The kilobase

Clever and beautiful constructions, bigger than two chunks
- Defense: killing biters as an art
- Castles, Throne Rooms, Decorations (comfortable living in the Factorio World)
- Main Bus Concepts
- Modular Systems, Factory Streets, show how all works together
- Megabases
Please provide us with blueprints or saves, if that makes sense of course.
Forum rules
Clever and beautiful constructions, bigger than two chunks
silverkitty23
Fast Inserter
Fast Inserter
Posts: 117
Joined: Wed May 11, 2016 6:52 am
Contact:

The kilobase

Post by silverkitty23 »

It's not good enough to be a Mega base, but it's about as decent as I can get it in vanilla 0.12. It wasn't always vanilla - you'll note the huge "explored" area, because I was running RSO Radar for a bit. I also had a bunch of nixie tubes showing quantities backed up for iron/copper/stone/coal/oil, and I had enhanced map colors, colored concrete, and all those sweet blueprint mods: string, flipper, upgrade planner, targeted deconstruction). But with 0.13 so imminent I decided to vanilla-fy my base lest I be reliant on a mod that doesn't work in 0.13.

You'll note that other than RSO Radar, I didn't change the game really - just added some color and display stuff. I wanted to see what the default experience was. The default settings aren't too bad except: 1) not enough stone - there's never a patch worth running a train to. 2) too many biters - it's a sea of red out there and it gets ted.i.ous.

I added RSO Radar much later and regretted it pretty badly in terms of filesize and depression over the huge sea of red. Around the same time I added colored concrete, which is awesome, and the blueprint mods, which really should be in the base game because life without them kinda sucks now, and nixie tubes, which are totally sweet. I also did "enhanced map colors" and despite its simplicity, it really, really, really enhanced my day - another candidate for the base game.

Anyway: on with the base tour:
Iron Unload
... copper unload is a duplicate on the other side of the smeltery. Someone once asked someone else (I was neither of these people): "why are all your trains sitting in stackers when they could be busy unloading?" I tried to address this by having 8 unloading stations and no stacker. But the experts (if any are reading this, which is unlikely) are already ahead of me and know what I'm about to say:
This plan was an utter failure. Because the trains always choose the nearest of the 8 unloaders, so I had 1 or 2 stations backlogged with ore, and 6 stations sitting empty, and 12 empty lines into my 16x16 balancer. Sadness.
In v 0.13 I can fix this with combinator tricks (I could fix it in 0.12 with the smart trains mod, but I just explained why my base is vanilla at the moment).
The 'real' fix is, of course, to do logistic bot unloading, so that I don't care if the trains prefer one station, and I don't have to sort them by ore (and I no longer have to worry about miners which sit on two resources out in the field), but that's a different base. Not the one I'm trying to make. In 0.12 I hacked around it by having my trains cycle stations, but that's a workaround I only used because 0.13 is so close.

Moving on:
Smelting
Just imagine huge amounts of this. Pretty standard stuff. The splitters are so I don't have as many idle furnaces at once - also the lanes are 36 long, so in theory those 72 smelters can compress my copper and iron belts. The stone paths are "reserved for beacons" space - though if I go beacon, the lanes will be too long and I will have to shorten them from 36 to X, where X is a number I don't know.

Next stop:
Basic Greens
Ho hum, the same 3:2 green build everyone uses. I didn't reserve space for bacon because if I use beacons, the ratio approximates 1:1 more closely and so I would need a total redesign, anyway.

Speaking of basic stuff:
Red Battery Blue
Pretty standard fare here, too. By now you may have a noticed a trend in my "designs", but I'll get to that in a moment. For now I can say: the blue circuit build would be wholly inadequate for multiple rocket launches. It drains pretty quickly if I build modules. Not nearly as quickly as the red circuit build, though. I have 64 assembly machines making reds (the setup you see there goes to a length of 16, thus 32. Then is mirrored on the other side of the main bus) and they empty quickly and refill slowly (not for lack of resources, though, just for being slow). It is sad making how quickly all my reds fly away.

And finally for the basics:
Sulfur Acid and Plastic
Zzzzz. Probably too much sulfur and acid for the batteries+blues, but whatever. That plastic build can keep up with the reds as long as I don't run out of oily.

I also have belt-based builds for ammo, which looks like batteries without the liquid import, and concrete, which looks like batteries. And a smaller smelter for stone bricks.

For the rest I went with your joe-basic smart factory. But we'll get back to that. The trend in my belt-based builds was this: ||-> belt, inserter. There is nothing else. No inserters loading from or to corners, or underground belts, or splitters, or even facing belts end-on. No mixed belts. No long-handed inserters. Sure, I may spend more resources loading up the belts, but it's a one-time cost. Even if I empty a belt and have to refill it, it means I used that much resource, so it would have been spent anyway. The "belt buffer" only has to be filled once. And one-time costs depreciate to negligible over time. But that's not why I do it that way. I just don't like/trust anything but ||-> . Probably an aesthetic thing. You may have also noticed (especially if you consider "standard" builds for green circuits) that I don't buffer-chest. Which brings us back to the smart factory.
Smart Factory
Why have my bots running around all over the base, when I can bring everything to them? My 'smart factory' is probably faster than the average 'smart factory' because my bots move less and thus recharge less (and thus I don't need a million roboports covering basic traffic lanes around the whole setup). And if I'm gonna have a 'central storage', then it might as well be convenient to the smart factory as well.

I have more to add about the refinery, oil load/unload, details about the combinators on the left side of the smart factory, engineering trains, and my walls, but I'm super tired and not sure anyone is interested, so I'll snooze for now.
silverkitty23
Fast Inserter
Fast Inserter
Posts: 117
Joined: Wed May 11, 2016 6:52 am
Contact:

Re: The kilobase

Post by silverkitty23 »

part 2

So now we get to the refinery, one of the only places where I considered ratios while designing. ...and I probably got the ratios wrong. Because that's what happens when I try to work out the maths in this game (yeah, I got a bachelor's degree in math, but that was 1992, eek. I've had plenty of time to forget everything I don't directly use in web design).
Refinery
It's probably overblown - and certainly the production/speed modules seem like overkill. But overkill is better than underkill? I think I can feed those 2 plastic lines and the sulfur line, and that's really all I need. In theory, I have it set up so the east two refineries (and not the other five) contribute to lube and fuel when I'm short on those (and also happen to be spending petrol on something else), but it's hard to tell because I don't spend much petrol at the moment.

Speaking of overblown. We have the oil unload stations.
Oil Unload Disaster
I once tested this by filling a train with 300 barrels (the max, because each car is half oil, half empty) and driving it into the station, and it unloaded all 7k oil into the pipes and returned half the barrels to the train in the 20 second I have the train set to wait. So that's good. Except I never bring in a train with 300 barrels in practice. Default map settings, insufficient expansion, and the oil-outpost was the very last thing I designed (my old base was on top of 20 oil spots) all conspired to make it so I get like 30 barrels when my train sets out. So I don't need the second loader, and I don't need most of the first loader. But, hey. Overkill is better than underkill?
In this "design" for an oil unloader you'll find one of the rare exceptions to my ||-> rule. I have inserters taking from the end of belts instead of the side. Sigh (the other exception being that I load ore onto corners a bit in all the iron/copper/coal/stone unloaders... plus, another big exception area I'll get to in a bit).

I guess while we're on oil, we'll step out of the base a bit and look at the oil loading stations (I have two, one hooked to the aforementioned 20 oil spots from the original spawn location/old base, and one hooked to 5 oil spots up northwest).
Oil Loader
You might think I would feel more confident of my designs as time progresses, but this, the last build I made for my base, is the one I'm least confident of. I know it's overblown for the amount of oil I'm getting, but it also has a persistent issue with half the chests filling up the other half not. I'm using Mad Zuri's "fill the chest" balancer, but the middle chests always get screwed out of content with it. I believe his design requires high through-put to actually function as intended.

Okay, back to base. This is the "map" of my base
Base Overview
The thing I'm proud of here is that my logistics bots can "mostly" live in the smart factory and engineering section. If I end up with some ore or coal or plastic on hand, they have to travel out to those unload-stations to drop it off, but otherwise the furthest they ever go is transferring barrels to the oil-unload station. Which is a lie. The real furthest they ever go is when I'm coming back home after blasting some biters, and they try to catch me at the corner of my base to restock my destroyer capsules while I speed along in a train going right back to where they started. I wish there were a way to 'turn off' my logistics demands temporarily (just until the train brings me back to the center). Other bits I like include: I cleared a path through the mainline for running around the base. I got a train to the center/control panel for myself (this is really for my own health and safety trying to leave the base - running on the tracks is a bad idea). I have exactly the minimum roboports to cover my area - even over/through the RR tracks (except in the smart factory where I have a bunch more). My mainline ends, but it's not "capped" by any buildings because I placed stone brick where it would go to 'reserve' that space, and my belts act like the stone brick is full of mainline even where it isn't (yet).

I'm pretty unhappy about the distance from plastic to reds, but then I'd need to rearrange coal and the refinery, and blah.

Speaking of the smart factory, let's take a closer look
Smart Factory Detail
I have 9 constant combinators which represent my control panel - here I basically dial the "demand" for various products... and by demand I pretty much mean buffer storage (ick!) but the numbers I tweak the most are for modules and capsules, so for them it's less about buffer storage and more about what I think I need to spend in the near future. The 9th constant combinator controls whether bots go into service or sit in chests (the indicated lights next to it are "on" if I'm dumping bots into the network).

Because in 0.12 there isn't a way (yet) to get logistics data to mix with circuit data, I have red wire running to every passive provider and storage chest on my base (which are all conveniently right around the smart factory, so it's not that bad). The 8 "controller" combinators come together to one pole, which feeds an additive combinator the result, which it multiplies by -1 and adds to the total from all the passive providers and storage chests. So that output is "stuff - demand". So all my smart inserters have one setting: "if thing < 0, go" - because if thing < 0, it means demand exceeds storage.

That used to be the end of the story for my combinators, but I recently added a decider to the mix "Each < 0, send Input Count" - that isolates only the things where demand>storage. It feeds another "each times -1" combinator which comes together to the pole (highlighted in the picture) right behind my control panel, where I can see Just the things that my factory should be making (in this case, though, it's not *really* making anything because it can't make fuel or lube on demand - they are byproducts of needing sulfur or plastic elsewhere).

Then, just to make life annoying for myself, I add demand back to stuff-demand to get "stuff" back on a green wire, so that a little bit northwest...
Percent Display
This section made more sense when I had nixie tubes installed. I had a display of how much ore I have on hand, and then I combine that with the ore my station could possibly hold (which is a number I calculate by hand and stuff into the constant combinator shown in the picture), and thus I can figure out the percentage of ore I have versus how much I could have. Which tells me either I need more outposts, or I'm doing fine. A string of lights (you can only see the stone one in the bottom-left of the picture) are on if the percentage is <33 or >66 (via those decider combinators) - which made more sense when I had the colored concrete mod, because <33 was red and >66 was green.

If I had red ore, then I'd need to make a new mining outpost. Since I'm not using mods, I'm not using FARL, which means I grab my rail blueprints, and a bunch of rail and signals, and wander off in my conveniently-right-there train to some corner of the map, lay out a new line to an ore patch using my personal roboport, and then... ugh. Who has room to carry all that mining stuff? And my PAX train has no storage. What am I to do?
Engineering
Good thing I have a half-dozen trains on stand-by, ready to deliver what I need. I just plop down a station, name it "Summon [need]" and here come the goods. So in the previous example, I "Summon Mining" (if I forget which wording I used, I can consult the map and look at main base to remember 'Mining' is the magic word). Place my mine, and then send the train back home, where the requesters fill it back up, and my logistics bots fill the requesters back up from the smart factory that is literally right next to engineering. So wonderful.

Okay. Tour is almost over. Let's step back out of the base again and look at the world I've lived in for 500+ hours.
World Map
Base, huge piles of solar north and northwest of it. Old base west/southwest of it - the last remnant of that is boxes of coal (I wasn't buffering coal, but there was a big coal patch in the middle of where I wanted my new base that I felt obliged to drain and move before I built on top of it). A couple mines (default settings. All the RSO people are groaning at how easy I have it) and walls. Oddly, the walls don't really get used. I've been practicing "the best defense is..." on the biters for a bit and don't really aggravate them with my pollution. But IF I aggravate them, they have zero chances because
Wall Segment
Trains deliver repair packs and ammo, logistics bots distribute ammo, construction bots repair turrets, guns do more damage than lasers when fully upped, but lasers outrange them so both my lines engage enemies (actually, the lasers strike first and the guns mop up the behemoths) and the worst that ever EVER happens is one gun turret needs a little repair. I've never lost a wall or a turret, which is why my restock trains don't carry them anymore. There are other ways to arrange the reload of guns, but I like this one because the biggest problem with gun turrets is that '10' is an absolutely stupidly low number to be their ammo count. If they run out, they probably need the replacement ammo now rather than later, so I don't want to wait for belt loading (no stack bonus) or for ammo to cycle from another turret (because the nearby turrets are probably also low on ammo from firing). So everyone gets a box. The roboport is arranged to cover the boxes only (to minimize danger of losing the port. Which seemed like a bigger possibility before I saw the wall in practical use for 100 hours or so). Anyway. No biters get by. Even when I hork off a huge base full of them, realize I'm out of destroyers, and run back to the wall dragging 10x as many biters as come in one of their usual raiding parties, the wall brushes them off like nothing. Regular raids don't hold a candle.

Okay, I think that covers everything. Here's the save file: http://silverchat.com/~silver/factorio/kilobase.zip . It has all my rail blueprints, which are the part the work best, and all my other blueprints, which are mostly scratch work and don't represent the latest state of my builds... Experts have little to learn here (though maybe they can offer me some advice/help. Except none of them are reading this), but beginners might have some tricks to learn, maybe?

Oh. Wait. Right. I keep mentioning I have one big exception to my "||->" rule.
It's science. Yeah. Sigh. In science you will find my only mixed belts and long-handed inserters and even a place where I side-feed a belt. Also, spaghetti. Ugh.
Blinded Me with Science
The biggest thing to note here is that I'm not doing 5/6/12, I'm doing 7/8/16. What up with that? You wouldn't notice normally, but somewhere around follower-count 13 or so I discovered that 20 labs with Lab Efficiency 4 will outpace the assemblers. Add about 25% assemblers and all fixed. So if I were starting a new game, I'd start with 5/6/12, leaving just that extra 7 spaces right of the green-science assemblers for expansion, and then add the other 2/2/4 assemblers once I do lab efficiency 4.

Okay, that's the kilobase. It's no megabase, but I suspect it's good enough to make a megabase with. Actually, I suspect parts of it (like the smeltery) are already as big as they need to be, but I'm not sure. Maybe.

This was my third game of Factorio, but really more like my 7th because of the number of times I scrapped and restarted my base. I've got some errors I think I can fix with 0.13 or mods, and I might have some signalling issues with the "archway" rail segments that feed the iron/copper unloaders, but this is as good a base as I think I can make for now.

Enjoy,
-silver
Qon
Smart Inserter
Smart Inserter
Posts: 2164
Joined: Thu Mar 17, 2016 6:27 am
Contact:

Re: The kilobase

Post by Qon »

silverkitty23 wrote:If they run out, they probably need the replacement ammo now rather than later, so I don't want to wait for belt loading (no stack bonus) or for ammo to cycle from another turret (because the nearby turrets are probably also low on ammo from firing). So everyone gets a box.
1 fast inserter gives 2.31*5 ammo per second. 1 fully upgraded turret uses about 3 ammo items per second. You can get a pretty compressed setup with chained turrets without any of them ever running out. At 0.13 the rapid inserter that grabs 12 items per move should make chaining turrets even more effective.

Would be nice to see production and electric grid stats.
My mods: Capsule Ammo | HandyHands - Automatic handcrafting | ChunkyChunks - Configurable Gridlines
Some other creations: Combinassembly Language GitHub w instructions and link to run it in your browser | 0~drain Laser
silverkitty23
Fast Inserter
Fast Inserter
Posts: 117
Joined: Wed May 11, 2016 6:52 am
Contact:

Re: The kilobase

Post by silverkitty23 »

So I went to the control panel and said "you know what, I need 200 more of each type of Module-3, 1500 more blue belt, 200 more destroyer capsules, 2000 more steel barrels, 2000 more solar panels and 2000 more accumulators, oh, and constantly pump new bots into the logistics network, and I'm also going to research Rocket Damage now. Kthx, bye."

A few minutes later, I took an electric snapshot at night (so the lights are on):
Electric Nighttime Active
Then waited for the solar to kick in and just as it was about to finish recharging the accumulators:
Electric Peak Recharge Active
and then a shot of production at that moment plus a few seconds:
Production
At which time the demand-ometer says
demandometer.png
demandometer.png (99.19 KiB) Viewed 11263 times
...that I still have a ways to go on solar, modules, accumulators, and capsules, but I'm almost done making belt and barrels, yay. I've done Rocket Damage 1 and half of Rocket Damage 2. And I have ~320 new logistics bots, and about ~300 new construction bots in my network. It's been about 20 minutes since I pressed the buttons to make stuff go. My character, incidentally, hasn't had to move at all, though I have run up the main bus a little to look at the red/green production centers.

As predicted in the comments above, my production of reds is wholly inadequate, which in turn slows down the blues. Despite iron peaking at >7k/s, the mainbus is practically empty from feeding green circuits and belt production.

Note that I started from "all the stations are full, including oil" - I doubt I could keep production this high before I can fix the unload stations in 0.13 and add a bunch of trains (and new mines, of course)... but, then, my current demands are artificially high. I never ask for this much at once normally. I suppose I should actually research rocketry and try a launch, but that was never really a goal of mine. I did a launch in game #2 :) Game #3 was for experimenting and learning.

edit: better production graph - 5s is near meaningless. here's over the last 10m:
Production 10min
I mentioned this isn't a mega base, right? Sigh. I have no idea how to make one of those. Even though I've seen them. I mean, I have ideas, but I also have questions. I know I can throw modules at my foundry all day, but then my lines will technically be too long for blue belt, but I don't know by how much. I have no idea how people get so much red circuit production - I can conceive of adding two more plants of it, but then I probably need to add at least one more green plant, too. It's crazy making.

-----------------------

Edit 2:

Sigh. This is messed up.
Problem Bub
Belt and barrels finished making (barrels came up short from lack of place to put them), so my iron lines aren't empty anymore. Reds are slow not from lack of resources, but just because they ARE slow... BUT: half my greens are sitting around contributing nothing, leaving the other half to do all the work. grrrr.

So maybe when I provided a save game download link and said beginners could learn something from this base, I meant in the "learn what not to do" sense.

...oh, egg on my face, 5 minutes later ALL the greens are effectively idle, because demand for them is way below capacity. They were just lacking iron because belt was eating it all. Once they got iron, they all caught up, but the bottom half caught up faster than the top half. Which is still weird to me.

----------------------

One hour (game time) after I kicked off all that production. I chose this moment so we could see everything spike up when the production started.
Production One Hour Later
Electric One Hour Later
Demandometer One Hour Later
Finished the belt and destroyer capsules, and 4500 concrete for kicks. Hung on barrels because of lack of storage space (I didn't have room for 2000 barrels in the passive providers attached to the assemblers). Finished the accumulators, but only 1250/2000 of the solars, finished 200 Productivity Module 3, but only 180 Speed Module 3 and 150 Efficiency Module 3. Added 900 logistics bots and 350 construction bots to the network -- because I wasn't using construction bots for anything, they filled their roboport and stopped making more :/. Researched rocket damage 1-5, rocket shooting speed 1 and 2. Science idled for a while because I forgot to put a requester for alien artifacts (I didn't always have 19k of them sitting around doing nothing, so I thought to control the science uptake of them).

Production 1 hour later, after the kickin:
Full One Hour Production
Qon
Smart Inserter
Smart Inserter
Posts: 2164
Joined: Thu Mar 17, 2016 6:27 am
Contact:

Re: The kilobase

Post by Qon »

silverkitty23 wrote: I mentioned this isn't a mega base, right? Sigh. I have no idea how to make one of those. Even though I've seen them. I mean, I have ideas, but I also have questions. I know I can throw modules at my foundry all day, but then my lines will technically be too long for blue belt, but I don't know by how much. I have no idea how people get so much red circuit production - I can conceive of adding two more plants of it, but then I probably need to add at least one more green plant, too. It's crazy making.
Step by step guide to megabase:
  • Decide on your challenge goals. No bots, belts, trains? Any specific creative factory style that might me ineffiecient but still cool if scaled up? Any mod that increases or decreases difficulty?
  • Get a goal for you megabase. A megabase needs at least some form of production goal or similar. Usually rocket launches, at least one per minute. It requires a bit of everything and has complexity and scale enough. If your challenge is tough enough even a smaller production goal can be considered a megabase. If you have a mod that gives you overpowered items then several rockets/minute might not be an impressive goal. Harder challenges, overpowered mod items and even balanced changes to vanilla makes your megabase harder to compare to others. Most people use something like RSO mod though, otherwise you will never be able to get enough ore to run your megabase and you will have to spend 100% time rebuilding mining outposts and 0% building you factory. It's impossible and not interesting or fun unless your production goals are very limited for your style of megabase.
  • Build a small base that can produce everything you need to build a base and learn the game. You are going to need something that can give you lots of belts, bots, trains, assemblers, beacons, modules and so on so that you aren't limited by building material. This base will abandoned later. You can't build a megabase without a "kilobase" to support you, but you can't build on top of a kilobase to make a megabase either. You are here. Finish up any necessary reseach. Maybe send a rocket before going to the next step.
  • Calculate what you need to get to your goal. Without maths you will always just run around corking leakages and uncorking bottlenecks without actually improving anything in any reasonable timescale and without knowing if the design you are builing on top of is possible to extend enough. Get your ratios reasonably correct. Learn the speeds of belts, bots, inserters, assemblers and so on.
  • Plan designs that can achieve the production numbers you calculated in the previous step. Get ratios within each design correct. Decide which materials should be bussed around and which should be produced locally to that subfactory, informed by the input numbers needed locally and globally for your goal. Decide how you want to split up your base in units. 1 base that does it all or several factories that handles one part each? Or maybe you need several factories of one type? Maybe you make your whole base as one unit that you then make several identical copies. Your calculations will inform you what kind of splits are reasonable.
    Flatland, Test Mode and Toybox are good mods used in a save dedicated to making blueprints without worry about research, biters, space, power usage, mining and so on. All have to be achieved but this way you can do one at a time without being limited by the others in the design process.
  • Plan land usage according to your designs made in previous step. Seize the land you rightfully deserve!
  • Place your blueprints. You should have the materials needed from your first smaller base. If your blueprints are tested and designed well enough then this could be the fastest part. Travelling, setting up train network and mining outposts are the biggest timesinks when actually creating the base since construction bots will build your predesigned blueprints for you.
My mods: Capsule Ammo | HandyHands - Automatic handcrafting | ChunkyChunks - Configurable Gridlines
Some other creations: Combinassembly Language GitHub w instructions and link to run it in your browser | 0~drain Laser
User avatar
siggboy
Filter Inserter
Filter Inserter
Posts: 988
Joined: Tue Mar 29, 2016 11:47 am
Contact:

Re: The kilobase

Post by siggboy »

silverkitty23 wrote:I have no idea how people get so much red circuit production
Your Red Circuit production line is very clumsy (and the Blue one, too). If you build that way you waste a lot of space and belt and it just makes things more awkward than necessary.

Red Circuit is extremely easy, you only need 3 ingredients and the ratios are simple (2:1:1). You feed cables on one belt and a mixed belt of plastic/green on another one and output to a third belt (with red inserters). You can make one looooong row of Assemblers that you just put back to back. It couldn't be any easier.

The cables you make in Assembler 3 with 3x Prod 3 and 1x Speed 3. You need 1 for each 11 Red Circuit assemblers (unless you change the ratio with modules).

Blue Circuits is pretty much the same (long row of Assemblers, the pipes are slightly in the way). Use Prod3/Speed3 on those.

Green Circuit should be made in a 1:1 ratio setup, use 3xProd3/1xSpeed3 for the greens and 2xProd3/2xSpeed2 for the cables to get an almost perfect 1:1 ratio.

Your making way, way, wayyyyyy to much sulfur, you're making enough sulfur to launch 2 rockets per minute or more.

Of course all this requires long handed inserters and mixed belts, which you do not want for some reason. Way to make it difficult for yourself :).
Is your railroad worrying you? Doctor T-Junction recommends: Smart, dynamic train deliveries with combinator Magick
silverkitty23
Fast Inserter
Fast Inserter
Posts: 117
Joined: Wed May 11, 2016 6:52 am
Contact:

Re: The kilobase

Post by silverkitty23 »

siggboy wrote:Of course all this requires long handed inserters and mixed belts, which you do not want for some reason. Way to make it difficult for yourself :).
Qon wrote:Decide on your challenge goals. ... Any specific creative factory style that might me ineffiecient but still cool
There we go. My challenge is to do away with mixed belts and long-handed inserters. Guess I should take care of that science build to not be an exception :)
User avatar
Azraelle
Burner Inserter
Burner Inserter
Posts: 16
Joined: Sun Apr 17, 2016 5:37 am
Contact:

Re: The kilobase

Post by Azraelle »

siggboy wrote:
silverkitty23 wrote:I have no idea how people get so much red circuit production
Your Red Circuit production line is very clumsy (and the Blue one, too). If you build that way you waste a lot of space and belt and it just makes things more awkward than necessary.

Red Circuit is extremely easy, you only need 3 ingredients and the ratios are simple (2:1:1). You feed cables on one belt and a mixed belt of plastic/green on another one and output to a third belt (with red inserters). You can make one looooong row of Assemblers that you just put back to back. It couldn't be any easier.

The cables you make in Assembler 3 with 3x Prod 3 and 1x Speed 3. You need 1 for each 11 Red Circuit assemblers (unless you change the ratio with modules).

Blue Circuits is pretty much the same (long row of Assemblers, the pipes are slightly in the way). Use Prod3/Speed3 on those.

Green Circuit should be made in a 1:1 ratio setup, use 3xProd3/1xSpeed3 for the greens and 2xProd3/2xSpeed2 for the cables to get an almost perfect 1:1 ratio.
That's great for when you finally get modules. Usually you have to set up green and red circuits before you get modules and level 3 assemblers, so the 3:2 ratio for green and 8:1 ratio for red make the most sense when you first set up, and by the time you get modules, re-arranging half your factory is time-consuming.
silverkitty23
Fast Inserter
Fast Inserter
Posts: 117
Joined: Wed May 11, 2016 6:52 am
Contact:

Re: The kilobase

Post by silverkitty23 »

Very good point, Azraelle, but he wasn't wrong to call me out on this base because I am (a) making it hard on myself with my "no mixed belts or long-handed inserters" rule, and (b) I did already tear down a module-capable base to make this base, so I could've been improving the design as I went instead of blindly using my old green/red/blue setups.

But, yes, in general, that's how a lot of people end up with "mid game optimal, late game suboptimal" green/red setups.

Although:

I'm not an amazing player with years of experience, so I'm not sure I'm in a position to give advice, but if I were, I think one piece of advice would be: "once you get bots and blueprints, *do not fear the teardown* - if you can make a better design, then go somewhere empty, make it, test it, and if it works, blueprint it and ruthlessly replace your old stuff. you'll be happier in the long run."
User avatar
siggboy
Filter Inserter
Filter Inserter
Posts: 988
Joined: Tue Mar 29, 2016 11:47 am
Contact:

Re: The kilobase

Post by siggboy »

That last piece of advice was in fact given by Qon. He even went so far as to suggest to use a sandbox map to actually design the blueprints in peace and quiet (I concur, it's a good idea).

About the red circuit factory: obviously you do not design that way in your bootstrap factory. The bootstrap factory uses what you have while you go. It does not matter, because it will be destroyed when you transition to your megabase.

However, if you make it your rule to not mix belts and don't use red inserters (again, why?) then it should not surprise you that mass producing circuits (or building into a megabase in general) gets a little bit harder.

Maybe build an unconstrained megabase before challenging yourself in the micro arena (you can still do that in your second megabase).
Is your railroad worrying you? Doctor T-Junction recommends: Smart, dynamic train deliveries with combinator Magick
silverkitty23
Fast Inserter
Fast Inserter
Posts: 117
Joined: Wed May 11, 2016 6:52 am
Contact:

Re: The kilobase

Post by silverkitty23 »

siggboy wrote:(again, why?)
mixed belts: personal aesthetics, and side-loading is a slowdown. red inserters: also personal aesthetics, but they are slower than fast inserters.

also of note: longer belts are only a one-time cost, which depreciates to zero over enough time. after that one-time cost is paid, the belt only makes room if you use items off the end, so the belt length no longer matters. in theory they cost updates-per-second, but in practice my base still runs at 60 for me on my computer. and I'm thinking when I go mega, I'm going full-on-logibo so then my belt length will be 0, anyway :)

Nothing you've said to me is wrong. I've only played 722 hours according to Steam (more realistically: closer to 500. sometimes I let it run in the background while I work or sleep to catch up on mining), which probably a small fraction of the time you and Qon have played. I'm actually reading your advice and considering it. But, I just like the discipline of ||-> . I've seen inserters struggle to pull things off of corners, I've seen bottlenecks on side-loads, I think it's unrealistic to have an inserter interact with a splitter or UG belt because those look like boxes that should be in the way of the inserter's operation. So I have a personal aesthetic. It's my base and my fun to have :)
User avatar
siggboy
Filter Inserter
Filter Inserter
Posts: 988
Joined: Tue Mar 29, 2016 11:47 am
Contact:

Re: The kilobase

Post by siggboy »

silverkitty23 wrote:
siggboy wrote:(again, why?)
mixed belts: personal aesthetics, and side-loading is a slowdown. red inserters: also personal aesthetics, but they are slower than fast inserters.
Not going to argue about aesthetics, but about those other points: merging two belts into a mixed belt is not a slowdown. The merged belt will be fully compressed. I think it used to be a bottleneck many versions ago, but not anymore.

Red inserters are slower than fast inserters, but that really does not matter if you unload Red Circuits. They're even fast enough to unload Green Circuits from an accelerated Assembler 3 if you unload into a buffer chest.

Oh, wait, you don't like buffer chests either. Sorry, my bad :).
also of note: longer belts are only a one-time cost, which depreciates to zero over enough time.
It's not about the cost of the belts. Your layout is very space wasting, and that means it will be more difficult for you to plan a factory layout for actual mass production. It will also slow down the game for you when you grow truly massive (belt surface is really bad for UPS when all the belts are actually moving).
It's my base and my fun to have :)
Absolutely. It's just that it seems you're struggling in a few areas and I was speculating why that might be.
Is your railroad worrying you? Doctor T-Junction recommends: Smart, dynamic train deliveries with combinator Magick
Post Reply

Return to “Medium/Big/Gigantic Sized Structures”