There's a problem we've been circling for two weeks now, and I want to write about it while I'm still inside it, before I have the tidy version with a happy ending.
I can't tell you what it is — you know the deal by now — but I can tell you its shape. It's one of those problems where every individual piece is fine. You can look at each part on its own and nod. Nothing is broken. And yet when you put it all together and actually feel it, something is off. Not wrong, exactly. Just... not singing. Half a degree flat.
Those are the worst ones, honestly. A broken thing tells you what to fix. This kind just sits there being technically correct and quietly disappointing, and it won't tell you why.
We've thrown a lot at it. Rebuilt it three different ways. Each version fixes the thing we thought was the problem and reveals that the problem was actually somewhere else, wearing a disguise. It's like squeezing a balloon — you press the bulge down here and it pops up over there. By version three you start to wonder if you're even chasing the right thing, or if you've just been staring so long you've lost the ability to tell good from bad.
What I keep coming back to is that this is the actual job. Not the parts that go smoothly — anyone can build the parts that go smoothly. The job is the two-week stretch where a thing that should work doesn't, and there's no tutorial for it, and no amount of effort
guarantees the fix, and you have to keep showing up to it anyway. That's the work behind the work. Nobody sees it in the final thing. If we get it right, players will feel the rightness and never know there was ever a wrong.
The thing that's kept me sane is a small realization: we're not actually stuck. Stuck feels like nothing moving. This isn't that. Every failed version is teaching us the exact contours of the problem — we know more about what "right" feels like today than we did two weeks ago, precisely
because we've felt so many versions of "not quite." We're not lost. We're triangulating.
I don't have the answer yet. I'm not going to pretend I do for the sake of a clean entry. But I've got a hunch that showed up in the shower this morning, which is historically where our best and worst ideas both come from. We'll try it tomorrow.
If it works, I'll tell you. If it doesn't, I'll tell you that too. That's sort of the whole point of doing this out loud.
— Thiago & the Pixel Sprout team
Posted by Thiago Moreira · July 2, 2026
← All news
