Getting past mid-game

Don't know how to use a machine? Looking for efficient setups? Stuck in a mission?
Post Reply
grouchysmurf
Inserter
Inserter
Posts: 29
Joined: Mon Sep 05, 2016 8:55 am
Contact:

Getting past mid-game

Post by grouchysmurf »

Hi. Long time lurker, first time poster.

I've been playing factorio quite a lot but I never managed to get past mid-game, which, solely for purpose of this email, I define as a point when I start assembly processing units. Usually - if not always - this is when I realize my factory is a mess and get bored with it, starting everything anew with a laudable of goal of getting everyting right, this time. If does not work out, as you can imagine.

My main concern is that in my games first part of the game is hasty race with bugs; while I keep my production ordered, I focus mostly of get things going, until I feel confident enough defence- and resource-wise. This happens when I max out green research and secured one or two patches of iron/copper. And when I connect the oil and start to crack it, then I realize my factory is neither expandable nor neatly designed nor optimal:

1. utility sections (inserters, turrets, railway) are mixed up with production (gree/red circuits);
2. science production, while I get it going, is no longer nicely laid out when I plug in section producing blue bottles;
3. production of green circuits does not meet up requirements of red/blue circuits and is not easily expandable.

I came to thinking that one-till-the-end approach is not viable; rather, you should start with a 'utility' or 'supply' small factory, which would provide all the items required for building your second and final base. A 'supply' factory would produce inserters, assemblers, tracks, engines and whatnot. This would enable you to lay out neat 'production' units making GC/RC/PU and all other things required to launch the rocket.

While I know there is no such a thing as 'right approach' in factorio, is that, what I stated above, the general approach you guys take? If not, what is it then?

I am a bit frustrated with all that reaching the point when I find no fun in expanding my factory. I'd like to win the game, at least once. Any hints, then?

User avatar
Philip017
Filter Inserter
Filter Inserter
Posts: 346
Joined: Thu Sep 01, 2016 11:21 pm
Contact:

Re: Getting past mid-game

Post by Philip017 »

i have played a few games, and after my first very messy game, i optimized a few things for my next based on what i learned,
i try to keep it small and keep a close eye on pollution. more pollution = more angry bugs.
i explore early, i want to see what is around me, i clear a lot of the fog so i know where the bugs will be coming from, and where some resources will be located at.
i build modular designs now, ex 3 copper wire per 2 green circuit factory
once i have robots and blue prints, i build more modular buildings, and once i have "enough" to subplant my old base, it is removed in favor of the optimized modular designs.
i have found the best way to have a functioning research machine is to spin the vials of research around the research machines, that way what is not used immediately revolves until it is needed, no longer starving the top or bottom when the factories cant keep up.
im just starting to tinker with the circuit network and trains, previously i just belted everything around - but the more resources you use the further you may need to go, so trains may be more viable.
Last edited by Philip017 on Wed Oct 12, 2016 11:51 pm, edited 1 time in total.

User avatar
siggboy
Filter Inserter
Filter Inserter
Posts: 988
Joined: Tue Mar 29, 2016 11:47 am
Contact:

Re: Getting past mid-game

Post by siggboy »

grouchysmurf wrote:I am a bit frustrated with all that reaching the point when I find no fun in expanding my factory. I'd like to win the game, at least once. Any hints, then?
"Winning the game" is easy, even with the most horrible factory. All you need to do is to make a rocket.

But it's not the point of Factorio.

You already say that you're not having fun, I guess because your factory doesn't feel right, because it's unorderly. I think that's perfectly normal, and a good way to approach the game, i.e. trying to make things more efficient and pleasing.

Here's my perspective on things:
  • I have a lot (way too many) hours in the game, but I still can't make an early factory that I would call "beautiful". At a certain point I just get lazy and draw some spaghetti here and there. It's not really a problem, unless it gets so messy that you don't have reasonable production any longer. For the end-game I usually destroy EVERYTHING and build a new factory that only makes things that I need for the late-game. You can also re-build your science facilities for better beaker output at this point. However, science is really not that important anymore as soon as you have researched certain key techs.
    Don't just start over, instead replace the parts of the factory that you hate with something better.
  • Install the mod "Creative Mode", and make a "sandbox map". Limit the map size (to 400x400 or some such), disable aliens, load it in the map editor and cover everything with a concrete floor. Then you can use this map with Creative Mode to practice building efficient designs. You can also export Blueprints of the designs you like most and use them in the mid- and late-game in your actual maps.
  • There are certain approaches that make the early- and mid-game easier. You should create a double belt (= 4 lanes) and put Gear Wheels, Iron Plates and Electronic Circuits on the belts; leave one lane empty, that you can use as an output lane. If you draw such a "main bus" you can branch off little mini factories quite easily and build most of what you need early on.
  • Get your science production down, go in your sandbox map (with creative mode), and make a science layout that you like, for practice. Then build it in your real game. Just leave enough space, that's the hardest bit.
  • Increase the size of your starting area to "big" or "very big" so you won't have to deal with biters before the mid-game. Reroll desert starts, to limit pollution spread.
    Or just disable biters, nothing wrong with that either.
Is your railroad worrying you? Doctor T-Junction recommends: Smart, dynamic train deliveries with combinator Magick

Hannu
Filter Inserter
Filter Inserter
Posts: 850
Joined: Thu Apr 28, 2016 6:27 am
Contact:

Re: Getting past mid-game

Post by Hannu »

I make always beginning base first. It has typically capacity of about one red belt of iron plates and one or two refineries. I have basic bus but I do not care much about layout or expandability in that phase. Typically I run that base until I get logistic bots and at least first speed and cargo upgrades. I build assemblers to almost anything I need and set stocks to quite large amount. Then I begin to build a new base nearby. I can go back to the first base and request full load of stuff to build next thing. I have a structure with production cells and buses and walkways between them. I bring raw materials by trains. I expand that base when needed. At current game I moved iron and copper smelting units away to get more room for circuits and steel.

grouchysmurf
Inserter
Inserter
Posts: 29
Joined: Mon Sep 05, 2016 8:55 am
Contact:

Re: Getting past mid-game

Post by grouchysmurf »

Thanks.

I am familiar with most of the game concepts - I am spoiled to the extent where I can't event think of any other base layout than main-bus. This is my current base:

Image

Biters are not a concern of mine - when fully upgraded, gun turrets can take down big biters and spitters with ease. In the current game I have AP ammo and biters didn't even evolve to medium level, so no worries here.

I will give a try to the CreativeMod - this is, I guess, what I've been for.

Right after sending the post, I realized it can be summarized as "how do I make my initial base expandable". It seems, that while it can be done, it's a hard task without having an upfront plan. It's good to know people do scrap their initial bases - thanks for letting me know. I used to think you guys have it all planed in advance.

Mehve
Filter Inserter
Filter Inserter
Posts: 318
Joined: Sat Aug 06, 2016 9:12 pm
Contact:

Re: Getting past mid-game

Post by Mehve »

God no. Usually my ultimate goal for the first base is to turn it into a research center/shopping mall that can be left alone to advance research and let me come back to replenish stock on building materials for the final base.

Experiences differ, but I really think the "Main bus" style of factory hits a wall as you work in earnest towards a rocket. Since each of the three ingredients for the rocket have quite distinct and separate processes, there's no benefit to throwing things on a common belt. Optimization and troubleshooting are also much easier when things are separate.

User avatar
siggboy
Filter Inserter
Filter Inserter
Posts: 988
Joined: Tue Mar 29, 2016 11:47 am
Contact:

Re: Getting past mid-game

Post by siggboy »

grouchysmurf wrote:This is my current base:
What you've shown is a very clean and organized main bus layout. My bases certainly don't look that neat.

You can make life easier for yourself if you don't bus plates and steel. Create a specialized bus with Iron Plates, Gear Wheels and Green Circuits. Only run belts with copper plates and steel to the few places where these are actually needed.
Early on you don't need Copper Plates and Steel outside of science.

That will make it a lot easier to have a "building store" that actually produces the items you need for expanding.

I would keep the science strictly separated from this. Design a little science factory that only produces beakers, and run it separately from your "building store"/main bus.
It's good to know people do scrap their initial bases - thanks for letting me know. I used to think you guys have it all planed in advance.
Unless you're speed running there's no reason to not scrap your base when your feel like it. Scrapping and rebuilding does not cost anything at all. If you deconstruct you get all the materials back. The only cost is your time.

Take your time, make things pretty so it does feel good and you enjoy watching your creation.

And don't worry about biters so much, it's really very easy to defend against them, with the flamethrower and the energy shield you can also be very aggressive and just clean out those nests that are closest to your base very switftly.
Is your railroad worrying you? Doctor T-Junction recommends: Smart, dynamic train deliveries with combinator Magick

Jupiter
Fast Inserter
Fast Inserter
Posts: 173
Joined: Thu Jun 23, 2016 2:38 pm
Contact:

Re: Getting past mid-game

Post by Jupiter »

Here is how I approached my first game. I watched a lot of let's plays before it so I already had an idea of how one might organize a base but that's not really necessary though it helps to have at least SOME idea.
I use a main bus but not like other people (in this thread) do with putting multiple things on one belt. I'd rather have multiple belts per thing. This approach got me right through to the end game and my base is now firing a rocket once every 2.5 mins. I have never built a second base although I had some preliminary production of belts, inserters, science (just 2 labs) and some other stuff to make my life easy.

The key principles I adhere to is to think ahead, think big and think spacious. Always make sure your production builds are expandable in at least 1 direction and that you build it in such a place that you actually have the space to expand (and not have to use that space for something else). You will have to pick a place for your bus where you can expand it in both the length and width (one of my circuit builds are about 75% the length of the bus itself! bit of overkill probably).

Remember that space and oil are truly infinite resources. A product build looking neat and compact doesn't really serve a function except for aesthetics. Make sure it can do its job first.

Leave a LOT of space between the different parts of your base. For example, you will want a gap of like 20-30 tiles wide between your initial smelter setup and the 'head' of your bus so you can always put in big, bulky belt-balancers.
Do not be afraid to redo entire portions of your factory. I have redone my smelter setup 2 or 3 times now and I am happy that I did.

Don't get lazy. If you want to do things in a certain way than stick to it or things become really messy very fast, even for the smallest and stupidest things.

Now some screenshots: http://imgur.com/a/0l9Ja
Smelting + science + head of bus (bit of a mess):
Image

End of bus + rocket production + 'end-of-bus' PAX station (it was there first, I've just built my bus right through it, no probs):
Image

EDIT: my point is that I was thinking on this scale from the beginning and so should you
Last edited by Jupiter on Sat Sep 10, 2016 12:27 am, edited 1 time in total.

nonstickfrypan
Long Handed Inserter
Long Handed Inserter
Posts: 53
Joined: Fri Apr 15, 2016 9:02 am
Contact:

Re: Getting past mid-game

Post by nonstickfrypan »

I thought I was the only one who gets to a certain point and then destroys their whole factory and rebuilds it all. I'm glad to discover I'm not alone. I do it this way mainly because early game I don't always have the production resources required to make things as beautiful as I might like. I'm often diverting resources towards defense instead of production and my production always ends up with "hacks" added to I can get such-and-such-item quickly. I think the best thing I "discovered" is to always have science production completely separate from the main factory with its own miners, supply lines and production -- it uses no resources from the main factory at all and that means I don't have to worry about science taking away resources from my factory. It doesn't sound like a huge advantage, but it truly is.

Hannu
Filter Inserter
Filter Inserter
Posts: 850
Joined: Thu Apr 28, 2016 6:27 am
Contact:

Re: Getting past mid-game

Post by Hannu »

nonstickfrypan wrote:I thought I was the only one who gets to a certain point and then destroys their whole factory and rebuilds it all. I'm glad to discover I'm not alone. I do it this way mainly because early game I don't always have the production resources required to make things as beautiful as I might like.
There is also practical limitations. At beginning I do not have exoskeleton legs and bots to clean forests. I can grow much faster if I do not produce long buses and reserve huge areas for future expansions. At my first game I made first spaghetti base and decided at some point that it is more practical to make whole new base and use old to product stuff. After that I have made it intentionally.

This is the first base in my current save.
http://imgur.com/a/ZkAYz

This is "official" base after couple of expansions.
https://www.youtube.com/watch?v=GUbHAHsxyBY

The first base is about 200 m north from north wall of a large base.

I want to make production cells with borders and walkways. The structure is aesthetic (IMO) but somewhat nasty to expand when reserve room are used. Construction bots are practically necessary because they can copy whole cell to another place.
I moved iron and copper smelting to separate factories and almost doubled my production capacity after that video.

grouchysmurf
Inserter
Inserter
Posts: 29
Joined: Mon Sep 05, 2016 8:55 am
Contact:

Re: Getting past mid-game

Post by grouchysmurf »

This is amazing, thanks for sharing. Did you have a specific goal in mind when creating the base?

I really like that you fancy gun turrets - personally I shun laser turrets and solar panels. Guns are fun, so is coal and steam.

As a sidenote: I installed CreativeMode mod and have to say it is even more fun than the game itself :-)

User avatar
Deadly-Bagel
Smart Inserter
Smart Inserter
Posts: 1498
Joined: Wed Jul 13, 2016 10:12 am
Contact:

Re: Getting past mid-game

Post by Deadly-Bagel »

Woah, that is some megabase.

I can't seem to get myself out of the central bus factory for a different approach, it's just too efficient. Four lines of copper and iron, two green circuit, one red and one blue circuit line, and a line for each of steel, coal, stone and bricks. Optionally a line each for plastic and batteries though currently these aren't used for much.

To start with, anything that creates a component that goes onto the belt goes on one side of the main bus, and everything else goes on the other side. This helps keep things tidy and then you're not trying to fit red circuits in an awkward spot because otherwise you'd need to send them back up the bus (though this isn't too big a deal) to use in bot crafting because you started building engines right up near the top.

Problems with this factory is it's not very space efficient making it hard to defend, it depends on the logistic network to move products around, and you leave a LOT of resources lying around on the belts, but it's very tidy, expandable and easy to implement (trains bring resources in at the top, products float around the logistic network).

The simplest alternative is a modular base connected with trains, sort of like Factorio Towns by Bentham. I intend on doing this with my 0.15 run once liquids become easier to manage. This sort of base relies heavily on trains which isn't necessarily a downside, and it's easy to expand and redesign. You have more control over your buffers by controlling the length of trains and how much each wagon holds, and hopefully 0.15 will bring some really awesome features like broadcasting circuit signals through radars, and the possibility for a vanilla train logistic network (otherwise consider the Smart Trains mod). We know liquid transport is being heavily improved which will also be helpful for this type of base.
Money might be the root of all evil, but ignorance is the heart.

Jupiter
Fast Inserter
Fast Inserter
Posts: 173
Joined: Thu Jun 23, 2016 2:38 pm
Contact:

Re: Getting past mid-game

Post by Jupiter »

Deadly-Bagel wrote:Woah, that is some megabase.
It's big but it doesn't have to much capacity for its size. Smelting uses steel furnaces and no beacons. I have in my base more smelting capacity (1.8 times for copper and about the same for iron) with just half the buildings.
700k copper/hour seems a lot but I've seen bases on YT that smelt 50k iron per minute and produce 24 saturated blue belts of iron plates (its the base from the Break the game series by Negative root). And in the comments of that stream I saw someone mentioning that he reached over 100k/min (or 150k, I can't remember).
Just to put things in perspective...

The base is very neat though and it looks waaay better than mine. I like the organisation but it isn't optimized by far and has greater potential in terms of capacity.

Hannu
Filter Inserter
Filter Inserter
Posts: 850
Joined: Thu Apr 28, 2016 6:27 am
Contact:

Re: Getting past mid-game

Post by Hannu »

grouchysmurf wrote:This is amazing, thanks for sharing. Did you have a specific goal in mind when creating the base?
My first goal was a base which use 4 blue belts iron. In the time when I made the video, I had made first expansions and used most reserve room. It had 8 belts for iron but used about 5. Currently my base can use 1.1M irons per hour.

I have personal restrictions, no speed mods and no beacons. It is just because I think they are overpowered. I like belts and large numbers of assembling machines. As someone said, my base is far from extremes. It is probably impossible to compete with those who use beacons and bot logistics. Even I could copy huge number of production stuff with construction bots, my computer could not run it and it would not be fun for me.

I am in some kind of dead end with that game. I do not have clear idea what I should do next. I feel increasing of production somewhat boring. It is just copying same rows of assemblers, smelters or miners. Larger base is just boring to run from one side to another. If I began to use speed modules, I would put rows of beacons between the assemblers, get larger numbers in production display and feel more bored to continuously make new mines. Maybe this game are quite "ready". Fortunately I like to drive by my work train all over my area and just watch how fully automated systems work.

I think that I will begin next game with Bob's mods and it will be a train world. I will make separate factories and haul stuff between them with trains. I made once such a game able to launch about 4 rockets in hour. It was vanilla except launch automation and mod which barreled all liquids.

I really like that you fancy gun turrets - personally I shun laser turrets and solar panels. Guns are fun, so is coal and steam.
I have also used lasers in my previous games. But I thought that I want extra production and logistics in this time. I was also interested to see are gun turrets enough to defend against endgame behemoths. Pollution and enemy activity have increased dramatically from the time of the video. I have doubled number of turrets and I consume about 10 k bullet clips per hour. Huge crowds of enemies make frequent attacks from every directions. They destroy occasionally couple of turrets and walls and kill about 50 construction bots per hour, but they have not been able to break through yet.

User avatar
siggboy
Filter Inserter
Filter Inserter
Posts: 988
Joined: Tue Mar 29, 2016 11:47 am
Contact:

Re: Getting past mid-game

Post by siggboy »

Deadly-Bagel wrote:Woah, that is some megabase.
It's large, but it doesn't produce that much. 12k plates/minute is respectable, but in my opinion it's not a megabase. The base looks great and it's very neatly organized, but the lack of beacons really limits the potential output. The oil processing is neatly grouped, but a lot of potential is wasted because no beacons are used. The same is true for all the production.
The steel smelting could be a lot more compact, you could easily produce twice as many plates in the same space with an optimized compact layout (without using e-furnace or beacons, mind you).
Jupiter wrote:Smelting uses steel furnaces and no beacons.
That's OK in my opinion. If you create a compact layout you get a very good output per squaretile. The 20% stretching from ore -> plates is not that important. Steel furnaces use a lot less energy than the full-blown e-furnace setup (if you use PM3, with efficiency modules e-furnaces are OK but then you might as well use coal smelting).
The simplest alternative is a modular base connected with trains, sort of like Factorio Towns by Bentham. I intend on doing this with my 0.15 run once liquids become easier to manage.
My current game is such a map, I spiced it up by creating my "towns" on artificial islands. I use a barrel mod and SmartTrains (obviously, because my train scheduler / logistic train network requires that).

I don't really think you should wait for 0.15 -- it's also not clear yet if we'll get a good train solution that will eliminate the need for SmartTrains.
Is your railroad worrying you? Doctor T-Junction recommends: Smart, dynamic train deliveries with combinator Magick

User avatar
steinio
Smart Inserter
Smart Inserter
Posts: 2608
Joined: Sat Mar 12, 2016 4:19 pm
Contact:

Re: Getting past mid-game

Post by steinio »

Important facts to be a megabase:
- you must use beacons
- you must not use all the free space

Otherwise it's only a normal base. :sarcastic:
Image
Transport Belt Repair Man

View unread Posts

Frightning
Filter Inserter
Filter Inserter
Posts: 807
Joined: Fri Apr 29, 2016 5:27 pm
Contact:

Re: Getting past mid-game

Post by Frightning »

I personally like to make a 'starter base' which I use to get down the tech tree and make the items I need for early and mid-game. Once I have certain key techs, most of which are red+green science, but a few take blue science (e.g. adv. oil, assem3 tech, and adv. logistics chests), I start work on what will become my endgame base, usually I eschew belts and their many scaling issues in favor of a logistic bot based design (logibots are amazing for nice, tidy, scalable base designs; and have pretty astounding throughput with all their upgrades done; I'm managing to run a base that smelts over 3000 iron and over 3000 copper plates per minute on less than 1500 bots).

User avatar
Deadly-Bagel
Smart Inserter
Smart Inserter
Posts: 1498
Joined: Wed Jul 13, 2016 10:12 am
Contact:

Re: Getting past mid-game

Post by Deadly-Bagel »

The advantage of bots is that if you need to expand something, you can whack it down anywhere. In fact it's favourable if you have production scattered about your factory because in general it means less travel time. One of the runs I'm planning for 0.15 is a no-belt run, I won't lay a single belt for the entire game. Will be nasty until I get bots but it's incentive to rush.
Money might be the root of all evil, but ignorance is the heart.

Post Reply

Return to “Gameplay Help”