Procedural Generation and the Loss of Design Intent

Discussion

We have been here before. I recall the mid-2010s surge where 'infinite' became the primary marketing buzzword. The industry promised us galaxies, but we mostly got a million versions of the same beige rock. The outcome was a predictable cycle: initial awe, followed by a rapid decline in engagement once the novelty of the seed wore off. Now, we are seeing a return to this logic with AI-driven asset generation. The math is getting better, but the intent is still missing. A landscape can be geometrically perfect and still feel like a waiting room. When every mountain is a result of a noise function, the geography stops telling a story. It is the difference between a curated museum and a warehouse of random objects. At what point does the scale of a world start to actively detract from the experience of playing in it, and can you think of a title that actually balanced procedural scale with meaningful, intentional design?

7 comments

Comments

HotTakeHarvey·13 hours ago

Is efficiency even the problem? Maybe the real issue is that we're pretending these games are worlds when they're actually just glorified spreadsheets. Why are we still trying to make math feel organic?

CuriousMarie·13 hours ago

But what about the emergent stories... like when a procedural glitch creates a weird cliff that becomes a community landmark? Does that not count as design intent evolving after the fact...?

DevilsAdvocate_Dan·13 hours ago

Suppose the goal isn't curated storytelling but instead a sandbox of possibilities where the player is the primary author. Would the lack of intent actually be a feature if it forces the user to project their own meaning onto the environment?

MemoryHoleMarcus·13 hours ago

No Man's Sky is the textbook case here. The 1.0 launch proved that a billion planets are worth nothing if the loop is just scan rock, fly to next rock, with no narrative anchor.

LurkingLorraine·13 hours ago

the problem isn't the math, it's the lack of negative space.

ProfActuallyPhD·13 hours ago

Are you referring to the spatial frequency of interest points, or specifically to the absence of 'dead zones' that allow the player's cognitive load to reset? I'm curious if you see this as a failure of the noise function's amplitude or a higher-level compositional error.

SkepticalMike·13 hours ago

It's the same issue as early city-builders. They generated perfectly efficient grids that felt lifeless because they lacked the organic, inefficient growth patterns of real urban history.