#math-and-meta
1 messages · Page 40 of 1
it would reduce the amount of items in a car and would increase the amount of RTT you have "available" thus the rtt is no longer too long and what i said before doesnt apply
RTT is "round trip time"
the trip time is the same (or potentially longer because train goes slower) when you add a car
yes, but the RTT you have available increses
not if theres traffic
choo choo
point being that "adding a car increases throughput of this line"
which y'all for some reason disagreed with previously
OF THIS LINE
choo choo!
you defined htis line as "the line whose throughput can be increased through adding more cars"
I defined "this line" as "one train with one car can't reach the throughput needed"
yes in this case adding more cars would help because it increases available throughput and reducing TTTRS would help because it would make the car not full anymore
I was playing a coop dedicated server and watched some kids spend 40 mins make a hyper cannon to somewhere
Then proceeded to use the exponential equation for hyper cannons to make one first try in 10 min
They where shocked
interesting, busy atm but I'll have a look later
Who says you don't need math...
who is the biggest math nerd here?
not me, but I had a friend take advanced calculus for fun. He taught me undefined lines are technically circular graphs with a radius of infinity. I could not explain it how he did, but he made perfect sense
ill ask @vapid estuary
K hit me
Kid is hungry early and there’s karate after, so I’ll be distracted but I’ll help out
sevrahn and i are having a discussion about how to calculate the ideal round trip time for trains. We disagree. Im not going to say who thinks what, ill just present both ideas.
We both agree that this is part of the right way
items_a_train_can_hold/beltspeed
however, one of us thinks you should add the reload time too, the other disagrees
should it be added or not?
the loading duration is a part of the round trip duration
nvm we came to an agreement
Lol
we were discussing for ages and i thought id get a third opinion
but that was rigbht when we came to an agreement
The round trip time, being the time for it to return to any given point on the route, includes two load and unload animations, so you can reason out the math assuming an instant load, but in reality you have travel faster than that to allow for the item transfer and get the predicted result.
items_a_train_can_hold - 2(Load&Unload) / beltspeed would probably work no?
This assumes you only have two stops, never delay for another train, etc
Add a buffer before / after train and you can ignore it?
Because over time it will average itself out as if it didn't exist
so 2(S x Load&Unload time) where s = number of stops.
And hypothetically this would be on a rail network where the train is solo, or you've got it dialed in so there's no train delays due to perfect signalling etc...
I don’t see why that would be so
👋
🤔
Sent the update to Ondar so he can update the wiki.
well the load and unload time counts as a delay on the trip time no? so however many stops is however many extra seconds slower?
Trip time just includes everything. Hence the variable RtD or RoundTripDuration.
From the moment you depart from point A until the moment you are departing point A again is all under RtD.
but were we not talking about the delay when unloading if there isn't a storage buffer at the stops?
I understand RtD.
that's why I mentioned S x load&unload time because each one is 1 or whatever seconds where items technically aren't being transported........
Buffers don't stop the lockout timer from affecting throughput.
to further model it, you'd need to know what you're doing at those stations
than why do people put storage buffers at their train stations? lol
it's not needed if you have 1 platform per item. but if you don't...?
with the way items stack in and out of containers I think it's funny that anyone would use the inputs of a freight station with different items...I just thought it was like "Here, input 2 mk5 belts for max transfer, if you want another item, make another freight station"
Because they help smooth things out when you're not dealing in terms of max possible throughput.
Which you rarely are.
I think by "funny" you mean "not sensible", but if you sink what doesn't fit, a couple storage containers buy you some flexibility while not requiring you to place 6 stations to make, say, an adaptive control unit
to me, six stations is not sensible
Automating Project Parts...
🏃 💨
choose your part, whatevs
And you would be right in thinking that, but I chose to use a more euphemism type approach.
I do plan on automating project parts, because I want to build the tokamak from Refined Power after I get my coffee cup
not to monkey-wrench the math discussion, it does seem to me like this is about right. It's from update 4, which IIRC has single-car trains. You can correct that by multiplying stack size by # of cars. https://www.reddit.com/r/SatisfactoryGame/comments/mm0l3q/analysis_everything_you_need_to_know_about_train/
The math is clean and doesn't have 4-character variable names
U4 is pre-signals pre-collisions pre-weight
effectively none of that applies from U5 onward
I'm going to sound like a real dope, but ... weight? I thought this was a joke concept
lmgt
There's a lot wrong with that post.
no the trains have weight
and i thought all the debate was excluding any question of signals
I'm not sure what you mean by 'debates'
I've been paying attention, this has been a topic for several days
but what specifically?
this is nerd central - there's been a million debates in the last few days
debate, noun, a contention by words or arguments
lots of discussion re train capacity
Max has been solved.
Had to update slightly because I had overlooked 1 variable in a single part of the original equations 😦
oh the thing with Laura? they just wanted a single formula for all situations rather than the multiple formulas on the wik
the actual capacities weren't the problem
there's two situations, so there's 2 formulae
whatever, i'm not trying to restart it, in fact actively don't want to
yeah but they combined the formula so you only needed one.
Throughput Equations: (1560 is used because it is the max of x2 mk5 belts, and all Times are in minutes.)
First you need TimeToFill. Which is reliant on Cargo Stack Size and Car Capacity.
TtF = ((StackSize * Car Capacity) / 1560) + 0.45133_
If TtF >= RtD (Round Trip Duration)
Throughput = ((RtD - 0.45133_) / RtD) * 1560
If TtF < RtD
Throughput = (TtF / RtD) * 1560
Maximums:
50 per Stack
-88.62s RtD
-1083.3 Items/min
100 per Stack
-150.16s RtD
-1278.66 Items/min
200 per Stack
-273.23s RtD
-1405.4 Items/min
500 per Stack
-642.46s RtD
-1494.25 Items/min
A wild Sev appeared 😛
were you around yesterday for hte formula chat?
Bolded part was the one thing I missed. Fixed now.
This is my first day back.
ahh ok, Baldur and Laura just combined the 2 equations on the wiki that were the case of 'platform fill time is longer than rtt' and 'fill time is less than rtt' into 1 formula
They didn't want to fiddle with multiple formulae
I think it was from last night so you wouldn't need to scroll too far up if you were curious. Lots of graphs so easy to spot where it starts
Oh. That's why it was nice around here. 🙃
...
XD
im a fan of sevrahn, i feel like we have some things in common
😩
Now if we can only get you to capitalize your I's...
😏
:)
Also, not sure what thr point of calculating by 1560 is. By your very own words that's not possible to achieveme.
Why not calculate at 780 over infinite time and ignore the lock out? With a 1belt > isc > 2belt >station and the opposite at unload the overall throughput over time will equalize out to 780.
This is a more real world situation, at least once people realize 1560 is impossible
and yo ucan get more than 780.
I can't be bothered doing that but you Can
im going for 960 items per minutes and it isnt difficult
You can replace every instance of 1560 with the variable "MaxBeltSpeed" if you prefer.
Why not calculate at 780 over infinite time and ignore the lock out?
Because then you aren't calculating for max potential throughput...
Ah. You want a theoretical
I mean, it's totally doable too
That is useless cause it's not achievable. Got it.
ah not the 1560 no
theres achievable ways to utilize those formulas too!
It isn't useless though.
lets say you want to transport 1000 itpems per minute, its possible!
You take the formulas used to solve for max, then flip them to solve for desired.
I think there was a train line where I figured I could do like.. 1260 or something pm per car?
its possible to do so reliably
It's a theoretical that will have little impact on actual factories being built by all but the very few 1%ers like yourself. So I guess... have at it, go for figuring out maximums. Knock yourself out.
Once you know what max is you know if desired is possible.
If desired is possible, then you can solve for the min and max RtD and as long as your route is between those times you know 100% you are getting desired.
you dont need to be a 1% to want to transport more than 780 of something in one car. if its possible, why not do it?
yeah I suspect most people will prob do the 1 belt per car which works with the vast majority of situations
you can transport 1000 plastic per minute as long as your round trip time is between 75 and 384!
thats totally achievable
If you're limiting yourself to 780 per car you're choosing to make all of your infrastructure take more space 🤷♂️
why would i use two cars when i could just use one by doing some research?
Sorry. I didn't say more than 780 isn't possible. I asked why not calculate a reasonable amount that the average person will work around, and in such a way that can remove the complications of lockout times from the equation entirely.
Something more people than the most tippy top of must hit efficiency people will be able to use
🤷
im not saying go for the tippy top
Never said you wete
you cna clearly see that 1000 is easily achievable
I prefer to keep it to 1 belt per car simply to keep the throughput calculated at each car simple. Like I'll know this car puts out 780 bauxite pm and just run the belt normally
if you input one belt into the station you'll run into problems 😼
as in the throughput of one belt 😛
i hope you mean one belt into an isc into the station with two belts
How is this being solved not helpful?
50 per Stack
-1083.3 Items/min
100 per Stack
-1278.66 Items/min
200 per Stack
-1405.4 Items/min
500 per Stack
-1494.25 Items/min
You can look at that, look at how much you want to move, and immediately know if you need more than 1 car.
It's not helpful because it contributes to the issue that max efficiency is the most important thing in this game.
But .
I'm not going any further than that cause I've already had it out with you over that conversation, and don't want to rehash.
Have fun!
it contributes to the issue that max efficiency is the most important thing in this game.
Knowing what the max is doesn't inherently tell you to push for it though...
It doesn't assign any importance, it merely exists as the known limit.
That being said the game literally tells you "Efficiency First" so... Devs would probably say pushing for efficiency is somewhat important...
2 mk5 belts is 1560. Freight stations have 2 inputs.
chat seemed to have not updated, and that was the most recent message I saw. Sorry if it was already replied to.
im taking normal calc for "fun" to ig bc im not doing college
im a junior in HS rn
mathematic
eh not in the slightest, had my math final today
dunno what i got on it yet eee
I love meth
same i mean.... what
*a
im taking AP chem and its scary how easily people can make terrible things
sends DM
🤨 😅
A love meth?
Math!!! The only fun drugs I ever had inside me were ketamine and carfentanyl. Now let's stop before someone complains about us not taking about math.
Some actual math stuff, does a track ever realistically limit throughput?
Like, in real game application
well track throughput is still theoretically infinite given that the train can be whatever amount of car long
I mean like, not having enough track for all the trains
Never a problem I would run into given fewer trains leads to better throughput.
unless you route all your trains through one section, I don't think you'll have issues
Many people lose throughput due to having too many trains on a given system.
So nothing I need to plan for? Good
I plan to go for a spider web approach instead of a NASCAR track

i just have lots of self contained singular lines instead of a complex network
I am in algebra HAH
Same but a few years up
hey everyone quick question. Building out my aluminum factory right now and trying to also build a nuclear factory pretty soon too. How much electricity does one nuclear plant consume?
Nuke plants make power, so I don't understand the question.
oh wait, so you don't have to hook them up to the grid?
you're right. that would be stupid if they consumed power and generated power, my bad
Or are you asking about the power consumption for all buildings leading up to making the Ur Rod?
no just the nuclear plants themselves. Like if I hook them up with a power line how many MW would they take. But it makes sense why they don't because they create MW not consume lol
Put Rod into Plant connected to grid, gain 2500 MW
yeah that makes more sense thank u I'm braindead
To earlier conversastion:
I just combined the calculations to get the max RTT for single train to maintain max throughput per platform when set to do full loads/unloads. And if the RTT is above that, you need second train.
I personally don't use any of that and just plan for the single pipe/belt per platform constant throughput.
And when that was going on, I was so out of it that I only realized what I got in the end afterwards, while analyzing what the new equation actually did, to see if it was correct or not.
.... was it correct?
Well, it does that max RTT calculation very well.
Simple: use the formulas I already wrote.
Automatically correct 😉
After I fixed the minute/second input mismatch.
I don't even remember. Was it a single formula for any occasion?
These?
ah yeah those are still split
they have to be 🤷♂️
I think Baldur merged them?
You can't. There are 2 separate conditionals.
You can write an equation that has different behavriours for different number ranges can't you?
If you want optimal max, then you pick this one:
If TtF >= RtD (Round Trip Duration)
Throughput = ((RtD - 0.45133_) / RtD) * 1560
Because optimal is the point where TtF = RtD
And that my equation is just that combined and solved for RTT, to get the "optimal" RTT for max throughput at specific stack size and belt speed etc.
RtD in that.
Once you have the equations that solve for max, you can adjust the variables to solve for any number under max quite easily.
How do i divide one 240 belt and one 120 belt (360 total) into 2*180 belts (i have only mk.3 belts unlocked)
you're either very smart or i'm very dumb (probably both)
Basic 2->2 balancer.
What's 120+60?
Yeah, ik, i just overthink and overcomplicate everything
Which is that split both and merge one from each.
And can be also done as 3->3 or 3->2 or 2->3
Could also be done as 13->2 or 47372904736->2
Rest need more thought.
And much more splitters and mergers, all those are still easy with 1 splitter per input belt and 1 merger per output.
another question is - do you even need that? can't you just use what you have on the belt?
i just wanna load balance for once, 180 going for 10 butter/min and the other 180 is splitting to two 90 for 20 beams and 60 pipes/min
if you want to load balance that's fine 🤷♂️ I'd personally just manifold though
i wanted to do that at first but then decided it would be boring to do manifold for every build
I mean sure - just keep in mind it takes a lot more work , space and can't be upgraded or modified, you have to rip it up entirely and start over if yo uwant to change it
totally in for it
Like a stupid amount more work and is insanely tedious if you're doing larger builds
Well gl then, but it'll take you years if you keep trying to solve every load balance issue with posts. Get good I guess? xD
still kinda new only like 100 hours or so, im not planning on making much of this save anyway
kinda treating the savefile like a training
182-2
exactly
no its (3! * 3^2) / (√36 * (2^3)) + 3 * (2^2) * 5 - 6! / (3 * 2) + (4^2 - 16) / (2 * 4)
i think 
thats what wolfram got
i was just messing around with chatgpt lol
i mean fact check that if you want
Chatgpt is bad at math 😩
oh just an FYI, chat gpt just tries to give you an answer that sounds like something a human would give, it doesn’t try to give you a correct answer
yes i know
using it for fun - not for anything serious
hashtag 'not turing complete'
I know, I love ai alignment problems
They should make satisfactory Turing complete
All good games are turning complete
like the AI help bot that turned racist because they fed it info from reddit
it can't do math or logic
It can execute code quite well
large language models are wisdom of crowd vs boolean logic
actually it doesn't at all
trust me, i've tried, lol
what it does is approximate the output based on whats in its knowledge base
a friend of mine is constantly all about how awesome cgpt is
and like its amazing in its own way
but i'm constantly breaking it for him to show him the limits of what it can do with some hilarity 🙂
like you can bait chatgpt into a syllogist logic argument that 1 = 2
and there's no consciousness in it to sa 'no-no that doesn't make sense'
its pretty funny to do
yea i had a very long conversation with it once about how 2 = 3
also copying back its own responses and changing a few words can yield some funny results
It's amazingly well for a language model
Sure, if you saw a human jump 20cm that's not very impressive, but if a house does it it's super impressive!
If a python interpreter interprets super simple python code correctly that's not very impressive, but if a language model does it that's impressive imo
the point being is that it doesn't, it just knows the outcome of the program its parroting
lok as in legends of korra?
you say four rather than going throught the inductive process of 1 + 1 = 2, 1+1+1+1=4
it cannot do that
Idk, tell people from 500 years ago about what chatgpt can do and they'd think it's magic
they'd dimiss it as a parlour trick
You should let yourself be impressed more often
Nothing is impressive if you look at it like that
hey, i'm an engineer in thought, the glass isn't half full or empty, instead its 2x the volume it needs to be 🙂
there's a reason i love the game 😉
You can think rationally while still allowing yourself to be open to wonder
remove coal from the game
Turning which direction? 🙃
boycott coal
uhm, why?
Why?
:p
should i switch to 2m foundations? my loc to car ratio is 1 to 4
It would make the ramp shorter for sure...
ive never actually used a train in game, would a 2m ramp that long be a problem for trains with a 1 to 4 ratio?
Steepest you can go is alternating 2m and 4m.
But most people prefer to just use 2m for a smoother ride.
1:4 isn't a thing anymore as of U5.
Dynamic weight eliminated it.
what does that mean for me?
Just don't do this, 1:400 yielded 13kmh with a slight uphill slope.
It means if the train runs at 1:4 you're fine.
It's possible you could do 1:5 or 1:6 though depending on what you're hauling.
For perspective, this is the train spread out all over the dunes.
Also making it entirely out of 2m ramps will not cause issues.
okay, thanks!
Repost from Tech-Talk:
Related to the Discord lowers Nvidia GPU memory clocks thingie:
Official Background on NVIDIA ‘CUDA – Force P2 State’
We asked NVIDIA about this feature and they replied:
[…] Basically, we added this p-state because running at max memory clocks for some CUDA applications can cause memory errors when running HUGE datasets. Think DL apps, oil exploration use cases, etc where you are crunching large numbers and it would error out with full memory clocks. These are the types of apps you really shouldn’t be running on GeForce anyway but since there are a lot of folks who do and were running into this issue we created this new mode for them.
So Nvidia cards are running on too high memory clocks by default and that was causing memory corruption?
And they just made this power mode that gets automatically applied if anything is using CUDA, NVDEC or NVENC?
So this might also be behind some of the weird graphical corruption and crashes people are getting in Satisfactory.
Because the cards will run at max while in the game, unless some other program in background limits the card to that lower powered state.
And that same VRAM access corruption might be happening.
Performance at any cost, correctness be damned?
Not sure why that was brought out of tech talk... but ok 👍
Because tech stuff related to Satisfactory is officially ruled out of there...
Not really channel for that.
does a little victory shimmy
just brought the first part of a big build online and like, conjuring dr frankenstein... ITS ALIVE!!!!
can i ask the room before i move onto the other stacks in the factory, if this is the right choice in how the input belts are routed? i'm sorta unsure if cable or rips shoud be on the middle belt:
"Right choice" is subjective.
What matters is that it works.
Also does it look cool as you walk/hover through the place.
its more of a subjective 'what's pleasing to others sorta thing
@prisma kraken I mean, this is how I route my belts.
lotta cleanup and building stuff still to do
i'm just so relieved that days of planning and math now have tangible output... feels good(tm)
How on earth does this work? Do you route the belts through each other? that seems like a pain to set up
Smart splitters.
10 Manufacturers can be fed by 1 belt in this case.
ahhh, never thought of using them that way
ah, and that's why there's a sink at the end
there's like this wierd point in a big uild where you're like not seeing anything productive and you start doubting yourself
Sink is a failsafe in case the game fucks up.
Once the manifold is fully spooled it should never activate.
i'm using a 4x, 5 mainfold into 2x3 even split
next number up from 120 manufacturers that's an even split is 144
and um, 24 more manufacturers,no thankies
120 was enough, lol
btw, that's only 30 that i've brought online, i have 3 more columns like that to proof and bugfix
the design is sound and working though
quite a nice morning to me at that fact 🙂
my personal wierdness
is being an engineer by trade, i expected to have it running sometime on this past sunday
my estimate was off
and i need to recalibrate and learn from that
i spent probably about 18 hrs screwing around with blueprint bugs
but i can't fault my mis-estimate solely on that
Woaah, that's looks amazing
there's still 28 hours i need to figure out
Do I need to fear for performance if I try that?
The smart part of the smart splitters scares me, I have a potato
Once you understand sushi manifolds... it's the only way to build.
I understand them I'm just worried that they'll wreck my fps
work in progress, tune in for final product, but like ITS ALIVE!!!!
does a snoopy dance
Try it and find out 🤷♂️
I don't want to build thousands of those in my world only to find out near the end it made my world unplayable 😩
If anything a sushi manifold will use fewer system resources due to it being a much less of a brute force system than having 1 belt per item.
It makes very little impact
they don't wreck your fps, laura
Okay, thanks
Also less objects rendered.
I thought maybe cus it has to look at what item it is yk
That it would be more work
hardest hit on fps is flood lights on lotsa machines
And that's localized thankfully
That is a lesson I learned hard! , but must have lights 😦
Time to sushi
In a cpu bound game, it doesn't make a huge impact
hads laura some chopsticks
I'm horrible with chopsticks
well not really 🤷♂️
I have the dexterity of a brick
it's a way to build
It was so embarrassing when I was in Vietnam 😩😭
I always resorted to stabbing my food with the chopsticks, I just can't use them properly.
Try that with noodles
😬
noodles are fine, rice is bs with sticks xD
You just hold them parallel and scoop with one hand 
you know, that's actually the modality of my thinking, but like it isn't numbers flashing dramatisized, it's more like you feel the numbers in your bones and the speak and animate you
i suppose for visual thinkers, its akin to that, but its very different for us mathic types
If I want to calc a master to produce half of my steel through coke and half through the base recipe is there a way to do that? Probably a super noob question.
Tools.
Yes using Tools is there a way to do that in the master?
What is "the master"?
Go to Tools. Enable alts.
Set target to Steel Ingots.
Turn off Oil.
Turn off Sulfur.
Set amount of Coal.
Add custom input of Coke at specific amount.
👍
I mean the plan that shows the full system.
I tried 🤷♂️
What I am asking about is if there is a way to show the two steel productions side by side in a full overview. Rather than in separate tabs. Perhaps I'm not understanding
you can add multiple products in the same tab
you dont need to generate a new tab
below the product you selected, just add another product
😂 🤦♂️
I was thinking, well I dont want to produce steel ingots..... lmao god damn
finally, this has taken me ages!
how on earth do you guys do the mammoth factories? 😮
this was only to feed 10 mans to make computers, lol
Very carefully. One step at a time.
and thousands of steps? lol
Eh. Each of my outposts takes about 2 weeks to build.
heh, I would love to see screenshots
perfect ratio ... except for the backlog waiting to be belted up
next time I think I will not power the whole thing until the last thing is connected
and after all my work I am only doing 10x2.5 computers a min, lol
you using roofs for your conveyors there?
I use a variety of methods for conveyors. If you're talking about the 2nd and 3rd shot, that was just too clean of an angle to not use.
Also @rustic patio Screenies!!
yeah, love it
Did an outdoor factory one time...
But this is more a #design-and-architecture convo now.
you don't make factories?
ive only made one 😼
for what?
I have been ensuring there are at least 2 foundations width between factory walls and machines ... I think it's time to bump it up!
Fiurst
Seucond
Thiurd
lol
phyeurst
sehkhohnd
theeurdh
now I am starting to need a translator! 😛
yuu dhonth rhilly nhied a thrahnshlahtore
chroux
The best way is how Sev suggested, but to elaborate
Remove Sulfur inputs
Switch weights to custom
change coal input amount (not weight) to the max you want to put in
Enable all steel alts
hit maximize
How do you switch weights to custom? I’m either blind, or very likely stupid.
are you on the beta site?
if you change to custom, it retains it's previous weights, right?
Yep. it just allows you to edit them if you wish
Why i can't load my train?
even if I turn auto drive, its only stopping on this station
is the station set to load?
The sushi gives pain
You take that back...
is the train docked automatically?
set the timetable and let the train drive on it's own
i have one station set to load and one to unload
everything is connected
and still train doens't want to load it
Is the station facing the right way?
That shouldn’t matter as long as there is power on that line somewhere
But if not then that could’ve been it
sec im gonna check it
still its not loading
xD
maybe i have something wrong configurated in train
Did you check station direction?
how can I check that
the train has to arrive so that locomotive is at the station
I think it is
Is it configurated good?
On station 2 its loading, Station 1 unloading
and there is settings
this looks like it enters the station but the platforms are behind it
------===>
= is platform
> is station
- is the track leading to station
They have like 2 stations directly after each other it looks like
oh i have 2 platforms xD
doesn't matter
But I think at least one of the stations is backwards if you approach station in the direction the train is facing in the ss
I just put random amount, you can have as many platforms as you want
And a platform doesn’t need its own seperate station
it looks like that for better view
why are there platforms on the other end
oh i thought i can give them there xD
No?
It has to look like this
oh
And the station is backwards
it has to face away from you on that screenshot
(on the previous one, not this one)
and the platforms closer to you on the previous screenshots are those that will be used
The spider web that will be the factory i'm making
first step, move these nodes around so my brain can comprehend this
I dont understand this lmao
is it good now?
Only platforms ‘behind’ the station will be used
and what about freight platform?
Is this a lizard?
where should it be
Those are the platforms
i thought "platforms" are normal platforms, not freight xD
what is "normal" platform?
empty?
you can use those too 🤷♂️ but they won't unload/load to that car
It refers to any platform
so when i will have this 3 platforms, witch i should plug conveyors to?
depends which car you want to load
from witch platform it will load
oh ok
first car = first platform
second car = second platform
etc
so if i have 4, i need 4 platforms or not?
you need platforms on the positions where cars will be
if you have 4 cars you want to load, you need 4 platforms
damn thank you guys so much, now its working
om nom
omg, how long will this take ya?
I want the option to shut down some fraction of my fuel generators and send the turbo fuel to be packaged.
My best thought on how to do this is if I make a header where the input to the generators is higher than the packagers. So that when the packagers are on the fuel will simply not reach the generators. Is there a better way?
You described a VOP, and I'm pretty sure that's the only way you could do it with a switch
VOP where Packagers have first priority.
No switch needed.
switch to turn the packagers on/off
What does VOP stand for?
variable output pipe
Variable Output Priority
or priority >.>
Gotcha
As opposed to VIP (Variable Input Priority)
Shouldn't be needed.
Packagers and their output eventually fill and then all fuel is sent to generators.
Packagers re-engage only when fuel is taken from them by the player.
Oh I suppose if you're wanting to do the production based on what's in a container and let it self regulate
Aye. He wants it packaged. Once container fills you're done. 🙂
yeah I can't think of a likely reason you couldn't rely on the container being filled as the trigger. Some people do just like manual control though I guess
my u6 weapons plant works exactly like that
it only flips on when i pull stuff out of consumables storage
rest of the time, its just pushing coke to steel production
i'm actually kind of proud of that design
for weapons, i didn't really bother doing any load balancing or calculations really, it was just how to shove all the machines into a space and not have it be complete spaghetti
i was doing so on top of the oil nodes on the other side of the canyon in NF, and just building in that niche was challenging
the other interesting thing about it is that it doesn't take a coal line as input, and instead relies on biomass->coal for filters and black powder
i didn't really see a point in tail racing a coal miner for the demand
which is why i don't hate on the biocoal or charcoal recipes
...they really do make an ample supply of coal for weapons/gas filters
i wouldn't rely on that for power production, but find it useful in consumable manufacuring
i'm not sure if that should be done that way in u7... the alien giblets are now better used for coupon farming
i did log into that world at some point post u7 release and took the ISC full of alien protein and sunk it for the dna points 🙂
i rarely log into that save anymore, but i'm well on the way to the golden nut trophy 😛
BRO.
What?
Oh that's nothing...
too long
can i somehow place this sign to fit to this beam?
Another one making their GPU driver cry in the future.
Im just making it for train roads
Mean those decorative signs.
how much train road do u want?
It's really not terrible
Not until you go over the limit and everything goes wrong: https://discord.com/channels/370472939054956546/1070603129290231838
Wasn't aware of that
almost 8000 signs.
So it just causes vram issues?
Yes.
Wow
Each signs is separate texture and object.
Geez that's a lot of signs
Ah, it should take me a while to fill a 20gb shared pool
But still that's rather impressive
Using in same type of decorative signs for a long road.
So yes, using signs is ok.
But using them as decorative elements like the image I was commenting on can cause serious problems at some point.
And yes, they do look nice.
@opaque oak this is the graph for fluid trains
896 is the max amount and the min amount is technically 0
because they can only hold 1600 their maximum throughput is a lot lower
Way to take away what I was looking forward to when I got up 😭
im sorryyyy
so basically, safe range is 850
i thought i was doing u a favor 🥺
or just plain 800/min
And they work better on shorter routes.
2 minutes for a full freight car
to reach a throughput of 850 you would need a round trip time between 92 and 112 seconds
800 is 81 and 120
Cutoffs for 600 would be the most valuable numbers here.
a bit more than 2 minutes
the machines simply use too much liquid, relatively speaking, for trains to make sense
dont tell that to sevrahn and his fuel train :)
I'm seeing their value go up quite a bit in contrast to both of you, lol
Think of how many people make short AF trains just because.
This makes that less bad if they are fluid trains.
Although half of a 2 minute RtD is just lockouts at both ends 😂😂
ye, the low inventory makes them rather difficult to use
they just wanna force us to use packagers
That's what makes them balanced with Freight though.
yea
i bad there was like one dude that spent weeks making the packager model just perfect
and now he convinced everyone to make fluid trains useless so that everyone has to use his glorious packager model
i mean i get it, the packager is very cool
Meh.
Anyway yeah. Thank you for taking away the one thing I had planned this morning 😭😭😭😭😭
im sorry 🥺
Just needs multiple trains per route.
When making a pipe fluid buffer, Does the buffer have to be higher than the refineries line?
Why buffer on the line at all?
Don't use buffers unless you're doing fluid trains.
Not necessarily, no. If you're trying to create a water tower to distribute pressure, then yes.
distribute pressure yes.
satisfactory does not have pressure
I hope they patch water towers tbh
I actually checked the basic principle with an IRL engineer. That's... actually literally how real-world water towers work. They're not there for water volume, strictly speaking. They're mostly there for pressure.
real water towers work in a way that you need pumps to deliver fluid to the high place
Still need it here.
satisfactory doesn't need pumps
?
You literally use a pump first to pump it up the first side. smh
you can put a fluid source at the top which gives whole pipeline headlift of that water source
which is what I want them to fix
IRL ones exist to allow using smaller pumps to do the lifting to the tower at constant speed.
And the tower then delivers for the bursts and keeps the pressure stable, that a direct pump cannot do.
Ah... I see, but also, that's how aqueducts work.
I'm aware how water towers work in irl
I don't want for them to work without pumps in satisfactory
Okay, that makes more sense.
i mean...... high source of water = free head lift
i wouldnt expect it any other way
unless you are in favor of capping the height fluids can go to after they go down
I think that [fluid source] should output water into pipeline only if there's space for water up to [that fluid source's headlift]
Means the buffer + 0-flow valve thing etc.
even if you assume that, people can just have a pipe running high above and then go down from there
that means fixing valves
its the valves that are the broken ones
if I understood it correctly, this is how you abuse it, right?
yes.
fixing this requires reworking how head lift works
so its no longer max(head lift a, head lift b)
so in this case all 3 extractors would still work
but I wish that in this case the bottom extractors wouldn't work as there's no "space" in their headlift range to put water
top one just keeps applying headlift to the bottom pipe cuz headlift is not the same as pressure and only measured up
yeah I'm aware, I'd just like to see this fixed 🙂
with a full pipe rework maybe lol
well if we're getting one in the future anyway
I don't think it's that complicated tbh. Even min(headlift a, headlift b) would probably work better
smallest head lift wins? odd
I'm not saying it's perfect but it would be less abusable
(and most "normal" setups would work the same anyway)
but i hope we all agree that a pump thats under 10000psi of pressure would need at least the same aomunt of pressure to pump water out
and would mean
Min(0, 10) results in 0
so theyd need to make it that buildings that are under headlift can push out fluid
is there a machine that provides 0 headlift?
yes, unpowered pumps and literal pipes without fluid in them
afaik all things that output fluid give at least some headlift
pipes only pass headlift tho, no?
which would make pipes way too complicated imo
a constructor offers no headlift... but also no fluid 😉
that would give unpowered pumps basically a way to turn off fluid input 🤔
would be bad to have no headlift, you might not have enough horizontal space to attach a pump... would look funny if you always have to go down first before you go up
(again tho, I agree that it's not a perfect solution, but I would prefer it over max(a, b) )
and ofc I'd prefer better solution over this
Pipes and buffers pass on headlift proportional to their fullness, unless they are full in which case they pass on system lift
yeah I don't think that would break the the min headlift, but w/e
it's all theoretical anyway
From my understanding, that's what allows fluids to climb slightly past headlift
The last pipe doesn't quite have enough headlift to fully fill, so it passes on a little bit of headlift, just enough to push the fliud a little bit up the next pipe
Depends what your factors for "best" are. The math is mostly all there on thr wiki too
I may do silicon circuit board (copper sheet, silica) and crystal computer (crystal oscillator, circuit board)
Seems like one of the best ways since I'm building the factory at the speed run spot
If I may recommend:
CtCB pairs better with Crystal Comp.
I'm not to sure tho
SilCB pairs best with CtComp.
Caterium?
Yes.
.
Personally I like Crystal Comp so I always use it an CtCB.
You need oil for Oscillators too, but if you're at Speed-runner's it is right across the canyon.
Ohh wait I forgot about that so I would have needed to anyways
Lol thanks
Yeah it's not to far
Do you just use default recipe for oscillator
Insulated is the only way you should ever make Oscillators 🙂
Okay good that's what I was thinking
The only way according to Sev, who is one person who feels the only thing thst matters is pure efficiency. If you agree, go for it. There are more ways to play than that though, and other considerations you might have. Play as thou wilt tho.
"Efficiency First"
isn't mentioned which efficiency 🤷♂️
Silica is one of those "need lots every here and there" resources, but quartz veins aren't too plentiful. I'm thinking for this playthrough that I'll make a 2000/min factory with multiple drone ports just to make it kind of simple to have silica wherever.
Now I'll just need to math out which recipe and location to use 😅
cheap silica, rocky desert
you dont have that many options when it comes to silica alts
theres base and theres the alt
Now that I'm looking at the wiki for usage, I could do without it via many alt recipes. Wondering if my plan is worth it after all
Okay, so now, a question:
Hmmmm...
This looks to me like a prime candidate for a straightforward High-Speed Connector factory.
I always use Silicon HSC.
I could use pure recipes, but that's added water and complexicity. I'm feeling that raw resources aren't an issue, and would prefer to keep it simple.
fused quickwire is good enough of a choice
If it's meeting the demand you need, it's a clean line.
Yeah, I was going at Silicon HSC at first, but now that I look at it, I could do without the Quartz.
if you can, you can try to use copper alloy (if you have iron near it for some reason)
not that much more complex and also might aswell
I legit almost never use Pure Iron as an alt.
I've never gotten close to using the 74k base iron on the map.
Honestly the extra water is just a pain and it's a ton of extra space
also, iron wire cable is like.... garbage tier efficiency
consider quickwire cable or insulated cable
Pure Ct is worth it.
Pure Fe is not (imo).
Pure Cu is fine, but usually I use Alloy.
Copper Alloy hmm. That would mean 50 more iron, and 120 less copper.
Copper Alloy is such a good recipe.
Honestly fused wire > default cable is better than quickwire cable imo
I think that's a good idea
both for even more boom
fused quickwire > quickwire cable
"Better"?
"better"
we did the math on this before
We're moving slightly out from the "straightforwad", but the saving in mind, I think it's worth it.
There was an IMO there for a reason
quickwire cable is ridiculously good but its slow
This.
so if you really want that reduced machine count if its bothersome for the build, then get insulated cable
I mean, the math part is easy. But there's a point where straightforward meets the math. 😄
100/min cable for just a bit of rubber is nothing to scoff at
It was basically Insulated vs. QW for "best" iirc.
I don't steer away from multiple machines, but when reducing machines comes with the cost of having to figure out belting more, then we get to the point where I might use more raw resources just to have straigher lines without shenanigans.
and for a "meh" sidegrade, coated
And it came down to "player choice"
yeah
I like QW Cable 🤷♂️
I love choice. And the other thing I love is knowing that in past playthroughs I used X, so now it's cool to go Y route.
honestly if it was a tier list:
S Rank: Quickwire Cable , Insulated Cable
B Rank: Coated
Coated trades some of the efficiency for a really simple setup
but transporting hor is a pain
so it has to be made on-site near oil most of the time
This is what the calculator would have me do in a perfect world.
I think that's not so great for 10/min. If I wanted to scale higher, it would make sense. But for the given output requirement, the added complexicity is perhaps a tad too much.
I decided on this.
Then again, it might be just an issue with how I visualize the plant. The latter is far easier to lay out for me at least.
for 27 ish extra quickwire and 18 rubber you could make 100 cable
and cut out the iron wire stuff
Do I really want to save a bit and add Rubber though?
since fused qw is so fast, you wouldnt even need more machines for that
Ugh. You made me question myself now 😄
could bring it in with the CBs
tiny mixed transport
CB?
circuit board
Ah, I don't have a dedicated rubber port, gosh darn.
those gotta come from somewhere after all
I mostly use the weird diluted & recycled combo to make recycled plastics.
All used for the factory, none sinked.
ah
Whatever I make, I make 100% efficient 😄
It might not be the most resource efficient, but I strive to not sink anything.
Thanks for reminding me, I have an 9800 surplus of Computers
I think I made this as a precursor for Super Computers a few weeks back, but I still haven't tapped it for anything. Making 10 cpu/min.
i mean technically the most efficient is for every non-power building to be built at 1% underclock and you build a hundred of them
but theres limits to how feasible some things are
Thanks, I hate that you just gave me a perfect idea for my next playthrough - minimum power consumption.
someone on the reddit attempted to do a playthrough with every building at 1%
which broke massively when the overclock rebalance patch came out lol
1% playthrough sounds horrible
why is it only 5 output if i have more than enough iron and coal?
I assume you've used maximise mode?
Useful numbers for T1-2:
You need 72 Qtz to equip Blade Runners.
You need 450 Ct to unlock Mk2 Power Poles.
You need 690 Ct to unlock Smarts.
yes
maximise mode makes equal amount of all items, until it can't anymore increase all items together
ahhh i didnt know that thanks!
My load balancer. Needed 240 and 90 from 330 coal. It was a pain
Is there an easier way to do this?
yeah, don't balance
manifolds can do the same
---S--S--S--S--S
| | | | |
Nearest coal for me to avoid doing this was 1.5km away
you don't need any extra coal for that tho
I already had a factory set up consuming 240 coal and needed 90 extra, so I put a powerslug and upped it to 330.
alright, and what's the problem?
Everything was set up to consume 240, I don't know how to split off 90
Which was why I built the load balancer
you don't, you just use single splitter and it will self-balance
(or you build a balancer, you can play in any way you want)
yeah but if one side needs 90 only and it gives it 175, then it will get more than it needs, fill up to the splitter and then only accept 90 because it's full
and rest will overflow to the 240 side
if the 240 side is important (e.g. coal gens), you can use a smart splitter to prioritise the 240 side so it will get all 330 and when that fills, the 90 will overflow
I always say that "if you feel like you need to do balancers, there's probably easier way to do it without them"
(not always better way, because definition of "better" depends on the person building it)
At least my factory needing 90 coal won't stutter now I guess
if you used single splitter, it wouldn't either. It would run at 100% capacity. The 240 factory would get less tho, so it wouldn't run at 100% until the 90 factory is filled
2 hours, wasted! This is unfortunate. Will keep your wise words under advisement
All I have to do now is pray FICSIT doesn't see how much time I wasted not being efficient
time spent learning is never wasted 🙂
That was two precious hours that couldve been spent on my new factory
Don't worry there is 300+ hours still to go.
:eyes the 60 + hours spent designing a beltway that is ... currently not hooked up to anything... on either side... and i dont have any desire to go back to that for now... so ive spent about 150 hours on other buildings...
yeah... couple hours figuring out some Perfect Split balancers isn't horrible. And now you know for next time how annoying it can be and you can save yourself the trouble with a smart splitter and an overflow option :p
Lol I feel you, I've spent a tremendous number of hours building my central storage facility and doing all the beltway for it with absolutely nothing going into it and a lot of the resources that will go into it not even being produced yet
Need to get that spaghett though lol
I'm pretty sure I spent like 6 hours at a pipe/belt junction ( 5 pipes 9 belts) where they had to meet going up and down and to the side to make it look pretty xD not that weird
aint that the truth
i have 600 oil at full flow, going into 20 refineries, 10 making plastic and 10 making rubber>
therefore HOR is 300/min, right?
going to 5 more refineries for fuel conversion - with 1 diverted to packaging
and then to 12 Fuel gens
for whatever reason I can't understand, 1-2 of the fuel refineries are not 100% and won't feed all 12 fuel gens
what am I doing wrong?!
First troubleshooting step: identify what machines are having issues.
Second troubleshooting step: check what is the issue with the machines
Third troubleshooting step: think about what can cause that issue specifically
Fourth troubleshooting step: attempt a fix
Fifth troubleshooting step: if there are still issues, repeat from (1)
So, to answer your question, you need to attempt (2). Right now you're are (1)
the only thing I've understood is the flow isn't steady for HOR - it ranges wildly from 45-295
Sounds normal. Still doesn't explain what issue the fuel refineries are showing
but the machines making the by-product are 100%
fuel machines aren't getting a steady flow of HOR thus never fill up
Example of issues:
Lack of input
One of the outputs is full
No power
none of the above, i've triple-quadruple checked
So the input pipe to the using machine is full? But the machine input buffer isn't?
sinks are in place on all outputs at appropriate rates
not even
Or some/all of the pipes aren't filling?
Now, let's move to (3).
Not enough HOR can be caused by either
3a) Bad math (wrong production numbers)
3b) Bad piping: the HOR is piling up at source and not reaching destination
3c) Bad production: the HOR refineries aren't actually producing at 100%
Which case you think you're in?
gonna check current machine fills, but everything is green until fuel prod, which is running 8 fuel gens steady, but not more, and according to the calculations, i'm producing around 160 fuel to the 12 gens, whereas I only need 144
might be 3c
... You just said you are lacking HOR - > the issue the machines have is "lack of input (HOR)" ^^
Check if the HOR refinery output buffers are empty or full? To see if it is 3b or 3a/3c.
i have 20 refineries producing @acoustic pawn% HOR
Bruh... That at 
so that's not it, the fluid just isn't being fed at 300m/min
Are the HOR refineries piling up HOR (Yes/No)?
no
i just figured if i'm outputting 300m/min across 20 refs, 10 plastic and 10 rubber, the flow should be perfectly balanced
Then the issue isn't in piping. If it was, the refineries would be piling up the "excess" HOR that isn't reaching the Fuel refineries. Instead, they don't have any of the "missing" HOR in them, which means we're in case 3c imo
regardless of whether one is 10 or one is 20 out
Your numbers are correct.
I blame the Plubber refineries not running at 100% and causing a starvation of HOR while you don't notice the (slight) reduction in production of Plastic and Rubber
The reasoning is above
so i switch off one fuel and suddenly 4 fuel refs are 100% but i'll get backed up on HOR feed
because otherwise 2/5 are idle almost every production turn
i just check every plubber (nice) and really zero issues, no build up of anything anywhere
So every crude oil input buffer was full on those?
And output buffers were empty?
yup
i'm starting to think it's a timing thing, not synced so flow varies based on that, since the math is all direct, production times might be the issue, since it just doesn't make sense otherwise
And the piping between those and the HOR to fuel refineries is now full?
How are the HOR input buffers?
With one disabled and with all enabled?
only the fourth one is slowly filling more than using
so 3 full and 1 filling
this is 100% but will eventually top out slowly backfilling the output buffers
So there was underproduction or the HOR to fuel refineries weren't primed, and now there is slight overproduction of HOR with one refinery turned off.
i've tried it with both, and at first when it's primed, everything is perfect, until it tapers off again
even tried with extra buffers, and still getting underflow, and I added a pump but it's not necessary for 4 meters
Question in that situation is, like Sushi Person tried to know, if in that situation the Plubber output HOR buffers are empty or full.
No external buffers.
empty
and naturally i would have seen my lines of products stop, but I rectified that with sinks
can you tell me what the general ratio is for HOR->Fuel->Gens is?
So underproduction on the Plubber side compared to the HOR -> fuel side
If it was transport problem on the HOR side, the Plubber refinery HOR buffers would be full.
And if it was transport/math problem on the crude oil side, the plubber refinery crude oil input buffers would be empty.
So you are using the Residual Fuel recipe, with 5 refineries?
but i'm thinking i should pop a switch on the Plubber and hard reboot so the timing lines up
yeah with 1 going to a packager and 4 going to gens
That should cause the output buffers to fill up when the residual fuel refinery input buffers empty.
If it was transport problem in the HOR piping.
We don't have to care about the fuel side at all, if the HOR to Fuel refineries aren't getting full output buffers and stopping because of that.
right
10 rubber for 200 total, 10 plastic for 100 total, so total of 300 hor, being fed to 5 refineries each using 60 HOR, total of 300.
But that should either work or have some buffer filling.
Count the rubber and plastic refineries again?
4th HOR->Fuel ref is staying balanced around 20-30m/min
10 of each
exactly like you said
i'm not getting enough HOR which is impossible
omg... count again he says
wowi'm sorry, i'll see my self out
8/8
thanks though, i understand clearly now >.<
Ah, even more missing than I thought.
But easy mistake to make.
Which is why rechecking is often important, and not just saying what you think you have.
Because when building lot, it is easy to make mistakes on exact amount.
I usually do multi-dismantle select for only the specific building type after building, to recheck the number is correct.
so basically adding the last 2 will solve my issue
2+2
Depending on piping work, you might get then problems with getting the full 300 flow.
But that will then show up as those buffer inbalances in the wrong direction.
Inputs empty but outputs full.
More probably on the crude oil side.
Short Mk1 pipings shouldn't care much about complexity.
If you have just two Mk1 pipes feeding 300 crude each.
but it doesn't matter if i use mk2 anyways, right?
Single Mk2 should be able to feed 600m^3, but again, depending on complexity it might not.
On the crude oil side.
If the HOR piping is Mk2, no problems.
something else i heard though, build mode can affect flow?
Build order might affect flow currently.
Best to build each segment starting from feeding direction, towards the usage.
It shouldn't, but at times bugs have caused it to matter.
And there might be such but in U7.
At least opposite build direction can get you free headlift in some cases.
right, i'm actually bringing the crude from the front to the back with the HOR going back towards the crude input, down 8m and back up 8m to the fuel refs
If you follow the rules a mk2 pipe can transport 600 per minute no problem
And not much reason to pre-care in this case when the piping is already built.
It will work, or it will not.
If it doesn't, then you need to look into it more.
But will most likely depend if the crude oil feed from the extractors will work right or not at full capacity.
Because I expect it is from the east coast island nodes.
you'd be right, i've got 3 feeding into 1 mk2 and a single into mk1 respectively, merging again back home
so to prime, i should let my ref buffers fill up, then power on the next section
Basically. More complete version is to block output from the next section until they and previous stage stop.
otherwise it's an eternity to level out
but since i'm using MK2 i should be good for a while once everything is primed
thanks again
You should feel ashamed 
Nobody makes mistakes here.
https://media.discordapp.net/attachments/456117300446232578/840230672713908334/Nuclear_Guru.jpg
As said, I usually just remove some strategic belts and/or pipes from the output side and wait until everything goes idle because of full buffers.
And then connect.
Can be done for just the last stage in production chain.
what's wrong with that?
You know, the fact that to make max amount of Fuel Rods you don't use standard recipes... 
@vapid gorge
this is the graph for fluid cars with mk2 pipes
this is the graph for coke
oh because belts are faster?
And the fluid cars smaller.
cobalt, what round trip time buffer amount are you comfortable with?
Doesn't that only kick in if the trips are taking longer than it is to fill the station?
It's that second line going down.
Limited by train interval with only 1600m^3 per fluid car.
I do 1 into buffer 2 into station. Big enough time buffer that way
So by placing multiple trains, you can keep it at that max.
the buffer amount is the amount of time your train is allowed to be late or early before you dont transport enough anymore
with a round trip time buffer amount of 1 second your trian would only be allowerd to be one second early or late before you dont transfer enough anymore
a good reason to not run it anywhere that close?
the lower the buffer amount the higher the risk, but the more stuff you can transport
my question was what buffer amount youd go with
at least a minute?
sure. I think all mine have way more than that.
let me do some math, ill be back in a bit
I'll be asleep 😛 it's midnight
where do you live? it’s 0812 here in the eastern us
just a minute
Australia or New Zealand?
if you want a buffer amount of 60 seconds, you can transport 1315 coke per minute
but only 570 oil per minute in a fluid car
so even if fluid cars were just as easy as solid cars, they would just be plain inferior
now packaged fluids vs fluid car is a different beast
because packaged fluids have a lower stack size than coke
those pesky australians
wonder what it’s like tmr
technically packaged trains have a slight advantage over fluid trains because of the higher belt speed
the only difference is the 780 belt speed vs 600 pipe speed
well, plus the packaging loop... which is just a pain on large scales, so no
ill stay with transporting coke
And I will be moving nitrogen gas in fluid cars, with 480m^3/min average requirement per fluid car, about 3.5km straight line distance.
But I'm intentionally crazy. 🤪
Need to calculate later how many trains do I need on that route.
122 seconds RTT
so if your round trip time is 1220 seconds, you would need 10 trains
But I haven't measured the actual trip time yet.
yes, you want the round trip time to be 122 seconds
So it will happen later.
so you add trains and make them wait to reach that round trip time
I know 😉
anything between 45 and 200 seconds will result in you transporting enough
if you go for 107 you could transport the highest amount, but if you go for 122 you will have the highest buffer
basically, i would gain 5 seconds buffer time, and for that i would pay with the entire packaging loop
so packaging is out of question for me, just like fluid trains, even if it wasnt incompatible with my philosophy
oh nvm i did the math wrong
packaged fluids actually only gains a buffer of 2 seconds
Hmmm
i think i fed the math up a bit more, but sitll, its one train vs one train
because when i ditd it again i got different numbers
so pleass ignore the screenshots above
Fluid Trains vs. Packaged and Ship is equal in terms of per-car when it comes to liquids.
Fluid just ends up needing more trains per route as we showed the other day.
You need 2 cars to ship 3200 either way.
Fluid is a bit more limited because pipe speed is lower than belt speed
Slightly.
But the main decision is "more trains per route vs. setting up packing loop infrastructure"
Definitely
packing loops are good for Nitrogen gas... 🙂

use N for an ingame calculator
the calculator will tell you.
what is wrong with it ? and okay
The fact that your generators are turning on and off
Water trouble