Are 2-color games actually fun, or is it just nostalgia?

For me the charm wears off around week three, and I've learned to plan for it. I spent a winter making a little puzzle game for a chip with 2K of RAM, and the first two weeks were pure joy. Every byte mattered, every trick felt earned. By week three I was fighting the thing, wanting one more sound channel, one more sprite, and resenting the hardware instead of loving it. What I noticed though is that the version I shipped was better than what I'd have made with unlimited resources. The constraints forced me to cut two mechanics that were dead weight anyway. So I think limitations make individual projects better, but they don't make a hobby sustainable, at least not for me. My routine now is to do a small project on something tiny, then take the lessons back to a board with more headroom. The discipline sticks even when the restrictions don't. I'd say if that shooter demo still sounds fun to you in a month, it was the gameplay. If you only remember it as 'that clever VGA trick', it was the novelty.

I've been on the other side of this for a few years now. I write little games for a PIC microcontroller in my evenings, 8K of flash, and I've never hit the wall the previous poster describes. My take is that the question only bites if you treat it as a playing hobby. Almost nobody is going to trade thier Switch for a two-color line buffer, and I stopped pretending mine were good games a while back. What keeps me coming back is the making part. Figuring out how to squeeze a decent jump arc out of a chip with no floating point is the actual game for me, and the playable result is just the receipt. So I'd say don't judge it by wheather you're still having fun playing the shooter after a month. Judge it by whether you're still itching to open the editor. If the answer is yes, the limits never become an obstacle, they become the whole point. If the answer is no, no amount of charm was going to save it anyway.

I do keep one modern board around for when I want sprites and actual sound, and honestly going back and forth between the two is what keeps both feeling fresh.

The second poster nailed it for me, treating it as a playing hobby is the whole trick. I built a little breakout clone on a tiny board last year and the fun part for me was never the game, it was getting the video timing right. I remember spending a whole weekend just figuring out how to squeeze the paddle drawing into the blanking interval, and that debugging session was more memorable than anything I've played on it since. So for me the limitation is charming when it's an engineering puzzle, less so when I'm actually trying to finish somethig playable.

The thing that kept me on tiny hardware was a spreadsheet, honestly. I had a column per scanline with the palette swap written out, and once I could see the whole frame on paper the two-color limit stopped feeling like a cage and started feeling like a puzzle. Filled that thing in on the bus, at lunch, whatever. The game I finished off the back of it was mediocre but I still open the sheet sometimes just to look at it. So my answer to your question is that the fun lives in the planning more than the playing. Does the shooter demo you tried come with source, or is it just a bin floating around?

The real test for me was showing it to people who don't care about chips. My nephew played my little two-color thing for ten minutes and asked when he could play the next level. No idea about the scanline tricks, didn't care. That told me more than any benchmark. Around the house I still judge my projects the same way, if someone who isn't into the hobby enjoys the result, the constraint worked. If only I appreciate it because I know how hard it was, that's when I start wondering if I built a demo instead of a game. Anyone here actually had non-technical friends play their stuff for more than a minute?

Tried porting a simple game I'd allready written onto one of these tiny boards and it was miserable. Felt like I was just subtracting stuff the whole time. Constraints only pay off when you design around them from day one, at least that's been my expirience. Now I'm curious whether that shooter demo was written natively for the hardware or ported, because I bet that's a big part of why it played well.

Update: tried the spreadsheet idea from the reply about mapping every scanline, palette swaps written out in columns. Filled half a frame in on the couch and yeah, the two-color limit felt more like a puzzle than a cage after that. Also sitting with the point about this being a making hobby, not a playing one. Question though: do you sketch the whole level on paper first, or only the tricky frames?