Since some time ago, I’ve been curious on how would a top-down NES-ish game (16×16 base sprites size) look and feel if it was rendered at 1x scale in modern displays.

I was a bit depressed with the state of indiedev in general so I decided to give the idea a go. It’s almost Halloween, so I dressed up and jumped into the bandwagon to see what I could create with a single session of Claude on its Free tier. (so no Claude code, just using it as an side-assistant)

What is this? it’s an exploration game, extremely simplified by design, you roam a procedurally generated forest looking for money. That’s it.

It certainly feels different, I think I managed to capture the feeling of “anthill”, a miniature world with a small living person on it you could control. At the same time it’s a huge (actually infinite) world, confined inside your screen.

One of the nice things I made is, I was adamant on not changing the scale, meaning it will look different depending on what display you use, but it will keep the overall navigation and rewards structure.

It’s also available for mobile where, I think, it plays much better because the rooms feel much less empty.

Is this really a game to enjoy? or more of a critique to 2026’s AI-powered indie gamedev? I don’t know… but I can tell it can be a bit addictive just wandering around looking for the next bill.

It can certainly be frustrating when playing at very high resolutions after the fancy effect wears off, which is why I think adding some zoom options would probably help.

Anyways, I still haven’t made it public… I think it’s worth considering a bit of QoL improvements and maybe a bit of variety on the procgen to fill the gaps in large maps with interesting or weird stuff.

In any case it has already being useful to experiment with gesture based controls on mobile, which I’ll likely put into use for some other more serious and deep projects.