> Can anyone explain what tricks it uses to become so addictive?
Good game design.
In particular, I see a few broad strokes of good game design here:
* There are always a variety of tasks to accomplish of varying scope and complexity. If a player doesn't feel like adding on the next stage of the factory, they can perform other maintenance tasks like cleaning up local bottlenecks or aesthetic optimization (e.g. straightening out a section of the power network).
* Almost every single problem is directly caused by the player's own actions, through a logical chain of events that's obvious once the player becomes familiar with the game. Why is the widget facility starved for inputs? Because the player previously split off 3/4 of the bolts for other production. The resulting challenges (see the point above) become meaningful because of the history, giving the game a level of intrinsic motivation that is usually reserved for sim games like city builders.
* There's no single "right way" to win. Some players treat the game as a size/speed challenge, to produce the most stuff in the fastest time; others look towards the most efficient or compact layouts; still others are content to reach the basic "win" condition (launching a single rocket) with a bare minimum of facilities and a lot of patience. The game doesn't condescend to the player to explain at them that their playstyle is wrong.
* The game separates "doing something" and "doing something at scale," but it makes the player progress to the latter eventually. As a more concrete example, the player can build the first few science packs (progression tokens) in their inventory, Minecraft style, but they very quickly need to set up automation to produce the ever-increasing required number in a timely way. Over a typical game, the most central aspects of production will go through three or four wholesale renovations as the player designs around different bottlenecks. It's a kind of intrinsic progression that I've seen no other game replicate -- even if you had access to all the whiz-bang shinies at the start of the game, the fixed costs of using them would still push the player to a "starter base -> big base" progression.
In some ways, it might be better to treat Factorio not as a single game that is 'won' or 'lost' through arbitrary rules, but instead as a hobby. "I've spent several hundred hours playing with model trains" doesn't sound outlandish.
Will Wright has a great quote on game design that I can't find, but it's something like "What makes games fun is when the player has as many choices as possible, and all of them are good choices."
I think factorio succeeds in this well: there are always many things to do which will improve things, so you have to try to optimize which is is the best decision to advance furthest with a given amount of time.
This is an excellent answer. When you read their blog you realize that all those points have guided the current design of the game. There was multiple iteration of the tech tree to invite the player gently into the different "phases" of the game. (see https://www.factorio.com/blog/post/fff-275).
They carefully thought about the name of things, the cost of things to communicate clearly to a new player what he is supposed to do next (manufacture new thing, ramp up the production of old thing, etc).
Yeah, when I went from copper and iron moving on belts to belts and inserters as items moving on belts I started to realize the rabbit hole was getting pretty deep.
Good game design.
In particular, I see a few broad strokes of good game design here:
* There are always a variety of tasks to accomplish of varying scope and complexity. If a player doesn't feel like adding on the next stage of the factory, they can perform other maintenance tasks like cleaning up local bottlenecks or aesthetic optimization (e.g. straightening out a section of the power network).
* Almost every single problem is directly caused by the player's own actions, through a logical chain of events that's obvious once the player becomes familiar with the game. Why is the widget facility starved for inputs? Because the player previously split off 3/4 of the bolts for other production. The resulting challenges (see the point above) become meaningful because of the history, giving the game a level of intrinsic motivation that is usually reserved for sim games like city builders.
* There's no single "right way" to win. Some players treat the game as a size/speed challenge, to produce the most stuff in the fastest time; others look towards the most efficient or compact layouts; still others are content to reach the basic "win" condition (launching a single rocket) with a bare minimum of facilities and a lot of patience. The game doesn't condescend to the player to explain at them that their playstyle is wrong.
* The game separates "doing something" and "doing something at scale," but it makes the player progress to the latter eventually. As a more concrete example, the player can build the first few science packs (progression tokens) in their inventory, Minecraft style, but they very quickly need to set up automation to produce the ever-increasing required number in a timely way. Over a typical game, the most central aspects of production will go through three or four wholesale renovations as the player designs around different bottlenecks. It's a kind of intrinsic progression that I've seen no other game replicate -- even if you had access to all the whiz-bang shinies at the start of the game, the fixed costs of using them would still push the player to a "starter base -> big base" progression.
In some ways, it might be better to treat Factorio not as a single game that is 'won' or 'lost' through arbitrary rules, but instead as a hobby. "I've spent several hundred hours playing with model trains" doesn't sound outlandish.