Remember that optimistic moment where I said I was entering the “polish phase” of Unthreaded?

That was adorable.

What actually happened was less “polish” and more “quietly dismantling half the game because it refused to behave like a reasonable piece of software.”

Moving Screws: A Simple Idea Ruined by Reality

At some point, I decided that screws shouldn’t just sit there looking decorative. No, they should move. Add a bit of life. A bit of dynamism. Make levels feel more interesting.

A harmless idea.

Except the moment screws started moving, they also started ignoring every rule they were supposed to follow.

• Screws snapping to the wrong positions
• Stacks behaving like they’d lost all sense of order
• Logic breaking depending on timing
• Perfectly valid moves suddenly becoming impossible

Turns out when your entire game is built around precision and order, introducing movement is like inviting chaos in for tea and then acting surprised when it burns the house down.

The Domino Effect of “Small Changes”

The real problem wasn’t just that moving screws didn’t work.

It’s that they broke things that used to work.

Systems that had been stable suddenly became unreliable. Level logic that had been tested and working started producing edge cases that made no sense.

That’s the fun part of development nobody talks about enough:

You don’t just build forward.

Sometimes you build backwards.

A lot.

Level Design… or Level Un-Design

While all this was happening, level design took a hit.

Because what’s the point of carefully crafting levels when the underlying systems are having an identity crisis?

Puzzles that were once solvable became inconsistent. Others turned into accidental difficulty spikes. Some just stopped functioning entirely.

So instead of designing new levels, I’ve been:

• Re-testing old ones
• Fixing broken interactions
• Adjusting layouts that no longer behave properly
• Quietly questioning past decisions

It’s less “creative design” and more “forensic investigation.”

The Long Road Back to Where I Started

Here’s the part that stings a bit.

After all the experimentation, changes, and “improvements,” I’ve essentially spent a significant amount of time getting back to where I already was.

The same point I confidently referred to as:

“Ready for polish.”

There’s something uniquely humbling about realising that progress isn’t always forward. Sometimes it’s a loop.

You try something new. It breaks things. You fix it. You refine it. And eventually, if you’re lucky, you end up back at a stable foundation.

Just… hopefully better this time.

Lessons (Reluctantly Learned)

If there’s anything this phase has made painfully clear, it’s this:

• Systems need to be rock solid before adding complexity
• “Small features” are rarely small
• Movement + precision logic = potential disaster
• Testing isn’t optional, it’s survival

And most importantly:

Just because something sounds like a good idea doesn’t mean it deserves to be in the game.

Still Standing

Despite all of this, Unthreaded is still very much alive.

The systems are stabilising again. The behaviour is becoming predictable. The foundation is being rebuilt properly instead of optimistically.

It’s not the clean “polish phase” I had planned.

It’s slower. Messier. Slightly more frustrating.

But it’s real progress.

Even if it doesn’t look like it at first.

Back to Work

So the plan now is simple.

Stabilise everything.
Rebuild confidence in the systems.
Then, and only then, return to polish.

No more “quick additions.”
No more “this should be easy.”
No more accidental chaos disguised as features.

Just steady, stubborn progress.

Because apparently, that’s what finishing a game actually looks like.

If you’d like to support development, here’s a coffee link ☕