math-and-meta

[true junco]

The issue with feeding up is that when the machine takes in fluid, it does so periodically only as the fluid is removed from the machine input buffer. So when this happens for the whole line, every vertical pipe starts to empty, and had to fill back up to be able to flow out again. So, its almost like simulating some sort of cavitation.

[proud lion]

Yea my coal source is pretty far away so it was a way of meeting in the middle

[deft lichen]

you CAN find coal nodes near water

just don't assume the closest nodes are best

[prisma kraken]

also, when running pipes upward, i really prefer running them vertically than up a ramp, its easier to know exactly where you need pumps

[deft lichen]

power generators can be remote, just run a power line back

[primal flicker]

Use that coal for steel, instead.

[true junco]

Well. You want coal and water next to each other for "Pure Solid Steel" set ups... but if you arent doing those, then yeah.

[primal flicker]

Or you can count 4m ramps to see where every 20 or 50 meters is.

[proud lion]

Yea the closest node is the one im using rn

[primal flicker]

Find coal that is near water. There are two clusters near enough to the grass fields start.

[proud lion]

Ill try just putting the generators on the ground for now instead of having them high up

[proud lion]

That is the closest node

[prisma kraken]

yeah, its not a right or wrong, its just simpler to figure out

[proud lion]

Let me go to my coal node and see if i can find water near it rq

[primal flicker]

Near Water >>> Closest

[deft lichen]

you have a map

[proud lion]

I forgor

[primal flicker]

V ping coal, then open map.

[true junco]

Share what you have revealed so far on your map. Probably best to turn off most map markers etc.

[frosty owl]

I think that unless specified, "double manifold" should refer to the humber of manifolds involved. Double-sided for the other one

[deft lichen]

if so, I'll have to update the diagram

thanks!

[gritty kite]

how would i go about spltiting 145 plastic incoming to 45, 40, 30 and 30?

[wind spade]

manifold

[gritty kite]

is that when u just keep splitting it and wait til they all fill

[wind spade]

yeah

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

[gritty kite]

alr i didnt want to do that but seems like the only viable method

does it matter the mk of the conveyors for that?

[wind spade]

it's the easiest one, there are other methods but this one just works everytime

[gritty kite]

as in, say you have a mk2 split into a mk2 and a mk1, will more ores go into the mk 2 side?

[wind spade]

mk doesn't matter as long as it's enough for the item amount needed

but personally I recommend to build it all from highest mk available, as it makes it fill a bit faster

but it's not really an issue to use smaller mks

[gritty kite]

ok

[wind spade]

as long as you don't exceed belt capacity in any section

[gritty kite]

ty

[sacred orbit]

I see using MK speed restrictions as possibly helpful. I say possibly since I haven’t tested it, but in theory if you’re using a MK X belt and splitting off with lower MK belts, then the MK X belt would fill faster as it’s bottlenecked at every split.

Now, the slower belts splitting off would obviously fill slower, but the belt supplying all of them would be full and able to feed another unit in asap

[wind spade]

it actually has the opposite effect - because it's bottlenecked every split, more machines start working, so you're processing more materials, so less materials are actually being backed up

[sacred orbit]

Huh

[wind spade]

this is one example situation, one uses mk1 side belts, one uses mk5 side belts. The graph shows how much it is produced. You can see the red line reaches the top capacity earlier (only a few minutes earlier, but it does)

[prisma kraken]

what's the integral of that look like?

[wind spade]

no idea, this isn't formula, this is simulation

[prisma kraken]

sum up the descrete data points and graph the sum

[wind spade]

I assume you want to know which way produces more items during the filling?

I don't have the data anymore, this is an image I shared a few years back

I'd have to write the simulation again

[prisma kraken]

i want to know the delta of number of items produced over the warm-up time

[wind spade]

well see above 🙂

[prisma kraken]

really i want to see the integral of it graphed 😛 quantity instead of rate

do you have that simulation code somewhere?

[wind spade]

🙂

unfortunately not

[prisma kraken]

you make me cry 😢

[wind spade]

I wanted to rewrite it anyway and make it part of tools, so I didn't back it up

[prisma kraken]

it would honestly be useful to have a simulator for arbitrary belt topologies

[wind spade]

the simulator was intended to be manifold-only

since loops and stuff like that are really hard to simulate

[prisma kraken]

yeah, i know, if it can be simulated from a definition that can be defined quicker than you can build something in the game though, it has use

a year+ ago, i had a manifold that i created to burst the ADS delivery for the elevator, and found the entire concept sort of ridiculous, the amount of high-value product i had sitting in machine buffers was a crime 😛

[ashen stirrup]

Is it more power efficient to run 2x smelters at 0.75 speed each, or 1 smelter at 1x and another at 0.5?

[static zenith]

is power draw wonky like that ? o.o

[wind spade]

iirc the first one is more efficient

[static zenith]

in the real world i would assume that 75% is gonna still draw near 100% pow er to produce less tthings.

so id go with one full power and one half power.

but in satisfactory theory thats another can of worms ..

[wind spade]

but you're saving... 0.12 MW

[ashen stirrup]

In the case of 2 smelters, sure... but distributing the general strategy across all production lines everywhere, maybe it adds up to enough to run another light or two 😂

[ashen stirrup]

Yeah, overclocking/underclocking production machines results in exponential changes to power consumption

[static zenith]

thats good to know

[sacred orbit]

That DOES in fact mean more machines at lower rates saves power overall. Even underclocking saves power

[static zenith]

i was thinking that .. just plan everything at .5

i dont mind building massive things ;]

[spice egret]

That’s not a bad idea either for early factories because if everything is at 50% it can quickly be doubled without building any new machines

I’m getting a little better at leaving myself space and just building more platforms but at first I had a bad habit of trying to cram stuff into tight spaces so I didn’t leave myself much room

[vapid gorge]

it doesn't save that much power because it means that you need 2x as many machines.

it's like if you fully overclock to 250% - overall it only uses 33% more power for the same amount of product.

also if you're building with power limitations in mind you're doing power wrong - theres easily more power than you could ever use on the map

[sacred orbit]

I’m planning my power around being toggleable in case I want to upgrade it without trying to power the whole factory at the same time

[vapid gorge]

that's just using a switch?

[gritty kite]

or even just deleting like one power line lol

[sacred orbit]

Yes

That means all power generation has to run through one point

[vapid gorge]

pretty normal. unless youre revamping oil to a different process? in general you can jsut build more of a power plant to the side and connect it up

[sacred orbit]

I’m just now getting to oil

[vapid gorge]

like if you want to double your coal power when you get a new miner or belt you just build more rows next to the current one and link them up, no fuss

[heady vine]

what's the reason for making screws as byproduct.
out of all possibilities...

[vapid gorge]

is that tools? you must have had some odd settings

[mystic moon]

It's probably the option that consumes the least power

And/or is the least complex

[vapid gorge]

just odd that it used the whole rod input rather than just the 2 the recipe needed

[fading swan]

Is there anywhere on the map thats ideal for heavy frames?

[snow dove]

no

[vapid gorge]

choose which recipes you want to use, how much of each resource, and look for a spot that has them. If 'ideal' is less importing

[heady vine]

It is

[heady vine]

not sure such option exists

[fading swan]

On top of modular frames, what else do I need to start mass producing? I just finished space elevator stage 3 and I'm kinda overwhelmed by everything, and don't know where to start.

[vapid gorge]

just go step by step - What I do is I tend to just mash lines of components together and let them fill containers. It's not 'efficien't but gets the job done and you end up with lots of stored parts. Good for early on and going up the tiers

this way most of the factory tends to go idle as buffers get full

[fading swan]

I'll start with heavy frames then make my way over to computers.

Actually im gonna finish my road network first

Or at least the main highway

[sacred orbit]

this should work as a barebones plastic/rubber factory, correct? 60 crude into the refineries, HOR into Coke into a sink, excess plastic and rubber merged into the sink too.

[vapid gorge]

in general I don't really recomend roads, at least early on. Takes a lot of time and often limited use. But if you like makign roads go for it

[vapid gorge]

sure

[fading swan]

If you have enough rubber, you can make circuit boards with coke and rubber

[fading swan]

Are they less efficient later on?

I could always set up trains

[sacred orbit]

only making 40 rubber/min atm

[fading swan]

ah

[vapid gorge]

it's more that your early factories are often not what you want/need later on and will every move or completely revamp them, or you'll want different logistics. Puting in a lot of early effort might be a time sink that you realise is near useless

[fading swan]

well almost all my factories are mk 4 logistics

Im at the point where the rest of my factories will probably be permanent ones

My motor factory is

Same with steel

Im gonna start my heavy frames, thanks for the help

Is there an amount per minute I should aim for with heavy frames?

[vapid gorge]

I mean if you haven't gotten a very final plan laid out you're working towards I wouldn't bet on it. You might change your mind at what your 'goal' is, or decide you lke doing logistics differently

happens a lot

[cinder silo]

I personally only produce 11.25 heavy frames per minute, 6 of those go to making fused frames later on.

[prisma kraken]

it depends on what rate you want to complete the phase 4 delivery, whatever your target rate is for making ADS's, you want to make 2-3x that rate in HMF's/min

i think bare minimum is probably 4.5/min, and more comfy is in the low double-digits (11.25, 22.5, 45)

i feel like it is useful to be making enough that you can build trains out w/o having to wait for more hmf's to get made, and that'll be a rate somewhere around 10-20/min

something else i'll just quickly toss out... if you carry pipes and rods with you when hd hunting, you can boil down a lot of materials at the crash sites into hmf's pretty quickly, and it is a nice supplement to whatever you are producing as you start to really need them for manufacturers and trains

[thick plank]

I wouldnt say thats a bad habit. Some of the most beatifull production lines I have are quite crammed. As long as you can keep it organized tiny productions are great (especially when using blueprints)

[fading swan]

Alright. I have a ton set aside from a smaller factory I made (about 4 per min), but i’ll aim for 15-20 per min.

[fading swan]

Actually I will decide how much I make based on the resources I have at hand.

[wind spade]

should do it the other way around 🙂

decide how much you want/need and then find a place with those resources nearby

[primal flicker]

Is it just me, or do a lot of recipes ratios benefit (with simpler numbers) from running machines and extractors at 90%?
Currently planning future fuel power infrastructure to support nuclear development, and turbo blend supplies a nice even number of generators with multiples of 27 crude being used, i.e. clocking oil extraction at 225% (or 90% of the maximum).

[spice egret]

I guess my issue is Im sometimes just cramming machines into a space they don’t really fit well in instead of just making the foundation slightly bigger so they can be organized at least

[deft lichen]

in a way, cramming is a skill

[primal flicker]

I am trying to learn a both/and approach.
Creative use of clipping that doesn't look cursed, but allows machines and logistics to pack more densely while maintaining a semblance of orderliness.

I have a bad habit of not leaving room between groups of machines as an aisle of sorts.

[frosty owl]

Cramming well is art

And it also goes well with sushi...
https://cdn.discordapp.com/attachments/553550313533997057/1036554475466391592/RDT_20221031_0837512722569835602576370.jpg?ex=65927515&is=65800015&hm=6d102a7f400c06a239b32e51f152b314105c44a5aa94399217858832fdff86d6&

[primal flicker]

So I'm looking at repetition vs complexity:
Would you rather build 60 GW fuel power capacity, or 60GW nuclear power capacity?

(With plutonium sinking)

[wind spade]

If I have power for it? Definitely nuclear, no reason to build worse fuel if I have access to nuclear

[primal flicker]

I'm toying with the idea, now, of skipping fuel altogether and going from coal to nuke.
Seems fun.

Setting up hundreds of fuel gens is tedious.

[wind spade]

and setting up double of that coal gens is even worse

[primal flicker]

Eh... Coal gens pack in very neatly. A bank of 16 occupied 8x9 foundations. And their build materials are less cumbersome to produce.

[wind spade]

need same piping but also belts

and it's double the amount for same power

[primal flicker]

I have a good pattern for the logistics. That doesn't bother me.

[wind spade]

then fuel gens should be easier 🤔

[primal flicker]

It's the HMF, computer, and motor automation hurdles, honestly. I can get parts for coal power much faster.

[wind spade]

tbh if not having enough materials is an "issue", then automate more materials 😛

[true junco]

Well. Fuel gens require processing crude into fuel (usually thru intermediaries) and disposal of byproducts, so you cannot discount that part of the footprint. But still. Fuel should take less space per MW than Coal.

[primal flicker]

Well obviously.
The problem is that I haven't automated ANY of the above three parts yet, in this save.

[wind spade]

you'll need them anyway, so I don't see a reason to not do them 🤷‍♂️

[primal flicker]

I'll do them later. But I still want to see how feasible it is to skip from coal to nuclear.

[wind spade]

sure, just saying my opinion 🤷‍♂️

[prisma kraken]

yeah, there's a lot of recipes that way.... ai limiters is another one

[prisma kraken]

it is quite doable, i've done so and seen others do so, you just have to be very smart about keeping your production power-efficient on the way there, biggest thing is to avoid pure recipes and save the refineries for what absolutely need them

the thing that's sort of ugly about skipping fuel is that you can fall into the trap of everything being a temporary build 'until you get to nuclear...' and you can end up with a bit of a speghetti mess

i think the worst mess that you can create with that attitude is in getting aluminum production going, there's a lot of stuff that goes into that production chain and the building of it just kind of ends up sprawling if you don't have a pretty good plan for it and treat it as a bunch of ad-hoc stuff

when i've done so in the past, i've gotten nuclear up and running and looked back on all the stuff i needed to rebuild and just kind of decided to restart rather than fix my sins 😛

[primal flicker]

Good to know! I plan to have the minimum aluminum automation necessary to get started with nuclear, but I do want it to be at least pseudo-organized.

[prisma kraken]

you always have to kind of do a aluminum-lite build to get it all up and running & get mk5 belts unlocked, but when that build then turns into casings+heatsinks+cooling systems+fmf's, it gets to be a mess, lol

[wind spade]

wdym by aluminum-lite?

[primal flicker]

Alu-minimum

[prisma kraken]

one thing that is worth considering if you're going this direction, is get a small default battery setup going just to get drones

[prisma kraken]

you can't do a proper aluminum build with mk4 belts, it needs a full 600 from a miner to build aluminum right, so you always need to do a bootstrap build with 480/min aka 'aluminum lite'

[primal flicker]

Yeah, I'll want to drone radioactive materials at minimum.

[wind spade]

it needs a full 600 from a miner to build aluminum right
why?

[primal flicker]

If 600 is an optimal number, then 300 should be just as optimal and half the size.

[wind spade]

^

you can clock machines to anything you want, so I don't understand why would I need 600 to begin with

[prisma kraken]

aluminum just likes being processed in multiples of 300... both the ingot making and things taking the ingots sort of like 300 input

[wind spade]

then clock everything at 160% and you need 480 instead of 300 🙂

[prisma kraken]

and yeah, you can always do a baux 600 build and clock it down

[wind spade]

you can do a baux 480 build 🤷‍♂️

and then later if you want, add a splitter after the miner to process the remaining 120

[true junco]

300 baux is really easy with sloppy electrode. Fits nicely into a BP at least.

[prisma kraken]

sure, just fit an extra couple of refineries into a small aluminum build, no problem just tacking those on the side, right?

that's what i mean by ending up with a sprawling mess 😛

[wind spade]

space is infinite

(almost)

[prisma kraken]

so is spaghetti

doesn't mean its desireable

[wind spade]

it's not spaghetti if it's two factories connected to one miner lol

especially since most of the time you can just plop down another module next to the existing ones

[primal flicker]

Sloppy electrode pure
That's my plan

[prisma kraken]

the other thing is its easy to get maybe 2/3 of those recipes when you unlock alum, all 3 takes a little grinding on the hd research

[primal flicker]

And if it's 480+120 then it might as well 4x120 + 120...

[prisma kraken]

usually i opt for sloppy and pure as the priority, knowing i can use coal at first

[primal flicker]

I'm not worried about that.

[prisma kraken]

i always get jammed up a little bit on waiting for the resaerch at the start of phase 4

[primal flicker]

I've been plugging them in as I work on coal power expansion and starter oil.

64 more coal gens laid out, and 90/min rubber and plastic rolling in.

Now I'll want to build a second floor, for 64 MORE coal gens... And a top level with the final 32.

[prisma kraken]

also, in thinking about it for a minute, for electro, i usually tap the crater lake oil which isn't accessible until you've unlocked 'adv alum processing'

[primal flicker]

Shop 🤝 Milestones 😆

But I think I'll be doing the Big Nuke Project on the east side of the map. Somewhere in the vicinity of Titan Falls (or whatever it's called).

[ashen stirrup]

Any tips for optimizing this? I'm trying to see if there's any way to eke a few more frames out of this without going to the wet/pure alts. I have all the alts available in phase 3.

https://www.satisfactorytools.com/production?share=vSrHQbwoUVd6sjCFLa66

[wind spade]

you can select all the recipes you have and hit "maximise", it'll show you max you can do, if you want to optimise for that

[primal flicker]

Are you optimizing production/resources or some other priorities?

[ashen stirrup]

I'm trying to get the most HMFs I can with the resources listed in the "Items, Input" tab

[wind spade]

then use the maximise function 🙂

and select all recipes you have, the tool will pick the recipes that lead to your goal

[ashen stirrup]

Yep, and it told me something I didn't want to hear 😆 so now I'm looking at belting another limestone node over

[primal flicker]

You only get 8⅓ without more alts.
What's wrong with wet concrete? It's in my top 10 most useful.

[ashen stirrup]

Ooh, I also just realized I had coal entered incorrectly

I guess my reluctance with wet concrete was just seeing people saying that the pure recipes are useless unless you're trying a crazy "max the world" build

[wind spade]

they are not

[primal flicker]

That recipe alone gets you almost 50% more frames...
What about getting more copper?

[wind spade]

pure recipes are "instead of having to get more ore, I just get water"

at the cost of more power ofc

but people that say the recipes are pointless are looking just at global availability, without considering local availability

[primal flicker]

I'm not a big fan of pure iron. Because it's so abundant and I refuse to use iron wire.
But wet concrete, quartz crystal, caterium, and even copper are worth using imo.

[wind spade]

obviously you won't run out of the resource, but do you want to belt/train/drone/truck more ore or just tap nearby lake to get more? 🙂

everyone has their own set of "good" and "bad" recipes, but basically anything is fine if it works for you, don't be scared to experiment, use whatever feels best for you 🙂 that's the whole point of alts - everyone gets their own experience

[ashen stirrup]

More copper would be a bit more transport than I want to try dealing with rn

[primal flicker]

That's the real question: would you rather build a water supply, or build more ore supply?

[primal flicker]

Just across the water to the east. Bring it down the cliff.

[ashen stirrup]

After putting in wet concrete, I'm up to almost 16 https://www.satisfactorytools.com/production?share=vj8cTHAuEnSEXuGtiqTH

[primal flicker]

There's a ton of pure nodes up there. Iron, copper, quartz, coal, some. Caterium....

[ashen stirrup]

Indeed there are 😄 I have a motor factory up there

I think I'll tap more of those nodes and the nearby caterium when I build computers, which will be my next project

[primal flicker]

If you put in your target production, then find what resources it needs, THEN find where those resources are reasonably close together on the map...

22.5/min

[wind spade]

Factory separation

[ashen stirrup]

My idea of target production is kind of flexible, since I don't have a really firm idea right now of what I'll need. Anywhere between 12 and 20 seems worthwhile if I can do it without too much inconvenience

[primal flicker]

Even something smaller is VERY worthwhile. 4.5 is a good initial production number, until you need more of them, more often.

[prisma kraken]

if you can mix in some plastic & rubber, this is a pretty good way:

(despite what the wiki lists, steeled frame is the least efficient of the mod frame recipes, what's listed there doesn't take into account adhered plate and coated iron/steel coated plate for some reason)

[primal flicker]

Steeled frames are my favorite just for factory density. Smaller footprint.

Steel coated plates and adhered plates bring the same benefit.

[prisma kraken]

yeah, they're really expensive comparitively, its cheaper to make more rips using the steel alts than to make the steel pipe to reduce the number of rips

imho, this comparison is something that i think should be adjusted in any balancing patch:

that's a big diff in steel cost

[primal flicker]

Using steeled frames vs default reduces my hypothetical HMF factory from 224 to 184 buildings (17.8%), at the cost of about 30 more iron and copper (8%)

[prisma kraken]

yeah, if you use stitched where you make the wire from copper or caterium, it becomes more appealing

[primal flicker]

Also, that power use difference 👀

[prisma kraken]

to me that isn't so appealing because i want to make a lot of pasta

[primal flicker]

Fused wire ftw

[prisma kraken]

the power actually isn't as much of a factor there... watch what happens when i add in the solid steel making:

it is more power hungry, i agree

[primal flicker]

OK I plugged in coated/adhered and using iron alloy and solid steel. Steeled takes copper/iron use from 354 to 403. Building count 192 to 169.

[prisma kraken]

i think for mf's, this is the best-of-breed:

[primal flicker]

As always, it depends on what you're optimizing.

[prisma kraken]

yeah, i'm sort of looking at a gestalt resource utilization with some experience from building big hmf factories... keeping the steel use for mf's down so you can make more pipe for hmf's just allows you to make more hmf's

mf's are either going to making hmf's or vers frameworks

both of them are steel-heavy things to build, so i think minimizing the steel for mf's is a win

by that token, you can go with all-iron mf's and do stuff like iron wire+stiched+cast screw+bolted

doing that type of thing though kinda ends up being a high machine count and forces you into pure iron

mf's just turn out to be something where any way you make them, its some sort of tradeoff that isn't very cut and dry

[primal flicker]

Honestly that sounds like good game design to me

[prisma kraken]

yeah, still, i feel the balance on it should be tweaked a little

it would be nice to be able to get 45 hmf/min out of a spot on the map with a lot of iron and coal with the mf's included in that factory, and not have it be a massive undertaking to build

[primal flicker]

Wait... Why pasta, tho?

[prisma kraken]

i like the loops

really you start making pasta at the end of phase 4 and start rounding up the available copper in all the biomes to make it, and the more loops the better as the endcap build for a world

[ashen stirrup]

doesn't hurt that particle accelerators are pretty

[prisma kraken]

as i said, i like the loops 🙂

they look pretty sweet en-masse

[faint atlas]

Thinking about using a lot of recipes with screws in them due to how simple they are to produce, such as bolted iron plate, bolted frame, and heavy flexible frame. Is this a bad idea? I don't mind having an enormous amount of conveyor belts just for transporting screws around the factory

I see a lot of people talking about how bad screws are

[wind spade]

they are not

what is bad with them is when you put them on the belts and transport them across the factory, because they are needed in high quantities

which is why you should always make screws directly in front of machine that needs them

[faint atlas]

yeah I plan to

[wind spade]

most recipes can be made to be 1:1 or 2:1 with some clocking

which is much better than making 50 belts full of screws

[true junco]

Have you heard about our lord and savior "Steel Screws"?

[faint atlas]

I just saw that

[wind spade]

I'll do you one better - steel rods 😛

[faint atlas]

Using steel rods makes the default modular frame one of the best right?

the other recipes use screws or steel pipes and thats a lot more effort

[wind spade]

well depends on your definition of "best" 🙂

practically best is "whatever you like most"

[prisma kraken]

steel rod+steel coated plate + adhered plate is about the most efficient resource-wise, coming in at 90 iron + 90 coal + 90 rubber + 30 plastic for 60 mf/min, but it also ends up being a stupid number of machines

a little kinder coming in at 100 iron+coal is the combo of bolted frame + steel rod + default screw with steel coated plate & adhered

[true junco]

Making Default Screws out of Steel Rods is the most screws per Steel Ingot, and the most screws per Iron ore.

But I massively prefer steel screws for many reasons.

[faint atlas]

thanks

[true junco]

Honestly. I should search for all my comments on steel screws and put them into one place, edit them up decently and just repost that from now on. Lmfao.

"The Steel Screw Manifesto"

[snow dove]

turn it into a book, publish it, then mail it to them

[prisma kraken]

steel screw is great when you are dealing with recipes that require multiples of 13, and i really prefer it for them, for anything else i think steel rod + default screw makes more sense

specifically in the case of bolted frame as a recipe, it eats steel if you use steel screw

[true junco]

I would only consider using something other than steel screw, when i need way less screws than steel screws makes.

Like if i had a build that could be done with 1:1 machines using cast screws, I would consider using them because i literally cant waste iron fast enough. Lol

As for steel screws vs steelrodscrews. The consumption difference is pretty small IMO. Its obviously there, but steel screw is already such a huge improvement over the 2 iron screw formulations.

[prisma kraken]

bolted frame is a strange recipe though, its 140 screw/min requirement seems to me to have been chosen so that default rip & bolted rip combined with it require 260 and 390 screws for use with steel screw, but it just doesn't work right with that all

its like once upon a time it was part of a really nice combo of recipes but something changed in a balance patch without updating the recipe

[true junco]

Well. That combo you told me about a while back works great...

[prisma kraken]

same thing happened more recently with fine black powder

maybe i'm just not looking at things holistically enough and there's something like oscillators that the excess rip output of the combo is meant for

[true junco]

I just like this little BP so much. 🤷 (note: all machines overclocked to 250% except doe the steel screws, which are clocked to feed their respective assembler's demands.

[prisma kraken]

i have a feeling that on a wall somewhere at coffee stain is a diagram of the 'perfect' minimal factory, and like all the strange recipes that exist are used in it

as a module, that looks like it would work perfectly with compacted steel

yep:

[true junco]

Hmmm. Thats interesting. Does that line up well with the manufacture of the compacted coal itself too?

Looks like CC is made at 25/min.

[prisma kraken]

no, that happens in 25/min

[true junco]

At a glance. It doesnt look like CC lines up particularly well with any of its products.

[prisma kraken]

lines up if you clock c-coal to 225 or 112.5 though

its kind of moot though, if you're using sulfur for something, you're goal is to use an entire sulfur node for that purpose

that's my thinking at least

[true junco]

Was working on this last week.

[prisma kraken]

but yeah, compacted coal always just lines up with its uses badly... turbofuel, heavy tf, and compacted coal all end up in icky numbers

[true junco]

Nice... even more reasons not to do turbofuels. Lol

[prisma kraken]

well, blended tf doesn't use compacted coal 🙂

only other use of it that i can think of is for fine black powder

which sucks now as a recipe after the weapons changes buffed black powder, but still has a niche use if you just want to import compacted coal and coal instead of coal and sulfur

(very niche use-case)

[true junco]

Fine black powder is one of the recipes i dont understand the appeal of. I think it was just left behind in another update.

[prisma kraken]

it was, before update 6, black powder was 2x expensive to make and they changed the cost of the base recipe w/o updating the alt

[true junco]

https://tenor.com/view/well-there-it-is-jeff-goldblum-gif-14058551

[near hatch]

I just set up a train station that stops at a pure overclocked oil node, the pipes are telling me that 600m3 is extracted each minute which is normal but the train station gives between 520 and 540 item per minute as outgoing transfer rate. Can someone explain to me why ?

[vapid gorge]

1. don't trust transfer rates
2. you probalby haven't set up buffers properly

[vapid gorge]

you want this at both stations #design-and-architecture message

[near hatch]

oh god I don't have any room left for this

thanks tho 🙏 i'm gonna squeeze this in

[vapid gorge]

for trains stations, you need some kind of buffer at both ends.

Because they stop transfer rate while loading you can't move 2 pipes or belts full with 1 platform

[vapid gorge]

and build it exactly like in the image

[near hatch]

a simple buffer is enough ?

[vapid gorge]

the small one like in the image? it should be

I think if it's a very long trip you might need the bigger one

[near hatch]

it's relatively short so I'm gonna build this one

[near hatch]

Do you know if the junction crosses are important (the ones at the same level as the buffer) or a simple 90 pipe is good enough ?

[vapid gorge]

yeah you want the buffer above

looks there's probably ways to tweak it so it still works - but unless you're very confident about pipe works in the game I wouldn't try

prob a good idea to make a blueprint of it too

[near hatch]

yea i'm gonna do that cuz it's pretty annoying to build temp fondations to place the junction crosses

[stuck ingot]

Would you guys say that 1000 Rubber/Plastic will be enought for the rest of the playthrough?

I just got into Tier 5/6 and will build my first ever oil proccesing plant.
It will produce the 1000 of rubber and plastic and 1600 Fuel for i think 120 or so fuel generators.

Is this to much this early
and should i build something smaller and upgrade it with mark V belts ect. or is it good to have something this big earlier.
I am trying to not have to build a bigger one later on.
Thx

[true junco]

There are weird things with pipes. Many of the devices in the pipe manual are only certain to work with the junction pieces. I gather its because of some inherent difference between junctions and pipes.

[true junco]

Its a good start on the fuel side for sure. Probably more rubber and plastic than you need immediately tho.

inb4 somebody says you should only make as much plastic and rubber as you need to make whatever you are using the plastic and rubber for

[stuck ingot]

I just dont want to have to come back later and build another one and would rather just ship the rubber with trains etc to where i need it

i really want to know if the more experienced players would say if in the entire playthrough 1000 R/P is enought to last me. Or if i should go bigger to have peace of mind

[vapid gorge]

eh look - at worst it just means you have to choose recipes that use less plastic/rubber

this game shines when planning final projects. It's a sandbox so it's best value is setting your own challenges. You can, absolutely, finish phase 4 space parts with that amount of oil product. But if that's your only goal for the game I would say you're missing out

[polar ridge]

there are other goals

?????

[vapid gorge]

yeah your own. It's a sandbox. For example I've gone through the tiers a few times and settled on automating quite a few of the space parts like 15/15/60/60 pm. Also part of the goal is to build very local high tier hubs so I have to import very few items

Also a lot of curved buildings. Make a goal around whta you like

[polar ridge]

I will make a huge dick in the sky

im unstoppable

[vapid gorge]

Sure 😄 whatever floats your boat.

I've got a couple other ideas for projects. Like only using the resources of the east swamp to build everything - quite involved

or using only drones to get stuff back and forth for the whole map

[near hatch]

This design is great but I think there might be one flaw. For example if you need 600 ressources/min on one end and the buffer is getting filled after an unload of your train, the buffer will take more than 600 leaving your pipe that need 600/min with less than 600 until the buffer is filled to a cerain level. I don't know if the solution I just built works but on the ends of the buffer I placed Mk1 pipes (or valves, I suppose it should work) to limit the filling process to 600.

[stuck ingot]

I mean this is my first playthrough so my only goal is to beat the game with nice looking clean factories.
I will experiement with other aspects/ challagens later saves

[near hatch]

I'm not sure of what I'm saying but that's what I understood from what's happening in my plant

It still isn't 600/min constantly but it's closer than with Mk2 pipes on the buffer, I'd say I'm averaging 585/min with Mk1 and before it was something like 550/min

[vapid gorge]

the pump going one way or another will be enough. It'll take a bit to stabalise and it relies on you using the pipes correctly to feed objects.

For a short trip that buffer system has been tested a lot

have you looped the pipe feeding machines?

[near hatch]

What do you mean?

The manifold pipe is going back to the start ?

[vapid gorge]

loop

[near hatch]

No I didn't

[vapid gorge]

Essentially. What you'll want to do after you set up the trains buffers ect, is set up the feed manifold like this. Turn on all but like 1 of your machines so alllll the pipes and buffers and machines are full

after you check the pipes and machines? then turn the last machine on

everything should be stable then

[near hatch]

Is it really something essential ? I mean, pipes have a limited volume so It's just gonna block the flow and feed the previous one over and over

[snow dove]

helps mitigate the issues that backflow causes

[polar ridge]

air will float my "boat"

[vapid gorge]

block the flow? don't use valves

And yeah - in most pipe builds you want to loop and flood the system

[snow dove]

and by helps mitigate i mean entirely remove the issues

[near hatch]

I'm not, I meant like once it's full, it's not gonna go further in

[snow dove]

gonna assume you replied to the wrong person

[near hatch]

Not sure if I'm expressing my point correctly

[vapid gorge]

share an image of how youre feeding machines from the train line?

[snow dove]

Backflow fills up the games limit for flow through a given pipe segment, by allowing a secondary unobstructed path you allow the fluid blocked by the decrease in throughput to travel up and towards the end of the manifold where there is throughput to spare regardless of any backflow’s affects

[vapid gorge]

ok 2 problems

1. you'll need the loop and pre flood
2. you don't want to split a manifold over 2 floors

it's not impossible to get a split floor manifold to work, but it's not something I'd recommend to someone who wasn't very pipe experienced

[snow dove]

keep pipe systems as simple as possible

that’s a good rule of thumb for anyone at any point in time

[near hatch]

Okay noted, but I really can't reorganize this thing in a single floor, or at least make it a manifold to the top

[vapid gorge]

over clock the lower ones to 200%?

[near hatch]

In the end it's just a manifold into another manifold

[vapid gorge]

make it a longer row?

[snow dove]

could also split it before the manifolds

[vapid gorge]

even splitting it before then is a bit iffy in my experience

especially dealing with train buffers

[near hatch]

aesthetically not pretty

[snow dove]

yeah not ideal, but should work

train buffers does add a degree of complexity to any fluid system tho, yeah

[vapid gorge]

sometimes there are sacrifes to be made when you learn new logistic issues 🙂 but consider you're early on and will probably want something different later anyway right?

[snow dove]

i always like thinking of the first play through as a long tutorial where i learn the skills and general knowledge for a second nicer playthrough

[vapid gorge]

if you absolutely can't over clock the lower machines

1. Loop both manifolds
2. flood the system as described
3. have powered pumps in front of each looped manifold

[snow dove]

550 hours of tutorial later, just started my second playthrough

[vapid gorge]

so like a pump here and here after doing everything else

[near hatch]

Indeed, it works nicely for the moment so I'm gonna leave it like this, it's just for power so it's not a big deal if it's not constant to a certain degree. I'll make sure to do thing properly next time

[snow dove]

also i’d use mk2 pipes for the entire system

having throughput to spare is always nice with fluid systems

[vapid gorge]

and look early on imo having a system only running at like 90% is fine? just get some stuff out right

[near hatch]

got it

[vapid gorge]

I tried something super fancy ages ago with a big power station - I could only get it to like 95% of expected output

[snow dove]

100% uptime is a nice ideal, but don’t stress too much over it

[vapid gorge]

It would have taken a massive redesign to get it to 100% which I couldn't be bothered with

[near hatch]

it's supposed to be constantly at 11,968

[snow dove]

you might catch some flak in this server for it tho

[near hatch]

I'd say it's good enough

[vapid gorge]

are you using coal power?

[near hatch]

Well, coal is generating the 1,968 but I planned dismantle it since I have this almost-10 GW oil based generator

[vapid gorge]

ah ok, so it's likely something unstable in the oil power that's got you a wiggly line like that. Not a massive deal, probably has to do with loops and pre flooding. Super common issue for newbies.

just means that if you ever want to get fine tuned set ups you'll want to work on your pipe skils at some point 🙂

[near hatch]

Yeah I noticed that it's not as easy as belts ahah

[vapid gorge]

It's not - BUT like the steps I mentioned basically solves all issues unless you try to do fancy stuff

[near hatch]

Even though I had my battles with balancing item flow to precise values

I didn't use any manifold atm, I'm pretty proud x)

[vapid gorge]

1. point A to Point B - no merges or splits
2. loop manifolds
3. pre flood

Load balancing is a choice - same efficiency as manifolds, but it will get much much more complex the further along you get as you have to balance each step

also you can't upgrade load balanced systems or tweak them, they need full rebuilds

just stuff to be aware of

[near hatch]

good point

[vapid gorge]

oh and no matter what anyone else tells you , you can never actually load balance fluids.

[near hatch]

I just thought that pipes as it's managing flow, it's just gonna spread evenly in the manifold

so it's balanced, but it's not really the case

[vapid gorge]

once a manifold stabalises with a belt you could say the same thing though.

only the right amount of items on each belt feeds a machine

But yeah, if you do keep load balancing keep clocking machines in mind and do smaller groups and batches, will make things somewhat easier

[snow dove]

and try to avoid prime splits at all costs

[near hatch]

after some consideration, I might switch to manifold in some cases, you probably convinced me x))

[near hatch]

prime splits ?

oh

[snow dove]

ex 1:17, 1:13

[near hatch]

yea I tried some, it's just hell

[snow dove]

you can always add an extra machine and adjust clock speeds to compensate

[near hatch]

that's what I'm doing

[snow dove]

1:18 is significantly easier

[near hatch]

I'll run out of space before i can do a big prime split

Would you recommend having a big factory ? Like you gather everything in one place and make everything there. Obviously you make floors and stuff to expand. But the idea of "one place" is great or should I have smaller factories ?

[vapid gorge]

depends what you mean by 'big' and if tha'ts your end goal

[near hatch]

performance-wise also it might be an issue, I don't really know how the game engine is managing the world

[near hatch]

well, produce most items in one place

[vapid gorge]

parts per min is is the important bit

[near hatch]

items that are essential to build machines or to complete tiers/phases

[vapid gorge]

like if you ahve a decent computer and you're only doing like 1/1/4/4 of hte final space parts? performance you're prob fine

You're biggest challenge will be design and layout

and I can almost guarantee you your first massive build will be a mess

[prisma kraken]

its better for any number of reasons, including engine performance in the game to keep things decentralized and spread out

[near hatch]

I actually played the game a while back but couldn't recover the save as a lot had changed so I am confident in my first big layout atm

[vapid gorge]

So for example this was my first big project, 9k plastic and 10k aluminum ingots per min. It worked but was hideous. Tore it down.

You could make all the parts in the game in that space but it'll be evey more complex exchanging a ton of different parts instead of a handful like this

[prisma kraken]

yeah, Cobalt makes a good point, in your first world, you won't have the skill to build big yet

[vapid gorge]

the other side of the coin - if you mean something that's a bit hap hazard and not perfect? this was my 'starter' base, made some of everything except nuclear material

oh and the final space parts, it could have though

[near hatch]

Oh god

[vapid gorge]

it's a lot tidier on the inside - the last layer with manufacturers and blenders got a bit messy

you can kinda see it here

it was walkable, and I could inspect sections as needed w/o fuss and access storage containers

purely utalitarian though

[near hatch]

Here is my first base
#screenshots message

[vapid gorge]

tier 6 ish?

[near hatch]

Idk it was 2019

[vapid gorge]

ah fair 😄

[near hatch]

I think the game stopped at tier 5

or 4 I don't know

[vapid gorge]

but yeah so as to the question of 'mega base' depends how big you're going and if you're planning on making it permanent

[near hatch]

got it

[vapid gorge]

I honestly wouldn't even try for 'good pretty and perfect' until I unlocked everything and got all the recipes I want

[near hatch]

I understand but it's Satisfactory, everything must be good pretty and perfect x)

that's how I try to play the game

[vapid gorge]

depends how much time and effort you have and want to give right? to me early stuff is temporary so I make it functional

[near hatch]

the moment I place a belt which is not at a 90 angle from another I just replace it

[vapid gorge]

Hah xD maybe I won't show you my build style

[near hatch]

ahah

I want to see now

[vapid gorge]

pinged in other channel 🙂

[near hatch]

10GW constantly, thanks (:

[tired viper]

is there any ficsmas planner? ^*

[cinder glade]

Mk1 Iron Farm, for starting out - Second Floor is for Tertiary / Quarternary parts (Screws / Reinforced Plates)

At 100% Efficiency, well under the Capacity (100MW, Drawing only 46 MW, totalling 68 MW with copper and Limestone added.

Ik its nothing compared to some of what happens, but I am pretty impressed with myself!

(Oh and it gathers at the fron in 3 storage containers for friendliness!

also yes thats on paper! I dont know anything good that i can plan and do the maths on,

[wind spade]

there are a few tools in pins in this channel (or in #welcome )

one of them is mine 🙂

[vapid gorge]

unless you start sinking them the load balanced set up will break down once the storage is full

[cinder glade]

yh

Its only temp

So I can produce materials to

[cinder glade]

Progressoo thank you

[wind spade]

(also most people don't really do load balancing and use manifolds)

[cinder glade]

Wdym?

[vapid gorge]

manifolds are probably your best friend

[vapid gorge]

!wikisearch manifold

[brisk shore]

https://satisfactory.wiki.gg/index.php?search=manifold

[cinder glade]

Oh

So it just does the load balancing

for you

[cinder glade]

What do u suggest for this

[wind spade]

first machine overflows, sending excess materials to next machines, etc.

[vapid gorge]

what you were doing manifolds and load balancing, what you were doing, are both efficient, one just takes a few minutes to spin up.

but load balancing gets more complicated quickly and you can't upgrade or tweak the system at all, it all has to be completely replaced

[cinder glade]

ah

[vapid gorge]

Don't load balance is the suggestion 🙂 or when you get the Awesome Sink you sink extra parts so the system keeps going

[cinder glade]

So just Sink stuff at every level

I gotchu

[vapid gorge]

If you like load balancing - go for it! just be aware it can become a LOT as you go on and it's a lot of extra work

[wind spade]

just final products

[cinder glade]

For now the plates rods and screws are final

I will implemetn the sink

I will probs redewsign with Mk2 Infrastucture

[vapid gorge]

and yeah that's an issue with load balancing, with manifolds you woulnd't need to redesign, just use extra space to the side and extend it

[cinder glade]

Thank you for your help

[cinder glade]

Yeah, when I was looking over it

I was like hmmm, this could get a bit tight especially with mk2 😆

[vapid gorge]

oh one last little bit - if you use manifolds and want to skip the time it takes to work at 100%? hand fill the machines first. Usually it doesn't take too long to spin up and isn't an issue but things that use wire and screws can take a long time since thye have huge stacks

[cinder glade]

Just fill them quick and then they'll kick in quick

[vapid gorge]

yup! though again imo the wait time is usually not a pain unless it's wires/screws being used in the recipe. And even then I just wander off and go build something

[cinder glade]

yh

[sacred orbit]

which saves the most power, a MK 1 miner at 150% or a MK 2 miner at 75%?

[brisk shore]

Official Satisfactory Wiki

This is the totally awesome community driven wiki for Satisfactory!! Come on in and check it out!
<3 @vapid gorge

[sacred orbit]

that link does not help me

[vapid gorge]

also if you're running out of power the solution is build more power

[vapid gorge]

look up miner in the wiki

[sacred orbit]

it's not a matter of running out, it's conserving what i have while i build more

[vapid gorge]

you're running off power storage I take it? then best bet is to shut off part of your factory.

[sacred orbit]

no power storage

why can't i just get an answer?

[vapid gorge]

because I imagine very few people know off hand what each miner takes initially in power, how much they take at different % and can compare them. The info is in the wiki

why would I bother remembering how much each miner takes when I can take 2 seconds to look it up

[sacred orbit]

that's crazy it doesn't

[vapid gorge]

you could eyeball it and guess mk2 and 75% is probably the same, or look at the power formula to find it out specifically, or build one of each miner and test it out

myriad of options

!wikisearch clocking

[brisk shore]

https://satisfactory.wiki.gg/index.php?search=clocking

[sacred orbit]

8.5 to 8.2

[vapid gorge]

huzzah

[wind spade]

heavily recommend to not use valves

[grave grove]

Pumps instead?

[wind spade]

depends, what are you using them for?

[grave grove]

Resource well trying to combat battery sloppy recipie

in short recycling the water into the same system

[wind spade]

ah yeah

[wind spade]

then use VIP junction from pipeline manual (link is in pins here or in #welcome )

valves suck at pretty much everything

they don't even have the exact values you set in them

[scarlet mural]

its worth it to invest into Catherium Wire, or the copper wire its okaish ?

[wind spade]

well, "worth" depends on you 🙂 if you find it worth, go for it 🙂

[vapid gorge]

Both have utility

Each alt essentially just gives you more design options

[scarlet mural]

i can push myself into 2400/m Copper ingots

worth it to swap wire production to Catherium ?

[wind spade]

again, up to you

[vapid gorge]

Why swap if you’re already making it? If it gives you a benefit? Sure

[wind spade]

do you have more extra copper or more extra caterium? which one do you have near the place where you need the wire? do you plan to do something else with that resource? do you have enough of that resource?

there's not any "better" or "worse", use whatever is good for you for your given situation 🙂

[true junco]

Do you mean "quickwire" or the recipe to make regular "wire" out of Caterium ingots?

[wind spade]

you can even mix them

[scarlet mural]

the "wire", quickwire its exclusive to Catherium not 100% sure here since i dont remember everything

[true junco]

Also, in my opinion, "Fused wire" and "fused quickwire" are "the best" ways of making either wire type.

[scarlet mural]

well, for sure where i'm i got several more Copper than Catherium, so i will stick with copper, arigathanx

[grave grove]

and I'm pretty sure the issue is water if all the math checks out and belts are correct etc.

[scarlet mural]

just need to push more logistically since copper are a bit all over the place, not centred into same area

Fused Wire seens interesting, arigathanx for recomendation

[vapid gorge]

Also assume your needs are going to keep changing as you learn progress, so don’t stress overly

[scarlet mural]

i think how i'm desing my factory i got a crazy way to manipulate and change here to there without much problens
thats why i like to check if there a better option

[grave grove]

Thanks for the help!

[true junco]

Both Fused alts are very productive and economical. They also have a couple of logistical advantages. 1st being because they are so economical you can transport a relatively small number of ingots to make large amounts of either wire... and that because they share ingredients it can be effective to make either wire from the same input manifolds.

[scarlet mural]

yeah, i notice that, fused seens to be gud, since i can make 2 gud spots to take catherium, and near i got several coppers Impure and normal
so it will be them

[civic tulip]

Hello there, need a bit of help with working out a design for a coal power plant.

I'm using 3 pure nodes, that's 720 coal/min with Mk.2 Miners which should be able to generate 3600 MW. That requires 18 Extractors (at 100%) which is 2160 m3 water and 38 Generators (at 100%) using up all that coal.

That can be reduced down to 8 Extractors [ 7 at 250% & 1 at 50% ] and 20 Generators [ 19 at 250% & 1 at 50% ] though I'd prefer a more neat approach of 16 at 250% and 4 at 200%. The only problem comes with the bottleneck of piping because the fact that 250% Coal Generators take 112.5m3 water.

[fierce ruin]

How much headlift does a machine give a pipe on it's output?

[scarlet mural]

i can push myself into 780/m Catherium Ingots, its worth it make such monumental effort ? cuz it will be 65 refinerys

[civic tulip]

Depends on the machine

[fierce ruin]

Refinery?

[civic tulip]

Water extractors have a 10m headlift

Oop

Mind if I dm you the wiki page on head lift? it's quite useful

[fierce ruin]

ah it says it

[civic tulip]

All I can think of is the use of valves but I don't have them unlocked yet and won't for a hot minute

[vapid gorge]

would need more of an over shot of the whole picture, but as it is yeah you're using a bunch of valves and splitting the manifolds over levels

like as much of a top down view of the whole set up as you can, build a tower if it helps

[civic tulip]

Would a cell of fluid buffers all connected to each other with a max storage that reaches the 2160m3 requirement with pipes going off the output linking with only 2 generators per pipe work?

[grave grove]

As I was building this I was also closing the build off, can't really get a top down view

[civic tulip]

It consolidates the input lines and means no overflow is worried about

Or am I wrong

[vapid gorge]

you're over complicating things I think - and fluid buffers are not the answer

[true junco]

General consensus is that valves and buffers are mostly useless.

[vapid gorge]

and add the option to cause more problems

[snow dove]

valves wholly useless
buffers only useful when just in conjunction with trains

[true junco]

Yeah. Often worse than useless. Lol

[civic tulip]

Really? In theory it should work despite the pipe bottleneck

[snow dove]

both very often cause issues and they actively detract from the idea of keeping fluid systems as simply as possible

[vapid gorge]

how? you can never move more water in even with buffers

[snow dove]

there is no way to surpass the pipe throughput limit within a given segment

[vapid gorge]

the solution to this is to make them in chunks.

typically the suggestion is for people to not over clock the systems and make water extractors: coal gens in a 3:8 as it fits perfectly.

you can still do chunks like that if you over clock things but you have to mange it slightly differently

[vapid gorge]

is waste water getting pumped into this as well?

[grave grove]

yes

[vapid gorge]

honestly most of this is hidden and it's super hard to tell what's going on - but I'd recommend a complete pipe rebuild

[grave grove]

yeah I have a 4 m floor that hides all it's busy but its organized but ya I didnt expect anybody to look at this and be like ya that's that

[vapid gorge]

OR leave it as it is if you don't care about your system being perfect.

if you see yourself changing goals and experimenting with stuff maybe just leave it since it seems to run mostly right

[civic tulip]

My fallback is to use 24 gens instead at 200% and use extractors at 225%, I just wish that it could work the way I am thinking because I feel like it makes sense,

[vapid gorge]

but if you want it perfect you'd prob need to redo the pipes

[vapid gorge]

make them im groups of 4 then? Are you trying to connect all the pipes together in a giant manifold?

[civic tulip]

I know what I need to do now I know that it's an impossible task to supply that original ratio of machines and how they were overclocked

I was just trying to see if there was any way it was possible, to reduce the amount of machines

[vapid gorge]

it's not - you could set them up at 250% burn rates and then figure out how much water and how to do it

it's not even hard

[grave grove]

ya the waste water flows to the bottom and then back up after merging with he Resource well water

[civic tulip]

So then why say that I was incorrect when that's what I was doing?

[vapid gorge]

buffers are not the way to do it.

but running coal gens at 250% is totally doable

[vapid gorge]

pain

[grave grove]

I recently added that to try and figure out what's going on with the system

[vapid gorge]

Well I'd be happy to take you through some steps to rebuild the piping if you like, if you feel it's good enough and you could jsut do better piping in your next factory that is also a thing you could do

[grave grove]

the 2 split the 500 wate to the correct ipes

[civic tulip]

I don't see why not, if you combine all of your input into one well that all the machines can in turn take from, that, seemingly, should work. The wasteful alternative would be to use and extractor for pair which would end up in using 10 extractors, but the whole point was to reduce buildings and to use the resources cleanly and perfectly

[vapid gorge]

there are ways to do fancy piping, and I'm not going to say what you're doing is entirely impossible (largely because I can't see it all to make that judgement) but in general, you probably want to make the pipe system as simple as possible

[vapid gorge]

how does the buffer help though? it doesn't increase throughput.

fluid buffers are essentially just a big pipe. Or you can look at pipes being tiny fluid buffers.

except that they fuck with head lift and act even more dumb when not entirely filled

the only real place fluid buffers belong are for fluid train buffering - everywhere else if a system works with a buffer it'll work w/o the buffer

[grave grove]

Overall I was sad after setting everything up and thought I knew enough about all the pipe mech to see it not at 100%

any tips etc is much much appreaited

[vapid gorge]

waste water and changes in elevation are particular challenges.
And you wanting tips as in you want to try to make this work?

[grave grove]

Yes

[vapid gorge]

would you mind if I dm you? it might fill the chat a bit

[grave grove]

Go for it!

[vapid gorge]

and if you want to draw a crude sketch of how you're trying to set up pipes with the buffers and feeding and all that I'd be happy to talk it through with you

[true junco]

Perhaps I am not understanding what you are trying to do.

You can only push up to 300m³/min thru a mk1 pipe. 600m³/min thru mk2. Buffers do not change this.

[civic tulip]

It's fine, I think I've worked through it in my head, I'm attempting the impossible by overthinking

[vapid gorge]

if you're sure - I've got a bit of time on my hands now and happy to help you work out layouts

[civic tulip]

All good, either way in the end I have to either create 2 more extractors or 4 more generators for it to work, the 20, 8 can't work with a Mk.1 Pipe bottleneck

[true junco]

3:8 ratio that many mention requires the extractors to be connected at different points along the manifold relative to the generators. Something as simple as 2 extractors on 1 end, and the 3rd extractor at the opposite end works well.

[civic tulip]

Well at the overclocking it is, it'd more-so be like 1 on one end and another on the other end of 4 others but yeah I think I get what you're saying

Though wouldn't something like this work? Maybe buffers aren't all that useful but they can at least seperate pipelines to combine excess

(Circles are extractors, squares are buffers if it isn't obvious)

[true junco]

It should work. But it would also work without the buffers.

[civic tulip]

But then you'd have a whole pipeline with a max 300m3 limit and you're trying to push 900m3 through it

😕

[true junco]

Not really. You have 3 pipes in and 4 pipes out. Make all the pipe connections youve shown but dont use buffers. Connect straight thru where you would have put the buffers with pipes instead.

[civic tulip]

Right because pipe junctions work as separation points for overflow, like the buffers

I swear the more I think about this game and it's mechanics, the more I unlearn

[true junco]

That is accurate. Lol. I have spent the last 12 years building a lot of piping... very little of that experiance applies to this game... which is probably for the best, but bothers me a lot. 😆

[civic tulip]

God, I wish piping could be more like conveyors, so much simpler to wrap your head around

Grargh fluid

And their silly dynamics

[true junco]

Meanwhile i want gases and fluids to work more like they do IRL. but then most people's computers would explode... 😆

[civic tulip]

If they did work like that then at least my earlier ideas would've worked

You could then load balance with valves

[true junco]

Maybe. Not really sure. The valves in SF dont exist IRL. so i dont have a good comparison. They are a weird combo of a check valve, a flow meter, and some sort of a proportional flow controller.

But almost all of the pipe systems i built were for large oilfield gas compressors (CO2 and Natural gas). And SF is a very terrible simulator of gas piping...

[civic tulip]

At least our gas lines can't explode in SF

[true junco]

I actually have ideas for that... but thats a whole other topic. Lol

[vapid gorge]

You can make a pipe system look like it's load balanced, but because they are bidirectional they'll never be truly load balanced

they would only be as 'load balanced' as a stable belt manifold where the belts are feeding what each machine is needed as there is space available

[sacred orbit]

so like this, yes? or should i have the two extractors on the ends be pumping into the ends of the pipelines instead?

[vapid gorge]

is it all clocked at 100% and it's 3:8 water:gen?

should work then - to double check and improve piping you can figure out if there's any throughput issues - assume each machine consumes 45 and do a bit of math, sometimes drawing a diagram will help

it's like a high school math problem xD

[sacred orbit]

Yes

Pretty sure this is the setup I used last time I played and used 3:8

[vapid gorge]

it looks fine 🙂

[cinder silo]

Test of @prisma kraken save

[prisma kraken]

oh yeah, that's better than i get, but yeah, you gots the stutter

[cinder silo]

Weirdly enough performance is worse in that save than in mine 🤷‍♂️ something is definitely up.

[prisma kraken]

there's something pathologically wrong with it where i'm just hitting unoptimized parts of the game code

i'm also seeing in that region 780 belts miss items

[cinder silo]

That said my tiny northern forest save had performance issues, still does to a degree and It was only like 500k and change.

[prisma kraken]

yeah, the HMF factory i have there is just fricking killing perf

[cinder silo]

Watch the footage here stall like a mofo, back when it was captured the save was only 466k.

[prisma kraken]

i think i need to take it back to formula and rebuild it in a way that is kinder to the game

that's odd, usually for me by the time i cross over into the bay area from NF, fps gets back to normal, just b/c there's less foliage

[cinder silo]

Yeah, it cleans up when I go over in to the rocky desert, just nearby the coal plant there and on the bluff where the space elevator is, it is dead jittery.

[prisma kraken]

i mean, in a fresh save, NF is fine for me

i'm not on the latest & greatest hw, so i run at capped 30 fps, and can enable lumen and stuff, its just a few large factories & perf goes poof

[cinder silo]

My machine is well on its way towards 2 years old now.

[prisma kraken]

i'm running an i7 7700k, rtx 3060ti 32 gb system + 12 gb vram

[cinder silo]

I can drop a 14900k in to my existing motherboard but at the absolute best, I'd gain on average 15% performance (optimistic)

[prisma kraken]

which isn't anything great, but isn't a potato either

[cinder silo]

And that is assuming I clock to the same 5.4 I have right now.

[prisma kraken]

yeah, i wouldn't drop the money on an upgrade just for SF just yet... epic still needs to fix some engine stuff

[cinder silo]

It is something I'd do if I could get the chip on the cheap in the future as it is a one part upgrade path, but not at more than £600

[prisma kraken]

yeah, that's money better spent on a nice bag of doughnuts

upgrades like that are things that you toss on your newegg wishlist and snipe if the price is ever right

[cinder silo]

I probably shouldn't have had 21 entrances on the test cannons, the speeds cause the engine to stall and waste time loading.

[prisma kraken]

i don't use tubes very much, really just that one you captured, i prefer getting around with the zipline

problem with the zipline is when you get an fps stall, it throws you off

[cinder silo]

Yay, invalid experiment ftw, see the problem in the test?

[prisma kraken]

did it chuck you into the wrong tube?

[cinder silo]

Yup.

I'm going to create some isolation in power to see if it fixes it.

[prisma kraken]

they've always been pretty janky for me, as such i prefer to avoid them for anything but elevators and slower convenience travel

[cinder silo]

I never ran two parallel cannons quite like it, was always one cannon and a return so that is an unexpected fault.

Gah, need some quickwire for power switches.

[prisma kraken]

btw, since u8 hit experimental, i've been noticing that i've seen a few times rails connect wrong

really hillarious watching the trains teleport onto another track 😄

[cinder silo]

I wish I had known my weird train accident was going to occur, it would have made for a wtf vid, I had a 2 engine 6 car train on the aluminium run simply vanish, I replaced it and went on about my day, two hours later it returned and caused a rail accident.

[prisma kraken]

i've never seen a train just disappear... trucks and tractors, yes, trains, never

[cinder silo]

Only ever happened to me the once, it's the reappearence that did my head in.

[prisma kraken]

that's bizarre, lol

[cinder silo]

My train line is as simple as it gets, a 1.2km run between two stations, no junctions or anything fancy, just a double headed train shuttling back and forth.

[prisma kraken]

i know trains in the game pretty well and have seen all sorts of wacky with them, but never what you describe, inexplicable

[cinder silo]

Here it is from far off, I followed the entire line with the games cursor.

[prisma kraken]

i can't even conjecture

in u7, i saw a lot of problems with vehicles getting teleported to the center of the map

but nothing like that with trains

only really icky thing i've seen is that in complex rail networks, after load, trains don't remember the signaling state and sometimes that can lead to deadlocks or collisions on a load

i've not seen that in u8 though, i believe css has fixed that

[cinder silo]

I've heard that, I'm keeping my rails simple tbh, my next planned rail link is to move packaged fuel from the blue crater to the dunes.

[prisma kraken]

i think i'm pretty confident that the signalling for rails is now fixed... it's probably been 1000 hrs of game time since i've had complex enough railways to see issues, and i haven't

in u7, about every other load, i'd end up having to fix something

[cinder silo]

Urgh, the terrain streaming makes a 21 stage experiment just not fly.

Urgh the stupid stream loading crap still snarls up at 15 stages.

It isn't looking good for accelerators.

[prisma kraken]

upgrade to UE4 when?

[cinder silo]

What's worse is it should all be loaded anyway, I have 64gb of ram ffs!

[prisma kraken]

it looks like where its hitching is on map tile borders

i've become intimately aware of where they are

[cinder silo]

I'll try at 12 then 10 stages, below that just isn't worth it.

12 stage is limited.

I'll see if there is much difference in a 12 stage unsupported,

Unsupported tube, 12 stages also.

Hmm, side by side I'm not seeing a discernible difference, it's looking like the supports aren't the issue.

supported trip took 26 seconds? ,unsupported trip took 28 seconds Meh I really need an on screen stop watch, relying on the footage timer feels off.

[sacred orbit]

so you can use hypercannons as just boosters?

[cinder silo]

Not very well, the travel times are just meh, so far barely worth it and I still haven't figured out the slowdown in the live cannons.

[prisma kraken]

can you discern whether the time is fps lag or lower veloscity?

[cinder silo]

I might need to add a loopback on the test cannons, so far both supported & unsupported are giving the same runtimes, but I'm not seeing the speed loss that I observed on the 18 section live cannon run that dropped all the way down to normie speed as shown here: #design-and-architecture message

[prisma kraken]

haven't they always normalized your speed at something like 48 kmph over a long time?

[cinder silo]

Nope.

[prisma kraken]

its been a while since i've played with them (like update 6-ish timeframe), but they always seemed to tend to slow you down to 48 kmph of a long enough distance, it was just if you could get going in them quick enough to overcome the slowdown

[cinder silo]

Even the longest cannon actually kept its 2k+ miles per hour to the end once it was up to speed, now it doesn't work at all.

13 seconds from central dunes to western grasslands.

[primal flicker]

What's a typical ratio of plastic/rubber use in the endgame?? Assume all alt recipes unlocked. Steel coated plates, adhered plates, insulated cable and oscillators, etc. Not planning on rubber concrete.

[wind spade]

depends on your goals and preferences, the ratio is anything from 0-100 to 100-0

[true junco]

Questions like that are so open ended its impossible to succinctly answer.

[snow dove]

Hypertube testing

(40 entrance hybrid cannon design, 1.2km of tubing)

1. [Supports, built in direction of travel] 8.550 Seconds
2. [Supports, built against direction of travel] 8.483 Seconds
3. [No Supports, built in direction of travel] 8.550 Seconds
4. [No Supports, built against direction of travel] 8.433 Seconds
   @cinder silo

couldn't seem to find any significant slow down for me

maybe it's tied to fps somehow?

or i didn't build enough tubing

i might be making an absurdly long cannonway

LOL, SCIM CAN'T SEE IT ALL AT ONCE

[snow dove]

boosting the size up to 5km, should be long enough

[snow dove]

ok setup is finally done, now i can do the proper tests

[snow dove]

Hypertube testing

(40 entrance hybrid cannon design, ~5km of tubing)

1. [Supports, built in direction of travel] 34.983 Seconds
2. [Supports, built against direction of travel] 33.833 Seconds
3. [No Supports, built in direction of travel] 33.967 Seconds
4. [No Supports, built against direction of travel] 34.133 Seconds

honestly the difference of the results is within a reasonable margin of error

full list of numbers for those that are curious

1. 8.54

2. 8.45

3. 8.36

4. 8.55

5. 4.700 | 13.250 | 8.55

6. 4.350 | 12.833 | 8.483

7. 4.750 | 13.300 | 8.55

8. 3.967 | 12.400 | 8.433

9. 3.867 | 38.850 | 34.983

10. 3.817 | 37.650 | 33.833

11. 3.900 | 37.867 | 33.967

12. 2.967 | 37.100 | 34.133

@cinder silo looks like how the tube is built doesn't affect it much

could brute force it by increasing cannon power

speaking of, gonna test if they changed how cannons gain speed

[cinder silo]

How far apart are your entrances, I'm still struggling to find a reason behind the speed drop in my active cannon but not during the test.

[snow dove]

2m spaced hybrid design i came up with a while ago

identical to this

i think they changed how they did the speed with the hypertubes

[cinder silo]

My inconclusive test used my classic design so something is definitely wrong.

[snow dove]

i'll try them again with the classic design

afaik they have nearly identical performance, but it could be having an issue rn

[cinder silo]

Though now it hitches hard when using above 12 entrances.

[snow dove]

gonna try it with my normal cannon testing setup

it takes like 2 minutes per test with the 5km testing setup

[cinder silo]

See here at 15.

Supported cannon.

[snow dove]

just a guess, but it could be due to fps issues

[cinder silo]

Just keeps doing that stupid loading thing when with the amount of ram I have installed it should already be loaded.

[snow dove]

you have a large degree of stuttering, whereas in my tests i don't

[cinder silo]

You can see when it freezes there is the animated dots in the bottom right.

[snow dove]

gonna have to upload it to youtube

my files at north of 200MB each

gonna take a while to upload, so i'll be doing tests in the meantime

[static zenith]

been trying to follow along with this hyper tube testing and theory.

love all the work you guys put in!

[snow dove]

this is what a strong hyperfixation gets ya lol

[cinder silo]

Hmm, how many entrances because the travel time seems to be very long, and dude don't del a vid when I'm in the middle of watching it.

[snow dove]

uh there was a certain detail in it that shouldn't be

lovely, youtube refuses to change my channel name

[cinder silo]

Odd, I've changed mine several times.

[snow dove]

ok there we go, just took forever to update

and now the preview hasn't updated

eh whatever

[cinder silo]

Hmm, that is the new channel name right?

[snow dove]

no, no it is not

it should be "irohlikestea" but it refuses to update

[cinder silo]

Take it down again lol.

[snow dove]

i hate youtube so much

[static zenith]

🙈

[snow dove]

https://www.youtube.com/watch?v=k1ucHG5E4sc&t=1s

[cinder silo]

But after watching the cannon test, it is very slow for the number of entrances.

[snow dove]

tested it in a private server first, but there we go

[cinder silo]

There we go, the name updated to irolikestea.

[snow dove]

not a big deal if that is a known thing, i'm just particular about what i show on the internet

[cinder silo]

I'm surprised a 1 minute clip is 200mb though.

[snow dove]

144fps at 1920x1080

uncompressed afaik

actually 218mb

[static zenith]

does satisfactory have a max frame

[snow dove]

you can set a max framerate

[true junco]

Wait... are you past the kill boundary?

[snow dove]

significantly

[static zenith]

but does it have internally a max

[cinder silo]

Travel time is far longer than a cannon with that many stages should be.

[snow dove]

unless the game engine has one, no

[true junco]

How?

[snow dove]

god mod in AGS

didn't think it'd circumvent it, but it does

[true junco]

I thought god mode protected against everything except the kill boundary.

[snow dove]

i'm very much beyond the kill barrier so it must negate it

[cinder silo]

My test cannon is only 2-3km at best, crosses from the dunes to the blue crater to the south along a temp sky bridge I used to supply power to an in-progress build.

[true junco]

Hmmm. Isnt there 2 boundaries. One that does damage, another that instant kills?

[snow dove]

mhm

i'm far beyond both

[true junco]

Thats crazy.

[snow dove]

i probably hit the kill barrier around the 3km mark

the other 2km will instakill if you aren't god moded

[cinder silo]

Speed wise it seems to be as fast as my 12 section run.

[strange matrix]

How are people loading very long trains without balancing, and maintaining 100% input conveyor throughput? Is it even possible without a priority/smart merger?
(long time Factorio player, and I am basically finding satisfactory... unsatisfactory, because of these design limitations)

[snow dove]

manifold

and just having enough throughput

[cinder silo]

I use an industrial can as an in-feed, with both outputs plugged in to the station so it can catch up after the lockdown phase.

[snow dove]

uh, cannon speed increase might be linear

wat

haven't finished the testing, but that's what it's looking like

[cinder silo]

Possible, but if so it is a colossal nerf that would need way more stages than I'm prepared to build.

[snow dove]

nearly done, then i'll graph it and see what it really is

[cinder silo]

👍

I'm sorta glad the supported vs unsupported didn't cause an effect because I'm not sure I wanted to review 40+kmof tunnels to remove thousands of supports.

[snow dove]

haha

only took like 3 hours of testing to figure that out

[cinder silo]

After comparing the two 12 section runs side by side the travel times were margin of error close.

[snow dove]

yeah mine were within a margin of eror

[cinder silo]

It was fairly straight forward since the skybridge already connected three biomes, that's why I have a floating cannon run that is at an odd angle.

[median heath]

Part of this also relies on what you mean by "100% input conveyor throughput"

Because if that means what I think it does, you cannot achieve it.

[cinder silo]

It might be worth me culling the runs to 12 sections and dumbing down the original supersonic ads in the facility because it is no longer the case.

[static zenith]

out of order signs xD

[cinder silo]

Pretty much.

[snow dove]

That's a linear function if i've ever seen one

(x axis is number of booster segments using hybrid design, y axis is number of meters traveled with my testing setup)

[cinder silo]

That does explain the massive loss of velocity.

[snow dove]

(line is y=30x)

[cinder silo]

That is definitely a deliberate change.

[snow dove]

absolutely

used to be exponential

[strange matrix]

in the sense that if I have X number of train stations, and Y Mk5 belts, there is no solution that will let me push those three belts into all of the stations without either a very large and ugly balancer, or backups on the MK5 belts that will mean their miner sources aren't working at 100%, even when the the number of stations is greater than the number of input belts.

[snow dove]

anyone happen to know the rate of gravity in satisfactory?

[cinder silo]

I'm unsure how to test it.

[snow dove]

drop test basically

you walk off a tall platform and time the fall

do that at a couple heights and you get a pretty solid rate

i can do it, just curious if someone already knew what it is

i can use the rate of gravity to calculate the output velocity which will give me a more accurate graph

[static zenith]

ive jumped off a few tall platforms ill tell ya

at various heights ;]

[snow dove]

did you happen to time them, preferably down to the millisecond

looks like i'll be doing that testing myself

[static zenith]

nah sorry ;p

[cinder silo]

Is there an always on-top stopwatch app I can grab which can be started with a hotkey (windows)?

[static zenith]

yea hook us up wih the tools

[cinder silo]

That way I could actually capture the timer within a video clip.

[snow dove]

i did it by recording via obs then going frame by frame

finding the start time, then the end time, then subtracting

[static zenith]

oh smart justt record then look at frames

much more precise then a stop watch

ever tried to rng shiny pokemon with a stop watch ? 60fps is near impossible to time lol

[snow dove]

i'll do it in increments of 100m

[static zenith]

well im curious to see what your gravity test turns out

[snow dove]

actually something useful for other tests too

[static zenith]

for sure

what about vertical boosters? can you drop into them and would it have a different affect

[snow dove]

i can, not a huge fan of em tho

[cinder silo]

I haven't tested verticals for a while, back in U7 they didn't work at all, just taking up space.

[static zenith]

just an idea ive never played with it.

[snow dove]

got the videos, now to go get the fall times

then spend a while figuring out how to do the math

or see if there's an online calculator

[median heath]

You're going to need to give specific numbers if you want me to grasp what you are talking about.
Because there are so, so many cases where X number of stations perfectly handles Y number of belts.

[snow dove]

infinite number of cases actually

lol gravity is linear too

y=39.7351x-64.2252
where x is the time to fall and y is the height fallen from

with a .3% margin of error

[strange matrix]

so lets say I had three belts, and five stations
top belt will flow completely into the first station.
middle belt will then priority merge into the top belt, to bring it back to full capacity, and bottom belt does the same for middle belt etc
as the stations fills up, this forward pressure moves from left to right, so that all three belts are being pumped into the stations and always moving, never backing up

in factorio, its basically

this, but i want to do the same in satisfactory, because I have 5 stations that have a max input of 5 * 780, and three pure nodes pumping out 780 ore each. I want to be able to fill those stations with the three mk5 belts as fast as I can without them backing up and reducing the miners efficiency
5 stations and 3 belts are example numbers.
I might have 8 mk5 belts, and 40 stations, I should still be allowed to have those 8 mk5 belts be working at max efficiency, because those stations have a higher input maximum, than the miners output maximum

[wind spade]

this is bad even in factorio 😄

but personally - just put one belt into one platform

[strange matrix]

bus balancing is dead, priority splits are now the meta

[wind spade]

yeah, but inverse

the leftmost balancer goes on the top

[strange matrix]

but when that platform fills up, the belt is now doing no work, while there are still stations that have input capacity not being met

[wind spade]

bus balancing was never meta

[strange matrix]

eh, pushing before the split ensures the split off always gets maximum available material

[wind spade]

pushing after the split ensures belts are pushed to the left

[median heath]

greeny pretty much got it, but yeah, Satis does not have prio mergers, and in like 99% of cases where you think a prio merger would be helpful, it means you should re-evaluate how you have the system constructed.

Because Factorio needs prio mergers given how the game operates. The systems that Satis has makes it a game where you do not need them at all.

[strange matrix]

anyway
i want this because I want to have a gigantic train, like 30 or 40 cargo wagons, being fed by only three pure iron nodes. Its going to take an ungodly amount of time to fill all the wagons, but i dont care, i want it

and i refuse to do a 3:40 balancer

[median heath]

In that specific case you are making an active choice knowing the consequences.
Which is fine. And a valid way to play.
But it doesn't mean the devs need to work on/change anything to accomodate said choice.

[strange matrix]

i didnt say they have to
and if someone tells me another valid way to do that usecase, then i will happily forgo wanting priority mergers

(valid that isn't a 3:40 balancer, because that is hell)

[wind spade]

if you want to do weird things then you're alone 🤷‍♂️

[median heath]

I mean, you want prio mergers because you're choosing to build in a way that they would be useful.

[strange matrix]

why is it weird

[wind spade]

why not 1 belt = 1 platform = 1 car

[median heath]

So the valid use case is... not building this setup.

[strange matrix]

longer trains reduce traffic on train lines, its a perfectly valid thing to want

[wind spade]

you won't have enough traffic to have issues

[strange matrix]

yeah not with 40 wagon long trains haha

[median heath]

Having train lines not connected when they don't need to be also reduces traffic.

[wind spade]

not with 2 car trains

[strange matrix]

yeah, i'd rather have a central train line rather than 50 A -> B isolated train lines

and i dont think thats uncommon

[median heath]

Which is, again, another choice you are making.

50 trains is quite a lot tbh.

And I'm guessing due to you having a "central" anything.

[strange matrix]

if i have 50 micro factories, then you're saying i should have 50 direct train lines between resource nodes and those factories...

or i could have one train line, with 50 offshoots to those factories

[median heath]

Uh..

1. There are things besides trains.

[strange matrix]

yeah im not having truck routes from bauxite to the north west desert...

inb4 "thats you're choice not to"

[median heath]

🙃

[strange matrix]

sorry, didnt realise satisfactory is a game where there is a right way to play and freedom is frowned on

[median heath]

You have trucks, trains, and drones.
Optimal logistics is using all 3 in concert where applicable.

No one is frowning.

My one and only point has been that prio mergers aren't needed 🤷‍♂️

Yes, there is an optimal way to do certain things.
But choosing fun over optimal is valid and encouraged.
And MOST things do not have 1 optimal path

[strange matrix]

doesn't feel very encouraged here 😂

[median heath]

Again, the only discouragement point I personally have made is that the devs do not need to spend effort on prio mergers.
Everything else I have validated and have said is completely fine to do so long as you understand what the choice you're making entails.

The only thing I would actively discourage as far as playstyle is people making uninformed choices about how they approach things.

[strange matrix]

okay, so is the only way to fill 40 stations from 3 nodes a 3:40 balancer
is there honestly no other way

[snow dove]

trying to calculate the rate of gravity using these numbers
400m | 11.683 Seconds
300m | 9.200 Seconds
200m | 6.683 Seconds
100m | 4.133 Seconds
can someone please help me out i'm struggling

[static zenith]

inconsistant lol wth
the further you are from the ground the less gravity affects you ?

[median heath]

If the choice is filling 40 stations from 3 nodes, you either make the balancer or do the manifold method that requires each of them to fill in sequence until the entire system is online.

[snow dove]

that's legit the in-game values

[static zenith]

oh i believe you! its just odd lol

[snow dove]

i'm just trying to figure out the actual value, so i can then use that to calculate the output velocity based on the distance traveled with my hypertube setup

[static zenith]

those results make me think they are planning to add different "planets" with different gravity

[strange matrix]

wait, let me make that graph not suck

[snow dove]

weird it's a linear equation

so using that rate,
could you use it to calculate the output velocity using distance traveled and height of starting point?

[static zenith]

thats over my head bro

[fierce cypress]

If it’s linear then the velocity will be constant

[strange matrix]

looks linear to me

[median heath]

@strange matrix if you know how much throughput you want per car, there may be a simpler way.

[snow dove]

given variable distance

[fierce cypress]

The distance won’t change it

[snow dove]

i mean i know the output velocity is different

[strange matrix]

how can I use the manifold method with more than a single conveyor line, that is the issue
if I use an injection manifold, then the miners will not be supplying every station, but only 1/3 of the stations, no?

[fierce cypress]

Then it’s not linear

[snow dove]

38m
66m
93m
120.5m
147.5m
174.5m
201.5m
228.5m

18m starting height

[wind spade]

why would you use that, if you can just use what you have on the belt

[strange matrix]

im not focusing so much on throughput, as three mk5 belts in = three mk5 belts out, but a longer train mean mean less trips/lower traffic/less time spent in the 27 second loading animation in comparison

[median heath]

Train will take the same number of trips away from this station regardless of 40 cars or 1 car because the time spent is a static value.

[strange matrix]

because i'd rather feed three mk5 belts into 40 stations, instead of one mk5 belt into 13.333 stations, three times

[median heath]

How many drop off points is this station feeding?

[wind spade]

or feed 3 belts into 3 platforms because it doesn't matter and you won't have train throughput issues

[strange matrix]

just the one
three pure nodes overclocked to fill a mk5 belt each
going into 40 wagons, so 128,000 ore
travelling 10 minutes to a dedicated smelting factory that consumes 2340 (780*3) ore a minute.
So the train will fill up in just shy of one hour, make a trip to the smelting factory, drop its load and then return to the miners and wait there an hour before they are ready to do it again

[median heath]

[strange matrix]

there is something like 70,000 iron ore a minute available across the map
tapping 2340 a minute of that is a drop in the ocean. If I tapped every iron ore available, then thirty 40 wagon long trains will become a traffic issue

[median heath]

So the simplest solution I can see is if you're willing to bump it to 42 platforms. @strange matrix

[strange matrix]

maybe im just too long a factorio player, expecting megabases to be viable in this game

[strange matrix]

:monkaHMM: go on... ?

[snow dove]

contacted someone i know who's significantly better at math than i am, hopefully they have better luck

[strange matrix]

lmao, copy and paste emojis doesnt work huh

[wind spade]

megafactories are horrible here

build factories near nodes and transport things to storage

[strange matrix]

im starting to get that impression lol

[median heath]

42/3 = 16
So each belt feeds 16 platforms.
(Again you can Manifold or very easily split 1 line 16 ways using splitters)

[wind spade]

just build miner -> factory -> transport -> storage

[strange matrix]

in this case, that works out really nicely
but its not everywhere that I am going to so perfectly fill 3 mk5 belts with 3 very close together pure nodes :/

[median heath]

This was the only case I was working on 🤷‍♂️

[strange matrix]

then im going to end up with the same issues when distributing from storage, no?

[median heath]

Nothing should be distributed FROM storage. It is the end destination of all things.

[strange matrix]

yeah no worries, i appreciate it
this is why i want prio mergers, because it makes the system much more dynamic

[wind spade]

no, storage is final destination

[median heath]

There are many dynamic things that can currently be built. 🤷‍♂️
Prio mergers will most likely only ever be mod-only, and that's completely fine.

[strange matrix]

only storage I use is the crusher haha

[static zenith]

o.o

[median heath]

Sink is what happens to everything routed to storage when the containers fill.
Until you personally take from containers, then they refill until sinkage begins again.

[strange matrix]

finishing the last space elevator feels like launching my first rocket in factorio
new goal is to use every ore on the map, to max out the ticket generation in the sinks

i have no use for storage :p

[median heath]

Your new goal isn't possible.

[strange matrix]

its been done before

[median heath]

By whom?

[strange matrix]

gimme a second, let me find where i saw it

https://youtu.be/PfnSlsqyhgA

amelieofthesea, theres a world save available (not checked it out myself, but maybe i should to see how she handles trains)

[median heath]

And to see if your PC can handle it.
Most players' hardware would melt before achieving all nodes tapped.

[strange matrix]

im no stranger to UPS being the limiting factor of a megabase :p

[median heath]

United Postal Service?

[strange matrix]

updates per second

basically, the only limiting factor in a megabase size in factorio is how stronk your PC is

[median heath]

Satis has updates per year if you look at the time between U6, U7, U8 😉

[strange matrix]

to win factorio, is to hit your PCs limits

or at least, keep going until the updates per second are so slow you dont want to play anymore

[zenith willow]

which?

[strange matrix]

satisfactory has a finite number of resource nodes
most players would say the devs made 100 or whatever it is iron nodes spread across the map so you can access iron nearby whereever you are
i say, that is 100 iron nodes waiting to be tapped, refined, process, and sunk for the juicy tickets

[median heath]

All.

[zenith willow]

haha, that's what I thought too

[strange matrix]

https://www.reddit.com/r/SatisfactoryGame/comments/16ugm5k/alternate_recipe_ranking_wspreadsheet_update_7/

[median heath]

There is no objectively correct answer.
So pick whichever you feel like.

[zenith willow]

Didn't see this one, I only found a update 4 one 😄

[median heath]

Don't follow it.

Every single "alt recipe tier list" is subjective as fuck based on what the person who wrote it thought mattered.

[zenith willow]

hehe I guess that's also true

It depends on what you need at that time too I suppose

[median heath]

That, and on what matters to you, personally.

[wind spade]

see #math-and-meta message

[mighty shell]

one megafactory per biom is working great for me plus drones

[wary trench]

I am in phase 7 what should I do?

[wind spade]

"megafactory" means "single factory with everything produced in it", so by definition you can't have multiple megafactories

[median heath]

There are only 4 phases.
And do whatever you wish. That is how sandboxes work.

[wary trench]

I am in phase four

I am playing with farsi language

[mighty shell]

forgot i was in the semantics channel. in the architect channel mega is any factory that pulls from 10 plus nodes needs a train or drains a biom

[strange matrix]

@median heath I figured it out

[median heath]

👍

[strange matrix]

left containers represent the miners
the top containers represent the train stations

you can extend the number of stations out to the right forever

the miners will output at 100% capacity all the time

when the left most station is full, the throughput moves right a station