So I was thinking about how you'd go about designing a factory that would complete a milestone the fastest. I was figuring that you'd want to make one that produces the required items at rates proportional to the amounts required while consuming less than or equal to the available resources. Is this the correct line of thinking/has anyon3 here done something similar before?
#math-and-meta
1 messages · Page 17 of 1
Do you mean a milestone or a space elevator stage?
Which milestone? And the early ones go through pretty quick, so....
Either/or really just thinking in general
You don't really need to create a factory specifically for milestones
For the space elevator, phase 1 and 2 I probably wouldn't bother either
Phase 3 and 4 would warrant automation, the decision you'd need to make would be how quickly you want the items produced and how long will it take you to build that factory (which is something only you can know since you know what you've already got automated)
You could theoretically build a factory that can produce all of the items for phase 4 in less than an hour, though it would take you a very long time to plan and build such a factory
I've made temporary ones, but that's usually like getting the stuff in there quickly or something.
Well yes that's more just rushing to unlock new milestone items. But typically you'll be wanting to automate the new components anyway so you can use them to build stuff. So you're not creating a factory purely to churn out your 1000 plastic you need for your tier 5 and 6 milestones but one that will supply a steady stream for your building needs (at the least)
Ah, yeah I was really just thinking hypothetically if that's the kind of factory you'd want to make if you wanted to minimize time to a milestone/space elevator goal. Cause yeah, it would be a pretty big waste of time of that was your only factory in an actual ongoing game
the last elevator parts are great for sink points
so they serve a dual purpose and a challenge
Watch the speedrun 29hours 100% to understand how to optimize the workflow. At the beginning loot crash sites, it really helps.
Where i can watch ?
I cant find it, im only see a speedrun package2
I assume they are talking about these? https://www.speedrun.com/satisfactory?h=4Package-Update_5-Solo&x=02qepjp2-kn0zgoo8.4qy2xo41-p8555q58.mln2850l
or youtube, search 4Package%
See box crafting technique. Is one of the best ways to get the phases done if you have a good storage room
Just wondering if this is correct or not because the mk5 conveyor can transport 780 materials total you can't overclock a mk3 miner (on a pure node) to 130 percent to get complete efficiency from it because the conveyor belt is the limiting factor.
That is correct.
It's been that way for years.
The 1200 from a mk3 at 250% on a pure was set before they knew mk5 belts would be the limit.
It's going to be addressed prior to 1.0.
No.
"mk5 belts would be the limit"
Game barely handles mk5s.
Mk6s would be worse.
Understandable
They had them in testing at one point but pulled them before they shipped to the main game.
How long have you been playing I remember joining at update 3 then had a break and joined back not too long ago but I can't remember my brain melting like this 🤣
Since before it was even available on Steam.
That’s long
I remember mk6 belts being available in old files even before update 2, including mk2 buildings back then
I still hope mk2 buildings make a return
probably not
Some builds look positively ridiculous due to building counts
i feel like that will be a mod thing but not a vanilla thing
i could see some mod that allows like overclock above 250% or mk2 buildings, or something like that that lowers building count
Should never need mods to solve vanilla issues
its not an issue
that's what overclock is. They replaced faster buildings with overclocking
Performance isn't an issue with large builds?
Also you're the one giving yourself problems with building count
I dont have issues with my massive builds and I'm on pretty old hardware
thats why people who go big or are doing demanding projects save edit in a bunch if shards, thats already 2.5 times less buildings
So you're already aware of the problem then
mk2 buildings would pretty much require a re-do of recipe progression
Not to mention entirely new models of existing buildings to accommodate mk2
Would they tho? Let's say they just operated twice as fast for twice the power, like how mk2 power poles have (almost) twice the connections
They also have a different model
Don't really think there'd be much progression interference there
but then thats just overclocking - the solution the devs have chosen to use
I mean... there's not much else to say to that other than "So what?"
more work and effort for something that has been addressed and moved past already
Well it's only a problem if you make it a problem right? You can get through the game with very small builds.
Tbh there's no response to that line of thinking
Like that's just personal choice
but either way, i still feel like this is the most likely outcome
exactly - people can deal with the consequences of their choices right? If you want huge builds with little computer issues build a rig like Tunnelrat's
he's a madman
It's work that could be spend elsewhere... On getting the game to 1.0 state
Yeah I don't agree with that, it's like saying "Why bother improving performance, just choose to build small"
They improve performance every update though
... it's not a performance issue - it's a user created problem. You can do pretty god damn big builds especially if you create hubs as it is
In fact u5 -> 6 did a lot of good for performance
I mean performance is involved but you can work around it unless you run a potato
yea, its not an optimisation
Like all things in the game there's give and takes.
Less resources? Prob needs more space
ect ect
Feels like you're both aware of the problem but your solutions are just "Don't do it", which is fine, but that's like, kinda pointless to discuss
Well it's a bit like complaining that staring into the sun causes damage to your eyes.
The solution is don't do it.
completely changing things up and removing huge chunks of logistics in games is a pretty big deal cause thats a game mechanic to play with
what we're saying is it isn't a problem, in any sandbox game you're eventually going to get to a point where your computer cannot handle the sheer quantity of buildings, the solution isn't to make buildings that are better, because then you are just delaying the 'problem'
and wanting faster machines is a fine personal choice - but it completely changes the way the game is played with logistics and planning
at that point - just use some mods if you really feel that way
exactly
Wdym? I do it on years old hardware and don't have an issue
Congratulations
To flip it around, if coffeestain were like, "Ok we're adding mk2 buildings that cost end-game resources", would you be here arguing that it's a bad idea?
I mean it's fine not being happy with a Thing - but when it's core gameplay stuff they'll never make everyone happy
ehhhhhhhh..... I can't imagine it being a thing w/o significantly hindering work on literally any other aspect of the game
and that would make me unhappy
If I want mk2 buildings there's a mod for it
there are mods that let you build nodes and just have everything in one spot if that gets your motor going
it depends on what the mk2 buildings actually do, if theyre just better versions but are balanced within the game then its fine, if they have separate uses even better, but then it just gets to the point where you start arguing "we need better buildings because mk2 still causes issues eventually" its a never ending cycle, and with the current buildings it is fine as they are, and im sure if you disagree you could find a mod to fix it
It'd be easier to just make overclocking have a higher cap if people really want that
if they do the same recipes? Just make have even faster overclock that guzzles power
Different recipes? Make them new buildings
which is exactly what i said 
I'm pretty sure there are mods that also do like 10x recipe speed?? there's a lot of options out there that radically alter game play for people who want it
the only problems i can see are belt throughput limitations
and mk10 belts ! 😄
machines that 'package' items into 'boxes' of items that get unpackaged later down the line to manage increased throughput
NO idea. Was never interested in those mods
Yeah teh boxing and unboxing one was interesting but didn't want to fiddle with it. Stuck to trains and keeping things local instead
They rather don't do Mk 2 because it effectively retires Mk 1.designs
simply slapping a mk 2 on it and not changing design wont do
as stated above, shards also help alleviate some of that problematic
not to mention alts
To be pedantic
Mk 2 buildings is not the only option
That's already the case with power poles
yeah, which are basic ass things
I mean, so are constructors
they dont feel like its warranted for production buildings
theres other ways to solve it
be that a stat boost or what ever the hell
but even that has problems
Mk.2 Buildings are the last option to fall back on if all the other stuff doesnt work out
Would you be surprised if they roll them out?
yes, very much so
Since twice speed buildings also clashes with belts
they already have speed issues
why worsen it
or even a 1.5x boost.
but, again, shards exist
I'll be honest I fully expect a total review of all belt outputs/inputs to solve the 780 speed issue anyway
Rather than a coal gen mk 2, theres fuel gens
Rather than a Fuel gen mk. 2 or whatever, you get nuclear
Sounds like the only way to fix it is to halve resource costs (and yes I know some of them can't be halved)
Yeah I think power is in a pretty good spot in terms of building counts
As for other stuff? They just gotta tweak recipes
easies way to cut down building count
and they wanna do a recipe revamp someday anyway
perfect moment to reevaluate all this
That does seem inevitable
yep, and its the perfect time to revisit this topic of machine count
Default Recipes, Alts and Power Shards
the trinity of balance
how many coal generators can i run on 180 coal a min
my belts can handle it
and my water i know how to upgrade that
take the amount a coal generator needs, divide 180 by that number
k thx
remember that your mk1 pipes only support 300m3/s in any given section.
yea i can add multiple
i can even do one water collecter per coal plant
that would be stupid tho
I see we had the mk2 building discussion again...
nah
how much can a mk2 belt move lmao
Up to 120/min
nevermind i read the description my bad i should have read that earlier
i need mk3
So go unlock it!
I'm not too confident they will. I hope they still plan to have a revamp done (in which case they would at least try to tackle the issue), but if not... this might be left unsolved as it is in Factorio. I heard that even that dev team didn't manage to avoid throughput issues at some belt junctions (bends), leading me to think that it might be too big a hassle to solve for the payoff
what is the "unsolved" issue in Factorio?
huh, never heard of that
I know this was an issue way back, but then they changed stuff and said it was fixed
also makes sense in the way that factorio pipes arent really made for max flow
you are always below the limit
so either it's old news or it's a new issue
or you need 5 million pumps
Compression loss on factorio belt corners was a thing yeah, still is afaik
factorio pipes aren't "rated" for a certain flow anyway, since factorio uses pressure-based system
indeed
(but also, in Factorio it's kinda rare that you need a throughput that would require you to put a pump every pipe tile)
if you consider that our pipes are also 1000 units for one, ours are just as fast
Does seem like they fixed the corner compression loss bug tho from a quick search, dunno if there's some other bug they were referring to
Us pipes bestest 🐒
our pipes have 6 times the theoughput of factorio
My hope for the similar issue in SF to be fixed too increases a bit
36000 units/sec
and suck 6 times more than factorio
vs factorios 12000
eh, slight stability issues
can be improved
Wait... Was greeny always blue? 
Am I tripping?
Or suddenly daltonic?
100/60
that's pump throughput, not pipe 😄
What do they mean, Mason?
What are the numbers?
ah.... wait
ok thank you
basically 3000 is what you want to count, as 6000 means "pump, pipe, pump, pipe", while in ideal world you want to do "pump, underground pipe, pump, underground pipe"
yeah got "promoted" by Baine (well, I was testing some stuff for him in #1038092680493801533 and he was giving me different roles to test link posting and after last test with dark blue role he said "consider yourself promoted" 😄 )
will needing .067 more cause a problem
the machine will stop once in a while, but shouldn't be an issue if you only need the production from 100 crude
set the number more precisely
👏 👏
So you got a pay rise, I see...
Too bad you never play to spend those Satiscoins 
well, this isn't 50 exactly either 🤷♂️
😭
it's 49.99998/min
aight so what im getting at is our pipes are ridiculously fast
NOOOOOOOOOOOOOOOOOOOOOO
so you’ll have an issue after 5 years of gameplay
i will wait eight months until i unlock wet concrete
I'd just do 83.3334% which is 50.00004/min
yes but perfect numbers
are not reachable always
after 50,000 minutes to be precise
no, more like every X minutes the machine will stop for a few seconds
if you count liters then sure. But I'd assume factorio uses m3 🤷♂️
eh, units
we use the same numbers of units afaik,
factorio does no division by 1000
im gonna use petrolium coke because i am bad at maths
after 50k minutes you’ll have 1 extra oil
assuming you’re feeding 50 oil a minute
uh, it's getting crude oil
whoops i read that wrong
that stuff is inefficient tho
fuel is better
and I was talking about the underclocked one, which won't have extra, but will not have enough
and needs no water
im going to sink it
ah i’m just completely failing to read, i need a nap
oh
off-topic but for the more dedicated followers - where do I find deep analysis of official videos?
looking back at this and comparing with real world fluid cars, it seems like factorio's fluid unit is indeed liter (25k liters is relatively normal volume for a fluid car).
well I'd assume they would be somewhere very deep
woah man
i will finshe gmaing pc tomorrow
yeah we use ridiculously large amounts
even if they used m3, we still use more units per sec/frame comparatively
factorio effectively uses 2/3s of a mk 1
and our machines need like way more than Factorio's, which makes the comparison a bit weird
but the transfer speed and pipe volume is way different
hm wait.... actually
transfer speed isnt the issue
its pipe volume. thats it
spees
Not sure what the point of that comparison is
But if you're including decimals, there's decimals in factorio pipes as well
no the comparison is we transport 12 times more stuff, thus the bigger issues
guys do any of you know how to calculate train throughput over a distance
you need:
- train loop time
- number of cars
- stack size of the item
then use it to a formula on wiki
oh great
well im yet to make any of the rail network so I'll have to get on that task now
So, is this wise? Looking to make recycled loops a bit cleaner.
Normally each refinery needs 30 fuel and 30 rubber/plastic. Noticed OC to 166.6667 makes the rato 50/50 and 100 output.
Now the debate if worth the extra power cost and shards to simplify the setups
tbh most of it comes down to "build a train path, put one belt into each platform and throw trains on it until they can keep up with production"
yeah that would be great if the cargo wasn't raw uranium
how does that change things?
I would use droned but I love my trainsbut
the ratio is the same no matter what the clock speed is
well i just don't want to make loads of filters to stop myself from dying when i test it
It is but easier esp with a 400 fuel pipe
although actually i guess i could just use containers full of iron ore instead
automate filters then 😛
well especially if you're lazy, automating stuff will save you from needing to craft them 😛
Been there done it 🙂
hahaha well look my tickets are getting me enough filters for the time being but i guess it would be cool to have an automated filter factory in the nuclear powerplant
spending tickets on stuff you can automate???
ikr im a madman
@sullen mulch greeny gave you the variables for the exact math, but if you need rough numbers the average train line people make has a throughput of 1200/min per car.
ah right, well i reckon i'll give the maths a try because I definitely don't want to deal with a radioactive train causing issues but ill jeep that in mind
🚙
For my Diluted Packaged Fuel production, i'm starting off with 585 oil/m
Then turning that into 780 Heavy Oil Residue/m with 19.5 Refineries
Then using that 780 HOR/m to turn it into 1560 fuel/m
And according to the wiki, each fuel gen takes 12m^3/m of fuel
so if i divide 1560 by 12 that gives me 130
So does that mean i need 130 generators, or is my calculation off
because it just seems a bit too much
looks correct
600 Oil = 133.3333- Gens on Diluted
So 585 being 130 is not "too much"
damn
There's a reason you rarely ever see a "small oil build"
that is a lot more than i expected
diluted fuel is great
How much did you expect?
maybe 20 lmao
??????
one of the reasons why turbofuel is often skipped directly to nuclear
HOR is broken as fuck because it INCREASES your oil conversion by a factor of 4/3
Diluted then DOUBLES your HOR amount into Fuel.
And you thought 585 was going to amount to 20?
idk didn't do enough research
just went by your recommendation a few days ago
i just found out about Diluted Packaged Fuel afew days ago
there's also diluted fuel (no packages, it's made in the blender)
diluted packaged fuel in refinery costs 30mj/item + 10mj/item for packaging the water + 10mj/item for unpackaging the fuel.
diluted fuel costs 45mj/item, so you save 5mj/item.
nice, but the first one can be done in tier 5 and the second one requires tier 7 which is a space elevator delivery apart.
i honestly would rather wait for tier 7 and try to get the blender recipe because building it is quicker and easier and takes less space. each "unit" of fuel gives 750mw in a fuel generator so 5mw difference is nothing you would even notice. which recipe you choose depends on how urgently you need/want fuel and how willing you are to hunt for HDDs.
kinda annoyed by you saying it gives MW, as it doesn't, it gives MJ 😛
I'm making a plastic factory about 1900 or more, using the spire coast oil
You'll need at least 2 nodes to accomplish that as 1 pure will net you only 1800.
looking at the other plastic recipes and I can't see a way to use all HOR up without dumping some to petro coke
Diluted fuel?
well that has its own issue
like what?
644 oil? Weird number?
He said "over 1900"
Ah
and thats following the 81 rule i assume?
Kinda weird but ok
1900 / 81 = 23.4567901-
81 * 24 = 1944
yup 
I can get closer to 1900 but the ask was "over" so I went with the simplest clean answer.
I think better question is why 1900
🤷♂️
Imagine not using Tools..
2,728 units/min of Plastic and 151 units/min of Rubber
2728 / 81 = 33.67901234567901-
No thank you.
Might as well make 3.6k and use left over for something else
I can sink the extra but thats the min I need
the 81 rule avoids recurring numbers in all machines in all oil setups?
I thought 1900 plastic was all I could get from spire coast so I went with that number first
The less clean math is that (2728 + 151) / 3 = 959.6666666666666666 Oil
Which is also just... bleh.
Recycled Loop
I mean just srtup two 1800 plastic loops and use your extras elsewhere
You can get 1800 from 1 Pure.
So you can get far more than that from the Spire Coast.
so I was going to try a different recipe instead of oil - plastic and HOR - coke - sink method
If you enjoy having less product per oil, sure.
Also if you're going to make that much Coke don't sink it.
Put it in Coal Gens and get free MW from it.
Issue is when you get to that level of plastic/rubber needs a recycled loop is the real only choice
but the idea I had of oil -plastic and HOR - Diluted fuel - to plastic meant the rubber made for that loop had its own HOR to remove
gonna try the recycled loop and change the 95 refineries I put down
so the 1800 loop is the best for even numbers of refineries?
No.
Its best for oil to end product ratio.
Recycled Loop is best for maximizing product per oil.
well I'm seeing the number 81 to get the clean clock speeds
Yes.
final output in multiples of 10.125
Tbh if you want OC the recycled refineries to 166.667%. Will get 100 plastic or rubber out of each instead of 60
60 * 1.66667 = 100.0002
100.0002 != 100
Well ingame its what it say soo 🤷♀️
Plus easier to work with 50 fuel and 50 rubber/plas input
What it say != what it do
Making what it say = what it do is the purpose of the 45-81 rule.
I get ya
is there a post that explains the math somwhere
The math for the 45-81 rule?
this reddit one I found doesnt help so much
reddit
5 votes and 5 comments so far on Reddit
Ouch...
Right in the feels.
What about it is unclear?
Given I wrote it.
@sonic wasp
where does 1800 or 3.6k come from
Like how do you turn 600 Oil into 1800 Product?
1800 divides into 10.125 as a uneven number
Aware.
which is why sev is saying to not use it
People who don't care about imprecision to 600 Oil to 1800 product.
the only reason 1800 was mentioned was because that was the max you could get out of 1 pure node, and therefore for 1900+ you need at least 2 nodes
If I am using 1 pure node I will do either 1782 product or at most 1792.125
So 3330 Oil total.
3330 * 3 = 9990
9990 / 81 = 123.33333
123 * 81 = 9963 product using 3321 oil.
only need around 3k total output
If you want to shave it closer:
9983.25 product using 3327.75 oil
ugh
3000 / 81 = 37.037-
37 * 81 = 2997 product using 999 oil
9989.57812 product using 3329.859373 oil
although thats probably cut a bit too close
but tldr any number of product that when divided by 81 gives a non recurring number is fine right @median heath?
When using the Recycled Loop, yes.
so 3159/39=81 so I'll go with that total
and I'll take a mk3 belt off of rubber in that loop for that rubber I need
all for 15 motors per minute
round up is good, round down is bad
Imprecision is bad.
https://satisfactory.fandom.com/wiki/AWESOME_Sink#Points_generated_per_item
and also see the recipe improvement ratios and farming points tips
Satisfactory Wiki
The AWESOME Sink is a special building that produces FICSIT Coupons for use in the AWESOME Shop by destroying items inserted into it, converting them into points based on their value or complexity, which in turn are used to print the aforementioned Coupons. Each successive Coupon requires more points to be printed.
The Sink can consume as many ...
Nah bro when I bought this game I didn't expect to need to do math 😢
i mean you don't need to 
True
you can try using a production line calculator if you need some help
I mostly just started so it's not horrible yet but it's getting there
Well if it helps - My method working up through the tiers is to make a factory production line for each basic part that has room to expand to deal with miners getting more ore out and then just linking belts together to form new production lines w/o worrying about math. Just let all the machines and containers fill up over time while you build and explore
I had a personal rule with gaming: If I need a spreadsheet to play I should play something else.
I broke that rule for this game
ive never used a spreadsheet for a single factory
help
oh mb
im running 16 fuel gens on 24 fuel p/m, im creating 400 p/m so i need a 17th gen to use the excess fuel right? i don't want any downtime on the refineries so i can't just let is sit as extra, any NASA scientists here know what i should clock my 17th gen to?
you have them overclocked to 24/min?
And clocked to 246.2288%?
mines at 246.2289% but it says it requires 24p/m in the chart thingy
246.2288 is more accurate
ight i will change that, but i still need a percent for the final machine right?
8 at the end is better, since it's slightly below 24, not slightly above 24
but tbh i would just use overflow for the last gen
the 17th should be clocked to ~145.351781 @twin inlet
thats a number
yup 
yeah the fuel gens seem to be in a bit of a mood
but space is my issue here so i had to
U7 fixes that thankfully
Space is never an issue, map is huge and verticality is a thing
thanks so much haha
whenever that drops
are you in a cave or what
kinda
Caveman 
for future reference this is the formula 
where operating rate is the desired percentage of the normal operating rate
also another question, if i was to just not set an overclock percent on the 17th machine and just set it to 250x, it would pull in too much fuel right?
no, because 100% < 145.3517%
so fuel would back up
250%
man maths is not my friend, how the heck do i enjoy this game
but if it was at the 2x clock speed like the others there would be overconsumption and the gens would just cut out for small periods of time every so often because there is an 8 fuel/min undersupply
just overclock the last one and let it turn off and on
its just a few weeks until U7 drops
then you can set the percentage precisely
👆 then gen clockspeeds will become linear
thank god
if i did that it would pull more fuel than i have right? my setup isn't load ballanced, i rely on everything slowly backfilling
it would, but why would that be an issue, nothing backs up
you can't load balance fuel gens anyway
but then the gens turn on and off
i feel like im missing something incredibly obvious
it doesnt matter if it turns off
it will fill and then turn back on
and then back off
and then on again
so what should i do, use the math mr tomato got me or just let it turn on and off
either should be fine
i would just let it turn on and off until U7 drops
it'll have to be redone at U7 either way
New 1800 plastic factory done, now to prime it and test!
I prime it by manually filling all the residual rubber with max resin and all the recycled loop hoppers with max plastic/rubber as well
Cool
Anyone have any technical info on the logic steps performed by trains at stations? Specifically around 'full load' state? I'm having an issue with trains getting stuck in permanent dock when using it. Seems that having some other items already in the wagon might be messing with it but I'm getting odd results in my testing.
Before any one asks, I've tried various and/or settings and times and those don't seem to be the culprit.
that logic is currently broken, theres an issue with multiple freight cars being ignored and other weird cas
Lame 🙁
Early Access game in a nutshell
Thanks for answering I guess. Has CSS aknowledged it at least?
Yeah, seems it gets reported every couple of months.
I was concerned one of the mods I put on was causing it, but it seems the bug itself is vanilla, though the mod may be causing it to occur more often.
Personally never seen the use for "full load" state. If someone could enlighten me?
train don't go BRRRR when train no need to
?
"Your train doesn't constantly move when there is no need to due to you needing items in low quantities and thus you can afford longer waiting times"
There's no real train logic in satisfactory in the first place though
May as well embrace it and just plan around your trains running on a timer
made a fuel tower to test bottom feeding....
surprisinglystable
then again, only using half the pipe capacity....
also modded pipe settings
Bleh, mods.
But when I do fuel towers I tend to sort of do this.
Pump it all the way to the top and then let it trickle down.
except its not that,
Just being pushed up, seems like.
I know, I can see the differences.
ye
so far, perfectly stable
which makes me happe, now i gotta turn off the pipe mod and see what happens xd
Lot simpler than my last setup, I'll give it that.
though im nowhere near capacity
it shouldn't break......
like this should work perfectly fine anywhere
I hear the qualifiers.
hmm.... seems to work so far
The 'nowhere near capacity' thing 😛 . Also I find more problem in unlooped systems if the machines are pulling in larger volumes per minute. Fuel gens have a comparitively low take
attempting to VIP junction.
Lack of pump make cause issues.
Does the bottom pipe, which is the recycled fluid, have a pump prior to the VIP junction?
just a valve
yeah, valve might cause problems I'd swap it for a pump
you NEED pumps for the VIP
☝️
fixing
iirc you need them before the junction, not after
That is before.
Flow direction is towards us.
That is before
then it's VOP? 🤔
No.
Recycled flows towards us.
Fresh comes in from the left.
You're looking at the back of the refineries.
The mk1 pumps are just not in frame
The three stack is the sloppy solution. The 4 stack behind is the scrap
ah then it's not just "flow towards us". Understood
hey so i'm making supercomputers and i noticed high-speed connecters and AI limiters are right next to each other. is it a good idea to use one drone to bring it to a place or should i use a train?
I use neither. Would build most components very close together or in same factory complex.
yeah but oil and quick wire aren't
both are valid methods of transport
trains are harder to setup since they require rail and signal infrastructure but once a network is established it can be re-used and expanded easily, trains also use electricity as fuel which can be useful since it is not a material item, and trains also are easier to transport larger amounts of items.
drones are quick to setup as the only infrastructure you need is two drone ports, however they consume batteries as fuel which can be a pain to set up and distribute, they also can't carry as much but are set up to be a quicker and simpler alternative
i was more saying same 1 drone or train
i don't wanna use 2 drones cause of batteries and it's tedious
If your gonna move stuff choose the lesser quantity. Like I wouldn't train quickwire due to volume
You might save yourself batteries if you set hte drones to only fetch more items when there's room to drop them off?
thoughts on this factory plan?
Seems fine? W/o context of location and logistics though not much I can comment on.
location is yet to be decided, but i'm going to use the central nitrogen well, so somewhere around there
alu will be trained in
and HMF and rubber either train or drone, most likely train
Well HMF can be made entirely with iron coal/coke - so you could probably move the hmf and rubber from the same location. With that in mind though if you use Electrode scrap then you could probably combine the oil production in one spot if you place it in a good spot
oh and limestone
ive already got HMFs, rubber and alu ingots produced at other locations
Ah fair.
I suppose you could always consider droning in the nitrogen and that would give you some flexibility on location?
but im not too limited on location since the area around the well has everything i really need - nitogen, water and copper, and with a mk3 miner i could get the 385 copper/min out of that south east normal node
Ah yeah, that's where I've got the foundations for world bauxite processing
only problem being i don't have mk3's unlocked yet
thats what this factory is for
oooohhh right. WELL. How many coupons do you have? I like either makign a super temp bauxite factory for unlocking mk3 ect or I buy the parts from the awesomeshop and unlock
Like aboslutely slap dash alum factory that I throw with existing production to smake a few things together and really get on with things
buying would be possible, i've got a few hundred coupons although i do find it kind of cheaty
I mean that's an entirely personal choice can't say youre right or wrong in deciding to use a game mechanic
but other than that the only other thing i wasn't sure about was whether or not i should make heatsinks, since they're only used for other processing not for the build gun
Though I'd either consider making a temp factory or build to be upgradable to handle mk5 belts and mk3 miners to save time in the future. But I liek reducing rebuild and retrofiting
Hmmm maybe make the heat sinks in a dedicated factory somewhere and ship it in? That way you don't have to trash it?
Just throwing ideas at the wall here - depends entirely how you like building and planning
and the other process that they're used for are cooling systems (made at this factory), plut rods - which would be a problem unless i use the pressure cube alt, and finally an RCU alt which i don't think i'll use
Maximizing plut rods CHEWS through alum yeah. Prob better ways to get points. Unless you want plutonium waste. Cause that's how you get plutonium waste 😛
have you seen the alt for the Fused Frame yet?
not planning on maximising plut rods
but will be necessary for a waste free nuke setup
base plut recipes are good for that
yea, and decided on the default, though its not too late to change
which means ill need to get heatsinks from somewhere, im thinking i can work that out later though and not produce them here since i think that will be easier
Yeah they made uranium waste recycling essentially need every resource on the map.
I think the only place to do waste recycling all locally is the east swamp?
i don't usually end up doing most things locally, actually maybe i do
i guess it depends on the factory - i made my starter factory locally, oil locally (duh) and alu ingots locally
Depends how you like doing things? Non locally means you're doing a lot of logistics planning. Locally means a lot of recipe and location planning. And you mix the levels of each
if you're interested I can pass along my resource pm by zone spread sheet once I revamp it
fair enough, i usually just do things that make logical sense for me, sometimes that means shipping in something else because it would take too much effort to produce on site, sometimes it means carefully picking a location so that i can use every node in the area
certainly sounds interesting
could be cool to have a look at
everyone's brains work diff! 😄
I spent 2 weeks making that sheet, then plotting 8 major hubs based on resources needed tweaking alts ect. I'm very proud my my max uranium rod factory that is almost entirely local production
Liquid fuel question:
I have 18 blenders making 800 turbo fuel/min powering 88 fuel generators that are overclocked. I have the generators set up in 8 vertical rows that are 11 long.
I'm going to split the blender outputs into two mk2 pipes, and pump the fuel up to the top left and right corners of the generators which will then flow down to each of the 8 lower levels, and then over to half of the generators in each row.
If I connect each row across horizontally, will that still work?
Mine isnt local exactly, but maybe regional.
Like my uranium cells and fuel rods are in the swamp
Bescons, crystal occilators and electromag fuel rods in the titan forest to the west of the swamp and supporting oil from blue crater
So many parts of that question hurt me...
teh variety of ways to do things is a thing i love about this game
Are they being fed vertically or horizontally? bit confusing
I setup shop in the hills north of the blue crater for aluminum and it sortof turned into a main hub
You need alum for so much it makes sense
Sorry if I explained it poorly, I'm pumping supply from the bottom to the top vertically, that will then flow vertically down to each row then horizontally to supply each generator.
right ok so - it could, technically work I think
Possible things to think about
Merging multiple floors of fluid manifolds can create a lot of issues with sloshing and back flow.
While this could work smoothly I would personally dedicate 2 or 3 blenders to a particular floor of generators and make a seperate manifold for each level
This will reduce sloshing and make trouble shooting and find flaws in the system much easier
My 2 cents
Fluids are finicky enough that my personal policy is make systems feed from Point A to Point B with no interactions with other systems whenever possible
I would actually break it down. Instead of one or 2 main out pipes split it even more but from the refineries them selves
Like have 200 or so in each set
So 3 blenders to make 100 fuel to supply each row of 8 generators in their own pipe network would probably work best if I understand
Exactly. You said turbo fuel right
yes turbo fuel
using the turbo blend fuel alt recipe if its important
... 'work best' is subjective - I would say it's the easiest to keep an eye on on and trouble shoot issues and best way to create stability
after a lot of pipe work I segregate every manifold. Always leave space for jesus in between them
Smaller segments even in a larger system usually is the best way to do it. Sortof modularity in a way
I only ever got this to work at about 93% efficiency xD learn from my mistakes
I have the turbo fuel production all set up, and I feel like I organized everything really well for my first attempt at it. Just working on the generator side of it now
I'd be careful on shard use for generators unless you're using a save editor to spawn more in. You're kinda locking those shards away forever until you build a new power structure
This doesnt exist anymore but my OLD turbo fuel base
Now replaced with nuclear in same location
New
I dont really use shards that much, but where would I get a save editor from? I've only had the game for 2 months and I'm still on my first run, not super familiar with mods and such for this game yet
theres a save editor on here https://satisfactory-calculator.com/en/interactive-map. on the right side of your screen there will be a button to upload a save to the editor. its super simple. i use it for power shards since theyre too limited. you can add them straight to your inventory with this
oh ive used that before, I didnt know you could just add shards like that, thanks!
Oh I'm sorry, I thought you said you had over clocked your generators
and there's a website that you can edit your saves in. Satisfactorycalculator.com . Though I would avoid using it for the first run or two if you like exploring and being surprised as it essentially spoils the whole map and location of everything for you
look at your miner and see how many coal per minute it produces, how much your belt can handle, and calculate how many generators you need to burn it
120 miner / 120 belt / 15 coal permin with no OC - so almost 8 generators per?
thanks
why only almost 8?
Mk2 miner, with non OC generators means 16 generators
Which is a pain in the butt with only Mk1 pipes
I think I needed 3 pipes
And it doesn't come out even I think
Note that you can set up miners supplied by pipes and belts in the middle, for two rows of 8
It helps if you pick up floor holes for pipes from awesome sink, and try to run pipes as much under the floor as possible
Which makes it easier to run your belts
Build at water and it's harly an issue
Feeding from above ias usually more reliable
I don't have problems feeding from below. But I have read that opinion a number of times
My coal generators seemed to have ran at 100%
I dislike feeding from above, I admit
I sometimes do it, but I don't like it. But I don't do it for coal generators
Can we take a sec to appreciate that the teaser box thing looks suspiciously like a blueprint bandbox
Even comes with 8m spaced lights
wth is a blueprint bandbox
if you have a band and the band owns a box and you make a blueprint of that box, you have a blueprint bandbox
slueprint sandsox
A bandbox is a type of theatre or stage. Bandbox, band box or band-box may also refer to:
Hi there ! I've been playing with trains for a bit of time now, and I have a question that remains unanswered : do we have any insight on the pathfinding method of trains ?
The more practical problem I have is : if there's a short path with a lot of congestion that results in a longer travel time than a long path with little congestion, which path will my train choose ?
they will always choose the shortest path
doesn't matter if it's blocked
Well it mostly means you have to have fairly strict logic paths for trains and provide space before stations dedicating as wait space if you ahve more than one train delivering/picking up at a particular station
Hence why a lot of people do 1 train per station
oh and it's why stations off to the side of main rail is useful (though station rail is calculated to be longer than it's actual length for the purposes of making it easier for trains to avoid passing through stations they don't use
Yeah it's just a very basic distance calculation with some bonus "fake distance" added if there's a station on a given piece of track
Reason for that being so that trains prefer to use track segments without stations
yeah I was already planning on having buffers at my train stations, but the reason I was asking is I'm building a refinery factory, and I have rails on multiple levels that are interconnected. Hence I wondered whether trains could choose the correct level depending on traffic
Well they'll choose the right level that their station is on?
If you want them to split levels you need to have stations on each level and specifically tell them to go there.
oooo yeah that perfectly solves my issue ! I'll simply put one station per level at the entrance of the factory
👍
do you think you can use algebra for satisfactory????
... in what way?
like basic algebra is just looking at addition and subtraction in different ways
but if you're doing multiple inputs you can jsut go something like if you have 200 parts and each machine uses 30 go 200/30 right?
i had a problem with a factory that took the some part in two machines. but i had only had that much of that part. i didn't know how to split it.
can you say that more clearly pls?
and manifolds are the easiest way to 'split' things https://satisfactory.fandom.com/wiki/Manifold in general
Satisfactory Wiki
Manifold, a.k.a. in-line splitting / merging refers to a type of building style where splitters or mergers are aligned in series (that is, one after another), usually parallel to the arrangement of buildings. This allows for compact building space and easier expansion.
It is the opposite fill method to the balancer.
English is not my first languid
thats ok, try that web page.
Your first language doesn't happen to be french is it?
worth a shot, it's my first language
i dont know French
Most people won't judge dyslexia tbh, hell my spelling is terrible and english is my first language 🙂
I'll scroll up and see if I can be of assistance.
as well as that web page I linked to yo uthis video is helpful https://www.youtube.com/watch?v=HwguwhizdlE
Hey guys and welcome back to another guide and this time we're going back to the basics in this manifold guide, the reason for this is that with the new layout series, I feel this will be a good edition so that new players can come to here to see how manifolds are built rather than having to explain it in every video.
In this video we cover wha...
ty
ok
If moving radioactive items , you'll be better served with load balancers as shown here: https://www.youtube.com/watch?v=G5d4foufHGE&list=PLQo-FPWWaZE-jn38NUxfxh5S2VyJNjnO3 , it is an entire playlist guide of different numbers of outputs.
Satisfactory 1 to 2 splitter | Tutorial Ep 1
✅ Merch: https://teespring.com/stores/random-gamer
🕹️ My Setup: https://kit.co/Random_Gamer/gaming-editing-work
This tutorial is about splitting one conveyor into two conveyors.
We need one splitter for this setup. Two of them while stacking of course.
🎬 Most recent videos: https://www.youtube.com/...
I've used this playlist as my go-to for specific applications 🙂
i have 700 in satisfactory so i can do input and output but ty
i was asking if i could use algebra in satisfactory
X and all that
1:2 splitter tutorial? 🤔
wtf is that
It's a playlist, not just one.
isn't that just "build a splitter"?
why would you need a video tutorial to 1:2 split lol
The 1-5 I used as a floor plan for my 1-100.
Making mistakes or missing something is not being dumb.
Thank you, I felt like I was gonna ask the obvious 
Yeah trying to ignore the fact I said at the start it was a damned playlist is just meh.
no matter if it's a playlist or not... there's still a video with 1:2 splitter
Not my fault or problem 🤷♂️
well I'm not saying it is. I'm just asking why do we need a video for 1:2 splitter 🤷♂️
@wicked tinsel not the best but definitely workable
might be possible to stack vertically tho
For that, best bet is to ask the creator, I merely saved the link to the start of the playlist.
I didn't ask you tho 🤷♂️ I was just writing here
Though I have encountered some people who probably need that 1-2 guide 🤦♂️ because some folks have been that dense.
possibly
now
unless you wanna fit 2 manus in one blueprint
you can put two of those with little shift
those are mirrored, but i guess you can make two blueprints
just gotta take this design and make it split-able
its a bit more work but really, switching out blueprints isnt that much of a turnoff, personally
and its easy too
3 refineries fit too
yeah
the problem is, they dont fit into double manifold, and neither should assemblers
so you can only really build one side and then need to connect other side manually or give up on it
what im curious tho
Attempting to make tiny blueprints?
is the blueprint maker surface snappable or not
from video it looked unsnappable, but shall see in practice
gonna be pain if there is no snapping 😄
might be toggle-able
each blueprint is offset by 10 cm
ok so for blenders and manufacturers, its gonna be a stack design if you wanna fit 2 into one blueprint
else you gotta make a modular design for one
i guess its ok
Why would you bother doing that though? You could just place the blueprint twice
btw, i hate when refineries are built like this due to their awful clipping :<
Eh takong a look not a total fan. Knew would be restrictive but doesnt really work with my design philosophy
Unless they can be snapped together modularially
Just because the bounding box is 32x32x32 doesn't mean all your designs need to be 32x32x32
on the bright side: this fits? and can be connected
knowing coding quality of this game, i fully expect there will be no validation in the blueprint file loader and it will just allow bigger blueprints if file is manually edited 😄
then its 2 refineries per blueprint
depending on if people figure out how to edit bigger ones... possibly
for assembler manifolds, it's gonna be far more valuable to blueprint the input side than the output side. I would just just BP a single row of assemblers so I could work that part in
Tbh I hope they increase the size of it, 32^3 is so restrictive that you can't even get a double manifold going without practically choking splitters down the constructors
in brighter news: it makes foundation platform building easier
just click 4 x 4 platform
I wonder if it allows snapping to world grid
i hope there's a way to mirror manifolds for this
unlikely, would mean mirroring buildings
true
you likely have to make "left" and "right" designs
fine for symmetrics but offset buildings (like foundries) would be an ass
so halves
depends how snapping and repeating will work with those
shouldnt it all snap fine?
not really
they do have grid alignment on foundations
We'll find out in experimental next week.
you likely need a 4 x 4 platform that fits it
like if you point on side of existing foundation with that blueprint
maybe there will be some way to mark something as reference point
that would be the easiest solution
like you mark one foundation as reference and then when building blueprint, it behaves as if you were building that foundation (and then spawns rest of blueprint once clicked)
Should be able to just rotate it, and yea, all questions that will be figured out in experimental.
and once again: if its not immediately as expected, QA site exists
I kinda doubt it's smart enough to know how to snap foundations
this is proof they read feedback on there
Sounds like coding hell
from the looks of the video it sort of is and sort of isn't
the video has no snapping at all
it snaps to the guideline
We finally reveal what the new machine coming in Update 7 is used for! Hope you'll find it useful!
⬇ Overclocking Info ⬇
https://satisfactory.fandom.com/wiki/Clock_speed
⬇ How to switch to Experimental Branch ⬇
https://steamcommunity.com/app/526870/discussions/0/5595176692476878404/
⬇ Links and Info ⬇
Website + FAQ ‣ https://satisfactorygame...
It has all snapping
You can see the blueprint stagger as his mouse moves
with the buildings you see the 1m snapping as he moves them along the foundations
and the 4x4 foundation grids snap to each other as well
where a 4x4 foundation block snaps to a 2x2 (say) foundation setup remains to be seen
https://youtu.be/9KBQyjy-a6g?t=338 this part is weird, it seem like mergers/splitters dont respect the snapping
We finally reveal what the new machine coming in Update 7 is used for! Hope you'll find it useful!
⬇ Overclocking Info ⬇
https://satisfactory.fandom.com/wiki/Clock_speed
⬇ How to switch to Experimental Branch ⬇
https://steamcommunity.com/app/526870/discussions/0/5595176692476878404/
⬇ Links and Info ⬇
Website + FAQ ‣ https://satisfactorygame...
but buildings earlier do, maybe bug
either way, i really like this, will be really useful, particularly for decorations/architecture
repeating stuff is super pain when building good looking stuff
Oh you meant placing on the pad itself
That's what I was thinking too, though Cobalt said that most of his stuff are on a scale too big to use blueprints for.
i did notice the mergers/splitters were acting like they were placed on bare ground, instead of how they would act on a foundation
Yeah I 100% guarantee someone will make a 3rd party tool to edit the values of blueprint data to get them all perfect
Just like SCIM lets you change the positions of stuff in your save files
yeah
i expect the first tool community will make is one that makes blueprints from existing builds
second (after the one that breaks the size limit)
I do hope there's no size checking
I don't expect to blueprint entire factories, but that's so smol you can barely blueprint constructor manifolds
problem is how big do you go
I find it a reasonable compromise between "blueprint entire factory and take the fun out of the game" and "no blueprints" 🤷♂️
i think 6x6 should be able to fit the two row refinery build 🤔
i guess the two row refinery blueprint is what i would consider "reasonable" size
especially since those are super pain to build without clipping
and pipes love to bug when building in constrained space
all I would ever want blueprints for are the repetitive tasks. I still want to fully design factories myself and integrate them into the space
but I like that whenever I find myself doing the same clicks over and over, I can take it to the Cube
I think an extra 1 single foundation size would make a world of difference
But yeah it's better than nothing
yeah, its better than nothing
They said they wanted to figure out a solution for if your blueprint has 1 minor error
I really hope that's an undo button lol
factorio had unto button for years now
its really shame there isnt one here
especially for deconstruction
Factorio knows people are really just here for interactive spreadsheets
I would rather retain the ability to alter/fix things within the blueprint after placing, than have it be all-or-nothing where I can only delete the whole thing
and my understanding is that it's currently the former
I think "minor error" here was like, building it slightly off from where you intended
Where there's no error in the blueprint itself, but you'd still have to deconstruct it all to get the position right
i really would like to blueprint this sort of construction, i made thousands of those
and they are like worst thing to build due to how game handles splitter building
but this is too big for that blueprint planner
wouldn't one half of this work too?
if you could turn it
hm wait...
Not really? Unless there's mirroring, all your splitters/mergers would be the wrong way
^
also, you still need to build all the lifts/ belts
since they dont auto connect
Yeah the more I think about it, the more I think 5x5 would be just perfect
btw, since im here, im going to do some small self promotion https://questions.satisfactorygame.com/post/6329fb4bca608e080351f150 :x
pls upvote
if you can BP the floating splitters and main belts in place, and then BP the assemblers separately, the lifts should be trivial if you've lined it all up right
The non-painful way to do that:
yeah, that only works if you keep your staff flowing 😄
Why would items stop flowing?
and you probably cant build that either, since that probably wouldnt fit anyway
Yeah you don't need to keep stuff flowing, just clip all the belts through each other
6x6 could fit a Nuclear Power Generator 
for record
this is how assemblers look like on 4x4 surface
even with Sevrahn's way, there is not enough space for output
you know what else doesnt fit into 4x4 surface? train intersections
and i sure as heck would love to blueprint those
That feels like SCIM blueprint territory tbh
ffs :<
Refineries could definetly use more than a 4x4 for nice blueprinting of them
My hatred for blueprints aside:
I'd be fine with this because I don't always know which way I want the belting to go.
So just dropping the buildings without the belts would be my primary method given if you want to change belts you have to delete them all anyway.
How does one even hate on blueprinting
Listen to Fucktorio people bitch about it for 4 years...
The hatred grows over time.
i had been playing this game for years, the worst part is placing rows of buildins and then making sure the spliters, belts, pipes & lifts are properly connected
it never works properly
Scoffs 
https://questions.satisfactorygame.com/post/61e228eb831c852052368c17 i have reported this before coz thats how annoying it is
you can fit 2 refineries in easily
just not facing each other
but connecting one half to a pre existing manifold line is a lot easier than building it all by hand
btw.... how are your vehicles themself holding up?
2 refineries and...?
well, they dont really work, coz those spikes are everywhere
and piping + belting
hadnt played U6 at all coz of how bad it is
i guess the main thing you can do with refineries is pre-attach the junctions/merger/splitter on each side. then you can easily tile and connect
there is no way to fix it either
i think i will start those blueprints by making bunch of building+splitter+merger combos
@oblique hollow random: Does a train station fit on the blueprint platform? 
It'll make a lot of pointless busywork like roads and rail pillars much easier at least
U5 station would, U6 station is bugged so probably not?
No train stations are 5x2x5
iirc
Blueprint for Truck Stops 😉
nope
there would be no point anyway, even if it would fit
I wish I could blueprint a assembler+manufacturer combo instead of a single manufacturer
stack it
4x4x4 fits 2 manus stacked on top of each other + floor inbetween
plus belting
You just know that code wise these are all going to only care that a building's centrepoint is within the 32x32x32 grid
I know. Not what I meant
Hell with any luck it won't even check for that
i doubt loaded blueprints will do any checking, its way beyond what this game usually delivers
and its also kinda the philosphy as always
"if you really wanna do it, sure, go ahead"
wanna take a dump on performance? yeah, load that 50x50x50 blueprint
heck i'd be surprised if the game didnt just crash from large blueprints and that being the limiter on their size
"nope, too big, system doesnt support it"
probably more an entity counter limit
With the way Snutt described them I believe the intention is not geared towards T5+ buildings.
Like the model is for building dozens of the single-input/output builds (Constructors/Smelters)
the other prolbme is, my save isnt very big but has absolutely abysmal performance nowadys, so its no longer really playable
i had sent save to Ben to check it, but it kinda went nowhere
Try blueprinting your whole factory and throwing it into a fresh save file
High sign density, by chance?
...i assume from vehicles? xd
But that's bigger than 4x4
no clue from what, i would need Ben or someone other to tell me
probably from architecture, there is a lot of pillars making structures, but eh
its nowhere near the way that should cause such performance drop
from what i heard, Ben's latest work improved belt performance by 30%
or rather, will improve
Architecture should just impact RAM usage. You're not out of that, right?
and some entity optimization
ofc not
Why do you seem to assume I should know how much RAM your savefile needs to run? 
game itself allocates 8gb ram when running
As far as I know, the game can easily demand >10 GB or RAM with savefile bigger than ~5 MB
@frosty owl about the 5x5 thing:
I wonder if they intentionally chose 4x4 because they know what people can do in 5x5 and that was the intention of they don't want the feature to be able to do that much.
I guess I should've picked up the challenge so that I could've shown why they should've made it 5x5 for starters 
its not completely unplayable, it runs at like 30-40 fps when in multiplayer
50-60
oh so with another player its worse
tho the whole combo of U6 changes, that delay, pile of bugs that keep showing up somewhat soured the experience
50-60 is unplayable?
the problem is, its not 30-40 fps constantly
there is some shit going on in render thread, so it randomly drops a lot
and that is annoying af
..lights?
like you rotate the camera and it freezes
might be lights, but im not too sure about it
check your lights setting
we dont have too many lights, also they have pathetic range anway
:
also, check particle effects
might be ultra 😄
Step 1 of getting better FPS: drop video settings
changes nothing looks-wise and runs worse
the thing is, its slow that way no matter where i am
Step 2: pray the difference is noticeable 
probably the factory itself is what lags
is it a mega factory or what
we did some reserach on that coz for my friend its even worse
not really mega
he doesnt play on ultra lights 😄
also check the VFX setting please
theres no notable difference (from what i could tell) visually between high and ultra
but ultra also performs a bit worse
i will try it later
there is also yet another bug with railroad i see
some train crashed, which should not be possible without bugs
ehh, maybe it will get better one day
I'm inclined to think a lot of signs in one place can cause weird (possibly similar) performance issues, even when far away.
Maybe something about how much info-per-pixel there can be when having a lot of them, with all their individual colours, texts, animations, reflections...
As a sidenote, imo signs are still in an early-mid stage, optimization is to be expected
so far the probability for all that is yes%
every issue ive heard or seen, at some point, is resolved
just not at the same time
and i think once content addition is done is when a lot more polish is done
had they fixed the lift off bug? 😅
the what
if you are using the flying thingy (hoverpack?), it tends to get stuck in going up form
some combination of flying & opening menus cause it
I think they fixed the bug with custom colored lifts loading black, recently.
Not sure if it wasn't just my extreme lag playing tricks
hold any key and open some menu > key gets stuck
ah that yeah, seen that before
there is also some obscure bug where parts of factory had cursed zones that refused to work with multiplayer
i never got down to what the heck causes that
there are zones where clients cant build belts/lifts
heard of that before, but if its cryptic to us is cryptic to the devs too
As someone who colored all of their beltwork and all with custom, that bug annoyed me a bit 
its 100% reproducible
how
there is just zone in my factory where clients cant build stuff, it just doesnt work
i never got down to what the heck causes that
server guy needs to go there and build from his pov
if you cant find out the cause how is it reproducible
there are few such zones, but its 100% reproducible in them
but no clue what caused the zone to be created
in those zones, but how are those zones made
like... if you heard "yo in mp sometimes some zones become cursed"
and you are there just "..... how the hell did that zone appear"
I'd have no clue as a dev what to do there tbh
load the save, go to the cursed zone, check wtf is happening there?
then figure out how it could had gotten into bad state
there is probably too many of something and it bugs on some artificial limit? dunno
well... can the save be loaded by others and does it exist in SP?
it doesnt exist in SP, its purely multiplayer bug, it only happens for client
there is probably some overflow on network layer and it refuses to sync stuff
so now the chance that, if i load the save, tell some other dude in my office "yo play with me" and i do not run across it, then like....
non the wiser
well, sure, it can always fail to reproduce for someone else
then you need a lot more test cases but you can't know how long it would take to encounter
but you know, ultimately, pile of such minor/medium issues adds up mentally
to the point where we were like "meh, its unplayable, fuck it" and hdnt really played it since little before U6
i know, but nailing this stuff down is also not easy
it took a lot of testing for US to find that dumb fluid loading bug
and then when we nailed it down did it finally get fixed
but that took multiple hours of tests
i think such multiplayer replication bugs will get pushed to the end of 1.0
Without even telling us, mind you 
as those are, frankly, among the harder parts to test
personally i would also push those back as long as possible
because "i got other issues i can more easily find and fix"
Do you think we could increase the chances of them fixing the throughput issues if we get more data on it? 
i mean its all pretty solid now
junctions and mk 2 backflow into junction
and mk 2 wanting to flow when full
Yeah, we could just crunch more precise numbers and make sure wether welded belts are immune to loss or not
a pure mk 2 pipe with no junctions has like... no isues moving 600/min
oh belts
eh, applies to both
Yeah, though I only test belts the logic could be extended to pipes somehow, depending on coding similarities
from what i can tell theres 0 similarity
inb4 Sev comes with "don't think of pipes in terms of belts"
Don't think about pipes in terms of belts.
just that outside the behaviour is somehwhat similar
YES
@wind spade 👋
i doubt pipes and belts share any code
they are coded entirely differently
the problems with them are probably two different bugs
ive been into modding and learned some more about pipes
the true inner flow math is still a mystery, but the outer layers are clearer to me now
belts are also somehow tied to FPS on server/single player
how wouldn't they be