#math-and-meta
1 messages ยท Page 270 of 1
you could also double the frameworks created which make them nicer, but that might be a bit of work to build
what do you mean 'cleaning'?
Just make the numbers more even
But ye, how would i split this, what im most confused about atm
either
- use a manifold (ie just chain splitters), which will eventually self-balance as long as input = output
- make two sets of Steel Pipe constructors that each make the relevant amounts
- in this case, smart splitter set to Overflow the excess
when one part's going into storage/sink you need to be a tad more careful manifolding since that part won't self-balance (ie if you send more than 26/min it'll happily accept it and not back up, and that backing up is required for manifold self-balancing)
True, heres more, this is probably what i find the most confusing in the game atm
same deal
these are universal
if you've got two miners, you could have one pull the 112.5 and one the 56.25
or for any setup, as long as you have mk3 belts, you could put it all on one belt and manifold that to the smelters and foundries
True, Which method do you recommend for the iron pipes?
I don't see any iron pipe
since you've got output to storage/sink there, I'd do a smart splitter with Any (or Steel Pipe) going to wherever the 54 is going, then Overflow to storage/sink
Oki
https://www.satisfactorytools.com/1.0/production?share=mlZrdsws3CuPN3hWRh2V heres the entire thing, any changes i should make?
looks reasonable
looks like the sorta thing I'd end up with, I also like to try and use round-ish numbers of nodes where I can
Feels l ike this was meant to be.
you don't need to merge it, so why split it
if decimals give you a stroke this isn't the game for you.
they are entirely arbitrary to the unit of time you use
How many fuel gens should I have with the 400 fuel?
19, and buffers?
no buffers
!wikisearch FG
The Fuel-Powered Generator is a power generator building that generates power by burning liquid fuel (Fuel, Liquid Biofuel, Turbofuel) or gaseous fuel (Rocket Fuel, or Ionized Fuel).
At 100% clock speed, one Fuel-Powered Generator produces 250 MW. The type of fuel does not affect power production, only...
how much does one gen use? 400 divided by that
the wiki is your friend
don't need buffers for anything (apart from train station buffering)
Okay
I donโt really mean decimals, i mean like splitting 57 in 1 direction and 139 in the other, instead of like, 60 and 120 etc
I love it when the numbers are round.
Only issue is I only have mark4 belts. But it's nothing splitting can't fix.
I mean you can do this with mk1s
Oh right lol.
practically anything can be done with mk1s ๐
Only problem now is making the factory.
It's still better than when I tried to make a big factory, idn't have enough resources nearby, and idn't feel like getting resources in or underloking everything
that's why you pick a place for the factory that has the resources nearby
I have 3 pure coal nodes in spaghetti belts. And 3 pure iron nodes are practically at the location where I'm gonna make the steel factory.
That is, if I progress sulfur to the point I can blow up that stupid rock.
manifolds split things evenly over time, so you don't need to manually split them if you have a fast enough belt
you could also make 57 in one group and 139 in another group
that is a lot of steel beams and pipes lol
if they're for personal use
Personal use for steel beam is 90, steel pipe is 60. Rest is used in heavy modular frame factory. And might use some of the materials later for other factories.
I'd recommend not building things in advance
So only build how much I'd need?
yea
you can't know how much you'll need in the future
unless you've planned your entire playthrough from scratch ๐
How much would I need to keep for personal use?
a few per minute is usually enough, depends on how fast and how much you build
Is it a good idea to have seperate factories? And that I just truck the steel beam and pipes to the fatory that will make the heavy modular frames?
if that's how you want to design it, sure ๐
or separate factories that don't have trucks between them
would this amount be good for personal use?
more than enough. You'd be fine with like 5pm
alright
activate your windows
nuh uh
@vapid gorge there's a pump there if that's what ur asking
yeah there its , sometimes it needs to be right before hand.
so I think something weirder is happening with your system that I can't see.
I'm not going to be home for hte next 40 some hours though as I'm heading out for the weekend. My recommendation is accept your fuel station not running at 100% for now and I can take a look at your save when I get home if you like
there's honestly something probably small and simple causing the hiccup
I'm heading to sleep though, it's 1am ๐
can we keep this conversation is dms
alr thx for the help regardless, have a good rest haha
Mysterious flow in manifold
Time for my 1st alu production :o
you can activate windows via the command prompt btw, dont gotta spend $100
im a proud non-activator
or just use free OS if you don't want to pay for it
how many meters is this roughly?
i wanna plan a factory that knows how long different conveyor belts and miners need to reach the factory, but getting the first converyor belt right is a pain
using the grid system and platforms to keep track of knowing how much distance i have
Map coordinates are in meters, so you get a direct readout of length in one axis (X or Y)
For diagonals, you can calculate with Pythagoras' Theorem (a^2+b^2=c^2)
foundations are 8 meters with snap points each 1.0 meter so you can also use those.. zoop 10 of one color, that's 80 meters, zoop 10 more of a different color etc.
For short distances, put a foundation and moving it one "notch" is +1m.
this is not trigonometry
thank you a tonnn
yee dats some spicy info XD lets seee if i'll get close enough to realtime with my calculations
nps
Joints might be a bit weird, and mixing different angles also weird ๐
prolly also gotta account for the time it takes for the miner to start mining and boot up to synch things
yeah it takes 10 seconds
yuss, still better than not caring at all tho, cuzz in extreme cases it can desync a factory
you might have to make some squiggly lines to get the lengths perfect like this ๐
they put them when a line B is shorter than line A, line A can't be made shorter, yet they have to be the same length.
yusss, i got dat alot lmao
with playing desert
oh prolly could also just measure in time to get an exact value
the squiggles are for stuff that is starting and stopping unpredictably
just so that they can send the same signal down line A1 and A2, and it will arrive simultaneously even though A2 is further away
huhh o.0
out of curiosity, what's the reason behind needing to know the length?
hmm im kinda trying to make seamless factories and sometimes have multiple sources merging into the same conveyor belts and its all messy and desynched sometimes
shouldn't that be covered by the "buffer" (which is space on the belt + internal machine buffers)?
like having a line of limestone constructors, and needing an input of 180 limestone, taking different sources that require different amount of time to get to the factory to form those 180 limestone per minute
like if you provide 180 in total, it shouldn't matter how far things are, it will eventually balance out
yaaa but it makes me kinda crazy xD i started atleast putting storage containers to kill overflow from conveyorr belts, and just try to make em even lengthed roughly
but measuring time is alot less pain than length for dis
dunno wat either is tho
ohhh das how you mean, mb bad language rn xD
for your example, if you have e.g. a 60 source close and 120 source away, it doesn't matter that the 60 goes there faster, the machines will for a while run just out of the 60, until the extra 120 arrives (all is per minute)
truee if you let it run for a bit
i'd say that plus storage containers could jus be enough, (to have enough resources already at the factory when power goes on again)
even without containers it will work ๐
but you do you, it's your save, just saying that it may be unnecessary ๐
its not reallyyy important but it looks better xD
hmmm just thinking about the pain about playing like months or years ago xD
i might just need to split/merge a little smarter
planning on doing multiple buildings that are specialized for one or two parts and then reroute to other buildings, and stack layers ontop of each other and use a ton of signs and lamps and what not and decorations XD
doesn take that long tooo tbh- i don wanna go tooo precise, jus roughlyyish
checkin that the material prices are the same on every lane to begin with and das mostly enough for my uses tbh
i want to conect this drone ports like this, how to do it
In A is my Fuel base where i craft the fuel for other drone ports to use and i want to deliver the fuel to other drone ports to use it
so far am conect each drone port with a single new drone fuel base A1 or A2.....
something like this
but if i want to use a new drone port and that drone port need fuel i need to put another drone port in the fuel base A3 and conect it to the new drone port somewhere in the map
Drones only go a to b
Generally itโs easier to have multiple drones stopping at one location and fuel just that one stop.
Easier because fewer points you need to fuel
is that posible?
something like this?
but in drone port A you can only put B por as destination but no C at the same time
This is fine yes
Thatโs because a drone can only have 1 destination
You could have A to B to C but that would require 3 drones
A to B
B to C
C to A
something like this?
Sure but you canโt do it with one drone.
And you donโt want to have the fueling drone carry non fuel
@pulsar lagoon There's some nice reddit posts describing the standard "many-to-one" style drone setup often used for fuel distribution. The setup is also useful for any item that has multiple downstream consumers though.
can you share the link?
@pulsar lagoon Google "satisfactory the ultimate drone guide". Then ctrl+f "many-to-one".
But generally you need to go in game and play around with drones transporting items a bit to really get a feel for them. They're pretty straightforward overall.
ty
I did some hard drive gathering. Still have 7 left to scan in maam.
Which ones should I keep. And should I rescan the rest, or leave as is?
only you can decide that
there's no bad choice with alt recipes, pick any that you like
you can get them all anyway
Actually I think it's fair to say Charcoal and Biocoal are objectively useless
But otherwise yeah it's up to them
It's not fair, I've seen people legitimely use those
what are things you want or need to make?
Let's say you have 1 piece of Alien Remains (any kind), you could somersloop that to 2 Alien Proteins
-> You then sloop those 2 Alien Proteins in to 400 Biomass
-> You then sloop that 400 Biomass in to 960 Coal (and any additional biomass you would like to feed in to 'stockpile' the generators a bit)
960 Coal is capable of driving a 4800MW plant, for the cost of 1 alien remain a minute
[Without slooping, 1 Alien Remain > 1 Protein > 100 Biomass (potentially 375MW of BioBurners w. Solid Biofuel) > 120 Coal > 600MW of Coal Generators ]
Now this is an extreme example, because it would chew up some sloops and be highly dependent on your belt speed in the alien protein > Biomass stage.. But that is pretty nuts for something you could easily get more than 1/min of haha
(I made the example as I have seen that exact one done, they didn't want to hike to the nearest coal node so they just completely made their own plant lol)
Now Alt: Charcoal however.. 1 Wood makes 5 biomass = 6 coal when using Biocoal. Alt: Charcoal turns 1 Wood into 10 Coal directly . So it is a big improvement if this is a setup that people would wish to use over their current biomass burners for any reason.
I won't say I'd do it personally, but it's an option for others haha
this is why this game is so good, there are so many ways to do the same thing
...you can type in the amount you want per minute?
you can even type in math
im flabbergasted
they really thought of every QoL feature
im redoing my oil once i get blenders, no point in it now
Yes, but there's still only limited precision, f.e. in the example above, you aren't really producing 80
oh
if you type the amount per minute you want, the game calculates the clock speed you need, and sets the clock speed based on that
so for your example, you need 133.3333333...% clock speed, but it's limited to 4 decimals, so you can either have 133.3333% or 133.3334%, which would result to either 79.99998/min or 80.00004/min
obviously you may say this is close enough and live with it, but some people don't like this, even though it's very minuscule difference
im irrational so, ironically, i dont like it
if i wanted to do a mandifold for that, how should i build it?
Seems like a lot of effort to make sense of a recipe that requires manual feeding on an automation game. This is niche in excess and still not useful enough because it is not unattended.
I'm not arguing that you can use the coal you generate but it's going to be limited and end without warning
Do note however, that alien remains and other biological parts can be made into dna capsules and sunk for way more points than the value of coal you get. It may be even worth slooping those for more dna points.
it's decent for e.g. ammo factory
hmm i could use 5 assemblers and send the overflowing encased beams to a sink
i think that works?
why not 5.625?
?
oh assemblers
well if you make 5 assemblers, you won't have overflowing beams
because you still have materials only for 4.688
oh right
oh heavy frames, how i wish you didnt hate me
WAIT
it only shows a rounding error there
its possible
it doesn't show rounding error. It shows exact numbers
sorry its just because i didnt put my mouse on it, it shows one at 68.75%
alright time to get the heavy frame alt
what is limit for grid id in the game?
grid id?
oh i already had it lol
this better be a good place for my heavy frames
Build them below the map and distribute them using drones so it's like they're ascending from hell
they belong there im about to redesign my whole factory cause of them
oh load balancing my blessed
Holly crap
Brb i'm gonna make diamonds out of spitters remains
The known "problems" with fluids have nothing to do with floating point problems, fwiw
And the majority of the problems in the past with fluids/belts also haven't had anything to do with floating point precision (though I do think there's been one or two)
Factorio switched to fixed point representation for fluids in 2.0, which fixed some of its own fluid rounding errors with floating point
From someone whoโs had to do fixed point math professionally, know that converting to fixed point isnโt simple. You have to do tradeoff analysis. Satisfactory devs are rightly leery of switching due to the risk of a brand new and different set of bugs appearing
Converting a fundamental thing like float to fixed point isn't simple inherently
You're directly contradicting a developer of the game here
It was a mistake to use floating point rather than e.g. int
we know that now, if there was a time machine it wouldn't be done, but it's a bit of a different matter to make a significant change like that in a patch with a game that has persistent worlds.
just found it funny to see basically my own argument but a year before, back in alpha. It would've been easier to change it then.
really proud of this early game oil setup, making little bit of everything for the future from all of the oil in this area - 2 normal nodes + 1 pure node > 400 fuel (just straight into gens for now), 400 plastic, 400 rubber, 360 resin/240 canisters (depending what i need, smart splitters <3) 1260 coke, also used some water and made 240 liquid biofuel setup that can feed 100 biomass gens that you can see right there
well its ugly but hey everything running at 100% ๐ค
The fluid system in particular was confirmed as Working As It's Supposed To
you may be confusing old floating point issue bug with pipes that was resolved already with current "issues" of pipes (that are actually features)
Nah, you just can't resolve floating point imprecision issues (and thus phantom creation or deletion of fluids) unless you simply don't use floating point, which is still in use AFAIK. It's fundamentally imprecise, and accumulates error.
Phantom creation + deletion of fluids is, AFAIK, not a known problem with the game
It is, just on a very small scale. It's impossible to do floating point math without it.
There was a problem with that back in the day, I believe, but that wasn't even floating-point related, I don't think
you just can't resolve floating point imprecision issues
sure, but you can make them very small and have systems around to even further make it accurate
Well, I suppose if you've seen a developer say recently that there are problems with floating point precision in the game, then I couldn't argue with that. Everything I've heard in the past year from any vaguely official source is that floating point is not the source of any known problem in the game, though
I think it was shadow fixed u6 or u7
most of the times where dev said floating point issues are in pipes were pre-1.0
you can, but it's additional complexity and doesn't completely resolve the error - while a system which doesn't use floating point is less complex and does not have the error at all. It's simply better if you have to pick A vs B architecture.
The reason that we're not using it right now is, AFAIK, because the devs who wrote it didn't know better at the time - and because any change to the simulation would require significant investment and could bring new and unintended bugs.
It's more just that people know that floating point values are imprecise, and stuff like fluids in this game exhibit behavior which seems erratic and imprecise, and assume that the two must be related (plus that there may have been some floating-point problems in the past). Everything I've heard from the CSS side is that no current problems are related to it, though. (Nor is the current pipe/fluid quirks, which are technically not problems by various definitions. :)
It's very intuitive to be like "numbers between 0 and 1, that's what floating point is for!" but it's also wrong, and harmful to accuracy.
Anyway, as I say, if you do have a source or whatever where a dev is saying that floating point is currently causing problems in the game, then I'll have no choice but to stand corrected. :D
The video from 1yr ago that the quote is from
i closed it a few hours ago though, it was Snutt talking about floating point imprecision injecting error into pipes.
Yes, but floating point has not fundamentally changed in the last year or even the last 10.
no but pipes did
Yeah, I remember Snutt talking about that too, but my impression was that those problems had been fixed in 1.0, yeah
It's literally impossible to fix them within the floating point format, they may have been reduced in magnitude.
There's always ways to get around floating point imprecision
and they specifically mentioned fixing the "floating point bug" with pipes
Anyway, feel free to believe what you believe, of course, but that pre-1.0 comment doesn't convince me. :)
Why do you think that the devs of a game have somehow fixed fundamental issues of one of the most common number formats?
I'm not saying there's no float imprecision in the game, but based on what devs said, current pipe system doesn't suffer visibly from any floats and most likely is not possible to have it become a problem (and it definitely has nothing to do with "pipe manifold issues")
(and it definitely has nothing to do with "pipe manifold issues")
nobody asserted that it did
for phantom deletion and creation of fluids, it's just a question of the magnitude.
Sure: make sure your simulation does not rely on being accurate beyond a specified decimal point
Round your values, like any other system that uses floats
I'm not saying they fixed float imprecision. I'm saying they fixed the bug that was affecting pipes stability (and was called "pipe floating point bug")
I'm not saying that's what they did but there's often ways to deal with it
Well, there may have been a bug involving floating point that was seperate. What i'm talking about is a fundamental inability of floating point to describe an exact and consistent number.
Yeah, I'm not suggesting they somehow changed the fundamental nature of floats
But a system like this could certainly be adjusted to not care that floats are imprecise
afaik game doesn't track units under liters anyway, so rounding would be a good way to fix it indeed
And since we don't exactly have access to their codebase, all we can go on is their statements that it was fixed
Not fully AFAIK
And since we don't exactly have access to their codebase, all we can go on is their statements that it was fixed
you are talking about a different issue
afaik game doesn't track units under liters anyway, so rounding would be a good way to fix it indeed
I know for a fact that it does, tenths of a liter are stored in save files and visible in the UI.
Anyway, I think we're at an empasse, and the doggo needs a'walkin'.
Agree to disagree, I guess. :)
if it's 1-5 decimal spaces (which would be around 1/100th of mililiters), decimal imprecision won't show there if you round
I am not gonna agree to disagree on the fundamentals of floating point, they are extremely well established and documented facts that predate our being born
Actually the idea of floating point is apparently some 4000 years old at least
so yeah, if you base your whole system on the fact that fluid amounts go to max of X decimal spaces (X being a reasonably small number), then you never suffer from any floating point issues
You still do because of accumulation of errors which are smaller than that, until they are no longer smaller
decimal imprecision happens (well at least in languages I deal with, but it should be similar in C++) around 8-9th decimal point. You'd have to have VERY VERY VERY bad code for decimal error to propagate three orders of magnitude higher
if you round each tick, the error doesn't stack
It does
because it causes you to incorrectly round e.g. up when you should have rounded down, and vice versa
you don't round incorrectly, if error happens around 8-9 decimal point, rounding to 5 decimals is safe
float errors affect the rounding of the decimal above them, which affects the rounding of the decimal above that and so on - through the whole number.
there is no limit to the error propagation
believe me, it would be enormous math and computer science news and like nobel prize level stuff if somebody fixed this
but it doesn't really make sense for there to even be a way to fix it - we have a finite range of precision given this amount of information, 32 bits. It can describe about 4 billion numbers with perfect accuracy, or more than 4 billion with significantly compromised accuracy.
there is a simple and easy way to avoid the problem, which is to just use integers if you need fewer than 4 billion discrete numbers. We want 1000 liter pipes? A 32-bit integer unit can be 1/4000000th of a liter.
Float throws away accuracy to make it possible to describe 0.00000002 or 200 trillion within the same 32 bits. That is not a requirement or even a benefit, so why take the downside?
Something can even be a pretty tiny error but accumulate over a million simulation tick iterations into a noticable effect.
I find it very weird that people argue:
A: that floating point error doesn't exist or can't propagate/accumulate all of the way - it's well proven by some of the greatest minds on the planet
B: that there is a significant benefit to use floating point in this case and therefore equivelant systems should be intentionally built using floating point, being aware of this problem
The only good argument that i've seen is that it's here now and change is hard, which is perfectly valid and agreeable. But i don't understand people who wanna argue about it supposedly not being an issue. It is an issue, a minor one, but that's how the math works.
I found it interesting that it was highlighted in alpha and not chosen to be changed back then.
it can propagate if you don't round ๐คท
and I heavily doubt you can get a single calculation of two rounded values that would differ in third decimal digit from actual result
Well, you can. And it does.
haven't seen such a case yet (and I've done A LOT of working with decimals). Best/worst case I could do was like 7th digit
I'm not defending their decision to use floats, but I'm defending their decision of not rewriting the entire fluid system to use integers instead, esepcially at this point in the development
I'd also like to point out that there's a difference between float arithmetic being fundamentally imprecise and float arithmetic's imprecision actually showing up as a noticeable impact ingame
The first is very true, the second less so
It's certainly true that using something else like fixed point representation or ints would have avoided all the issues with float arithmetic in the first place, but the imprecision of float arithmetic generally doesn't show up as a very noticeable effect a lot of the time in gameplay
I will just put this here
That should've been the core of the convo, imo.
Saying that everything works "fine" doesn't imply that there's absolutely 0% errors in the game's calculations, just that they're tiny enough that "enough" people don't notice.
In other words: pipes are imprecise, but not to a degree noticeable by 99.something% of players
is there a way to automatically import factory plans from satisfactory tools into satisfactory modeller?
no, because modeler sucks
and if you're already using a much better planner why modeler? break up your tools plan into smaller chunks , other tabs, if that's what you want
my reasons for choosing one over the other are irrelevant for this question But i see there really is no way hm.
I want to use modeller for instances where my factories are slightly different from the plan and so i have to adjust calculations with subtle changes without losing the entire configuration
Your solution of using multiple smaller tabs is fitting for those steps
well may modeler's lack of functionality keep screwing people over
this pure node is 300/min at 250%, yet the other two pure nodes at 600/min at 250%, is that a bug or true
teh smaller ones are prob well presurizers
oh they are
ill still build my proper fuel plant there though
seems like a good spot
presurizersa nd dnoes aren't the same....
yeah
you don't actually need loops
lot simpler and uses like 200mw less power
if you're using electrode circuit boards you're not looking to save on oil anyway
oh im not saving oil
and you can use recycle recipes w/o loops. Just don't loop it
send the plastic/rubber to another set of refineries that uses fuel.
recycled recipes worth it?
more output per oil at the cost of more complexity, buildings and power
Recycled recipes give you the maximal yield of plastic/rubber per oil (when combined with HOR and diluted alts), at the expense of complexity and power
ew i need quartz, yeah im scouting for more hard drives
The recycled recipe puzzle is cool. I enjoyed playing with that for the first time.
id make more than that/min just for depot purposes tbh, theyre the easiest of all of the conveyor materials to produce at mass relative to the stage
and youll want to make casings for the same thing too - blenders use casings to build
yeah i only did it so i could get the blender, i dont really know where to go from here lol
and yeah definitely try and get more hard drives - sloppy alumina is another great recipe as it removes the silica byproduct (and im pretty sure is more bauxite efficient)
wait so turbofuel doesnt make more power, it just lasts longer?
fuel gens always burn the same amount, different fuels burn for longer
kk
combine this with the electrode scrap and pure alu ingot recipe and you get 1:1 bauxite:alu ingot ratio
Fuel gens have a constant power production but variable fuel volume/time. So better fuels generate more power.
i have sloppu, pure, but not elec
only issue (maybe) is it uses crude oil. just build near it and your good to go
you could even use water byproduct if there is any to make resin products e.g rubber/plastic/fabric
In effect you can build more gens for the same amount of fuel. A fuel generator always produces 250 MW at 100% clock regardless of fuel type
Same is true for other generators as well
ok so im going to: finish temp comp setup > make baux setup > make new fuel plant (place in pic below, hopefully good) > do whatever
temp comp?
temporary computer setup
why temporary? theyre used for a few machines so youll need them for depot anyways
because im building it at my current oil area, which i initially rushed and ended up being a nightmare
blender, signals for trains, radar tower and programmable splitter
plus im building my new power plant here, so ill tear down what ive made
good area
i have my adaptive control units set up there as well as other oil stuff
and i have modular engines there too
you can make a nice train station and make lots of parts there
yeah
oh man
@opaque quartz @twin shore my build going so far, so gens gonna be like this +3 floors up
geez
Overclocking would cut down on how many gens/floors you need to do
A fully OCโd fuel generator takes 50 fuel/min and generates 625 MW
well, if i overclock 200% ill have to build one floor up
i have exactly 80 shards on me tho
one connected gen you see there just burns 20 fuel from 30 oil, just for the build to take 300 instead of 270 xd
Honestly blue crater has enough space to build on one level
You can go off the edge since your platform is floating too
thats seems plenty for personal use
yeah, buts it only going off of 1 pure bauxite node
then i get a lot of stuff to sink as well
sure, looks good
that's overkill xD
in the future ill probably make most of it be used in production lol
when connecting fuel gens vertical floors how should i do it, pic1 or pic2 - is there a difference? (view from side)
im only asking bc its gonna be full 600 pipe, so i expect some issues lol
not #2, and dont use 600 pipe
okay thanks so 300 for two floors and 300 for another two, like pic1
why? is mk1 immune to having issues when maxed out?
@twin shore same
made the same number but seemingly a lot more was used
This is mine, alts made it much more resource efficient
i think i forgot to enable water, so it wouldnt work with wet concrete lol
Only issue I can see is that there's no good way to split up the iron ingots, so it just has to be one super long manifold
It's probably gonna take 10+ min to saturate
just clock the ingot producers into groups
More specifically i mean don't try to use 100% of a pipes capacity for anything but the most simple layout
@civic bronze "okay thanks so 300 for two floors and 300 for another two, like pic1"
split into 150 per floor, so a pipe doesn't have to go up or down floors after
oh ok
Too much effort imo, easier to build the machines in groups and then just manifold them all together rather than doing all the math to match refineries up with groups of machines
alright i think i can do that ๐
I agree, using 100% of a pipe doesn't seem to work for me most of the time
There's always some sloshing
i managed to do it with pure oil node on my oil ref
I had exactly 600 turbofuel which you would expect to work, but even making a looped manifold it still didn't reach 600/min flowrate
So I just split it into two pipes of 300 each, never had a problem after that
Fluids are weird
exactly that
100% - split into 2 of 300, with one mk2 pipe it was at 95-97%
idk if valve even does anything there lol
heard to never use them but hey it works
you can't prevent backflow through a junction if you use a basic manifold design
because of the way manifolds work with preferentially filling pipes earlier in the manifold, and because of there being no such thing as a valved junction or pipe (the valve is a seperate part, and you have to connect pipe junction pipe valve pipe - you get backflow from the pipe junction pipe bit)
you can entirely prevent backflow if you use a gravity-graded load balancer which intentionally has some pipes underfilled at all times - a feature that is not possible on a basic manifold. That feature prevents backflow because no liquid is high enough to flow enter the junction on the downhill side, but it is on the uphill side. They require more knowledge, time and space to implement.
On pipe to pipe connections with no junction, a simple valve > pipe > valve > pipe > valve configuration will entirely prevent backflow, so that's an easy build to move something from A to B.
Given the pretty long list of constraints that have to be followed for 0 backflow with junctions, it's a lot easier to just make a pipe which is tolerant of a certain level of backflow
by not trying to force 100% of its flow through it 100% of the time in a specific direction
damn this is very well explained
Best way to deal with fluids is to not try to perfect them lol
Just give them the ability to do what they want
they want to be turned into cool things for pioneer at 100% efficiency (they told me) ๐
but like, on my pic each of mk1 pipes uses 300/min so why does it work
It can be done, they're just very easy to break.
The fluid simulation also depends on some information that isn't conveyed in a simple screenshot or mspaint drawing - the exact way that you connect the pipes and make a pipe out of smaller pipe sections affects the simulation, so two different builds which look identical can behave differently.
larger pipes are more likely to have issues due to larger amounts of fluids moved per tick
Pls donโt pay attention to what I have to assume is their valve nonsense
Hard to tell what youโre actually trying to do though
Are miners and resource collectors the only things you should be overclocking?
You can always make more crafters, but the resources available are very limited
depends on you
I was just not sure how to split 600 turbofuel in my 4 floors x 20 fuel gens
and resources are indeed "limited", but you won't really run out of them realistically
create 4 groups of fluid producers. send 1 pipe to each floor
while it's possible to branch manifolds it's often a pain to do so
10x the pain over multiple floors
What you are saying is if i have 32 turbofuel rafis, connect 8 with each floor?
precisely. Or however you want to clock/group things
with fluids you want, as much as possible. to go from Point A to Point B. No splits merges or elevation changes within the manifold
it's not impossible to branch manifolds, even over multiple floors. But it's fiddly and prone to failure. Like balancing rocks on the beach
I get it, thank you @crimson moat gave me same idea
I don't particularly care. They once tried to argue that valves have a use with 'if you actively clock all your machines wrong in this step, and every step after this, you can use valves to limit the flow' and then was done with them
I turned it on couple minutes ago, its still spinning up, for now i did 16 into 2 floors x2 just to see how that will do, but can change to 8x4 like u guys said no problem
put powered pumps after the split goign to each floor if you want to try to make it work
and flood the system by down clocking a few gens
you do not want elevation changes within a manifold
16x2 was not working, changed to 8x4 its all good now getting flat 20GW ๐
if you're branching manifolds , powered pumps right after the junctions is often important
glad it's working though
after every junction??
after teh branches going to different floors
also possibly right before the manifold on their particular floor
Didnt have someone find a use for valves some days ago? How did testing go?
Yes i did
Its the process of adding aesthetic elements or embellishments to an object, space, or event to enhance its visual appeal or create a specific atmosphere
Very useful for it
It was said that there would be no need for loops then. Sounds different
mcgal found that sometimes putting a valve in front of a manifold did something
areyn showed untested wonk with no further evidence of reliability. Which isn't surprising because it was just valves on every section of hte manifold, which is something people do randomly trying to fix pipes w/o understanding them so I'm not surprised.
Wonder what people use more often, valves or jump pads
valves
I think it would be cool if it kinda worked like smart splitter, for example when theres 300> fluid in pipe (mk2 obv) before the valve, it opens, if 300< then its closed, and u could tweak it
you can do that with overflow humps already
Theres really no point in preventing backflow if u think about it, theres literally gravity for it
Oh, cool
backflow is why manifolds stutters
I think maybe jumpads, because: even if few people use the pads frequently (unlike valves), everyone has to try the pads at some point for... gameplay reasons
(unlike valves)
people are more likely to spam valves constantly while jump pads are a bit of a niche gimick
Imo it just shouldn't if all pipes are leveled, thats just pipes being weird
Factorio has great pipe system, but i guess its way easier in 2d
no opinion on it, if a perfectly flat pipe has an empty space behind the main flow, it'll flow backwards.
I hear you, but like 600/min from oil extractor on one pipe shouldn't have any issues with it, if all machines use 600
Yet somehow it does
It pumps fluid constantly creating flow inside the pipe thats just dont stop like that in real life
because a machine in the middle of hte line can create an empty space free of oil. It's very simple
In Sf, yeah
Because fluid in sf doesnt have any kinetic energy/flow, and just tries to balance between whole system
Dont think i got any gas before
basically teh same as fluids since yo uwant to keep pipes fuill anyway
Tried looking it up, found pic from 5 years ago, this is what you mean, still works?
Im confused how this would work
Looks like it would create a siphon
because grav priority
it won't go up the hump until the pipe before it is completely full
Factorio basically doesn't have a pipe system since 2.0 lol
2.0 fluids are op
and if satisfactory turns pipes into covered belts that will be a sad day indeed
yeah they simplified it, but i prefer this to the bizarre behavior pre-2.0 lol
That's true
It's also way nicer for overhaul mods where you need to deal with a billion fluids
i just hate the pipe length limit is all
I'm trying to plan out my first nuclear power setup (wanna build plutonium fuel rods as well so I can long term store the plutonium waste rather than uranium).
Planning to use the Uranium Fuel Unit and Infused Uranium Cell alts, but wondering if I should go through the trouble of adding oil as a product and using the Plastic AI Limiter and Insulated Crystal Oscillator alts...
I don't wanna have a whackton of nuclear power plants. (thinking 2 on uranium fuel rods overclocked to 200%).
up to you really, will adding oil improve it for you?
personally I left oil out because I could, but if the gains outweigh the downsides for you, go for it
"up to your preferences" is the common answer for most "should I do X" things in this game
Oh I know, I was more looking for opinions on how others have done it, I guess... I don't have that much time to game, I rather spend a little more time on setting up and preparing and then building it, rather than just going for it, spending 10 hours building it and then realizing "oh shit, this isn't optimal"
the trouble with "optimal" is you have to define in what regard, resource efficiency isn't a one-size-fits-all thing
if you'll have oil near where you plan to build most of the factory, then it's more worth it to use oil than if you'd have to ship it in from miles away
oh no for sure, more of a "yeah by introducing oil and these and these alts, you get to cut out those and those resources"
yeah makes sense, I still need to find a place where I wanna build it...
Looks like north of the grass fields with the uranium from the cave and quarts/sulfur from the southern forest might make most sense for me. I've started my base in the rocky desert and only have some Ficsmas stuff there, so not too concerned with radiation etc there.
Although I still need to get caterium too ๐ค
Maybe I should just double up my rocket fuel generation...
oh, didn't know they changed it, didn't play for a while
2.4 heavy modular frame/min be like
no meme, that's literally all it does (dont mind yellow lights just turned it on)
Well tbh i leave a lot of gaps, is it just me who separates smelters like that? Its kinda dumb but i dont like how they go off grid
if i was home iโd show you my 6/min one lmao
Im ready
I slooped mine for 4.8

i turned mine into 12/min 
Oh you said "if i was" idk why i read like "when i get home ill show you mine" well now know im waiting 
thats the best way
hmmm i just came home, only 10 stacks so far ๐ฆ and its working at 92% efficiency, whyy... it was so simple, well gotta eat first
did you.. build in the same spot?
if you wanna save space, definitely conesider splitting up very very big arrays of machines (like the 2 billion smelters) into like 3 or 4 smaller rows
right by those 3 pure coal nodes in rocky desert?
it looks way better, grapics setting and colors and concrete do a lot damn XD
i love concrete
in terms of building itself, it's pretty much the same though ๐
just bigger
ik ik, i should go vertical too
and that lol, i didnt bother doing it here because i had no intention of caring abt aesthetics for this one build
oof my save is littered with unfinished projects that i never abandoned, i just do alot at once ๐ฅฒ
i was just like - i made 24/min in sftools and was like ay this doesn't look too bad, ill just make quick 2.4/min for early game it will be so smol
and yet...
24/min might be a bit... much... ๐จ
this is HALF of that, if you sloop it
keep telling yourself XD nothing is more permanent than something temporary that works
but numbers turn out so good ill show you
that is a fair point
7x pure iron nodes + 3x stone normal nodes, thats gonna be my later project, when i get mk3 miners and mk6 belts
8400 iron, huh ๐
thats <10% of whats there ๐
that's what you get for not using alts
im fine with it ๐ซก
@magic dock why would you go 100/min when u can go 96/min so clean 
Default recipe is 2/min so 100/min is clean
Fair but you dont have to split the nodes
finally figured out stupid quartz ratio, so from 480 ore you can make 300 silica and 180 quartz crystals
which means just place 8x constructors for both at 100%
Could anyone help make this more efficient cos its for power and im not really good with pipes
Im bad at explaining, and since we cant see your whole setup im just gonna pull up what people said that helped me understand pipes a little better
i fixed itself in the end thanks for the help tho
are those modded junctions?
Looks like they might just be clipped into the floor?
huh, if thats the case that looks clean
@dusky dust btw what are you up to now? new big project to finish up the save? new playthrough? having a break just lurking here from time to time?
It's slightly offtopic for this channel, honestly, but one main bit of advice would be to feed fluid from above, rather than below. Fluids work best when fed from above
And i asked if someone tested it
Heh, also offtopic, but at the moment just enjoying being finally through my 1.0 save. :D I'd resolved to take my time when I started it but had no idea it'd end up being nearly six months. :D
yea
i dont get it wtf 8400 iron ore
how do u even put that into 1 conveyor
almost 300 smelters??
200 constructors
das how i dont understand factorytools website, that just looks like an insanely big factory
like thats big enough to go out of your render distance
doesnt need to be but is this seriously how u build the factory?
Its imaginary tool, helps with math
Yeah, solvers like sftools don't give you the implementation details, they just give you the overall plan that you can then implement however you like
You can create small factories no problem
youre really gonna use like almost 1k buildings?
Later, maybe
rn im just building a computer farm over the course of a few weeks
cuz ive ran back to my base more than 10 times by now
i log in, run back to base 3 times and then log out
i refuse to use the to do list
For now i built 10% as im in early game, i used same ratios and recipes and made 2.4/min with 840 iron ore
i built one thats like 6/min
close to 6 more like 5.6
w limestone coal iron ore and copper ore
im gonna make a new world soon w a friend whos never played a game like this before, any tips?
Thats great, you can always double when u need more, with sloops
how do u implement it differently than it says in the tools?
why would you put it on one conveyor?
tools don't say how you should implement it though
they just give you the numbers
it does tho
no it just gives you amounts of buildings
yeah that's just basically "total clock speed"
how you reach that is up to you
Like in the example you gave -- you're asking how we'd put 8400/min ore on a single conveyor. You wouldn't do that -- in reality you'd have multiple miners extracting ore, each going to separate banks of smelters, etc.
u need to reach the power speed of 280 smelters?
3.5 machines can be 4 machines (one clocked at 50%), or 2 machines at 175% clock speed, or whatever
it's really up to you
Or maybe you'd even overclock/underclock machines, so your machine count doesn't actually match the diagram exactly
okay lets say it says 60 smelters
Yeah, lots of options
(and the arrows don't represent belts either)
you build different amount that in total has 6000% clock speed
30 smelters at 200 clock speed
resource flow
Like, it could be a collection of belts, or it could be a truck route, or a train route, or a vehicle route.
no, it says that "all the smelters make ingots which are needed at those 4 groups of machines"
yh i have no clue how to use the routes
Or maybe you even split the whole factory up into multiple parts which all do the same thing and just combine at the end
you can build 28 modules with 10 smelters each or whatever you want
You'd have some smelters going into Cast Screw constructors, some smelters going into Iron Plate constructors, and some more going into Iron Pipe constructors, etc
Different banks of machines outputting where they need to go
whats a bank
Just a group of machines
And it's up to you how to arrange them; which belts to merge together, how you want to lay out the factory, etc.
you could have the same plan without the arrows and it wouldn't change anything
id need the arrows to know how much to put into each thing that i need to craft
the boxes tell you that themselves
hover over one of them
it tells you how much that group of machines need and how much they produce
(in total)
how many of what?
like u can make alot of cement but not alot of like motors
how do i know if i select 1 2 or 5
just build small amount for personal use and you'll most likely be fine, later if you keep running out of something, build more
or 10
(and don't reuse existing productions for your new factories, build things from raw ores)
should i even build farms earlygame for space parts like space plating?
*factories
and it's up to you
idk i had 1 coal miner going into 2 different coal generator farms + 1 automated wiring farm
again, pick some small number to start with, don't think about which nodes you have, just pick a number and calculate the production from that. Then find a place to build it
99.9% of the game is you making your own decisions, there's no "best" or "good" way to play (so even the suggestions I give you are really just suggestions)
probably not, because it's dumb. Test it yourself.use at least 20 machines in the manifold
This confuses me.
Do I only need to care about the "per minute" or does the volume matter as well?
99% of the time, the "per minute" number is the only one you have to worry about
The other number is the amount that's consumed/produced in a single cycle of the machine -- technically those are the numbers that the recipe uses; the rate is just computed from those
It's often nicer to use those single-cycle numbers to see the exact ratios of material, to get a feel for how efficient a recipe is for various resources
(Obviously you can use the per-min rate to get that too, but there's generally far less fraction simplification if you start with the counts, and you also won't have decimals to worry about)
Thank you so much, very helpful. I got scared for a bit, I forget the numbers in my head when I'm flying from one side of my base to the other, no way was I gonna keep a double set.
XD
pro-tip: use the "todo list" feature to make notes for yourself
pro-tip 2: the cOdex has that info
Memory like swiss cheese like sam from quantum leap.
The per-cycle number's also useful sometimes when taking the internal buffer into account. Like for instance if a recipe is generating 6 mยณ HOR with each cycle, and there's 44.1 mยณ already in the output buffer, that's when the machine will go into idle, because it can't fit the 6 mยณ HOR into the output buffer.
But eventually you can dismiss the input buffer right?
Not sure what you mean exactly, but so long as your numbers all match up and there aren't any belting/piping problems inbetween groups of machines, yeah, all the input/output buffers should settle into a state where everything's running at 100%
Yeah thats what I meant @dusky dust ty.
While we're on the topic, I suppose, the particulars about those numbers also explain other behaviors which might seem rather mysterious
Like it's nearly always extremely useful for us to think about stuff in terms of per-minute rates, but it's good to know that that's not what actually happens
Which is the cause of a lot of the problems folks have with pipes. Like something's not consuming a steady rate of 60/min HOR -- rather, every N seconds it consumes 6 all at once
So you've got this sudden void which needs filling in, which is one reason why fluids can end up sloshing around a bit
And it's also why the particulars of overclocking works the way it does -- the percentage of overclocking applies to the time of the recipe
So a recipe might ordinarily takes 12 seconds; when you apply a 200% overclock, it'll take 6 seconds instead. The individual cycle consumption+creation numbers remain the same, though. It's just going through the recipe faster
That results in an average per-min change which changes the same way, of course, but it's just sort of good to know about
Though, again, 99% of the time you can just ignore it and only pay attention to the per-min
Good stuff.
Just started working with pipes, I'm sure I'll have no issues whatsoever.
keep them simple and you'll be ok.
I messed up really bad
I'm making a big fuel factory in the northern island area and I thought canisters get reused when you use packaged water, but they don't
so to make use of the diluted fuel stuff, I have to make 1800 canisters/min, which ultimately means 1200 crude oil to make the plastic and then the cans
but there's only 1500 crude oil I can use atm with 900 already in use, so I'm actually cooked for this design
?
packaged diluted fuel doesnt consume any canister as they loop and sustain themselves
you dont need to continuously produce empty canisters
@wooden jasper
oh wait maybe I double don't know how to read
hmm what's a good way to split this
60 output per packager into 50 input at each fuel gen
I guess conjoin 300 and split into 6 ways
Just manifold
They do
should I build 90 gens to save on slugs or just 36 with 3 in each
I have this tucked away where it probably won't render most times, so I can probably handle having it be so large
1800 fuel is a LOT
It's all your choice
the forbidden ladder
What if you want to expand your factory? Like I dont understand how some people make enclosed factorys, how do yall deal with it
you just build what you need as part of the factory that needs it
eg if your new factory needs iron plates, you make iron plates there, rather than expanding your existing iron plate factory, wherever that may be, and shipping items across from however far away it is
ah yes
think you're confusing this channel with #screenshots
yeah like meindratheal said - if i need more becausec i need new thing ill just build it in new factory that needs it. i like it, i tried doing one huge base on my first run even tho i knew it wasn't recommended and suprise - it didn't go well for me
Mmmh
Its a bit strange because I made the opposite
So I first did what you did and it was a total mess, but now I am doing like every material on its own and its a bit more managable, at least in the beginning of the game
Let's see how it turns out later on
mostly just planning and practice
i guess both ways can be fine if done correctly
its easy as long as you have more space than you actually need, when youre done with the factory, u can delete the extra space you dont need
ex, if your final factory size is 50x100, just build 100x200 size
yeah, planning - what u saw on picture before is a part of my power generation - it was planned and is complete, takes 300 oil, 480 compacted coal in, gives 20GW (well 20.250 to be exact because theres one gen just burning fuel from 30 oil) and theres no reason for me to expand it, anything i need ill build a new factory @stoic quest
Planning 
besides, its not even that enclosed, maybe the other pic gave you the impression that the whole thing is like that, but i could if i wanted tho
when you turning on new big factory and it doesn't even do much on your power grid
@rain lichen another one done - this time had to go little vertical, it makes 20 computers, 10 oscilators (+10 > computers) and 24 high speed connectors /min, uses 900 iron, 780 quartz, 300 copper, 180 caterium ore all locally (train doesn't go far, just enough to not have a belt), for now tho im feeding circuit boards into containers - but will automate them when i get drones, will have to make 54/min with just exactly 300/min of oil and a help from 1 or 2
๐
i put 1
in this factory too, because i was making 270 cables and needed 280 and i already use 300 copper - the whole overclocked node i have here, its so useful for situations like this just gotta tweak some clocks right, pre 1.0 this would be pain ๐
wait nvm i use iron for wires xd well i still have 300 ingots left here but just gonna leave it like this
oh my god
what is that factory for
to progress... why โ
yeah but what is it making lol
i said in the message
a mess
OH
how is that so
you can get more with normal turbo fuel recipe
Diluted fuel uses no sulfur and no coal ๐
let me just do a few more ones to see if i should do turbo fuel or diluted rq
Either one is fine, really. Though I wouldn't go out of my way to get coal+sulfur. If I've got them handy then I may use Turbofuel for power, but if not, I'll generally just go Diluted
(And really there's only a couple of places on the map where you've got oil+coal+sulfur all in close proximity)
thats like 3x less power produced compared to my turbo fuel
No, Turbo vs. Diluted is nearly identical in the end, when using the most oil-efficient alts.
360 oil -> Fuel-via-Diluted gives you 960 fuel, which is 48 gens (or 12GW)
360 oil -> Turbofuel gives you 384 Turbofuel, which is 51.2 gens (or 12.8GW)
is that using the turbo blend fuel alt
I mean, yeah, in your diagrams above you're combining both Turbo and Diluted, which will obviously give you more
Which is certainly also an option, if you don't mind all the fuel gen spam
Personally I tend to just use one or the other, though
eh?
I mean, in the end you've also got plenty of coal + sulfur, so it doesn't really matter
Few players are ever gonna max out any resource (with the possible exception of SAM, if someone's doing a big Ficsonium buildout)
now im really thinking
he already got blenders, why bother with packager loop, its the same recipe
you can definitely get way more than 384 turbofuel out of 360 oil
I just turned on Maximise and went with whatever Tools spat out
Only by adding Diluted in as well
Which, again, is certainly an option. I'd thought the original question was Diluted or Turbofuel
In which case there's hardly a difference. If you combine 'em then of course you can get even more
ah yes, I read it as max fuel from oil vs max turbofuel from oil, don't see why you'd kneecap your turbofuel by not using dilutey juice when talking about maximising output per oil
i might just go with regular diluted then?
To save yourself endless fuel gen spam, mostly. :D
lol fair
(Though, as I've said in the past, as a Pure Copper apologist, I probably can't actually complain about fuel gen spam)
But yeah, if you do really want to maximize your power-per-oil, Diluted+Turbo (+Rocket, once you have it) is certainly the way to go
all of this is confuddling my brain
alts = powah
Lol can you explain for us outside?
if you make a ton of fuel-based power you need a ton of gens
I mean "pure copper apologist"
Welcome to the fuel gang ๐
I wanted something small as well, turns out, with only 240ish raw oil i need something like 100ish fuel gens
at least overclocking gens is linear
Heh, yeah, Pure Copper is a refinery recipe to make Copper Ingots which is very ore-efficient but power/number-of-buildings heavy. Just a notoriously slow recipe which ends up requiring a ton of machines to run, and adds in piping as well. There's a nonzero number of folks who dislike it since there are other alts which aren't quite as good ore-efficiency-wise but are a lot easier to deal with, but I still like Pure Copper. My most recent Nuclear Pasta buildout has like 160 Pure Copper refineries
Eh i built 80 and it wasnt bad
I just noticed its ratio, ew ๐จ ore>ingots look great but damn thats scary recipe xD
I might actually go alloy for my next factory just so I don't need so many refineries
You worked it out pretty nice 2400>6000 im gonna note this for future xd
Yeah, I use the Alloys a lot as well. For copper they're nearly as good copper-ore-wise, and have the advantage of being a lot easier to work with (at the "cost" of the most common resource on the map. :)
yeah I should be less stingy with iron really
I'm a sucker for resource efficiency and water's basically free
but for numbers I'll realistically do, so's the iron
I basically always use the Alloys when I've got a factory which needs both iron and copper
Just so convenient since you're already bringing those together anyway
Pure copper is like purpose built for copper powder/pasta IMHO. Both playthroughs is where I used that for
720 ore -> 1500 powder -> 1.5 pasta (IIRC)
Would you build that again in third run? Lol
Idk probably? Maybe give leached a try next time
I really like leached just looking at it but kinda hesitate on using sulfur, theres not much on the map and i have no idea for its later use
Sulfur is basically used for higher level power (turbo/rocket/nuclear) and explosives. Thatโs about it. Edit: forgot about instant scrap aluminum alt recipe too
If you do diluted fuel for your main power for example you can avoid involving sulfur at all
Doing only diluted power would be crazy, so much oil and gens to match it, and imagine without augumenters
huh
same number of gens as rocket fuel
Are you doing diluted only
Oh, well
133 gens @ 250% = 100gw
Yeah, the power output of the gens is the same regardless of the fuel, the only difference is the oil efficiency and thus how many gens you can run for a given amount of oil. Diluted gets you more fuel using water. Turbo gets you more using coal and sulfur, and rocket adds nitrogen on top of that
anyone know a good HMF blueprint?
I mean... how much do you want to make? what recipes? what clock speed? etc
at that point just make it yourself, since you're gonna have to make all the decisions anyway
rocket fuel is also much more efficient with coal and sulphur than turbofuel is, hence the all-or-nothing approach with basic fuel to either rocket fuel or nuclear. Turbo's in a weird middle ground where it eats loads of coal/sulphur for not much in return.
with nuclear, dont you have a first few using uranium and then turn their waste into the subsequent fuels?
and __th____en feed that fuel into more nuclear
you can, but you can make 630gw on uranium
god damn
You can also just store the uranium waste rather than processing it if you really want. A big stack of ISCs on some corner of the map
@digital root for the alt recipes:
- do not ask questions in #screenshots , we can't reply there
- there's no "good" or "bad" recipes, all are useful in certain scenarios. If you like some, get those and use them. If you don't like any, don't pick them and keep them for future. If you don't know, flip a coin and pick one, you can get them all anyway.
So this is a late reply but it depends what you want when you say 'good':
If you want it to all fit inside a Blueprinter (no stack mods, etc), is Clean & Compact , has a moderate concern to resource consumption (some alt recipes to help improve how many resources in to HMF's out), then I personally have used this design https://www.satisfactorytools.com/1.0/production?share=66waoKweqpS1GZtnUGpr . Now just because I use it doesn't mean there are dozens of other ways to achieve the same thing better, just the way I approached it for this particular BP ๐
It keeps the amount of machines low all things considered. I have a screenshot of my current BP using that entire setup but with heavy powersharding to reduce the overall size further, so I produce 125% worth of materials to drive a manufacturer with the Alt: Heavy Encased Frame. Two blueprints chained together (has an input/output manifold that can be chained across BP's) = 1x 250% Manufacturer, so an incredibly small footprint build for 7.03125 HMF's/min. (Manufacturer gets put at the end of a 'line' of BPs manually). If I lowered the clocking down to 50% materials for a manufacturer (no powershards), I'd need 5 blueprints chained together to make 1x250% Manufacturer. Still small even then to be honest..
But this is certainly not 'good' for the lowest resource consumption nor power consumption, purely just in a regard to space, # of Machines, and ability to fit in a blueprinter very comfortably. Without the decor/railing I can jam that thing in a Mk2. Blueprinter comfortably w. no extra verticality needed
designed great factory for my current tier of belts, all 480/min (i will just sloop one smelter to not need 500)๐ฅน now time to build it (yes i need it, at least 25/min framework for continous elevator mod)
btw what do you think about this way of planning? i mean i put 7.5 reinforced plate in the input to make use of the iron node whole node and sink it, might as well get some points
i mean when im gonna use a node, i wont bother coming back to split like 100 ore/min from it
in that particular case, why not just make fewer RIPs to cut your iron usage
no matter the alts i can't cut it down to one iron node
instead of using 980 ingots and having to sloop, just cut the RIPs down to only use 960
what's your belt speed atm?
mk4 480
yeah i could do it tbh
that would be tough. But pure iron ingot + changing the bolted recipes
you could also consider solid steel
The way you planned it works great for what you have at hand, so I reckon you've done awesome!
Sure there are always other ways of making the same products for less input materials, but you've got a simple factory design/layout that works!
Personally with what Cobalt just said, with the Pure iron ingot recipe I reckon you could definitely get it within 480 Iron, 480 Coal, <300 Water, but at the great expense of factory room (bulky damn pure ingot recipes!) ๐
yeah, it's a reasonable plan as-is I'd say
oh and Iron wire + Stiched Plate
all those changes should get it down a LOT
This doesn't include iron wire & Stitched, but still meets your requirements if you wanted! ( Let's just call that 480 iron in ay lol )
how's the difference in power consumption ๐
daaamn whaat, half the iron and no limestone for same output
Terrible, hook line and sinker lol
got a link for that?
guessing the main tradeoffs here are a large increase in space and power
was gonna say water but the original used wet concrete
oh sorry yep! have a play with it ๐ https://www.satisfactorytools.com/1.0/production?share=r2eIVHO8McW2AWIQwQ07
Like Meindratheal mentioned above, the giant downside to this different approach is how physically large the factory has to be and the power consumption (I think almost triple)
if you play around with the output of R.I.P , you could get some nice clocking numbers to work with if preferred. I think 6.666 RIP's/min will make everything pretty much bang on except the Pure Iron Ingot refineries ๐
you guys are crazy the way you can analyze so many trade offs of different alts, takes me 5x the time in the planner itself
not too much analysis on my side tbh, I know pure recipes are chonkers for both power and space
if it feels like you're getting something for ~free then the cost is probably space, power, or both
but yeah water and space is basicly free ๐
ah, and building count (which usually implies space)
So this was the best I could do without using more iron or any limestone, 14.43 RIP (that actually is 480 iron ore/min in then) w. iron wire and stitched iron plate rather than the Cast Screw/Iron Pipe/Steeled Frame method ๐
Satisfactory is a sandbox of endless creativity and potential ways to do things! Alt recipes really open up an incredibly large variety of options. Some have tradeoffs some don't like, others can be hands down loved by all, but choosing the ones that work for you always matters most! I'd still say your original design is fairly elegant and simple, but using tools like this really can find the best ways to tweak it up
im speechless, im definietly gonna use solid steel ingot > steel beam instead of my steel ingot > molded beam that just cuts down whole limestone, and leaves everything the same
i swear i was tweaking different alts trying to cut down to just one iron node but couldn't do it, you just did it, and made more output with it
think i skipped over pure iron ingot subconsciously, guess im a sucker for even numbers ๐
personally I probably wouldn't have used any of the refinery alts for this production line since they feel a lil heavyweight for what would otherwise be a pretty simple factory.
but can't deny their effectiveness at reducing ore costs
for bigger factories I do overuse those alts
i feel like its something you grow out of evetually in SF, every newer players tries to keep numbers clean and then after playing so long you just dont care and know it doesn't matter and the important thing is you use way less material
meh, alloy recipes are great and saves yo ua ton of nonsense
i can't find the blueprint , how do i download the file?
This comes with experience. The more you play with different recipes, the more you get a sense of their strengths and weaknesses and applicability
Build it yourself. The planner output tells you what you need ๐
My second playthrough I deliberately tried to use different recipes than I did the first time around just to see how well they worked
on your first you only used standard recipes?
Nah I was using alts even in my first playthrough, but not until later in that game once I understood how they worked
In my second playthrough, my entire electronics factory (crystal oscillators, AI limiters, HSCs, RCUs) I did without any plastic or rubber, using the silicone and caterium alts
I could have easily supplied the oil-based products there by adding another train platform but didnโt want to deal with it ๐ฌ
haha, personally i love electrode circuit board
Yeah itโs good too. I wasnโt producing pet coke anywhere tho
Didnโt touch electrode scrap until the very end game in my trigon factory that I built on the Gold Coast oil islands
"clean numbers" are very arbitrary ๐คท basically pointless to focus on
I mean i know that, yet brain see full number brain happy
exactly
Wait what i did something wrong 1 / 3.333 cant be = as 10 / 3, what
Im too sleepy for this
But yeah good point of view, thanks greeny
well yeah I didn't check the math but the concept is clear ๐
close
0.444?
you have a calculator ๐
I dont even know how to get make an equation
Press N in game for the Nalculator
3.333/1 min = 10/3min = 1/0.3min =
๐ง
2.5/1min = 10/4min = 1/0.4 min
1.667/1min = 1/0.6min
1.111/1min = 1/0.9min
7.1423/1min = 5/0.7min
Huh
#screenshots message @torpid briar hey if that makes u feel better your belt carries 1675 items / 13min 
do dimensional depots stack in terms of upload rate?
yes, they each upload at full rate
baller haha
i now have it set up that i physically cant out pace the replenishment of my concrete
Clearly you need to build bigger with zoop. Have 5 depots filling at max and still drain it in minutes.
Jealous!
why aren't the top items going into the bottom storage?
because you're feeding the bottom to the top
u need to attach a belt?!
ah didnt see behind the dasher
this is the outgoing end ,...
yeah , my bad , but it doesn't work in the blueprint designer also ... is that intended?
Yes, very possible. 24x depot for my concrete. 240 x 24 = 5760 per min
I have 10 max depots filling haha, I canโt even use up all the concrete if Iโm using my 400 concrete blueprint for my railway
Beutiful
I assume this chat okay with asking questions cause I just started playing last week and I cant seem to find anything to help with the oil power plant I want to make.
What do ya need help with?
I just got to oil refinement last night. I mainly need help figuring out how much fuel/m I should be making and how much packaged fuel for my jetpack I should be making. I'll also be making rubber and plastic here, I was wanting to use the Heavy Oil Residue Alt, and the Diluted Fuel Alt. The spot I'm at can produce 1500 oil/m if everything is maxed out overclocked which is what I kinda want to do.
I struggle with liquids in this game
Don't try and do everything at once, or try and be perfect out of the gate. It's ok to destroy entire plant set ups and rebuild later.
where is 1500ppm oil you're using? West coast?
how much fuel really depends on how much power you want to make
I think it's called the golden coast?
It's a small group of islands south of the rocky desert
you won't need much at all for a jetpack, it just doesn't burn fast enough even if you're using it all the time to need more than a few packages a minute
That's good to know cause it devours the solid biofuel I've been using
Here ?
I'm also not solo, I'm playing with a friend. We're both brand new. I was thinking something like 60-80 packaged fuel/m should be fine
Yeah
That's the spot, I can only use a few of the spots since I don't have the fracking machine yet. I have 2 pure and one normal spot.
The Satisfactory Modeler is great for this, I can try and put something together rq
I've been playing around with Satisfactory Tools, it's helped with a lot of stuff
Tools is ok also
as a kinda biased person, I don't really like Modeller tbh
Modeler was a little confusing for me, which is why I opted to tools
Instead of solid or regular fuel try liquid biofuel.
It's the same strength of regular fuel but last way longer and you definitely don't need much of it, a container linked to a dimensional depot is more that enough
what the best recipes for bauxite
"best" is relative
Best depend on what you want to achieve
There's no "best" unless you define a very specific metric which you want to optimize for
I've got like six aluminum-processing areas on my 1.0 save and I think I'm using four different recipe chains between 'em
idk i just got all of it and i have not pland what to do next
for the main recipes there's a sliding scale of how much quartz you want to use vs how much aluminium you get per bauxite
If you want the most Aluminum Ingots out of the least amount of Bauxite, then there's only a couple of options for the chain you take
then there's instant scrap in its own lil bubble
But assuming you are starting with bauxite, sloppy alumina and pure aluminum ingot will reduce complexity to bauxite, water and coal
But few people are gonna legitimately run out of Bauxite, so optimizing for that is rarely actually necessary
ok thanks
and the tradeoff there is, less quartz used means less aluminium out
that why i got it all cus i know its a rare thing
Honestly unless you're very specifically trying to max out a resource in the game, very few people are ever gonna even approach exhausting a resource
pick a recipe you like, use it. Or check all the recipes you have and decide what you like the most
o
The only real exception (IMO) is possibly SAM, and only for folks who decide to do the Ficsonium nuclear chain
I thought about that, the only part of that I don't like is the feeding the bio matter to get the process going. I'd much rather just package fuel and let it build up and sink the overflow to avoid the plant shutting off.
On my 1.0 save, where I had those six separate aluminum chains and had automated literally everything in the game, I'm only using 33% of the bauxite on the map
o cus im 50/50 if i want to do the new Ficsonium nuclear chain but thank for the help
Believe me, you don't use that much and a container or industrial container goes a long way, plus you can sloop every step of the chain if you want to generate a lot of fuel from a piece of meat for example
IMO Ficsonium is quite fun to build out, though it's quite optional. The main challenge there really is just the SAM; it eats a lot, and the available SAM on the map puts a pretty tight cap on how much Ficsonium you can produce
Optimized Refinery Setup for 1500 Oil/min
Goals:
โ Maximize power using Diluted Fuel
โ Ensure excess Polymer Resin is processed into Plastic/Rubber
โ Create Packaged Fuel for the Jetpack
Step 1: Heavy Oil Residue Production (Alt Recipe):
For 1500 Oil/min, youโll need:
50 Refineries (1500 Oil/min โ 2000 Heavy Oil Residue + 1000 Polymer Resin)
Step 2: Convert Heavy Oil Residue to Diluted Fuel (Alt Recipe):
For 2000 Heavy Oil Residue, youโll need:
40 Refineries (2000 Heavy Oil Residue + 4000 Water โ 4000 Fuel/min)
Step 3: Power Generation
4000 Fuel/min โ (4000 รท 20) = 200 Fuel Generators
Power Output: 200 ร 250 MW = 50,000 MW
Step 4: Handling the Polymer Resin
Polymer Resin is a byproduct of Heavy Oil Residue, and it must be processed to avoid clogging the system.
Options:
Residual Plastic (Alt Recipe)
60 Polymer Resin + 20 Water โ 20 Plastic (Refinery)
Residual Rubber (Alt Recipe)
40 Polymer Resin + 40 Water โ 20 Rubber (Refinery)
For 1000 Polymer Resin/min, you can do:
16.67 Refineries โ 333.3 Plastic/min
25 Refineries โ 500 Rubber/min
OR a mix, depending on need.
Step 5: Packaged Fuel for Jetpack
Take a small portion of Fuel (e.g., 60 Fuel/min) and send it to a Packager:
1 Fuel + 1 Empty Canister โ 1 Packaged Fuel
Requires 1 Packager per 40 Fuel/min (So 2 Packagers should be plenty)
Empty Canisters are made from Plastic or can be recovered by unpacking fuel later.
Final Summary
50 Refineries (Heavy Oil Residue) โ 2000 Heavy Oil Residue + 1000 Polymer Resin
40 Refineries (Diluted Fuel) โ 4000 Fuel
200 Fuel Generators โ 50,000 MW Power
Plastic/Rubber Production from 1000 Polymer Resin
2 Packagers for Jetpack Fuel
ok thanks a lot
I didn't even think about sloops
don't forget to dump 100% of the plastic and rubber into a SINK, simply peel off what you are able to use or need to use, else it will clog the system and shutdown your power gen
you can also dump 100% of the poly resin, then add the plastic/rubber later - to get your power gen up and running asap
Beautiful. I appreciate the help, I'm also going to play around with the biofuel thing too just to have something while the fuel plant boots up.
Not sure if this qualifies as "math and meta", but here goes:
How much storage is typically neccessary for a main hub? In my save, we have a big warehouse that has 3 of the double-stacked containers for each resource. Will this be enough in the late-game?
The meta nowadays, if you can call it that, is to lean on Dimensional Depots
technically 0 if you do dimensional depots
assuming this is purely for build gun use, depends on the item. Most things, one crate's probably fine, you're not building a billion train signals in quick succession for example, but you might need tons of concrete
Once you have them fully upgraded they do basically completely replace any kind of "local" storage
For some resources (like concrete) you'll end up wanting more than one uploader for the resource
In terms of local storage, though, two ISCs should be more than enough for basically any storage system you manually withdraw from. For nearly all items, you'd only really need 1
(assuming that you're replenishing them at a constant rate, anyway)
the typical Satisfactory Discord experience, 3 somewhat conflicting answers ๐
welcome to "it's up to you really" world
I only want to use those for the primary building materials: concrete, plates, rods, and the like.
I like the idea have having a central hub that things get delivered to via truck and/or train. The traffic looks pretty.
How "big" are those depots? How much can they hold?
There are enough Spheres on the map to build ~200 Depots after fully-upgrading the system, btw. You can definitely use them for literally every item in the game without too much trouble (though admittedly it's a nontrivial amount of work to collect all the spheres)
The Depot uploaders are basically the same size as a regular Storage Container
The "cloud" storage that the Depot uploaders send items to depends on how far you've upgraded it in terms of capacity. The max capacity is five stacks of each item. (It starts out at just 1 stack, I assume?) Upload speed per Dimensional Depot likewise depends on how far you've upgraded it. Default is 15/min, upgradeable to 240/min. Remember that if you can have more than one uploader for the same material, so like at max upgrades, if you have four Concrete uploaders set up, your Concrete in the depot cloud will replenish at 960/min.
So, the actual storage capacity is up to 5 stacks per depot, or 5 stacks per item across the entire network?
Like I said, I want to limit myself to only using these for the primary building blocks (those mostly used for road and rail building). To build factories across the map, I'll load a truck or train with the rest of the required resources (plastics, AI limiters, rotors, etc) and ship them to the site for use.
I know, that's not very "meta", but I like the logistics and transport side of things. With that, I'll have a physical warehouse. It is currently 3 double containers per item, and the building is tall enough for 5 of them to be stacked. I am wondering if my building is too tall, or if I need to add more to reach the top, or if I need to go higher.
By the sounds of it, people using these depots might actually have fewer items available at any time than mine has capacity for.
At the end of the day, I'm sure my current warehouse is "big enough", but I figured I'd might as well check if it needs to be bigger before I build the roof.
5 stacks per item across the entire network. Each Dimensional Depot you build is just kind of an intake machine for the cloud storage as a whole
What's generally done with Depots is to have storage containers in front of the Depots providing the replenishment "pressure"
Someone making heavy use of Depots might have as much material as you available to the Depot, but yes, the replenishment is indeed limited by the upload speed of the DD, times number of DDs you have set up for that item
Like on my save I ended up with six DDs for Concrete, so I could use concrete at 1440/min before outstripping what the depot could give me
(Which, admittedly, with some largeish blueprints I was able to do, when building out my nuclear power. Though I never had to wait for more than like 10 seconds to get building again)
I enjoy building Central Storage too (I maintained one all throughout my 1.0 save), but I admit that by the time I hit Phase 4 I literally never had to take anything manually out of it
Huh. Alright, so I think I'll settle for my tripple-stacked double containers, then. I'll eventually link some to dimensional depots as needed. Thanks.
You should read up more on how the dimensional depots work. They are a game changer (literally)
!wikisearch dimensional+depot
can anyone explain why my mk2 pipes stops when it get 600m3/s? Its between oil extractors (input 600m3/min) and refineries (consume 2,5x set of 240m3/s), i build enough of pumps and even buffer
The usual advice I give out for folks looking to send 600/min via pipes (which can also help out with pipe problems even before that point):
- Keep the system as simple and short as possible.
- Loop your manifolds (so: the input goes into both sides of the machines you're feeding)
- Feed fluid from above, so gravity does part of the work for you
4a. Avoid valves entirely (they've been improved for 1.0, so this one might not be as important, but you still don't actually need valves)
4b. Avoid fluid buffers entirely (except as buffers for train lines, where they are rather necessary) - Prefill your pipes! Full pipes are happy pipes. Wait until the system's thoroughly saturated before turning machines on.
- Place junctions before pipes. If you do snap junctions onto pipes, dismantle and rebuild the pipes afterwards.
See #screenshots message for an example of 2+3 specifically.
600/min in particular can be pretty fiddly; none of that advice is always required, but it often helps, and can be a simple It Just Worksโข type implementation.
fluid in pipes reaches dead end (either end of pipe or a full machine), and bounce back. If that happens on a manifold side pipe, the bounced-back-flow comes to the junction and some of it flows back to input, reducing it from 600 to less-than-600
I suppose one other thing to mention: even when following all that advice, some folks do still encounter problems. Little variants in how people build can lead to discrepancies and such, and it's possible that there's other factors as well
So don't feel too bad if that's not a magic bullet for you. In my experience it works most of the time, though
Please help me choose, also, what does rescan do?
if you don't know, flip a coin
(and "rescan" will give you a different two options)
Aha, is the iron plate a good option?
there's more drives than recipes, and all recipes have their advantages and disadvantages, so it's purely up to you
But do you use this/like it?
depends on situation ๐คท
it costs slightly more screws per plate, but has faster production rate, so you need less assemblers to make same amount of plates, basically trading resource efficiency for space/power efficiency
whether you like that or not is up to you
Does have the multiple of cast screws so thatโs great
multiple?
Cast screws is 50/min, this needs 250/min
As in 5 constructors of cast will make this for you
I mean you can clock machines to make the amount you need, so that's hardly an issue
there are other screw recipes as well that make screws faster
So I just realized something, how are you mixing the water and oil to make fuel without packaging the water first? Would this require the blender?
the image from the terrible planner shows a blender
That's rough, I don't have it yet and the plastic output wouldn't be enough to make enough packages. Guess I'll have to unlock it or find a way to get 4000 empty canisters for the 4000 water
https://www.satisfactorytools.com/1.0/production?share=Q3nlikfJ6waDjoKe83sS then do a quick recipe swap
oooooooh wait, nvm I'm so dumb. The packages come from the unpackaged fuel.
You do need to seed the system with empty canisters, of course, so it'll help to have a stockpile built up for that
you make a few packages at the start, adn then loop them
Keep in mind that you can fit a "closed" 2x Packager + 1x Refinery loop in a single 4x4 blueprint, and the blueprint can even include the empty packages inside the input buffer
So it becomes a little mini-blender, basically
also the loop is 1:1:1, so easy blueprint (including the pre-filled canisters)
I recommend 1 packager -> refinery -> unpackager -> back to start
keeps it simple
You should only need like 20-30 empty packages inside the buffer, I think, which could cut back on how many you need to build it out that big
I'll try to figure that out, I am brand new to the game. I just started last week.
(I generally just have a full stack in my blueprints for that, but that's a bit overkill)
is this something valuable
should i keep it or sink it
im early mid game
oh theres a wiki page it says the sink points
It's a resource you'll eventually automate. I don't remember if any crash sites require FMFs to open, but you could hang onto it for that purpose, if you want
Otherwise sink 'em
there is somewhere yeah
I'd be worried the 'storage' could be read as the shortest path and block traffic
oh yeah i didint think about that
if it's just for looks just don't connect both ends
yeah you were right :(
either alter the shape and length so it's objectively longer, or just don't connect both ends ๐
no big deal either way
i just liked how it looked :(
as long as u dont connect the rail its fine
still looks good. I've seen some people do interesting things with pillar feet sideways to make it look like a proper train stop thing
ill see if i can find an example, thank you :)
i am definitly going to try to mimic the look of that!
i got a hitchhiker
keeps saying station unreachable,even tho when i switch manual i can drive up,any help?
@vapid gorge Made one :)
maybe end the rail in some pillars or beams, something to make it not so abrubt?
i have a 3.5 item/M input line , i want to split it to one 1 item/M output and another 2.5 item/M output
how do i get that ratio?
kidna hard to tell, do you have 2 engines? 1 facing each way?
not too sure what you mean
or maybe end the rail a bit sooner so it's in the buffer thing
a top down view would be good, can't really see how the train is going in
if i am aiming for realism, i was thinking that this is a sliding buffer
wydm top down?
look at it from higher up, and then look down, to see the actual paths
like this?
I didn't know that was a thing ๐
no the thing is if i switch to manual i can drive up
but it wont auto go therr
yes, it sounds like you have an issue
and that image is from the side.
go above the rails, look down, take an image
alright
i am a huge city nerd so i know so much useless stuff that kinda relates to cities haha
the tree kinda blocks it
up more, or draw a diagram
wait someone found the issue in #satisfactory
ooh ooh! or use the satisfactory map
well I did ask if you had an engine facing each way
oh haha
but yeah that's a common issue
oh mb,ty for the help
no stress
i love looking at these maps haha
?
try cartography mod, it allows you to see the factory from ingame map like scim