Future of Combat in Factorio

[Syrchalis]

Why
Combat is secondary in Factorio, it's not an RTS, the combat cannot be too complex or hard, but that does not mean it needs to be bland and boring, it just needs to be elegantly designed. The general spirit of the current implementation can be kept, but some big changes are necessary for the combat to be elevated from what we have now to something truly engaging and fun that is challenging for all types of players.

>The goal is that you don't spend more time with the enemies than now (you need your time for factory building) but to make the time you spend dealing with them FUN and WORTHWHILE<


TL;DR

• Very few nests on the map
• Nests are difficult to kill
• Biters attack regularly from nests, large scale attacks, but with long intervals inbetween
• More and more nests spawn, requiring better defenses or the player to take some out
• Whenever a nest becomes active it is marked on the map and the attack route lined out (and regularly updated) so players can prepare defenses that - if properly built - will stop the enemies without interaction from the player
• More enemy types, enemies and defenses have distinct strengths/weaknesses

Details
Selected Difficulty
Very important - the difficulty scales significantly with the player options. We don't want to force players to play a semi-RTS if they want to focus solely on factory building. There should be at least 2, preferably 3-4 difficulty types, think "Easy, Medium, Hard, Insane".

• Certain enemy types don't spawn on lower difficulties
• The attack waves are smaller
• The nests are less fortified
• The stats of some enemies are weakened (e.g. armor of big/behemoth biters)

Dynamic Difficulty (kinda optional really)

• Player Score = Production and Research increase player score
• Active nests and evolution factor increase alien score
• Evolution is still determined by pollution, time and nest-kills
• Alien evolution slows when player has a lower score and raises when he has a higher score than the aliens

Nest Mechanic

• Starting from minute 15-30, depending on difficulty setting a nest becomes spawns within a set distance of player structures
• The nest is heavily fortified and takes resources and time to take out
• It keeps periodically spawning attack waves that follow a pre-determined path and attack regardless of pollution
• The nest location and the attack path are marked red on the map, so the player can prepare - factory is all about preparation and planning right?
• After some time another nest spawns
• If enough attacks are carried out, the attack path can change so the player has to build new defenses (harder difficulties)

Strengths and Weaknesses

• Biters and Spitters, along with turrets get an overhaul so they have distinct weaknesses and strengths (e.g. all biters have high flat armor against bullets)
• New enemy types and turret types (0 on easy, 1 on medium, 2 on hard, 3 on insane)
• Nests spawn with modifiers (e.g. "50% more enemies, 33% less health" or "spawns only large units")
• Nests give information about what they spawn (unit composition, modifiers) for the player to prepare accordingly
• Modifiers + Enemy Types = Tons of combinations = Stays interesting

Economy Rebalance

• All turrets are a lot more expensive
• Upgrading turrets now increases their cost

Some context - if you know where enemies will attack you don't need 1000 tiles of wall and turrets, so in order to avoid people just casually building 500 turrets and slapping down however many they need at the attack point they need to cost, A LOT. Upgrading needs to make turrets more expensive, because with every minute your factory produces more things. If the cost stays the same you will just end up with 500 turrets to slap down once again - however, the upgrade has to outweigh the cost increase otherwise spamming unupgraded turrets becomes the meta and we don't want that either.

[lindenk]

I like the idea. I was also thinking about how to fix bitters, possibly by adding a new radar that can predict when the next attack is to avoid forcing the player to drop what their doing to deal with it.

[Syrchalis]

I like the idea. I was also thinking about how to fix bitters, possibly by adding a new radar that can predict when the next attack is to avoid forcing the player to drop what their doing to deal with it.

Sorry I rewrote the entire post. The same ideas apply. Now I fixed many more of the glaring issues (especially the "endless walls and turrets along the base").


Okay, why is the walls and turrets spamming an issue? Well it's impossible to balance enemies. If you need to defend 1000 tiles of wall the waves need to be pretty weak. If you know where they come and can create a chokepoint the enemies need to be A LOT stronger to present any challenge. Also if their attacks are focused on a few spots you can make defenses expensive (we are talking like 10-20x more expensive than right now) - which avoids many other issues.

[Veden]

viewtopic.php?f=33&t=37652

Had some good ideas with this theme in mind.

I'm currently working on making a combination of what you are proposing and the above post.

The map.png is an in-progress screen capture of biter nest main cluster spacing. Showing that the further from the origin the default base size gets larger.
the red circle is shown in the picture after the map.png
The green lines represent tendrils of nests that will reach out toward resource patches to put a type of mining colony (not planning on any resources being used up) create a network for bonus and have clusters merge.

map.png (891.15 KiB) Viewed 2161 times

This is an enlargement of the top left red circle in map.png.
I have added spacing in-between nests themselves to prevent the pathing issues that currently plague the vanilla perlin-noise approach.
Each base is generated on the fly.
The green circle is the hive, multiple hives can exist in a cluster. If all of the hives are killed the cluster stops growing.
The red circles I'm thinking are going to be specialty structures that when destroyed removes some of the bonuses or special units.

topLeft.png (5.22 MiB) Viewed 2161 times

Since there isn't an easy way to make adaptive global modifiers, I was thinking of just making lots of varying prototypes that can be swapped in to give the effect you are expecting.

[BlackKnight]

Nicely structured proposal Syrchalis
I like the ideas. I'm not a sure that all points would be a perfect fit for the game but there certainly are some concepts worth integrating - and the rest are on the right track.
The core of what we are discussing here I think is, "What is the point of Aliens now we don't use them for science."

ALIEN AI
Drawing on this, I think improving the AI of the aliens would certainly add an immersion to add to why players would want to keep them in game. They wouldn't be predictable and therefore would be more interesting to combat. So for example instead of them randomly being triggered to attack the same walls every so often, they would develop evolving tactics, not just evolve physical abilities to combat this...

HIVE DENSITIES
I also think the density of the spawned aliens should be contingent on proximity to the player/pollution. Its often bothered me how not just the local aliens "evolve" (and every so often attack) and grow in density.. but so does every other alien hive on the map -- evolving (slightly more plausible) but also increasing in density -- why/how come every possible hive is growing? Obviously aliens have to be more challenging later in the game when you wipe out all the most local hives.. but there has to be a better solution from the current setup. Maybe they have somehow shared their genetics.... Hmmm.. There has to be some solution anyway that balances these issues.

PATHING
I have seen on an older map, a really large attack group of aliens actually destroy a path through the trees on their way to a wall, which is great. They may "like"? the trees' anti pollution properties but dislike and be enraged by the pollution even more. Point being, I like the fact it provides a sense of destructible terrain. Even better, if they wore down the terrain the more frequently they took the same path, it would also help to predict where to focus on defenses/what bases to eliminate first.
Related, the Scorched Earth mod changes terrain type to show where aliens/alien bases are killed and where polluting factories/rail lines have been in the past. Would be so great if type of permanent landscape alteration was implemented into vanilla!!!!!!