Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Jabor
Jul 16, 2010

#1 Loser at SpaceChem

Dirk Pitt posted:

I am starting to mass produce modules and my green circuit production is going to poo poo and can't keep up with demand from red circuits. Is the best way to scale green circuits to just keep plopping down two circuit assemblers and three copper wire assemblers until demand is satiated?

Pretty much.

But keep in mind that at some point you'll be limited by how much material your belts can move - especially on the input side. If you want two full belts of circuits, you need two belts of iron and three belts of copper dedicated just to circuit production.

Adbot
ADBOT LOVES YOU

Collateral Damage
Jun 13, 2009

My red circuit factory has its own green circuit factory so I just truck belt iron/copper/plastic in. You only need one green circuit assembler per 6 red circuit assemblers.

Demiurge4
Aug 10, 2011


That's good enough to reroll my game, goddamn.

GenericOverusedName
Nov 24, 2009

KUVA TEAM EPIC

Collateral Damage posted:

My red circuit factory has its own green circuit factory so I just truck belt iron/copper/plastic in. You only need one green circuit assembler per 6 red circuit assemblers.

Blue circuits, on the other hand...

Jabor
Jul 16, 2010

#1 Loser at SpaceChem

GenericOverusedName posted:

Blue circuits, on the other hand...

Funny story ... with how long processing units take to make, it only needs to be 1:1 there (not counting the ones that go into red circuits)

Collateral Damage
Jun 13, 2009

I do the same with blue circuits actually. Feeding 10 blue circuit assemblers requires 12 red and 12 green circuit assemblers. (10 feeding the blue circuit assemblers and 2 feeding the red circuit assemblers). Then just multiply from there until you're producing enough units per second.

e: I guess the lowest common denominator would be 5-6-6

Collateral Damage fucked around with this message at 13:04 on May 18, 2017

RyokoTK
Feb 12, 2012

I am cool.
Yeah completely self-contained manufacturing plants for blue chips are actually pretty simple to set up. You only need four ingredients: copper, iron, plastic, and sulfuric acid, so it's not like the logistics are complicated, and the ratios work themselves out pretty smoothly.

Tupper
Sep 1, 2007

Fun will now commence.
A bit off-topic but the same is true for modules. They only require all three flavors of circuit, so you can just train them in. You can continue extending this idea for a lot of end-products.

The idea of lots of smaller factories linked by train is actually what MangledPork/Bentham does in his Towns series on YouTube. Good series, as long as you don't mind watching 0.14 stuff.

almost-fake-edit: If you guys aren't using Helmod for factory planning once you get into post-initial/staging base, you should check it out. The UI is badly/not documented and it takes a bit to figure out, but I managed to plan literally every factory building with modules and beacons.

seravid
Apr 21, 2010

Let me tell you of the world I used to know

Dirk Pitt posted:

Is the best way to scale green circuits to just keep plopping down two circuit assemblers and three copper wire assemblers until demand is satiated?

No. With prod 3s you no longer need three two copper assemblers for each green, one is enough.

seravid fucked around with this message at 15:52 on May 18, 2017

rockopete
Jan 19, 2005

To whoever posted upthread about wiring belt sections to prioritize one splitter output over the other, thanks, you have saved me god knows how much sanity and iron that would have gone to complicated multi-splitter arrangements. I am also now acutely aware of how ravenous my factory is for iron and copper, mainly for green circuits. Converting from bus to rail is definitely on the horizon. Anyone have pics of a rail bus, out of curiosity? Which may be wildly impractical, who knows.

Also, are processing unit and high end module assemblers best given speed modules, or productivity? Or productivity + speed beacons? The build time is just so drat long that it's tempting to go all speed + speed beacons.

Truga
May 4, 2014
Lipstick Apathy
My suggestion to everyone is to put productivity modules into everything that can fit them as soon as you research productivity 1, as they'll triple your pollution and power draw to make your factory build stuff *slower*, which is great because you can then build a bigger factory for even more power draw and pollution.

Tuxedo Catfish
Mar 17, 2007

You've got guts! Come to my village, I'll buy you lunch.

Chev posted:

I'm surprised, I play on default settings and biters always leave me alone for hours.

Your taste in maps makes a huge difference. I'm pretty new myself but I've made it to the midgame without even seeing a biter on heavily forested + watered maps, and I've had attacks literally as I was setting up power for the first time, both on exactly the same settings.

GenericOverusedName
Nov 24, 2009

KUVA TEAM EPIC
Unless they've recently changed this, the different terrain types absorb different amounts of pollution as well. Trees absorb a bunch of pollution. The nice lush green stuff absorbs a decent amount. The dry desert areas absorb barely any at all, it may as well be concrete. So it can depend on what terrain you start out on.

Qubee
May 31, 2013




rockopete posted:

To whoever posted upthread about wiring belt sections to prioritize one splitter output over the other, thanks, you have saved me god knows how much sanity and iron that would have gone to complicated multi-splitter arrangements. I am also now acutely aware of how ravenous my factory is for iron and copper, mainly for green circuits. Converting from bus to rail is definitely on the horizon. Anyone have pics of a rail bus, out of curiosity? Which may be wildly impractical, who knows.

Also, are processing unit and high end module assemblers best given speed modules, or productivity? Or productivity + speed beacons? The build time is just so drat long that it's tempting to go all speed + speed beacons.

what on earth is a rail bus? the same as a belt bus but with rails instead? why would you even do that. just have a big unloading station dedicated to one resource for the least amount of headache possible. Or have one of those fancy filtering depots that can manage everything a train throws at it.

zedprime
Jun 9, 2007

yospos
I almost always start with a fair bit of trees and water but I've still noticed a trend of later and later biter attacks starting in .14. I just wrapped up blue science in my current game and just had my first attack. It was one single biter.

Dunno-Lars
Apr 7, 2011
:norway:

:iiam:



Loopoo posted:

what on earth is a rail bus? the same as a belt bus but with rails instead? why would you even do that. just have a big unloading station dedicated to one resource for the least amount of headache possible. Or have one of those fancy filtering depots that can manage everything a train throws at it.

I think it means dedicated railstations/outposts to produce things. So the railbus is pretty much trains shipping basic materials around, and shipping finished material to where it's used.

Mine all over, smelt one place, make green circuits somewhere else, maybe smelt and make gears another place, one for red circuits, one for blue. And one factory that make all the bits and pieces you need.

Qubee
May 31, 2013




For some reason I stupidly thought he meant he'd have rail tracks going down the middle of his base with trains running up and down them.

Filthy Monkey
Jun 25, 2007

That sounds fun too. Somebody should do that. Replace as many belts with trains as possible.

M_Gargantua
Oct 16, 2006

STOMP'N ON INTO THE POWERLINES

Exciting Lemon

Filthy Monkey posted:

That sounds fun too. Somebody should do that.

I do loops. Raw materials on the outer most loops, picked up from smelters by trains on entirely separate rail networks (saves on pathfinding UpS a little), then you get toward the interior where it's trains of red, blue, plastic and steel etc, with some belts carrying more specialized stuff inward

ToxicSlurpee
Nov 5, 2003

-=SEND HELP=-


Pillbug

Filthy Monkey posted:

That sounds fun too. Somebody should do that. Replace as many belts with trains as possible.

I've done that. You just have a big loop snaking wherever it needs to go. Well there's a ton of ways to do it but you have a bunch of trails just stop at all the stops in sequence. Filter stack inverters grab what they need while the train is there. You can set cargo cars to only hold certain things so output stops stuff them with whatever is there.

I don't have the last save I did it on but I vastly prefer trains to a belt bus.

Mercrom
Jul 17, 2009
What's the maximum throughput on a railway compared to blue belts?

Tenebrais
Sep 2, 2011

Mercrom posted:

What's the maximum throughput on a railway compared to blue belts?

A railway cannot exceed the throughput of the belts that supply it. But it can definitely keep up with blue belts, and for much less resources and much more flexibly.

rockopete
Jan 19, 2005

Loopoo posted:

For some reason I stupidly thought he meant he'd have rail tracks going down the middle of his base with trains running up and down them.

Well I actually was curious if someone had done that as a gimmick or challenge. I figured it would probably not be optimal or efficient.

The big loop concept sounds interesting, I may do that. Before moving away and making a new factory, anyway.

Evilreaver
Feb 26, 2007

GEORGE IS GETTIN' AUGMENTED!
Dinosaur Gum

Tenebrais posted:

A railway cannot exceed the throughput of the belts that supply it. But it can definitely keep up with blue belts, and for much less resources and much more flexibly.

A little more accurately, they have a maximum throughput onto belts of 1.5 or so blue belts per car on the train, assuming your destination station always has a train unloading (it should) and that you're using stack inserters to unload.

Also it takes about 5 seconds to load or unload an oil car, so that's 15,000 oil per sec per car per platform. :getin:

Evilreaver fucked around with this message at 22:25 on May 18, 2017

ToxicSlurpee
Nov 5, 2003

-=SEND HELP=-


Pillbug

Mercrom posted:

What's the maximum throughput on a railway compared to blue belts?

Way higher. Each rail can run a bunch of trains. You also don't have to worry about underground belt running to get poo poo across the bus. Once you run your rails somewhere you can just be all like "yo train, get this poo poo here, then get this other poo poo, then haul it to this other place 30 miles away, tia." If there's a rail and a station you're golden. It also doesn't take fifty trillion gears to make long rail lines. I haven't sat down and done the testing or the math but let's just say I've also never felt the need as I've never managed to completely load a rail line to capacity.

ShadowHawk
Jun 25, 2000

CERTIFIED PRE OWNED TESLA OWNER
How much mining drill productivity research do I need before drills outputting onto blue belts is a waste and they instead need to output directly into chests?

KillHour
Oct 28, 2007


Way too much. Each one is only 2%, so you'd need hundreds to saturate even a yellow belt.

ToxicSlurpee
Nov 5, 2003

-=SEND HELP=-


Pillbug

ShadowHawk posted:

How much mining drill productivity research do I need before drills outputting onto blue belts is a waste and they instead need to output directly into chests?

I don't think that's all that relevant as ore piles are finite anyway. If you have a few drills that don't output anything because the belts are clogged it doesn't consume full power. Eventually some other drill will run out and the stopped ones will start producing.

Thel
Apr 28, 2010

ShadowHawk posted:

How much mining drill productivity research do I need before drills outputting onto blue belts is a waste and they instead need to output directly into chests?

OK so a blue belt by default can cope with ~1200 items per minute per lane, or 20 items per second.

As per wiki, the formula for items/s is Mining time / ((Mining power - Mining hardness) * Mining speed) = Seconds for one resource item. This works out to (EMD on copper/iron) 2 / ((3-0.9)*0.5) = 2 / 1.05 = 1.90476 seconds per item, base.

If you put three speed 3 modules in, speed increases by 150% to 0.761904 s per item: (2/1.05)/2.5

So then we need to know how many mining prod levels, at 2% prod per level, would make either of those cases drop to 0.05s per item. Since I'm actually bad at math I'm just going to do steps of 50 levels of mining prod:

NB: theoretical limits only. Given that Factorio works in discrete steps (the simulation runs at 60 steps per second), it's highly likely that actual results may differ

code:
MP
level    Base EMD    3x S3 EMD

0        1.905       0.7619
50       0.952       0.3810
100      0.635       0.2540
150      0.476       0.1905
200      0.381       0.1524
250      0.317       0.1270
300      0.272       0.1088
350      0.238       0.0952
400      0.212       0.0847
450      0.190       0.0762
500      0.173       0.0693
550      0.159       0.0635
600      0.147       0.0586
650      0.136       0.0544
700      0.127       0.0508
750      0.119       0.0476
800      0.112       0.0448
850      0.106       0.0423
900      0.100       0.0401
950      0.095       0.0381
1000     0.091       0.0363
So, with 3x S3 EMDs, blue belts become a bottleneck (for a single miner) somewhere beyond the 700th level of mining productivity. Good luck.

Ignoranus
Jun 3, 2006

HAPPY MORNING

Thel posted:

OK so a blue belt by default can cope with ~1200 items per minute per lane, or 20 items per second.

As per wiki, the formula for items/s is Mining time / ((Mining power - Mining hardness) * Mining speed) = Seconds for one resource item. This works out to (EMD on copper/iron) 2 / ((3-0.9)*0.5) = 2 / 1.05 = 1.90476 seconds per item, base.

If you put three speed 3 modules in, speed increases by 150% to 0.761904 s per item: (2/1.05)/2.5

So then we need to know how many mining prod levels, at 2% prod per level, would make either of those cases drop to 0.05s per item. Since I'm actually bad at math I'm just going to do steps of 50 levels of mining prod:

NB: theoretical limits only. Given that Factorio works in discrete steps (the simulation runs at 60 steps per second), it's highly likely that actual results may differ

code:
MP
level    Base EMD    3x S3 EMD

0        1.905       0.7619
50       0.952       0.3810
100      0.635       0.2540
150      0.476       0.1905
200      0.381       0.1524
250      0.317       0.1270
300      0.272       0.1088
350      0.238       0.0952
400      0.212       0.0847
450      0.190       0.0762
500      0.173       0.0693
550      0.159       0.0635
600      0.147       0.0586
650      0.136       0.0544
700      0.127       0.0508
750      0.119       0.0476
800      0.112       0.0448
850      0.106       0.0423
900      0.100       0.0401
950      0.095       0.0381
1000     0.091       0.0363
So, with 3x S3 EMDs, blue belts become a bottleneck (for a single miner) somewhere beyond the 700th level of mining productivity. Good luck.

I think a blue belt is 20/lane, which is actually 40/second total.

RyokoTK
Feb 12, 2012

I am cool.
I think most people directly input into chests so you can save a tiny bit of space and have a really easy blueprint stamp for ore fields of any size, powered by bots. Belt bandwidth isn't the issue.

Thel
Apr 28, 2010

Ignoranus posted:

I think a blue belt is 20/lane, which is actually 40/second total.

That's correct - blue belt capacity is 20 per lane per second, which is 1200/lane, or 2400/belt. But each miner can only output to one lane, and the question was at what level of mining productivity does a blue belt become a bottleneck.

Duzzy Funlop
Jan 13, 2010

Hi there, would you like to try some spicy products?
It turns out I've never really played Factorio "the right way" after having watched a few YouTube tutorials for poo poo I wasn't quite sure of.

I've never used a "bus", even though some main arteries for raw materials / intermediate products have reoccurred in my setups.

https://steamcommunity.com/sharedfiles/filedetails/?id=754378586

Is that tutorial for buses something to go by? I'm mindblown by all the intricacies and mechanics, and I've never considered using splitters to balance loads, so all of this is pretty confusing to me. That Katherine of Sky has been popping up a lot for me on YouTube, but her blueprint design for green circuits also struck me as really weird before someone shat on it in here, so I'm not sure if she's a good source of information.

Any basic advice, or is that guide solid?

Sininu
Jan 8, 2014

Duzzy Funlop posted:

but her blueprint design for green circuits also struck me as really weird before someone shat on it in here, so I'm not sure if she's a good source of information.

They were complaining about blueprint in the shared Google docs table that was also linked in that post.

Duzzy Funlop
Jan 13, 2010

Hi there, would you like to try some spicy products?

SinineSiil posted:

They were complaining about blueprint in the shared Google docs table that was also linked in that post.

She appears to have the same one in her shared GoogleDrive blueprint library.

https://drive.google.com/drive/folders/0BwVHGs2mds_XRUVCekwxYnNldnc

Is her bus tutorial solid otherwise?

REDjackeT
Sep 2, 2009

Duzzy Funlop posted:

It turns out I've never really played Factorio "the right way" after having watched a few YouTube tutorials for poo poo I wasn't quite sure of.

I've never used a "bus", even though some main arteries for raw materials / intermediate products have reoccurred in my setups.

https://steamcommunity.com/sharedfiles/filedetails/?id=754378586

Is that tutorial for buses something to go by? I'm mindblown by all the intricacies and mechanics, and I've never considered using splitters to balance loads, so all of this is pretty confusing to me. That Katherine of Sky has been popping up a lot for me on YouTube, but her blueprint design for green circuits also struck me as really weird before someone shat on it in here, so I'm not sure if she's a good source of information.

Any basic advice, or is that guide solid?

It seems like a good starting point and most of that is what I generally do. I personally bus gears as adding central production feels easier to me then tweaking onsite, and I'll partially bus sulfur (not acid), stone and/or brick to get them to the area they need to be in, like a military construction zone.

seravid
Apr 21, 2010

Let me tell you of the world I used to know

Duzzy Funlop posted:

I've never used a "bus"

And you don't have to!

ShadowHawk
Jun 25, 2000

CERTIFIED PRE OWNED TESLA OWNER
If they add that start with bots mode there should be an achievement for launching a rocket without placing a conveyor belt.

KirbyKhan
Mar 20, 2009



Soiled Meat

Duzzy Funlop posted:

She appears to have the same one in her shared GoogleDrive blueprint library.

https://drive.google.com/drive/folders/0BwVHGs2mds_XRUVCekwxYnNldnc

Is her bus tutorial solid otherwise?

Oh yeah, she's fabulous. Very Midwestern cat lady that feels right at home with a sewing circle making lord of the ring themed embroidery. Her sperg level is not as high as Xterminator and his crew, so that's why I watch her. Her info and layouts are rock solid, very well documented and they work at a high level of efficiency.

Adbot
ADBOT LOVES YOU

Toast Museum
Dec 3, 2005

30% Iron Chef
Is this the green circuit blueprint people were complaining about?



What's the problem? Is that not the right ratio with expensive recipes? Not compact enough?

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply