I'm making a ZC and ZQ ripoff for school, and my team decided that we were going to scrap paletted arenas so we'd have better tiles.
Well, it sucks a lot.
When I want to have more than one tile for a different c-set, i have to import it as a different value. Just a small group of overworld tiles has cost me about 7 times that of a similar quest in ZC, and with no real benefits. Sure, I can rip tiles from other games directly ( can't use them cause this is a real project 😞), but it wouldn't matter anyway as these are all 32x32 tiles.
I say if we re-do palettes for Zquest, we just allow for something like 64 color palettes or even 256, but not let the user have a full range of colors to work with.
It's not all bad though. The transition to 32x32 tiles is simply amazing. You can sure put a lot of detail with that size.
Why 32-bit palettes don't work that well
If/when the palette system gets overhauled, the user would be able to choose between paletted/unpaletted tiles. They would even be mixed and matched.
and really, that's the way I wish I could do it. I'm not drawing the screen per pixel, so the concept of C-sets doesn't apply.
oh yeah, one more ZC/ZQ programming question. are tile pages loaded one line at a time, or are they done one line of one tile at a time? are the tile pages we see when we boot up the program anything like it is laid out in memory?
Basically, one line of one tile at a time. The tiles are currently laid out in memory as a series of bytes. Each byte represents 2 pixels. So, there ate 128 bytes in a tile. Each set of 8 bytes is a line in a tile. The tiles are in a series in memory, too. They are just laid out on screen in a 20x13 fashion.