Early Building Store

[Breith]

That's funny... The reason I don't want to go in too much details on the maths is because I'm also a scientist. Do that every days at work and for the game would be annoying... It's funnier to just try and see .

So, well, I agree with one point: this design should be used until we need red circuit, so we need to include Green Science production. So I did and actually it works pretty well. The number of assembly machine should probably be increased as well as the number of labs if we start to use this early base for something more complete than just assemble basic stuff but it's not really an issue. The iron output is still too low, but I switched off the long/fast/smart inserters production line by inverting the direction of the splitter after the second inserter assembly machine and I turned all gears inserters away from the belt aside of the yellow belt assembly machine. Now, the gear production is enough and actually I'm probably stacking some. I could easily either stole them for my personal use or directly feed other assembly machines manually in function of what I need (e.g. long arm inserters).


Just an advice for the next step: do not try to bus gears. It's a waste of time and iron, belts (all colors) do not have the throughput to sustain the gears usage for a whole factory. It's way faster to assemble some locally.

PS. I checked. The gears outflow is not enough, due to a lack of iron. But if you shut down the 2nd pair of green circuit production, the iron supply is enough. I probably need more time to be sure of that, but I assume it's a good alternative: with all inserters and belt depend products not assembled anymore, 4 green circuit assembly machines are overkill.
PPS. By changing ~ 5 stone furnace to steel furnace, the production of gears is stable and the assembly machines that make assembling machines can be turned on without consequences on the belt/inserter/science production. The production is however really low due to the lack of efficiency of the inserters. I could increase their number, but I fear it is going to bleed too heavily the gear line, negatively impacting the belt production and thus the green science production.

[Qon]

I tried red/green sci on the same production line but it bogs down the store production a lot. It does bleed off some of that darn copper though

Don't try that with that design. It is clearly not made to produce anything beyond green circuit. More classical smelting area + bus is way more efficient. If you want your design to be effective, you need to accept that it won't survive the 4th hour of game.

It can probably last for much longer. But you will have to make a dedicated science factory so this design isn't starved. The things you produce here don't require that many assembly machines, and putting them on a main bus later might not be needed and will make it less compact. Just upgrade the belts and you can probably also use it with some red circuits later.

Just an advice for the next step: do not try to bus gears. It's a waste of time and iron, belts (all colors) do not have the throughput to sustain the gears usage for a whole factory. It's way faster to assemble some locally.

Gears are twice as efficient to bus as iron though. Iron plates takes twice as much space since one gear is produced from 2 iron plates.

[Dry Hairy Tree]

OK. So I'm trying to take all this in and produce an environment that works in and out of the sandbox. The original post was aimed at a building store but it has become obvious that what occurs post building store (smelting, science) is important for longer-term game dynamics and shouldn't be scrimped on, plus defense should be (relatively) immediate.

You need to read the three photos from the bottom up - I have to go so no time to edit them in order (silly website reversed my upload order). More pending...

[Dry Hairy Tree]

Now I think the system can support a store, I can automate a few things, upgrade and expand smelting, automate some more, by then the oil researches will be through. I do wonder if I shouldn't have automated belts right at the start and then replace where they're made later.

I do like your smelting design Breith, it was just beyond the scope of what I practically needed that early. As it stands I found it very resource hungry to put up fast. More hand crafted iron! I'm certainly due for some smelt-retrofitting soon.

I think another thing to consider for an early store is belt storage. Say you don't want 50 mining drills made (one stack in a restricted box), you could store 3, 6, 9 on belts... (don't quote my numbers) and a couple in the assembler itself... This might be very useful if you want a wide complement of products without such a drain on a pre-upgraded system. The problem will be with aesthetics.

[Blurb]

The throughput of ore is one mining drill to one smelter - this is double what it needs till I upgrade smelters then it works just fine. As for being insulted, you can be insulted if you want.

This is the first attempt at the concept and it works great, I've since improved it a fair bit but I'm still running ore on single lines and when I want more I'll add it. When I empty all the boxes yes, the iron is low for a while, but I just scored 50 buildings, 200 inserters bla bla - I don't need it to be instant. Sheesh.

EVERY build I make differs, so if you got some perfect recipe you follow - completely sad - to have a creative game and then force yourself to a regimen. Come on, tell me more how I should play the game.

You might as well cover your ears and scream LA LA LA I AM NOT LISTENING.
Your opening post states that other existing builds are wasteful of both space and resources - then you throw a fit when I point out the glaring flaws in your proposed alternatives.
Not only do you come off as a hypocrite, but blankly dismissing criticism is further detrimental to any improvement of your design ideas.

Here's a picture to illustrate my proposed smelter setup.
Note that I have only shifted around existing parts (lamps omitted by arbitrary preference) and added two splitters - and in doing so doubled the potential output.
An additional selling point is that one does not need to later separate copper and iron plates, as they're placed on their own belts right off the bat.

I hope the following is obvious to all (but better foolproof, right?); the splitter on each output should be placed so that there's an equal number of smelters before and after the splitter.
The attached picture only shows 4 furnaces before and after the splitter, but yellow belts could easily support more than 40 total stone furnaces (making for 20 furnaces before and after each splitter).

[Dry Hairy Tree]

So, in the first post you say you're insulted. Then you post again to inform me I'm a hypocrite.

Check your approach, better still check it at someone else's door. This is a game, but you are not fun.

[SpeedDaemon]

For your circuit production area, you can actually eliminate the smart inserters to allow building with very basic tech. Move the underground belt entrance one tile east, and place a splitter north of it. The underground belt will block the copper and let the iron through. Obviously, this requires the iron to always be on the far lane.

[vanatteveldt]

For smelting, I would probably merge the coal with the ore, so you can get rid of the two coal lines. Since you have only one output lane per plate type, you also only need one input lane (as 1 ore -> 1 plate), and even less for coal.

I have to admit I don't automate most early production. I automate gear and circuit asap into a box of one or two stacks, but if you have those in hand making most things is really fast. When I build green science I just add an output box for inserters and belts so I can pick those up quickly, and when I start with oil production I set a pipe plant somewhere (as had crafting 100 pipes is just not fun). Anything else (signals, assemblers, blue inserters etc) are extremely quick to hand craft if you have the base materials.

Of course, in the context of a 'no crafting' challenge or playstyle, setting up an iron+circuits line makes a lot of sense. I normally don't add a gear line, I just produce gears where they are needed, as they are generally needed in very high quantities, but that is a matter of preference.

green science

inserters

Note that this is a death world, so space is at a premium until I get the tech to claim some more territory, hence the overly compact design... As you can see I chose to keep iron and copper together, but until blue science it might be fine to keep them together, maybe I'll try in my next attempt

[Breith]

OK. So I'm trying to take all this in and produce an environment that works in and out of the sandbox. The original post was aimed at a building store but it has become obvious that what occurs post building store (smelting, science) is important for longer-term game dynamics and shouldn't be scrimped on, plus defense should be (relatively) immediate.

Err... No. Just no. With this design you're losing everything that was cool with the original one: have everything produced in a line, which make is easy and quick to create. Even if you do not used the advanced production (long/fast/smart inserters, underground belts, splitters) at the beginning you can easily switch on the production by increasing the iron output from the smelting area. With the new design all of that is abandoned and your design has nothing really interesting anymore. It's just a basic production line.
But yes, the coal and ore lines should be merged. I usually do that, I have no idea why I didn't think about it here.

[Breith]

Double post \o/ .

I wasn't happy with the smelting area, so I redesigned it to merge coal and ore lines. I was then able to add an additional copper furnace column (I'll come back on that later). This is way better with only stone furnace and way worst with only steel furnace. In the first case, the iron plate outflow is almost enough to provide iron to the whole assembly line, while in the second case yellow belt can not provide ore to all furnaces... But too much steel furnaces should not be needed with this new design, so I guess it's fine.

However, I realized that with the whole assembly line running at almost full speed, the green circuit output is not enough. So I added another green circuit assembly machine complex... and my iron plate supply was then too low. I chose to increase the number of furnace instead of upgrading to steel furnace, it's easier in early game and I have the additional space, so...
However, at that point I realized that even if this design is not destined to survive for too long, the lack of scaling ability is annoying. With the help of my dears robots (I already said I'm lazy...) I then moved some stuff. This is the new version. It does not really change anything, I just reorganized the green circuit and gears productions. Now, the green circuit output can be easily increased by adding new assembly machine and same goes for gears and red science (follow the black arrows). Even the number of labs can be easily increased (follow the yellow line).

Let me know what you think about it, but I think it's one of the most compact design we can do. Of course the horizontal space between main assembly line and lab can be easily reduced. I do not think that 2 lines of Labs are useful. I didn't do it here because in case of I needed that place later while working on the design.

[Dry Hairy Tree]

Good call Breith. I thought it was winding up just looking like a main bus too. It amused me to play along and watch my creativity blend into the homogeneous thought of people who 'do it right'.

I do like mixing the ores and coal, makes a lot of sense. The wee splitter turning the lines double - love it, so tidy and simple. Simple is doing it for me lately.

Had no game time a few days got a presentation next week and I barely know how to talk to these people as the field (biogeography - am actually a plant/fungi/microbe/ecology person) is rife with jargon and silly models. As it's also evolutionary theory it gets worse.... So I'm trying to simplify a ton of horseshit to say - What drives evolution. Yeah, it's a biggie.

But all this thinking about the presentation of evolutionary process brings us right back here to Factorio - where purifying selection almost knocked out a unique mutation.

The sympatric encroachment of extant alleles almost overwhelmed the apomorphy arisen at their periphery.

See - I've been practising horseshit!

[Breith]

My next attempt will be to move the assembly machine line a little bit further south and to reduce the vertical spacing. We never know, it might be useful to increase the number of furnaces and it's really easy to do so by expanding east (copper) and west (iron). I realized than even if you want to go to main classical "bus" production, it might still be useful to keep this production for a while. After all, aside of the space it takes, you do not really need to take it down until it's really not useful anymore. Yellow belts are quickly not useful, but it's not the case of long and fast inserters. It might just be a good idea to just shutdown part of the line.

Good luck for your presentation, I know how stressful it is... especially when you do not know what you're going to talk about and/or when you realize that your results are wrong (I am doing Arctic Physical Oceanography).

PS. Ahahahaah when I started my tests, the game clock was at 8h. Now I'm at 12:30.

[siggboy]

That smelter setup is giving me an aneurysm, and tauting it as good is downright insulting.

This is exaggerating. It's not an efficient layout but it does not perform any worse than an optimal layout would. The worst thing he does wrong is using separate coal lines (and one too many, even), wasting a few belts which might hurt slightly early on, but that's about it.

You can do away with the coal belt entirely by mixing each ore belt with coal (creating a 1:1 coal/iron belt and a 1:1 coal/copper belt) and feeding those on both flanks of the furnace column. Then simply output iron/copper plates to a mixed belt in the middle.

Two belts saved and the setup becomes two tiles narrower. You don't even need splitters to create the mixed belt, so you can do this really early if you create two separate coal belts somehow.

Also with such a layout it's enough to have two inserters per furnace (one for the input side and one for the output). Anything more does nothing at all and is pure waste.

[siggboy]

Here's a picture to illustrate my proposed smelter setup.

That setup is not much better than his.

[Dry Hairy Tree]

I understand there is some (unwritten) call for efficiency, but really, the obsession with X amount of resources per minute is only justified if you are actually using said resources.

Here's a week of my life I'll never get back 50 hours of pure creative fun. I have one way and two way train systems working in conjunction (the 'civilian' trains never enter the military zone at the bottom but the military trains can go anywhere). One hold-up in 50 hours (that I caused). Robots run my military (mostly) but everything else is yellow belts - yellow!

It took 30 minutes to assemble/launch a rocket, it's good enough.

I had a bunch of aliens stuck on islands. most excellent rocket practise!

[Breith]

I started a new factory with 0.13 so I used the layout we discussed here as a base for my factory. I made several changes in the layout to be more practical. I took some screenshots, I'll show you that as soon I'm back from work.

I really think that this begining was one of the best I did since I started to play Factorio. Of course it wasn't perfect, and the layout for the Green Circuit should definitively be changed to get rid of the filter inserter (new name of the smart inserter). Otherwise, the lack of iron is not as much an issue as I though previously (based on 4 columns of iron furnace, not one). Of course, we do not have enough most of the time, but by a reorganisation of the assembly machine you can prioritise what you want first (basically belts & inserters). It means that you'll need a few more Assembly Machine, but it's not really a big deal if the compensation is to be sure that you'll never run out of stuff.

I know this post is a little bit empty, but I wanted to let you know that since a lot of people are probably currently starting a new factory. More details as soon as I can post the screenshots. I wanted to do that yesterday, but when I looked at my watch it was 1:30 in the morning. For a week day it was clearly not clever to spend 30 more minutes to write a fully detailed post .

[Qon]

Also tried this out a few days ago before the 0.13 patch. I implemented my own suggestions and it is working out really well.
With my changes it's fast enough for blue science at steady pace (not doing science on the building store!) and I'm almost not crafting anything by hand at all because it is so easy to add a new machine.

It's organised enough that I've researched the first blue science upgrades to bots but haven't crafted any because I haven't needed them and they don't seem that important when I already have machines crafting anything I need so quickly, so organised and so compact that it is comparable to a bot factory.

A little bit of spagetti is still there because I added red circuits and other mid game stuff ad hoc because I didn't think it would last this long. If I planned this out with all end game products in mind then I think it is good enough to supply you up to rocket launch at least. I really prefer this over a main bus now. Those just waste space like crazy when you try to do everything on them and it's just a bit too much work to split of resources from the bus to each production facility.

How to make a successful building store (applies a bit to main bus as well):

• Don't do science on your building store. Do a dedicated science build like this instead: https://www.reddit.com/r/factorio/comme ... k_compact/
  True for all the things you make tens of thousands of or more. Solar and accus, science packs, bots for bot factories, blue belts for a mega belt factory. And intermedidiates like all circuits and gears.
• Don't waste your building store belt space with copper.

[Dry Hairy Tree]

Most excellent replies good people. New patch huh, what's new? Guess I'll go take a look. I don't even know the official release date but - I can't wait anyways!

Yeah for my playstyle a building store is way better than a bus. I'm not fixing to get achievements, I'm fixing to have creative fun. If achievements are involved along the way, all good.

Yes, it does appear science has to be separate unless you are building a main bus. Luckily we're all pretty good at putting red/green together fast by now. I hate crafting science packs

Arctic Oceanography Breith - I have some interest there in that I'm a fan of the extremophile communities found in polar/deep sea habitats. Microbes kick ass. Hey I've just discovered Elephants (but not hyrax so occurred after last known branching of Afrotheria) have a missing codon in their cyt-B gene. No idea what it means except I found an elephant specific (and mammoth, mamut) marker.

This game needs a boat - you can examine the fish and I can shoot rockets at onshore aliens. Winning!

[PetrGasparik]

What a great idea!
I am just learning this game (and trying to educate others by making videos, in Czech) so tips like this is precisely what I need.

Thank you, sir!

[Dry Hairy Tree]

Thank you Petr.

Alternately, you can just put red science up and start on this if you want earlier belts and inserters etc on demand. I actually smelt my first iron (from an electric drill) to go straight into gear->belt assembly then I break that down and add it to the building store later when I get to it. Belts are very handy automated very early game, my highest demand.

So now I make smelting, automate belts, and then red sci (with room to spare for green), then more smelting and a building store (move belt assembler if i need to, make inserters etc), some bricks and iron... then I go back to finish green sci, and continue expanding store (underground pipes, assemblers, long inserters, etc).

It flows a bit easier then. At that point you should be able to build relatively fast with stacks of things in your inventory. Don't forget to limit your chests or you tax the lines too heavily and might have way more than you need of some things and nothing of others...

The secret to the success of the store is the gears/circuits and the order in which you make things. Assemblers should go late on the line or they'll tax everything else till smelting is upgraded. How many gears and circuits you have depends on how much stuff you want to make/can make according to where you're at in the research tree.