very clean build, this is REALLY easy

[vast jungle]

just a belt to the side or up or down every X refineries when its full... maybe with a smart-splitter after X+1 refineries if its not exactly full after X

[static rock]

Umm... Kinda?
2 secs

and for reference i dont know exactly how clean you would call it 😛

[oblique hollow]

in the most simple case, you would only need to merge 12 refineries onto one line with just mergers

no overly complicated compressing required

[static rock]

already done
But the problem is someone might not have made all 12 in a row be active at once.

[oblique hollow]

well that sucks

aaaactually, you might just be able to get away with JUST the one line of merger-splitters

[static rock]

#screenshots

All of the ins and outs go through the pillars to the sides of each floor

Each floor has 24 refineries, 10 floors total.

[vast jungle]

so every floor produces 24x65 iron ingots/min?

[static rock]

yeah, or 2x 780 lines.
But i cant confirm they will be full

Hence im looking for a massive belt compressor haha

[torpid robin]

i mean seriously the link i posted to the screen shots up a bit would do exactly whatg you want

its wahts i use them for all the time

[vast jungle]

who cares if they are full or not? You need the 20 belts anyways if they are "nearly full"

you cannot compress them down to even 19 belts...

[vast jungle]

@oblique hollow your "780 stable" assumes that the first two belts have more than 780 together ^^

[oblique hollow]

actually this is just a bandaid for poor factory design

[oblique hollow]

stability really decreases consistently the further down you go. thats the thing here. the very last ( right) belt will NEVER be stable

[static rock]

Yes.
Yes i did a poor factory 😦

[vast jungle]

maybe you should just work on optimizing so that every floor produces full belts?

[torpid robin]

doesnt always work like that though

[vast jungle]

better to fix the issue than to fix it in the next step

[oblique hollow]

just increase water and iron ore input

[torpid robin]

he went for design 1st. then worked belts after. i understand that

[static rock]

Its mostly the fact that the inputs are using overflow gates.
Per line thats fine but on a much larger scale it starts to get a little jumpy at the end

otherwise this thing would be massive with the loadbalancers

[oblique hollow]

you have a smart splitter manifold?

[oblique hollow]

prefill all the refineries and just use a normal manifold

you will have to turn it all off for the prefilling, of course

[static rock]

2 lines of input at 420/line i think it was?

ofcourse 420 is a user, sorry for tagging you 420

[oblique hollow]

35 per refinery 🤷

so 420 per 12

thats actually pretty easy to fix.

refineries still take input when power is off, right?

cuz then all you need to do is flip a power switch

[static rock]

not that im aware

[river night]

afaik the input fills when they are off, but when they dont have power at all then it doesnt fill, or something like that

[oblique hollow]

point is: prefill em, then you can use a normal manifold. smart splitter sometimes get too greedy and wont fill the last machine

thats because overflow always keeps an item or 2 inside

[static rock]

huh, i didnt know that

[static rock]

ok, so i think your stable line manager, preflooding the system, and normal splitters are gonna do me the trick

Thanks mate, thats a lot of help 😄

[oblique hollow]

ive seen the issue on another users refineries before. was doing diluted packaged fuel. last refinery never got any water bottles because the smart splitter just preferred to feed the machine before it

[static rock]

yeah...
If thats the case all of a sudden large sections of my...
Everywhere
Are worrying 😕

[vast jungle]

I don't get why everyone is manifolding the dpf loops anyways

Keep the containers in self-contained loops

No splitters, no mergers, no storage boxes

Just three belts

[topaz hedge]

Speaking of fuel and turbofuel.. I guess my advice for large setups is overbuild the setups so they produce more fuel/compacted coal than they need. These 1-5% losses in belts and pipes are killing me

[oblique hollow]

this was the water bottles going in, not the fuel canisters

it all was a giant loop

[topaz hedge]

Lovely

[vast jungle]

Doesn't matter, don't manifolds the belts of a dpf loops... None of them

Makes your life a lot easier

[oblique hollow]

how else would you feed the machine? with a balancer? the bottles need to come from somewhere

i get what you mean, but its still a manifold

[vast jungle]

Like I said, no splitters and mergers at all for the loop

[oblique hollow]

i dont get how you are supposed to feed the machines then

[upbeat tide]

When you need to do that 40 to 50 times it becomes crazy. Better to manifold the canister loops

[vast jungle]

I disagree

Just put in 40 or 50 containers in each loop and you are done

[upbeat tide]

I setup my dpf in groups of 7 refineries. Works well for me

[upbeat tide]

Not even close to enough.

[oblique hollow]

henning please show a damn picture because i cant imagine what you mean at all. this might just be a misunderstanding

[upbeat tide]

You need at a minimum 400 canisters per refinery, packager, unpackager combo

Each machine can store up to 100 canisters and a bit left over

[vast jungle]

https://photos.app.goo.gl/6XFpkdAVoSwaMy4g9

Unpackager directly attached to Packager, Packager directly attached to Refinery, Refinery directly attached to Unpackager

[oblique hollow]

so seperate loops for each combo

[vast jungle]

100 empty canisters into (Water) Packager, done

because there are no mergers and splitters, even 50 canisters might be enough (depends on the length of the belts)

I know a few people who do it with 30 canisters per loop

the whole manifolding makes it easy for the canisters to become unbalanced... which is the reason for the whole mess... so don't mix the solid stuff from each loop...

[oblique hollow]

or just use DBF

and not deal with this whole mess at all

[vast jungle]

when you use self-contained loops there is no mess at all... they just work ^^

[oblique hollow]

packaging and unpackaging is a hastle though

[vast jungle]

thats true... but even if you want to keep the "normal" TF creation, there is a "Dilluted Fuel" recipe for the Blender too... which is as effective as the DPF loop.
I use DPF because I am still on EA

and I need more (turbofuel)power right now...

[oblique hollow]

with the blender, this would all just be.... Refinery (HOR) > Blender > Blender > there you go its turbofuel

dpf is.... Package water, Produce HOR, HOR + Water to Packaged fuel > Unpackage > Make Turbo

[vast jungle]

yeah... but IF you use DPF, please don't manifold them 😉

[upbeat tide]

this is how I setup mine, and I have had zero issues ever with this method

[vast jungle]

the issue with this design is that you need enough containers in the storage to make sure that the "upper" DPF is getting enough of them... I would guess you have more than 400-800 containers in your storage, right?

[graceful storm]

enough but not too much, unspecified atm

[upbeat tide]

yes, but that is a known need. Plus when you do this 6 times in one build this becomes a way to keep complexity in check

[vast jungle]

thats why I like the small loops... they stay sane, regardless of their number ^^
and you need less containers

in my new TF powerplant I am even thinking about not manifolding all the created fuel... to keep as much of the factory separately from each other.

unfortunately the rates for the HOR creation don't fit to the DPF input... maybe worth downclocking them to 75% so its a 1-1 too

[oblique hollow]

using a sub-floor to manage the belts and lifts could actually make this very pretty

[vast jungle]

thats what I am doing...
when you look at my image the upper thing is the unpackager, the lower one is the packager (with output towards the Refinery)...
the packagers output belt is going below the refinery and into its input on the other side.
unpackager output and packager input are directly connected with a belt lift... as are the refinery output and unpackager input.

[wicked tinsel]

technically, since it uses 60 items/minute on the branches, i think if you use crap belt it wont stack items inside the packagers 🤔

[vast jungle]

yes, but its still something easy to get wrong and difficult to debug

[wicked tinsel]

well yeah

i just threw container worth of canisters at mine 😄

[vast jungle]

I like to keep my power production as simple as possible ^^

[glacial hemlock]

15 stacks for 5-loop manifold seems fine for me. If you go for individual loop, you can get them run with lesser canisters

[vast jungle]

I am thinking about building completely independent slices...
each slice "1x HOR production, 1x DPF loop, 3x Turbofuel creation"

this should scale very well and should be easy to debug

[frosty owl]

Ye, I underclock the HOR. Unless you're particular about how much oil the setup must consume, it's easier to handle the extra oil rather then some extra HOR imo

I still think that manifolding at least one of the products is the easiest way to have a handle on big setups for debugs and such ^^
Also I don't see the big issue of having many canisters in the loop (unless one really runs low on them for some reason)

[signal matrix]

What is the blue thing that comes out of the refineries when I transfer crude oil into fuel?

[tiny sentinel]

thoughts? I'm out hard drive hunting right now.

[signal matrix]

At the moment, Im just dispening it into the Sink, is this wise?

[tiny sentinel]

that's resin. use it to make plastic

[signal matrix]

Is it more efficient to build more machines, or overclock them

[tiny sentinel]

you could also skip resin altogether. make crude oil into plastic/rubber and Heavy Residual oil, then refine the heavy residual oil into fuel. you'll need the plastic anyway to make fuel canisters for packaged fuel.

if you have the power shards to spare, overclocking them can be more efficient. fewer machines the better when you get into late game and the save starts to lag.

[signal matrix]

Ive been holding off on making Nuclear power because of the infinity problem (the problem thant Nuclear waste cannot go into the sink) should i wait for the stable build to come out, because Im not sur eif it is going to keep the Plutonium fuel rods as going into the sink

[tiny sentinel]

you only really have to worry about lag when you've built a mega base and you're approaching the object limit

[frosty owl]

Just don't bother about the waste. Not worth holding yourself back because of some radiation ^^

[tiny sentinel]

most people are pretty sure they'll still be sinkable

[signal matrix]

Its not the radiation, it is the problem that it has to be strored indefinitly.

[vast jungle]

it doesn't really matter... Plutonium waste will coming out only in small numbers... so you can easily game for months or years without running out of space to store it in a remote location

[frosty owl]

I'm for the unsinkable side. I bet all my coupons on that

[tiny sentinel]

they wouldn't have had a point value assigned to them if sinking them was a mistake.

[signal matrix]

Im personally deciding to wait until that fact is set in, because then I can set up my power plant at the edge of the world and can eliminate any need to visit it

[frosty owl]

It doesn't. You can still just process it later after U4 and either sink it or compress it to 1/5th or less.
TLDR: you have plenty of choices to deal with it, why care about a few ISC of waste? ^^

[signal matrix]

Maybe they want to know how people would react to this change. Maybe people like the challenge of the problem

[signal matrix]

Have you SEEN ImKimbitz. He makes 25,000 waste an HOUR

[tiny sentinel]

i think it's a fair pay off, given how much power it takes to process the plutonium and how much power you're giving up by sinking it.

[vast jungle]

plutonium takes a lot of Nuclear Waste and only create a bit of Plutonium waste

[frosty owl]

ANY well done nuclear plant should follow that idea. Regardless of U4 or not.
Except with U4 you may want to touch your waste again (which you would still do in the same area, probably)

[vast jungle]

and the waste still stacks to 500

[frosty owl]

Have you seen the specs needed to produce such amounts without having your PC die? ;)

[signal matrix]

I havent started on the waste.

[vast jungle]

just build a drone-port to fly it to the processing area 😉

[signal matrix]

Im just being careful.

The reason im doing this, is because of a VERY simple reason

Lag

If you store something Indefinitely, then Lag will start up

[frosty owl]

Btw, since waste stack in 500s, that's still about 90 HOURS to fill one ISC

[vast jungle]

stuff inside an ISC doesn't really produce (much?) lag... because there is nothing to transport or render

[frosty owl]

You're better off worrying about machines, really ahahah
Eg: if your machines back up, the number of items idling around easily dwarfs the amount of waste you can have. Unless you've been playing with nuclear for 200h or something

[vast jungle]

build a "nuclear waste storage" from 100 ISC and even at ImKibitz rates you would be good for 9000 hours... or (lets say 10 hours a day on average 😉 ) 2-3 years of constant playing

and 100 ISCs is not that much

[signal matrix]

I know, but still, it is just not a problem i want to deal with

[vast jungle]

fair... but I think this was (and should be) the tradeoff for nuclear power

[frosty owl]

Imagine being killed while walking with no care just because a drone filled with waste passes overhead, sucking away your 3 ticks of HP in a moment

[signal matrix]

IS there any talk of Mk 6 Conveyors, because mk 3 overclocking 250 just wastes energy

[frosty owl]

You can overclock to 780/min now

[vast jungle]

its a performance problem I think... similar to the Lights before U4...

[calm flax]

there is talk but don't let belt speed limit you. Its only a problem when a single machine need more than a single belt can do

[vast jungle]

as is the case for miners

[frosty owl]

Tbf, Mk6 MIGHT be possible, if you disable its interactivity
Eg: Mk6 belt kinda like mk5 but with a "glass cover" so you can't pick items or be transported by the belt. Only visual rapresentation needed

[calm flax]

Yeah its a problem for miners. They should have multiple outputs imo. But if you have a mk5 belt @ 780. Whats next mk6? 1440 or something? In which case you just hit the next limit up

[vast jungle]

a new MK3 miner with 2 outputs would be a nice solution

"MK3b Miner"

[calm flax]

just build an infinite manifold which refresher points

[frosty owl]

If you want, you can set up portable miners on the node and unload them I to a container, treating the container as a miner

[upbeat tide]

There was a mod that added multiple outputs to a miner but it didnt last long due to lack of updating it.

But a dual output miner would solve the cant get to 1200 a min issue

Plus 600 a min belts are much nicer anyway

[frosty owl]

Would also let us make more pasta

[upbeat tide]

More aluminum!

[calm flax]

i normally run stuff at 720 rather than 780. That way when you drop below about 60/min left on the belt you can merge another 780/min belt into it

Sometimes as low as 600

[upbeat tide]

I prefer 240/480/600/720/750 depending on circumstance

[calm flax]

it generally depends how much the machine that being fed needs. eg for bolted plate it needs 90/min which is a 720 belt for example

so after 8 assemblers you have an empty belt 🙂

[frosty owl]

Generally I like the 720/750 too :)
But I use 480 often to load freights (I keep 480x2 as the max)

[calm flax]

i also use mk1 for machines where suitable cause you know the gap size from splitter -> machine so you know if its working correctly

also manifold populates better

[frosty owl]

Bruh. I use so much MK1 ahahahah

Curiosity questions. What are your belt percentages on SCIM?
Eg: 10-10-20-20-40 means 10% MK1, 10% MK2, 20% MK3...

[wind spade]

Manifold does not populate better with mk1s, it's the same

[upbeat tide]

I use mk1’s ocassionally but mostly forlooks. Like on my DPF loops between machines, etc

[frosty owl]

With 60/min inputs and MK1 belts you get the perfect: Balanced manifold

[timid yew]

i've a problem or it's maybe me who are idot idk i want produce 3600 m3/min turbofuel
so i did that
but when i look on the refinery and do calc from the recipe i got 2250m3/min crude used

[calm flax]

@wind spade Actually it does. I am talking here like mk5 -> mk1 on the splitter -> machine only. So when you do this the splitters for the 780 is going to immediately overflow the first 12 splitters. So you get to about 80% population almost immediately. So the first 12+ machine start building immediately rather than sequentially until the buffers fill up. So if you have a manifold -> manifold -> manifold kinda setup you populate the downstream ones way faster than pushing a 780 all the way.

[timid yew]

75x30 = 2250
bc 100 heavy on 30 refinery = 3000

[calm flax]

The main difference isn't so much that you populate the manifold faster. But you get get a greater level of paralellism faster

So technically if your pushing conrete at 780/min down a manifold at the start and you have < 12 assemblers the manifold is populated instantly.

[wind spade]

over time the saving is zero. You still have to fill the same amount of machines with the same amount of items, so e.g. if you have 10 machines with 100 stack size item, you have to fill 800 items (as the last two don't get filled), so your output will be delayed by 800 items anyway. If you use fastest belts instead, it sure does start less machines, but overflows them faster, so it balances out. I've done the math and the simulations, if you run both setups at the same time next to each other, at the moment when they are both filled to max, they have produced the same amount of items in that time, meaning there's no real gain.

[calm flax]

Ahh filled ad when it starts producing items are not the same thing.... It also does matter if you turn things on / off or re-route resources

[wind spade]

I know they are not the same thing. But while you fill more machines more early, I overflow them faster, resulting in same time to fill up.

[calm flax]

But when they are full is not important. When all machines are at 100% production rate is

[frosty owl]

@calm flax I think you're referring to a balanced-manifold rather then just a manifold at that point ^^
There is no NEED ot populate that

[calm flax]

@frosty owl No I am not

[wind spade]

that's the same thing in manifold. all machines at 100% production rate is when all machines are filled to max (with exception of last two)

[calm flax]

nope....

they are at 100% production rate when they have enough sustained items to produce at 100%.

[wind spade]

explain how you can have 100% production rate BUT some machine is still filling up

[calm flax]

i just did.... you didn't nderstand it?

[frosty owl]

Aren't you talking about a manifold with 60/min outputs?

[wind spade]

I don't understand really. You have e.g. 100 input and 10 machines eating 10/min each. If all machines are producing at full speed, it means they are all eating 10/min, so there's no overflow items left, so no machine can be filling up

[calm flax]

@frosty owl No mk1 inputs with a 780 backbone

[wind spade]

example from my manifold tool:
5 machines @ 20 input, 100 total input
see how the "time to fill" is same as "time to 100% production"?

[calm flax]

yes 17 minutes... but you can do that in 2-3 seconds....

[wind spade]

yes, with a balancer 🤷‍♂️

[calm flax]

without a balancer....

[wind spade]

no

[calm flax]

yes.. for the first 12 machines.....

[wind spade]

only if the inputs are mk1s and input per machine is 60/min

that's very special case and at that point it's kinda like a balancer anyway

[calm flax]

yes which is the case for a lot of machines.... except like screws

[frosty owl]

So... A manifold of mergers with 60/min inputs and 780 merging line feeding a manifold of splitters...?
Are we using "input" in the same way here?
By manifold's "output" I refer to the belts connecting the manifolds OUTPUTS (the splitters outputs) to the machine's INPUT

[calm flax]

Like a good example for this would be encased beam with concerete with a stack size of 500

[wind spade]

I'm talking in general about any manifold

[frosty owl]

That's a whole meme for me to make xD

[wind spade]

for those manifolds, all possible machines are filled at the point when the manifold starts producing at 100%

and for those manifolds, putting mk1s to input belts doesn't help you at all

[calm flax]

it does for the initial fill.... but you not thinking of how because your only seeing "my way is the only right way"

[wind spade]

initial fill is kinda pointless 🤷‍♂️

[calm flax]

I agree having mk1 doesn't help you or make any difference after its been running for 20 minutes. It des however make a difference int he first 5 seconds.....

[frosty owl]

I just call those "balanced manifolds"
So you both ARE referring to those 🤦‍♂️

[calm flax]

initial fill is pointless until you turn things on / off... then suddenly it becomes important

[wind spade]

if you're caring about immediate output, then idk why are you doing manifolds, balancers are for you 🤷‍♂️

[wind spade]

wdym by that?

which things?

[calm flax]

Like the factory building your in

you know switch off the power plug cause you don't need it running. Then suddenly when you need it running again you switch it on again....

[timid yew]

someone can help me ?

[calm flax]

Which suddenly makes it massively important because you don't want to wait 20 minutes util it is producing materials at its respective rate

[wind spade]

I don't really understand, are you referring to the factory that you're building, or a factory that you're supplying with the manifold?

[vast jungle]

Standard U3 Recipe?

[timid yew]

no

[calm flax]

Can when you have that manifold feeding a downstream manifold with what it produces now takes 40 minute to filll

What its actually suppling isn't import the point is you want to go from empty -> 100% as fast as possible.

[vast jungle]

3600 HOR from 2700 Crude Oil sounds fine... 3 to 4

[wind spade]

in the end your factory will make the same amount of items/power/anything, so it's only relevant if you REALLY need a few items from that factory in the first 2-3 minutes or so

at which point it's just easier to handcraft them

[calm flax]

yes... but you have to wait 20 minutes for it. So... well you can keep waiting 20 minutes for it. I will keep waiting my 2-3 seconds until it gets to 100% production rate

[wind spade]

you don't get 100% production rate at 2 seconds from the whole manifold lol

[frosty owl]

@calm flax , despite me loving the idea of not taking forever to load machines (thus why I usually balance) I think greeny is in the right here.
Unless your machines require exactly 60 or 120/min there is no convenience in using low tier belts as the manifold's outputs (belts feeding the machines) as far as time needed to load the machines is concerned

[calm flax]

you do for the first 12 machines.... which is what your not understanding...

[timid yew]

for the mistake i don't understand why it look like don't use 2700m3 crude
i got with the recipe 100m3 heavy but it take 75m3 crude
so to get 3000m3 i just done 30 refinery at 250%
but if u look closer 75x30 =2250

[wind spade]

I'm understanding that, stop putting words in my mouth. I'm saying that while you get 12 machines running at 100%, I get e.g. 4 machines running at 100%, but very quickly overflow to more machines, eventually catching up with you and going even faster than your setup

[vast jungle]

but you don't need (according to the chart) 3000 but 3600... 2400+600+600 = 3600

[timid yew]

i need to watch better xD

[calm flax]

well you don''t go any faster you get to the same speed but finish later actually....

[timid yew]

it seem maths burn my brain xD

[vast jungle]

three arrows with HOR is a little bit annoying, right?

[wind spade]

graph of what's happening

(rough estimation, probably not to scale)

[timid yew]

yea i missed the last part

[calm flax]

i see you inverted your items produced axis

[vast jungle]

so ready to build the 90 Refineries on your quest for MORE (non-nuclear) POWER? 😉

[timid yew]

power shard are ready to help me xD

[wind spade]

basically the integral of both "functions" (those that would produce such a curve on the graph) has to equal to same value, since both manifolds will have to produce same amount of items at the end (and produce same amount of items in the process)

[calm flax]

yes but i am not talking at the 17 minute mark. I am talking about the first 30 seconds.....

[vast jungle]

I am building my own (U3) second turbofuel powerplant at the moment, so I know the pain of "LOTS of Refineries"

[wind spade]

sure, but what use is that you produce more items in first 30 seconds? if you really need the items that fast, go for balancer or handcraft them. The whole point of manifolds is that they take some time to work, so if your factory has to go 0 -> 100 fast, then you shouldn't build manifolds

[calm flax]

there is certain behavior you missed in the calculation because what i am describing acts more like a sets of steps on a graph :/

[timid yew]

yea i do that b4 do a LOTS of factory to store at base

[wind spade]

I can simulate how the graph works if you want to. I have the code for it, I can plot it exactly as it happens 🤷‍♂️

[calm flax]

or you just run it in the game both ways and see which one finishes first

[graceful storm]

what is it comparing? the belts that run to the machines?

[oblique hollow]

wth is this belt stuff about

[wind spade]

what do you define as "finishes"? starts running at 100%? if yes, then mk5 belts win

[frosty owl]

I bet on greeny of you test it xD

[calm flax]

@graceful storm ITs 780/min manifold backbone with mk1 from splitter to machine mk1 vs mk5 for the splitter -> machine connection

[wind spade]

manifold side belts, mk1 vs mk5

[calm flax]

so the mk1 gets up to speed sooner because it it will overflow the splitter to more machines faster for up to the first 12 machines.... since the 780/min has nowhere to go

[graceful storm]

so it's a waste of setup time to use mk 1s good to know

[wind spade]

unless you like the looks, both setups will work the same and have the same "delay" at the end.

[frosty owl]

Machines will START faster, but not up to speed

Unless they require 60/min ofc

[graceful storm]

but 780 overfills faster so I guess it catches up

[oblique hollow]

aah, so this?

[wind spade]

yeah we're not talking about manifolds that require 60/min, since they are very special case that works instantly

[graceful storm]

do 60 below not work instantly?

[frosty owl]

Eventually, it reaches 100% a bit sooner since it doesn't try to fill the last 2 machines as much as the MK1 one (thus filling the others sooner)

[oblique hollow]

then what is the use case because this stuff ONLY works if the machine requires less than or equal to mk 1

[wind spade]

yeah, it even goes faster for some time, see the "paint graph" that I posted, rough estimation, I'm working on simulating and plotting the whole thing

[wind spade]

their reasoning is that more machines start working at 100% faster 🤷‍♂️ idk why is that relevant if the end result after a few minutes is the same

[oblique hollow]

not necessarily faster, but at the same time, this only works if they all need less than mk 1 though (or equal)

stack size might affect this.....

[calm flax]

It makes sense if you turn stuff on / off to have fast production sooner..... otherwise (as i have said previous) it isn't important

[oblique hollow]

oh god im having flashbacks to my monster equation

[wind spade]

well yeah, you don't get any benefit from putting mk1 belts if you have 90 input 😄 but we can generalise this to "putting slower belts on sides"

[calm flax]

@oblique hollow yup stack size does effect it

[oblique hollow]

duh, so does any manifold

[wind spade]

if you turn stuff on/off while manifold is filling up, then you get no gain as well. Since the mk5 takes more time to empty. There's really no gain

[oblique hollow]

Let me put it this way:
as long as the manifold is shorter than 4 machines, mk 1 will have no effect, and mk 5 will be just as efficient

this is because splitters split in half

not exclusive

[calm flax]

OR split in 3

[oblique hollow]

the 4th machine would get 1/8 of the input at first

of a mk 5 beslt, thats 97,5

so still faster than mk 1

[wind spade]

it's essentially like a buffer. mk5 fills more of the buffer but produces less items at start. But if they stop getting input, mk5 has more buffer so produces for longer (or has faster start up time when it starts getting input again, because it has more left in buffer)

[oblique hollow]

Using slower belts is only beneficial when its a long manifold

then it will actually reduce startup times

dont bother with short manifolds though

[wind spade]

for a few machines, not for whole manifold

[oblique hollow]

yeah, the machines further back. it gets horrible at around 8 machines for normal manifolds, assuming a large stack size

this topic is way too complicated to dumb it down to "if you do this then it works"

its all too variable

@calm flax your case would start to fail at large machine numbers (around, like, 13) too

there is no universal solution

best you can do to decrease the lag of large manifolds is to use injection manifolds

[calm flax]

But thats the assumption if you have a splitter. feed it 780. The first 2 splits left/right are going to take 60 each so 120. So you still have 660 flowing in the middle. So 60/60/660 isn't exactly "half". So the middle gets a step effect like 780 -> 660 -> 540 -> 420 -> 300 -> 180 -> 60. Where a mk5 connected goes. 780 -> 260-> 86 -> 28 -> 9 so a mk5 belt initially feeds something usable to the first 7-8 machines max. The mk1 feeds the first 12 (possibly 14 depends on recipe requirements)

[wind spade]

I'm working on making my manifold tool export csv data to be able to import them to google sheet and plot the data. I can run it for a few different setups and you can check the results yourself. Speaking of which, can you come up with a good set of tests to cover all the cases?

[oblique hollow]

exactly. that case works best above 4 machines and below 14 machines

above 14, it fails too

"fail" here means "doesn fill at roughly the same time anymore"

if you simply prefill all machines by hand, though.....
you can manifold it all with mk 5

[calm flax]

yeah so for things like concrete that matters because the stack size is now 500. So for something thats 12-14 long or so. The mk1 works better initially for population of the manifold because all machines get sufficent quality of materials to work @ 100%

[oblique hollow]

shut up Floof xd come here if you wanna debate

[graceful storm]

why would you manifold concrete large scale?

[calm flax]

@graceful storm encased pipe / beam?

There the mk5 belts even after the 3-4 minutes it going to still be seeing materials flow of things like 9/min

[iron prairie]

Also, if you have Smart!, Area Actions, or similar, you can literally blow thousands of concrete with just a few clicks.

[oblique hollow]

this isnt about building though

[graceful storm]

I don't and still do

[iron prairie]

I was answering "why would you manifold concrete large scale?"

[calm flax]

yeah which has absolutly nothing to do with building :S

[oblique hollow]

this is about manifolding concrete into machines

not building foundations

[calm flax]

or anything with a large stack size so this effects things because we have stack size increases everywhere

[graceful storm]

but fair, I just set up local concrete places, by the time I need concrete it's there, But encased beams/pipes is a valid reason in hindsight

[calm flax]

So when you start to have a manifold feeding another manifold. This effect is cumulative. Since if the first manifold take 5-10 mintes to get up to speed rather than in the first 1 minute. Its going to take even longer as well

[oblique hollow]

thats why you mix balancers and manifold for injection manifolds

and maybe that belt thing too, for the ultimate injection draw-reduction manifold

[calm flax]

@graceful storm I have a wet conrete factory on a pure nodes for a build connecting to a train station so i can have it delivered anywhere on the map at a rate of about 880/minute or so for large scale building (same for plates)

Smart mod + concrete train = empty train

[graceful storm]

thought about something like that, but still setting up proper train infrastructure

[iron prairie]

If you're talking about injected manifolds: I'm not sure you need balancers per se? If it's all unidirectional, bringing in a fresh line resets any manifold effects.

[graceful storm]

maybe before, so all the belts are more distributed evenly compared to overflowing many manifolds

[calm flax]

@oblique hollow What do you mean by injection manifold? Feed from both ends? feed from the middle? or overhead feed into the middle?

[oblique hollow]

anything! injection means feeding from multiple points, feeding from the end only works if you reverse the direction half way through or something, since these arent pipes

[graceful storm]

I just add a merger in the same direction

[iron prairie]

Usually, to me, injection manifold implies "I need more input than a single belt can provide, so I feed in a fresh belt every so often".

[calm flax]

yeah thats what i mean by feed from both ends. Where you end up with something thats like a 2:1 input ratio when it makes sense to feed one input from one direction and another input from both ends for example

[iron prairie]

I mean, you could split a single belt to make an injected manifold, but that makes the manifold more complicated for little real gain IMO.

[oblique hollow]

the splitter reduction ratio after the 5 splitter would be 1/32

so this way, you can reset this reduction havlway through or something

its basically breaking it up into multiple parts

[iron prairie]

The issue there is it's not really all that much help. Sure, the ones towards the far end of each sub-section get a bit more input at startup, but it slows down the process of flooding the machines prior.

[oblique hollow]

thats why mixing it with mistralol's method could help too

unless your ratios arend multiples of one belt

[glacial hemlock]

inject manifold may be used for multiple saturated belts feeding into a long manifold

if your idea is to split a belt into multiple belts then injects them, that's not the full picture

[calm flax]

Depending on their being a buffer in the system when its turned off like an industrial storage container or just the buffer in previous machines from supply of materials. Then yeah if you double belt it you can probably fill 12+12 machines

[glacial hemlock]

container may be useful, but is not a necessary component here.

[calm flax]

So 24 long manifold in thery could come up to 100% working speed in 1 minute

@glacial hemlock yeah thats why i was saying using the source materials buffers and a double belt from previous factory

i just realised my concrete train can now deliver 128,000 concrete for building supplies after stack size increase to 500 :/

[glacial hemlock]

that's great news.

[calm flax]

seems a bit overkill now tbh

[glacial hemlock]

that also means trains are now buffed 2x for common items and 5x for compressed items

[calm flax]

they increase slot size in trains?

[glacial hemlock]

no, but the golden 4m6sec is going to be 8min12sec, which is a great boost

[calm flax]

yeah for some items

[graceful storm]

not motors, rotors, or RIPs though

tragic

[calm flax]

Some of the numbers add up more like wire used to be like compressed cable in trains because stack size = 500 (250 cable)

[graceful storm]

interesting never thought of moving products based on stack size, just thought of reduction ratios, or convience

[calm flax]

oh you have a whole new level of optimization to think about lol

[graceful storm]

lol greeny and his linear programming is at least 3 levels above me

[wind spade]

ok, I got the exporting thing to work properly now. I can plot the graphs for any manifold. Testing now for 780 input, 30 per machine, 26 machines, 100 stack size

[graceful storm]

30 demand or 30 input?

[calm flax]

same thing?

[wind spade]

30 demand per machine, manifolded

[calm flax]

can you calculate one ofr 200, 500 stack sizes as well?

[wind spade]

sure, let me just check if it's working properly first

[calm flax]

cable, rod, plate all 200 now

[wind spade]

and setup a spreadsheet

@calm flax chart for the case I shown, blue = mk1, red = mk5. Vertical axis is production at a given time. Time steps are 0.1 seconds, so may not be 100% accurate but it's good enough imo

[calm flax]

something interesting to add ot that would be total items produced 😄

[wind spade]

here you go

[calm flax]

see told you mk1 wins 😛

[wind spade]

it's not a win

they produce the same amount

[calm flax]

blue line > red line lol

[wind spade]

and mk5 is filled earlier

[calm flax]

in the early part

[wind spade]

it's not a win if you produce the same amount of items 🤷‍♂️

[frosty owl]

If you only care about the early part, you may want to just halve the number of machines and keep them running xD

[wind spade]

if you care about early part, you don't want to do manifolds, but balancers

[frosty owl]

If power is the reason you turn them on/off

[calm flax]

can you recalulate with stack size 500 and 14 machines?

[wind spade]

input per machine?

[calm flax]

yeah like concrete input say 20/min 500 stack size and 14 machines

[wind spade]

so total input 280?

machine input 20

[calm flax]

feed it 780

assume the manifold is well in a manifold

[wind spade]

that would require me coding in overflow 🤔

[frosty owl]

That kind of goes out of the scope of the point, I think

[wind spade]

exactly

[calm flax]

but actually what it really needs is initially fed 780 for X items followed by 280 (sustained rate) to basically simulate the source machine buffers being emptied eg an initial 9300 at 780/min followed by 280/min

I guess thats a PITA for the maths :S

[wind spade]

that's super specific case. Most of the time your manifolds are built to match the output anyway

[calm flax]

well that is actually the common case eg 14 assemblers fed by 18.6 concrete constructors

[wind spade]

if we take buffers into consideration it's all over the place and basically depends on the specific build setup rather than manifold efficiency

[calm flax]

So the common case would be the concrete is stalled and all buffers full so 500 * 18.6 = 9300 initial burst of items dropping to 280/min sustained

even then there is another variable which is the stalled concrete on the bus

[wind spade]

I'm assuming you have the previous factory connected to sink, so it's running at 100% and not buffering up

which is the only relevant case which can be decided mathematically

[frosty owl]

Buffering is bad for your FPS

[upbeat tide]

Its unavoidable sometimes tho. Looks at you trains

[frosty owl]

You have them working in overflow or by letting them empty-out the sender? ^^

[wind spade]

stack size 500, 14 machines @ 20 input, 280 total input

[graceful storm]

that is a close line

[wind spade]

yeah, the difference is pretty much nothing

[frosty owl]

Can you calculate a graph for a balancer too?

[wind spade]

at max difference, you have 8 items more with mk1 than mk5

[graceful storm]

but setup time is slightly

*build time

[wind spade]

so if you really need to have 8 items more a bit earlier (but less items later), you can justify adding mk1 belts.

here's your graph for difference in amount of items produced

so yeah, in first 20 seconds, you have ~7 more items if you use mk1 belts instead of mk5. But after two minutes, you already lost that advantage

I'm gonna need to add this to my tools, it's so much work to generate these 😄

[calm flax]

yeah i guess there is

i wonder if there is a generic maths solution to it 😛

[wind spade]

I doubt so

[magic shadow]

tl;dr manifolds suck

ducks bammer

[wind spade]

the data simulation is easy, I just can't be bothered to automate creating of graphs and import of data to excel 😄

[wind spade]

no, manifolds are still great, there's just pretty much zero difference between placing mk1 and mk5 belts to the manifold sides

[magic shadow]

oh so my balifolds suck also?

unless i split them further from that

[wind spade]

what?

[magic shadow]

example of constructor taking 30/min
manifold of mk1 belts split in half to two constructors

manifold flow-limited with balancers
or balifolds for short

[calm flax]

building 26 feed balancers sucks more 😛

[wind spade]

since it's basically a balancer and works 100% from start, I'm not including these in my calculations

[magic shadow]

purist

[wind spade]

nah, it's just irrelevant to calculate something that you can answer without calculation

[magic shadow]

yes

[lethal shell]

Does anyone have a graph of the cost of points for the awesome sink? I'm having trouble wrapping my head around the formula

[wind spade]

it increases every 3 tickets

[lethal shell]

Does it increase exponentially?

[wind spade]

yeah

[lethal shell]

Yikes

[wind spade]

it's something along the lines of 500*(tickets/3)^2 + 1000

(tickets/3 rounded up)

[lethal shell]

aye

[graceful storm]

isn't that quadratic?

sort of

[lethal shell]

I forget the definition of quadratic

[graceful storm]

ax^2+bx+c

b=0 in this case

but it's cieling of a group of three so it's not a true quadratic, because it has "steps"

[lethal shell]

ah

[graceful storm]

exponential would mean tickets would become impossible to get very very quickly

[lethal shell]

do they eventually become impossible to get?

[wind spade]

no, it'll just take A LOT of time to get them

[lethal shell]

alrightyy

[graceful storm]

quadratic keeps it viable for a short time, late game it's a dumpster

[wind spade]

well considering that you can get over 100m points/min

if you have that amount of points, you can still get 1 ticket per minute at 1500 tickets

and considering there's only finite amount of stuff in shop to purchase, I think you're fine 🙂

[lethal shell]

Watch me spend all my tickets on Color Cartridges

[graceful storm]

why flowers are everywhere

[lethal shell]

I'm in the middle of a desert

and I was 85% joking

Thank you to whoever got rid of that

[graceful storm]

^

[frosty owl]

Finally someone who gets it
How does "manibalancer" sounds?

[graceful storm]

balifold sounds better

[oblique hollow]

semi-fold

[torpid robin]

Who needs a bashin !! Who thinks manifold sucks . They suck

[graceful storm]

no wonder my vacuum cleaner doesn't make straight lines

[young aurora]

If I was to transport items long distances what would be better conveyors, trucks, or drones? If I couldn't use trains

[oblique hollow]

belts

you dont even have access to drones yet

[young aurora]

Then why do people use trucks?

[oblique hollow]

usually no because the truck ai is bad

[torpid robin]

Cause they want to try them out and they are available . But yea belts is the better option

Belts won’t cause you problems. It’s set and forget

[young aurora]

Performance wise like with fps, is conveyors better?

[pulsar edge]

mostly for fun and aesthetic reasons

and they are fun to mess around with, and done right, they can also look super cool

[full raptor]

How is the poly resin rubber forced into recycled plastic machines before using rubber from recycled rubber?

Trying to figure a way to make these machines make more of rubber or plastic, whatever is needed

I guess just mk1 belt is a good idea

[oblique hollow]

so.... input priority

[oblique hollow]

if you use recycled rubber and recycled plastic, you dont need any polymer resin rubber

they feed each other

[full raptor]

if you dont use that rubber, dont the crude refineries just turn off

[oblique hollow]

produce something else

im just guessing, its not like i know your production setup

so i can only throw wild guesses out

if you were to describe it, then maybe i and others could make better suggestions

[iron prairie]

With recycled rubber/plastic: you'd just hook them up in sort of an X-shape, where rubber feeds back into the recycled plastic line, and plastic feeds back into the recycled rubber line.

[oblique hollow]

then they only need fuel from refineries to keep their act up

[iron prairie]

With 50/50 rubber/plastic, you get 1 rubber and 1 plastic from 2 fuel.

Each refinery you switch from recycled rubber to recycled plastic increases plastic production by 90/min, while reducing recycled rubber by 90/min.

[full raptor]

there two are what im doing, but i dont want just 900 plastic or 300 of each

[iron prairie]

You will have to have a mix. Even if you want 100% plastic output, you still need one recycled rubber refinery per two recycled plastic refineries just to keep the recycled plastic refineries fed.

For now, I'd just ignore the residual rubber: that's mostly a side gig.

So, at 1/3 rubber refineries, 1 fuel -> 1 plastic. At 50:50, that's 1 fuel to 0.5 rubber + 0.5 plastic. At 2/3 rubber refineries, 1 fuel -> 1 rubber.

By changing the fraction of rubber refineries, between the limits of 1/3 and 2/3, you can thus alter the ratio of rubber to plastic as net output.

The residual rubber lets you get away with slightly breaking that 1/3-2/3 limit, but with the HOR/DPF/recycled rubber+plastic alts, it's a mere 1/9 of total output, so that's not a huge effect.

[full raptor]

Thanks a lot, appreciate it

[iron prairie]

When I've done recycled rubber/plastic setups, what I'll usually do is have ISCs for the inputs and outputs, with a total of four smart splitters on output lines.

Splitter 1: Input: Recycled rubber. Priority line back to the rubber input ISC (feeding the recycled plastic refineries), with overflow sent to splitter 2.
Splitter 2: Input: Splitter 1. Priority line to the rubber output ISC, with overflow sent to an AWESOME sink.
Splitters 3-4: Same as 1-2, except replace rubber w/ plastic.

[upbeat tide]

My current rubber recycled plant is very steaight- forward

• 6 yoyal pipes of fuel, 400m3 each
• 300 injected rubber from residual rubber
• I have a dedicated section of recycled rubber plants feeding my recycled plastic machines with the remaining 500 rubber need

Each output belt has a smart splitter overflowing into a sink for never ending cycles.

I do not have ISC’s, but I do buffer the fuel and HOR

[frosty owl]

I think performance wise a truck might beat conveyors, if set up properly (enough distance to make it worth it)

[upbeat tide]

Trucks can he interesting. I used to have a truck highway between blue crater and the grasslands. Went south into the void and crossed over that way.

Sure was interesting. Sometimes I seore they would self eject into the ether

[upper harness]

So not sure if I'd want to put this into here since I am on experimental, but here we go!
SO I've recently been trying to min-max my factory.
Steel is a problem child right now, and I am stuck on which recipe to use.
I want to use Compacted Steel Ingot recipe but am worried I wont have enough sulfur later to do a max nuclear setup (with trashing the plutonium into the sink so I won't have to deal with radiation build up)
Looking at Max Resource mining rates, I feel as if I need to value sulfur a ton since its the 3rd rarest resource in the game, capping at 6840/m
Any tips? like a better steel ingot ratio? I am not worried about Iron usage, just the other resources used.

[magic shadow]

solid steel alt

[frosty owl]

Yep, I'm with the flully one.
Solid steel all the way. Use compacted coal only if you're SURE you can spare the sulphur (sounds like a waste to me, since I don't think you need that much steel, but not the point xD) @upper harness

Oh right... If everything else fail... there's still the Coke Steel you can rely on :D

[upper harness]

hahaha ... no

[upbeat tide]

Solid steel ingot
Solid steel ingot
Repeat after me!
Solid steel ingot

[frosty owl]

Don't underestimante its usefulness. Unless you use tons of it for something in particular, you can "waste" a little oil here and there, whether it's on steel, coated plates or whatev ^^
Let it "lubricate" your production, if I give the idea

We got more oil after U4 after all

[upbeat tide]

Gonna drop this here

[frosty owl]

Just 10 foundries for the coke xD

[upbeat tide]

Still, solid steel > coke steel > vanilla > compcted steel

Just because far better things to use your sulfur in

[upper harness]

I get it, the main reason I got stuck, I saw that Compacted Steal Ingot use a bit less coal.
And since black powder isn't used for anything but weaponry, I thought why not.
But then I remembered batteries and nuclear and brain got broke

[pulsar edge]

like mass producing weapons?

i think that's neccessary

[upbeat tide]

Nope. Turbofuel, nuclear, nuclear pasta, etc

I do automate munitions but not alot

[pulsar edge]

nah we came here to destory the ecosystem and i intend on doing so

[upbeat tide]

Munitions look eeh on belts tho

[iron prairie]

AFAICT, nuclear pasta doesn't require sulfur. It's just copper and pressure conversion cubes, with no sulfur anywhere that I can see.
That said, as of U4, the demands for sulfur have only gone up. Batteries for drones, nuclear is still sulfur-heavy while producing less power, meaning turbofuel is more important, etc.

[frosty owl]

The order of what needs it the most (from an end-game perspective) now seems like: nuclear, batteries, turbofuel, steel, weapons
Just gotta divide it in those, but once you get good nuclear you can rule out turbofuel and probably have way too many weapons too xD

[upbeat tide]

My base already sucks up 3780 sulfur 🙂

2400 -> turbofuel
1200 -> encased uranium cells
180 -> munitions

[graceful storm]

why 180 in munitions?

[magic shadow]

all of the boom.

[upbeat tide]

Left over sulfur.

I use 3 normals and a pure for my turbofuel plant.

The left overs plus tad left over compacted coal = enough black power to automate some munitions

[frosty owl]

What will you use for batteries?

[graceful storm]

improvised sponges

[upbeat tide]

Still got 3060 unused sulfur on the map. Plenty

And now my 2333.33m3 TF plant is more important

[frosty owl]

Right, I forgot how much was the total sulfur

[upbeat tide]

6840

Thats just not cool

Sulfur need for max uranium cells in U4

Why not straight 1200? Or something?

[iron prairie]

You're also forgetting max plutonium, which IIRC drives the sulfur consumption up to something like 3.5k.

[upbeat tide]

Well, I already have 1200 sulfur used in uranium soo its not a huge deal

2441.5 specificlly for max plutonium rods, 52.5 rods to be exact

But only 300 for normal recipes. Which maxes at 10 rods

[frosty owl]

Are you using the rich uranium recipe?

[upbeat tide]

There are no alts for uranium cells in U4...that work

But in U3 yes im using the encased uranium cell alt

[frosty owl]

I mean the one that uses uranium and waste, don't remember the name

[iron prairie]

Maybe he's referring to fertile uranium? Which I see as an alt of last resort, as it really drives up plutonium waste production.

[upbeat tide]

Yea its not a good alt for sure. Ans Snutt said today alts are still in a not finished state...thankfully

Just hope they have decided that most of the alu alts are good as it, besides instant scrap

[iron prairie]

I... like fertile uranium as a gameplay thing, I just wouldn't use it unless I absolutely needed those last scraps of power.

[frosty owl]

I don't quite get it, how do you get 52 plutonium rods out?
Isn't that like over 2k waste/min? :thinking_helmet:

[iron prairie]

Based on the math I've done, you can actually get 88.42 plutonium fuel rods/min with use of fertile uranium.

2100 uranium/min, 16/19 of which becomes uranium fuel rods.
100:75 uranium to the uranium waste w/ nuclear fuel units, so 1326 waste/min.
Then, you have 15:1 uranium waste to the plutonium rod w/ all alts, so 88.42 plutonium rods per minute max.

[upbeat tide]

what Greeny's tool tells me with all alts on

[iron prairie]

How much uranium waste are you feeding it?

Hm. I may have, in my initial math, messed up on the uranium to uranium fuel rods ratio. Now I'm getting 26.53 uranium fuel rods and 44.19 plutonium fuel rods.

[upbeat tide]

787.5 is absolute max atm

[iron prairie]

If I did my original math right, which I may need to question (as I originally overestimated total uranium waste), you'd need to multiply that by 16/19 to account for uranium going straight to fertile uranium rather than uranium fuel rods.

[timid yew]

if someone can send the prod line from greeny to plutonium rods? bc it don't work for me

[upbeat tide]

https://u4.satisfactorytools.com/production?share=QRcTNU147wezRA8Hr94N

[iron prairie]

You need to go to "Items, Input" and manually add uranium waste as an input.

[timid yew]

got rekt by by the prod line 😒

do u use power shards on lines ?

[iron prairie]

On miners and oil extractors, almost always.

[timid yew]

bc greeny prod line made me sick just at look what i need

[iron prairie]

On anything else, almost never.

[timid yew]

oh i see

bc when ulook how many building u need to be efficient look like that line will take so much time

[iron prairie]

To be fair, that is max nuclear power.

[timid yew]

yeah but power are always no enough xD

[glacial hemlock]

That's true for U4

[timid yew]

i just take like 10h to make maths and build for rework my turbofuel with the blender

[frosty owl]

Nah, nuclear can cover for pretty much any max production up to 1.2 OC I think ^^

[timid yew]

just for looking nice and no spaghetti (or at least not something ugly)

[frosty owl]

I'm pretty sure you go down to 17 if you don't use the fertile uranium alt

[torpid robin]

Isn’t that using too much uranium . I thought when I was looking at it you could only use 320 uranium or something for the plutonium

[frosty owl]

The amount used in that recipe is based on the max possible amount of waste you can produce if you used all uranium so that sounds about right

[iron prairie]

(3/19)*2100=331.6, so "320 uranium/min for fertile uranium" sounds about right.

[frosty owl]

I'm pretty sure you're using that alt in your calculation though 😅
No other way to get over 20 rods/min I think

[tiny sentinel]

opinions?

[frosty owl]

Which you lack now?

[iron prairie]

Usually, those are asked in #screenshots , with numeric emote responses serving as the voted answer.

[tiny sentinel]

i'm thinking I get the steel coated plate, then later get the bolted iron plate since they math input and output

[frosty owl]

Good thinking

[tiny sentinel]

silly me, thinking to ask alternate recipe advice in the #math-and-meta channel

[iron prairie]

But, I'd go for #1 (steel coated plate). Bolted iron plate is OK, but a bit of a sidegrade: saving a marginal amount of power at the cost of marginally more iron input. Steel coated plate does take a small sip of...

Yeah, no, enjoy the blocklist.

[tiny sentinel]

um... anyone gonna finish what they were saying? I was actually interested in the constructive answer.

[frosty owl]

If you back up the Pic with your reasoning, I think it's fair to come here. It means you're looking for discussion rather then just a vote ^^

[tiny sentinel]

indeed. I'm not new to the game, but I'm not 1000+ hours deep with all the alternate recipes memorized by heart. I honestly don't know which recipes might be good or bad and the math/meta behind those conclusions.

[frosty owl]

By "good thinking" I meant that I think your choice of coated plates is the best bet, to then go for the bolted plates

Their numbers match up really well

[tiny sentinel]

the community seems to have a good handle on it, so I'm looking for educated discussion from veterans on which recipes are good and why.

[tiny sentinel]

that was my thought. I know eventually it won't matter since I'll end up with all the recipes in the long run.

[frosty owl]

That's why I suggest to always get the one you need the most atm

[snow maple]

As a starting point, the Wiki has a list of all the alternates with some basic advice on how good they are.

[tiny sentinel]

I've got up to T6 unlocked and I've neglected to hunt down many hard drives, so I'm out on the hunt now getting caught up.

i have checked that out a bit.

[frosty owl]

Don't worry if you get your head bitten off every now and then. Lots of people here, you're bound to "bump" into someone sooner or later xD

[tiny sentinel]

I'm mostly looking for the oil related alts.

[tiny sentinel]

no harm no fowl, some folks might just be having a bad day. I'm good.

[iron prairie]

I, for example (going off Vencam's response), respond really poorly to sarcasm in response to me noting that a particular style of question usually goes in a certain channel.

You're still not getting off my blocklist.

[frosty owl]

You already know about the "portable MAM" trick, right ~

[tiny sentinel]

ok, I couldn't have known that. If I'm blocked why are you still engaging in discussion with me?

[tiny sentinel]

yes, plop it down, start research, deconstruct and go hunting for the next hard drive.

[frosty owl]

Ye, be efficient! 👌

I usually end up analizing half the drives by the time I get back ahahah

[tiny sentinel]

it does seem kind of cheaty, but the only other alternative would be for them to allow parallel research of multiple drives at the same time.

[frosty owl]

Eh, vanilla enough to not be cheat imo xD

[tiny sentinel]

the portable mam trick only works because all the mams are linked

i think

[frosty owl]

You can think of it that way. Though, I'm not sure you even need to have 1 MAM on the map in order to use the trick

[tiny sentinel]

I'm pretty sure the caterium computer is the obvious choice here, or am I'm completely wrong?

[frosty owl]

Yep yep. Unless you need quick canisters for some reason

Tbf, compacted steel is nice if you have the compacted coal lying around, but that's a bit of an edge case

[tiny sentinel]

yeah, I get the sense that some alt recipes are meant to solve different problems

rather than being a straight all around upgrade

[muted crypt]

which recipe is better overall base recipe computers or crystal computers? I don't have caterium computers rn otherwise I would probably use that... my guess is that I should use crystal computers because no screws?

[frosty owl]

It's either screws MK1 or screws MK3, really

You could say screws MK1 are more abundant since you can make them with iron

The MK3 though... Those churn out pretty fast too

[muted crypt]

are you talking to me

sorry I'm a bit slow tonight

[frosty owl]

Happens. I didn't bother add a reply since the message was above.
See, me learn :happy_hannah_2:

[muted crypt]

also as I was asking that I got caterium computers from a hard drive

so now I have all three options available to me

the question is... which do I use

[frosty owl]

Oh, I got confused too it seems xD
The oscillator computer is exoensive AF but easy to make. Great early on

[muted crypt]

I don't exactly have plans for mass producing anything else (at least in this factory) with quartz, and two pure nodes are nearby

[frosty owl]

Lemme check recipes online, don't remember the base one well

Resource-wise the quickwire one is probably the best. The standard can save you a BIT of quickwire, but is heavier on oil and circuit boards.
The third relies a lot on the oscillators (which are expensive) but is otherwise incredibly fast and easy to make

[muted crypt]

I don't mind the oscillators one here ngl, might do that

[frosty owl]

It is pretty damn convenient if you got them automated

[tiny sentinel]

these all look so good. I think I'll go with the flexible frames for the moment, unless anyone has convincing arguments for the other 2.

[torpid robin]

Silicone circuit board

[tall stag]

HMF aren't used in many part recipes, just buildings. Meaning you don't need that many of them and there are better alt recipes for HMF. Circuit boards are used in a lot.

[torpid robin]

They are in u4 they are used a lot

[frosty owl]

Quite

That is also my personal favorite recipe :D

[tall stag]

https://satisfactory.fandom.com/wiki/Heavy_Modular_Frame
Adaptive Control Unit and Fused Modular Frame are the only 2 parts.

[frosty owl]

And they're quite important ones. Also, power generators and trains take heap of those, so getting at least 10/min can be a good minimum

[tall stag]

Heavy Encased Frame is my fav HMF recipe. Oil is annoying (which is the other reason i like silicone circuit boards).

[frosty owl]

Too much steel in that for my liking :P

[tall stag]

Interesting, you prefer oil over steal?

[frosty owl]

Quite, in some amounts

[torpid robin]

I agree with the encased frame . But it does eat steel very quickly

But I think a combination of both Coke steel and solid steel is good

5.5k oil or something like that can make the same amount of steel as 25k coal

[fierce palm]

is the turbo blend fuel alternate recipe a better alternative vs using the refineries?

[frosty owl]

I explained it a bit here @tall stag

[torpid robin]

So say you try to max the end game items out . You run outta coal very fast . And it’s due to a couple hundred hmf

[glacial hemlock]

Everybody: is blend...
"yes''

[frosty owl]

In an end-game perspective, oil is pretty abundant so you can use it with some flexibility and generosity

[torpid robin]

Yea it really is . Along with quartz and cat funnily enough

[frosty owl]

To fill km the gaps pretty much
Need a tad more steel? Get some coke. Need more plates? Get some plastic there... And so on

[torpid robin]

It’s bloody coal . Copper and bauxite that stuff you up

[frosty owl]

Those actually run dry pretty fast as long as you try maximizing more important things first (allu and it's products mostly)

[torpid robin]

From the plan I have set up in the calc . Using certain alts . I managed to spread the load on them

I ran out of copper and only used like 6k cat and quartz

And 22k coal

That’s to make 60pm of every item . And 40 on of the elevator parts

[gray flower]

i am waiting for that turbofuel in blender recipe on my EA game

[frosty owl]

Take it to 60/min on each SE part and you'll run out of cat and quartz too or close to it xD

[gray flower]

How much does the blender make with the blender

[frosty owl]

Damn circuit boards and speedy connectors suck on both silica and quickwire

[torpid robin]

Ran outta copper

The copper powder is a big ouch

Cause I tried . And ran outta coal too

[frosty owl]

Nah, you can do it. Up to 61 something pasta, so 60 still leaves you some copper

[torpid robin]

Ye Ye know you can

[gray flower]

hai yeti

[frosty owl]

Don't add Lowe tier materials too, it's easier to calculate for a bigger number of high tier and just cut down on them later

[torpid robin]

But that’s one item lol. I like to look at an overall what I’d like

Not just maxing one item out

[gray flower]

right idc about math cause i am on EXP but when i am on EA i will

[frosty owl]

Sure, just saying: if you keep producing everything equally (60 pasta and so on), copper is not the biggest limiter. Quickwire and quartz get there soon enough

[torpid robin]

It’s how I plan my base layout . It helps me to have everything there as an overall to see what I can save on . Then when I come to the line I break it down then

[gray flower]

i will be doing that to yeti

having decatied area and make it bigger number is better than small

[torpid robin]

Tbh I think bauxite is . Cause you need like 3.7k or something for nuclear ? That’s a big chunk right there

[frosty owl]

Yeti, you're being rude, not even saluting back

[gray flower]

i will be testing on EXP first then EA on U4 day when it comes

[frosty owl]

Well duh, I thought finishing uranium and bauxite was implied

[torpid robin]

Need better belts

[gray flower]

i am Doing math later just checking out new stuff and testing

[worn galleon]

Does anyone know if the vertical build limit is still 2200m?

[pulsar edge]

I didn’t know there was a build limit

[torpid robin]

there is a height that you will take damage at. its just below 2000m

[cedar mica]

Given that some parts of the map is below 0m, 2200m of total build height, might be correct

[vast jungle]

has someone already done the math how much Power and Materials you loose when you throw the Plutonium Rods away?

[sullen cloud]

If the max of approx 32 rods/min is correct that will be 32/0,4*2500=200 GW of lost power

[gilded maple]

So, steel screws being 260 is fantastic, unless you’re using Mk.4 belts, now it’s just a nuisance

[torpid robin]

yea i think this is how it works

[vast jungle]

Resources necessary for removal of 100 Nuclear Waste:

• 1 Blender (75 MW): 25 Silica, 15 Nitric Acid, 30 Sulfuric Acid (+ 75 nuclear waste) => 50 Non-Fissile Uranium
• 0.5 Particle Accelerator (? GW): (+50 Non-Fissible Uranium, 25 Nuclear Waste) => 15 Plutonium Pellets
• 1.5 Assemblers (22.5 MW): 30 Concrete (+ 15 Plutonium Pellets) => 15 Encased Plutonium Cells
• 2 Manufacturer (110 MW): 8 Steel Beam, 3 Electromagnetic Control Rod, 5 Heat Sinks ( + 15 Encased Plutonium Cells) => 1 Plutonium Fuel Rod

Power: 75+22.5+110+? MW / 100 Waste

Does this sound right?

(so this is the overhead for 20 nuclear powerplants)

[oblique hollow]

250 - 750 MW for 100% clock speed on the particle accelerator (for the pellet recipe only) .... now to underclock......

[oblique hollow]

should be 82 MW - 247 MW for the PA

sooo. at an average of 164,5 MW for the PA.....
75+22.5+110 + 164.5 /100 = 3,72

so 3,72 MW per Waste, i guess

[vast jungle]

even without underclock you loose nearly nothing of the power of a nuclear powerplant

the resources might be a bit more interesting, but its not that much too... so waste removal is nearly free

[oblique hollow]

you forgot to add the refineries for the sulfuric acid

and the nitric acid

and all the other products xd

[vast jungle]

yes... I just concentrated on the primary chain... because there are too many options for the rest

still, the amount of power is not really that high... I would have thought it would be higher

[oblique hollow]

underclocking of the PA is overly strong because of the non-linear nature of overclocking and the high output rate for plutonium pellets

[vast jungle]

underclocking for something that is just a few MW per nuclear waste ist not really worth the effort... a powerplant only creates 5 waste per minute
so there is not much power to save...

2500 MW production compared to 20-50 MW for waste removal?

[oblique hollow]

i mean, on full power its 250 to 750 MW

[vast jungle]

yes... for 200 nuclear waste

[oblique hollow]

and at 50% you straight up have 1/3rd the usage

[vast jungle]

lets talk about "power savings per nuclear powerplant"

in this category I think the Particle Accelerator doesn't matter, because it's not the largest power cost contributor

(you need only 1 accelerator for 40 nuclear powerplants)

[oblique hollow]

the endless amount of refineries / manufacturers / blenders is the thing here

[vast jungle]

even then... when I go to Greenys tools they give me an average use of 475 MW for 1 plutonium fuel rod... plus 1/2 accelerator means ~ 640 MW per fuel rod

thats including production of all other items

so its 6.4x5 = 32 MW on average per nuclear power plant...
1.5% of the power output

thats nothing

I think CSS did drop the ball on the numbers... they looked at the Accelerator and thought "thats a lot of power" but didn't consider that one PA is enough for 200 Nuclear Waste per minute

[river night]

they did say just yesterday that they are monitoring and tweaking the plutionium/uranium situation

U4 is far from "finished"

[vast jungle]

Its easy to see why this happened... I didn't believe it until I did some raw calculations myself and was wondering "thats all? did I make a mistake?"

the interesting question is "what percentage of the nuclear plant output is a good tradeoff for eliminating all waste?"

at what point would you consider "build a storage facility" or "use the plutonium rods for power and store the new waste" ?

[river night]

cant forget that its not only output from the plant that you waste, you also dont use the rods to make power, so have to include that in the math

[wind spade]

rods are way way way way way more radioactive than the waste

[river night]

but who cares about that if all they do is exist briefly on a conveyor and then go into a sink

if you use them to make power, they would exist in the same fashion and just go into a power plant

[vast jungle]

and each fuel rod gives 130k Points in the awesome sink

[swift robin]

i used to assume using coke for steel making was wasteful but dang does it beat getting more coal over 1km away

[vast jungle]

you have 3 options now.

1. Nuclear Power with Nuclear Waste (the old way)
2. Nuclear Power with Plutonium Power and Plutonium Way (the "new" way)
3. Nuclear Power without Waste (WTF?)

and option 3) is clearly the best...

option 2) gives you what? 20% more power than option 1) ?

[river night]

the best according to what factors? just not having waste at the cost of energy? Efficiency would make my "best" :p

but as said above, they are going to balance the energy difference between those 3 to make it a more meaningful decision

Snutt flat out confirmed that yesterday

[vast jungle]

yeah...

most important thing, they should increase the power output of 3)

[river night]

you mean 2?

[vast jungle]

yes (sorry)

what they should do is decrease the amount of nuclear waste you need to make Plutonium Fuel Rods... this both makes option 2) better and option 3) worse

[river night]

so you would need more PAs to get rid of all the waste right

[vast jungle]

and you will be able to fuel more Plutonium Powerplants with the same group of Nuclear Powerplants

[wind spade]

for now

[vast jungle]

the points are just icing on the cake... the "no waste" and "not enough more power" is the real problem.
I don't care for the points at all

they could reduce it to 0 points and the "no waste" option would still be awesome

[river night]

and he was wrong

he said yesterday its intended

[vast jungle]

So Snutt has more common sense than the one who overruled him 😉

[oblique hollow]

there was something interesting i heard on Kibz' stream: what if sinking rods costed you points?

[wind spade]

what if you couldn't sink them

and the waste would actually have a purpose to be ingame

[oblique hollow]

Plutonium Refinement 2: Electric Boogaloo

[vast jungle]

you mean like "only way to make nuclear pasta is from (among other things) nuclear waste? or maybe even plutonium waste?"

so you must balance the production of nuclear power against some other production (and side-costs)?

[wind spade]

no I mean the "it's something you have to store"

[vast jungle]

ok