A few of my friends asked me to do a couple devlog-type posts... this seems like a good place to do that? I don't really have a blog anymore so I'll devlog here for now.
Anyway, yesterday I did some informal "user testing", had a couple friends play through the game while I watched. It was a really fun and fascinating experience, I have pages of notes now. It was tough to keep my mouth shut and not give hints when they misunderstood things but I'm really glad I did. I wish I could do this once a week with new people each time. Maybe twitch should make that a service.
They managed to break the 3rd puzzle, the screen that introduces toggle switches. It was fun but also a big problem, because that means they didn't learn how toggles really work and floundered a lot on the 4th puzzle, which relies on that knowledge and also introduces bumpers. So I'll have to come up with a way to disallow the easier solution.
For some reason in my head I've always found it easy to remember what I've wired up to what, but they were constantly getting confused about what they'd wired, so I've got 3-4 ideas about how to clarify things, by changing the color and shape of wires and connections depending on state.
I'd also love to implement some sort of "debug view", where you can see (but not modify) the robot's internals while it is running through a puzzle, rather than having to rely purely on your memory of what you wired up. I'm still trying to come up with a good way how I might fit that into the limitations of pico-8 though.
We also came up with a great idea on how to better explain everything on the and/or/not logic gate tutorial screen.