What Actually Matters In Procedurally Generated Games

As we enter the final week before the launch of No Man’s Sky, online chatter around it get’s increasingly heated. The posts from supposedly the sole person on Earth outside Hello Games playing a retail copy haven’t dampened things in the slightest. One source of all the skepticism is the fact that the game uses procedural generation and how it doesn’t force players to an end goal. This new trend of never-ending procedurally generated games seems to be getting some pushback from those who prefer static level design and endings. What actually matters when you’re playing the former type of game?

I haven’t gotten into these indie procedurally generated roguelikes that much. Off the top of my head I can name basically two I’ve tried, bought, and enjoyed enough to keep returning to: FTL and Rogue Legacy. I never even got into Minecraft. The only other games I play a lot that use procedural generation are space games similar to NMS: Elite: Dangerous and Space Engine.

Thinking about why it’s those games in particular led me to the conclusion that the most important thing about these games that rely entirely on procedural generation is their very most core gameplay systems. You have to really like those core systems in order to like the game at all, and I think that’s what’s tripping up a lot of critics of this style of game.

With conventional games, if you don’t really like the core loop or the controls, you might still like some of the other elements. A lot of people hate the combat in The Witcher 3 but the game gets heaps of praise for its story, graphics, and how its world is designed in general. Those things provide all the motivation some people need to play the game. There are whole groups of people who primarily play games for riveting stories or well-written characters. These procedural indie roguelikes usually have none of this. They tend to at most have a world and some gameplay systems, and they are the only motivation for playing.

This probably happens because the small development teams behind these games have put together some controls, some systems, some gameplay mechanics, a loop, but don’t have the resources to craft levels or a huge pre-built world. They take those systems and randomize their elements to create a game. That’s great if players really like those core systems, but if they don’t there’s absolutely nothing left. That’s why I couldn’t really get into Spelunky — I didn’t really make a connection with its systems or how to engage them.

The issue with NMS is that some people are already sold on the idea of seamlessly flying to and between realistically-sized planets in a systemic sci-fi universe, and some people aren’t. I already know I’m into that because of the aforementioned other space games I’ve been playing. What I’m mostly anticipating is seeing how Hello Games has balanced all the different systems. The Reddit leak seems to at least be positive about the controls and flying action.

If you’re the kind of person who needs all that other stuff to be drawn into a game, think about all the games where you enjoyed the core controls and mechanical feedback loop so much that you repeatedly played the game after finishing it. Did you really need an end goal to get through that if you just kept on playing after the end? People kept downloading more maps for Doom and Quake for decades after beating them, would those people want procedurally generated maps?

BULLETS:

• Something crazy happens to Jupiter’s moon Io for 2 hours every day – The Washington Post washingtonpost.com/news/speaking-…
• A look at the racial divide of Pokémon Go in D.C. – The Washington Post washingtonpost.com/news/local/wp/…
• So… humpback whales and orcas are having a planetary whale war right now mnn.com/earth-matters/…
• There’s little evidence flossing actually works usatoday.com/story/news/nat…