Read on Twitter
when discussing game projects with folks, i often see them struggle with the same problem: how do you know what to build first, and in what order? how do you know "how much" game you need for a first playtest?
one answer to this can be a minimum viable product (MVP).
one answer to this can be a minimum viable product (MVP).
a MVP is the smallest possible "core loop" of gameplay that feels representative of its final intended direction. this means you strip away any features that aren't absolutely critical to understanding what the basic game experience looks like; you only build the very foundation.
note that i said a MVP is *representative* of the intended final direction. for perspective, on the left is a MVP of a Burger Experience™ and on the right is the full Burger Experience™. left burger is the smallest possible amount of work which validates "burger taste ok."
what about for a game?
let's reverse engineer a couple examples. let's start with Dark Souls. why not.
let's try to think of what the smallest possible amount of work could be that makes a player feel like they are playing a "standard" section of Dark Souls.
let's reverse engineer a couple examples. let's start with Dark Souls. why not.
let's try to think of what the smallest possible amount of work could be that makes a player feel like they are playing a "standard" section of Dark Souls.
we probably need some basics; a playable character-- just one class, the Warrior perhaps. we need its movement to feel correct for the game's intended pace-- slow, deliberate, precise.
death is core to DS, so we probably need a health system allowing you to take hits and die.
death is core to DS, so we probably need a health system allowing you to take hits and die.
we probably want to think about space a bit (DS tends to have fights in super tight, close spaces that are pretty vertical) so maybe we build a chunk of hallway/arena that feels like a "standard" Dark Souls area. But just ONE chunk. Not a whole map.
and we should build an enemy that is emblematic of a "typical" Dark Souls enemy. not a special fight like a boss or something that dies fast; something in the middle, where you get to engage in + experience the feel of DS combat. like a giant skeleton! https://darksouls.wiki.fextralife.com/Giant+Skeleton
so the MVP would be:
- you navigate through a couple tight corridors and one arena, fighting a few giant skeletons
- you can take hits and if you take too many, you die
- you can dodge, parry, and attack
feels like the core of Dark Souls, right?
- you navigate through a couple tight corridors and one arena, fighting a few giant skeletons
- you can take hits and if you take too many, you die
- you can dodge, parry, and attack
feels like the core of Dark Souls, right?
of course, there are so many things this MVP doesn't cover. the bonfire system, upgrades, different classes... but all of those things are "more." they make the game better and replayable over the long term, but without them, you could still have something closeish to Dark Souls.
what about a totally different kind of game? what about like, a narrative adventure game? say, Life is Strange?
Life is Strange has a mechanic where you rewind time to certain points to try interacting with characters in different ways and changing story outcomes.
Life is Strange has a mechanic where you rewind time to certain points to try interacting with characters in different ways and changing story outcomes.
it looks like the devs started their MVP in a place that makes a lot of sense: build the time travel mechanic first, then the story around that.
the biggest urge when MVPing a story-focused experience is to go hard on the plot, characters, etc first. but a MVP of Life is Strange could probably have zero percent of the long-term story worked out and you'd still feel like you're playing LIS if it was done right.
here's how i'd do a MVP of Life is Strange:
- pick a high stakes dramatic situation, like 2 NPCs breaking up
- go through 3 scenes back to back. you talk to both NPCs in each scene, learn new info, and generate 3 rewind points
- the puzzle "solve" is helping them avoid a breakup
- pick a high stakes dramatic situation, like 2 NPCs breaking up
- go through 3 scenes back to back. you talk to both NPCs in each scene, learn new info, and generate 3 rewind points
- the puzzle "solve" is helping them avoid a breakup
the MVP approach works for PvP games too. often the strategy here is one map, one character. i.e. Overwatch famously pulled Tracer from the cancelled Titan and built the Anubis map as the first "proving grounds" for the game experience. https://en.wikipedia.org/wiki/Development_of_Overwatch#:~:text=Initial%20development%20of%20the%20game,with%20similar%20time%2Dmanipulation%20abilities.
it's worth noting that it's much easier to reverse engineer a "good" MVP for something released because we already know what the final game looks like. when you're mid development it's harder to know. sometimes the MVP helps you realize you DON'T want to make that game, or can't.
the way i approach it in my work is like this:
- make a list of all the features your game should have.
- cross out anything obviously nonessential.
- cross out anything important, but not critical.
- when you have only critical things, prioritize them.
- the top 3-5 are the MVP.
- make a list of all the features your game should have.
- cross out anything obviously nonessential.
- cross out anything important, but not critical.
- when you have only critical things, prioritize them.
- the top 3-5 are the MVP.
i sometimes see the argument that "for big content-heavy games, you can't do a MVP, because it's about the breadth of stuff to do."
i am dubious of this claim. i think if you really feel your game can't be MVP'ed, that might mean it doesn't have a stable foundation or thesis.
i am dubious of this claim. i think if you really feel your game can't be MVP'ed, that might mean it doesn't have a stable foundation or thesis.
this old extra credits video goes into the topic pretty well if you're looking for more info, or additional examples:
i hope that if you're struggling to get the "spine" of your game together and aren't sure what to focus on or prioritize, looking at this problem the way studios do is helpful to you!
