Trains, Outposting and other things that annoys me in Factorio

[OldFart]

I just got the final achievement in Steam and, after 500h ingame, I want to went some emotions. Please, comment, what I'm doing wrong

Trains

Trains are huge. Train networks in megabases usually take up a lot more space than production buildings.
Not only train stops are huge by themselves, but they also require train buffers to ensure the high throughput of the train station.
Even then, sometimes unloading chests became unbalanced. Buffer for one wagon drains earlier than other and that imbalance amplifies itself. I don't know why, I always have a belt balancer right after the station, but imbalance happens anyway
Loading-unloading stations are tedious and repetitive to build.
I usually avoid trains and belt everything till the bitter end.

I played krastorio2 recently. Mod changes the stack size for ore to 200. That greatly increased the throughput of unbuffered stations and the train network took much less space than usual.

Outposting

Every so often, too often, in fact, I'm interrupted by the shortage of resources. The algorithm is already known:

• Find a suitable ore patch
• Route rais network to it
• Build station, don't forget about buffer for trains - paste blueprint
• Build Belt balancer - paste blueprint
• Paste a bunch of mining drills - bots will finish the job
• Connect belts
• Configure station name and train limits
• Add more trains to the route
• Return to the previous task for a minute
• Notice, that there is a shortage of other ore
• REPEAT

My main complaint is that in the late game, the ratio of miners to production buildings is way too high. Mining productivity helps a bit, but it takes a lot of Space science to feel a difference. We need miners MK2. That may be fixed in Factorio 2.0 and I hope new miners will not be too expensive
Game flow is frequently interrupted by boring and repetitive task

Biters

To be fair, I have exactly 0 ideas about what can be improved there. I usually play in peaceful mode or at least without biter expansions.

The most annoying part is to maintain the defenses until bots are researched.
Next is building a new line of defenses each time the factory must be expanded, and deconstructing the old one. Again, that's interrupts the game flow for me

Biters are not higher on the list because I can tune them down. And I usually do

Mics

• Iron ore in the recipe of concrete is just annoying. Concrete is the only recipe that uses iron ore and is the only reason all iron ore can't be smelted on-site.
• Red / blue belts are too expensive in mid game, especially blue
  Amount of iron gears can be cut In half. It will not make any noticeable difference late game, but will make red belts a good option in mid game
• Red belts are of 2x yellow belts, blue belts are NOT 2x of red belts
• Belt balancers - they look nice, but they are big and cumbersome and essential to maintain throughput. Every balancer, that is not 4x4, is usually a blueprint made by something else.
  We need 1x1 splitters, that, if built one next to another, share the input and output lanes and balance all of them automatically
  And we need a way to unload train wagons perfectly evenly
• Lane balancer - we need the ability to do that, because inserters take preferentially from one side of the belt

[Tertius]

Not only train stops are huge by themselves, but they also require train buffers to ensure the high throughput of the train station.
Even then, sometimes unloading chests became unbalanced. Buffer for one wagon drains earlier than other and that imbalance amplifies itself.

If you have a stacker in front of unloading stations and slightly more production than consumption, the buffer chests can be kept down to 4 chests per wagon, because the next train has to travel just a few tiles to replace an empty train.

4 chests per wagon and each 4 chests resulting in one output belt can be reasonably simple balanced, and having a stacker in front of the station means all chests will be full before the train is emptied, which reduces imbalance. The crucial thing for balancing I found is to balance the train-to-chest inserters of an unloading station with the madzuri algorithm. Not the chest-to-belt inserters. This way the most empty chests are filled first until they are balanced, only then all chests are filled equally.
Such an unloading station has 1 output belt per wagon, so you need a regular balancer after the station. Lane balancing is done by the madzuri balancing within the train-to-chest inserters and wagon balancing is done by the regular balancer after the station. In my bases, this results in 100% perfect balancing over days of operation, no matter how imbalanced the factory is pulling items.

In the early game I use 4 wagon trains, so they have 4-belt balancers, in later stages 8 wagon trains, so they have 8-belt balancers - the biggest sane balancer.

Outposting

I prepared a single blueprint for an ore outpost. It contains a mining drill field to cover a 60x60 tile area, a loading station, both connected by a balancer, and a rail circle. I just need to stamp the blueprint over a resource patch and connect the rails manually.

For oil resource fields, I use the mod Well Planner and a blueprint with an oil loading station plus rail circle.

Biters

I'm not the combat guy as well, but in my recent map with everything vanilla and default settings I accidentally started in the desert, which results in a big pollution cloud and early and frequent biter attacks. I avoided biter expansion into the starting area by placing single cheap items, for example single belts or pipes, as near as possible to biter nests (expansion will not happen near player buildings). This doesn't avoid biter groups attacking the base if the pollution cloud touches the nests. It just avoids expansion of nests into the starting area near the base - nests that have to be cleared first otherwise.

Place gun turrets at choke points and where biters frequently travel. Supply them manually with up to 50 ammo, usually lasts until the next visit. And place them randomly near production buildings to catch biters that directly attack your factory while you're away.

Then, as soon as they are researchable, research flamethrower turrets and flamethrower. One flamethrower turrets near chokepoints where enemies often approach, supported by 1-2 gun or laser turrets to finish them off.
Prepare a tileable lightweight defensive wall blueprint with a mix of flamethrower turrets and laser turrets and deploy this instead of manually created defenses as soon as you researched construction robots.
With this approach, even my desert start was neither too tedious nor painful. Flamethrower turrets are crucial. Laser turrets are the best addition to them, because both don't need extra ammo and ammo supply infrastructure. A single oil well can supply all flamethrowers of a whole megabase with crude oil.

When I need to expand my factory in the early game, I go with a flamethrower and destroy the nests in a way I can use water and cliffs as natural barriers and build the smallest required defense manually. Instead of walls, you can also drop single belt or pipe pieces to avoid enemies expanding back into the cleared area, but you need to make sure remaining nests are outside the pollution cloud, because as I said above this will only prevent nest expansion, not attack groups.
This goes as deep into enemy territory as possible to have a long time until it is required to expand again. Look at the minimap: destroy all nests within and near the pollution cloud and create good space between the nearest nests and the current pollution cloud.

Some of the things you mention are addressed in Factorio 2.0, but other things are just the new challenges that pop up if you grow your factory beyond some size.

[mmmPI]

Please, comment, what I'm doing wrong

I'm not saying you do anything wrong, but i think it's always possible to improve, and highlighting the problematic area is already a first step ( i know this) and asking for help is the second ( allegdly)

I have a few questions to help me understand better the situation :

Do you use productivity modules ? ( level 3 everywhere ? not in miners ? no productivity modules ? ).
What is your average science per minutes when you start to feel those problems ? 60 ? 600 ? 6000 ?
Do you use the default settings for ressources / less biters /rich ressources / train world ?
What size of trains do you use ?
how many "belts per wagon" do you usually attempt ? ( unload only on 1 side and 1 red belt per wagons, or unload on both side and 2 blue belts on each side ? are quite different problems ).

I think those are all impacting the "tedioussness and repetitiveness" of outposting. If you go for only 60 science per minutes with productivity modules level 3 everywhere, rich ressources, no biters you will have easier time.

The size of train i found make it less annoying to expand when they are "small" 2 to 4 wagons, versus 8 or 16 , then using a "lower" number of belts per wagon, 1 red only 1 side for the unload, will "cap" the outpost reducing its yield but increasing the amount of time it can provide the lowered one. It may sound like nothing, but it really compound with the research too. I also tend to spread mining drill apart on a ore patch for the same reason. This makes it easier to ramp up the number of mining drill without feeling like needed to "replace" the depleted one.

For the balancing of train stations, there is a "known" way with combinators to balance out the chests, this is done by summing all their content , connecting them with say a red wire, and dividing that total amount by the negative of the number of chest ( -6 if you have 6 chests ). then send this result to all the inserters with the same color of wire, so they read as value the average with a negative sign. And with the other color of wire, each inserter should be connected to its own chest. This way if the result is "positive" it means the inserter guard a chest that contain more than the average, and thus its condition should be "active when something is positive". If the chest contain less than the average, it will yield a negative result, and make the inserter "wait" for the other inserters to unload the chest that have "above the average". You can use blueprint for this ,but understanding the way it works allow to reuse it for many purposes or to make custom trains station everytime. Helps a lot !

For the balancing of belts i think it helps too, because if the chest balances themselves, you only need to allow ore to flow from any belt/lane to every other belt/lane, you don't necessarily have to create something that balances out with splitters, only allow things to reach their destination from any entrance of the "fake-balancer".

This and using splitters priority too, instead of balancer, give the highest priority to the outtest lane and a bit less to the second one, third and the last one should have only splitters with priority toward the third one. The logic is a little different but it required much less room/splitters/materials per blueprint which may makes it faster/lighter to expand.

Game flow is frequently interrupted by boring and repetitive task

That's the worst when you plan to do something and you run out of ressources or feeling like constantly interrupted, either for defense or outposting. I enjoy more biters when they are really hard,( not boring for me nor repetitive because every little tech change has immediate impact) or when there are none/ peaceful (no repetive defense building), for the repair, i generally try to rush blue science and construction bots and/or keep my firepower strong enough to kill biters without letting them damage my defenses when possible. I don't repair anymore past construction bot.

For outpost, i think it is part of the game to manage those timing, i have used mods for infinite ressources of various sorts, oil-like ore, ore coming from satelites, ore coming from the ground, ore from water filtration and so on. I think those are defining the type of game i am going to play, shaping the balance of it. They are the challenge to overcome, (when they deplete) or the objective to reach (sustainability).

There are mods like https://mods.factorio.com/mod/YARM too who are designed as "tools" to help managing outpost lifetime, i'm don't feel like needing those but i know it is a popular mod i see a lot on screenshot. I used it in the past, to avoid having my game flow interrupted, so i could anticipate when i would need to build new outpost and make sure i don't get ready for something else just seconds/minutes before it as that's frustrating.

Those are two different directions to change the potential negative feeling without changing the way of playing compared to the earlier points hopefully that's enough tools to ease with the complaints while waiting for the expansion

[NineNine]

You don't have to do "outposts". I run a simple border around everything.

If you feel like you have too many miners vs assemblers, you can put production modules in all of the assemblers that you can, and that'll change that.

I don't care about how my trains load or unload. It doesn't effect my system.

Almost everything that you describe not liking can be changed by how you choose to play the game. Nothing you're describing is inherent to the game. Try experimenting with some different ideas and see if you can't resolve most of these things for yourself.

[FasterJump]

Congrats on the achievements!

I play in "train world" settings, and I feel like ore patches are big enough. I just have to maintain 2 early game, 3 mid-game, and maybe 4-5 max in a 1k spm base (iron/copper only). It's a bit repetitive but blueprints helps. Also, there are massive train network coming in 2.0 so expect things to be much easier to automate and much less repetitive. There are also new miners, quality, etc. which means that mining outposts will last a very very very long time. Bulk inserters should also help with loading/unloading stations.

Factorio 2.0 will add a lot of QoL, but will also increase the scale of bases. Hard to tell if it will lead to more or less repetitive tasks overall, but I'm sure it will be fun.

Ore stack size increase and lane balancers could be a good surprise in 2.0, even if there are workarounds for that.