Genre Analysis: Factory Builders

Defining Games

automated resource extraction and refinement
logistical challenges driven by spatial constraints
optional optimization of existing systems (generally optimizing resource throughput)
gated technology progression
problems/puzzles/modules flow into each other (as opposed to puzzle games or zachlikes where they are explicitly independent)
partially self-directed goals
challenge can be overcome by spending time waiting
highly moddable
procedurally generated world

puzzle like, but complete solutions are easy to achieve
optional optimizations on top of said solutions
satisfaction of watching your contraptions run
satisfaction of taming the wild
interesting, unexpected failures due to unexpected interactions or subtle bugs in the player's design
gameplay can be extended, enhanced, or changed through mods or built in options
hits on "engineering creativity" in a way that other games don't
- engineering creatvitiy is more about coming up with solutions within constraints. It's not a blank canvas. It's based on analytically evaluable metrics
scratches similar itches to programming, but with clearer gradients and fewer frustrations
- great debugging: can see what's working and why with exceptional visualization
- basic solutions are straightforward
- very weak syntax constraints, very intuitive syntax
- solutions generally do not need to be as robust as in the real world
- tech progression doesn't break previous (natural) assumptions
- good quality of life tools and docs

This list is roughly ordered by the closeness to the genre.

Zachlike puzzles (automation heavy puzzle games)
- Opus Magnum, Incredible machine
- Greater focus on additional (secondary) optimization constraints (which are provided by the game)
- Micro-optimization as an entire game
Base and city builders
- Sim City, The Sims, Cities: Skylines, Dragon Quest Builders
- Build out a complex and delightful area, but with no / little automation
- Most of the actual functioning of your constructions is abstracted away
Colony sims
- Rimworld, Dwarf Fortress, Oxygen Not Included
- Balance complex systems, but with a focus on the needs of your workers, not scaling production
- More about the emergent push and pull of life in the colony. Less about steadily pushing towards some goal.
Idle games
- Cookie Clicker, Candy Box
- Automate everything, but without the complex chains
- Additional mechanics and optimization opportunities unfold over time
Train simulators
- Open TTD, Mini Metro
- Focusing almost entirely on logistic networks powered by trains
Real Time Strategy
- Starcraft, Warcraft III, Red Alert
- Command units, build bases, and extract resources
- Very little automation: resource production is almost entirely abstracted
- Much more time pressure!
Level builders
- Y'know, like Mario Maker
- Focused on creating interesting experiences and, in some cases, creating fun and interesting contraptions
Tower defenses
- Dungeon Warfare 2, Bloons TD, Gemcraft
- In high quality games, there are strong optimization and contraption elements

These games are all factory builders combined with the genre(s) in brackets.

acquire goal (generally something to produce)
determine recipe
(optional) unlock tech prerequisites
find relevant inputs (raw resources or previously produced items)
(optional) if necessary, rework existing parts of the factory to produce relevant inputs
build prototype
(optional) optimize and scale prototype
integrate with larger factory
reap fuits of your labor
(ad hoc) discover and fix issues with previous designs

player-driven optimization
determining production chains
physical layout of factory elements (making sure they reach each other and all fit, etc.)
transport logistics (moving things around the factory)
refactoring previously constructed parts of the factory as new requirements (or better methods) are discovered
returning to your factory after a break, and needing to refresh yourself on how it works
(optional) combat disrupting operations
(optional) temporal variation disrupting operations, often by creating instability in resource supply or demand

What goals of the genre are in tension with each other? Note that many of these design tensions are not unique to factory builders.

mechanical complexity vs meaningful richness
helpful innovations vs trivializing the game
- unlocking new tech can sometimes remove interesting challenges instead of creating interesting opportunity for improvement
- (insert roast of Factorio's logistics robots here)
disruption forces interesting changes vs frustration and sadness to see your stuff get wrecked
long play time vs tedious waiting around
giant automated bases vs computational limits
joy of optimizing and creating big factories vs uncomfortable implications of industrialization and colonialism
accurate simulation of reality vs clear and interesting puzzle situations

mediocre combat
- poorly integrated
- frustrating
- snowbally
- thematically incoherent
- difficult to balance
- limited depth
poorly managed complexity in UX
poor tutorialization
bland, unoriginal aesthetics
unoriginal contraption mechanisms
- everyone uses inserters and belts
boring and low-impact environmental variability
copy-paste of optimized designs
treadmill-style tech progression: same mechanics, but with a different coat of paint and "numbers go up"
- driven by lack of end products
terrain modification that makes the world less interesting
no meaningful penalties for overproduction of resources and mindless expansion
frustrating control schemes, especially in games with a player avatar
- also results in difficulty seeing the whole factory at once
missing QOL features