I'm not sure exactly how to describe this but here goes. I'm working on translating a game (well, it's already translated, I'm just adding the final touches on the "hacking" side) and there's an issue with a certain string of text.
Now, the 1K of VROM that contains the number tiles (0-9) (I'll call this chip A) is swapped in and out with another 1K of VROM that contains other graphics (chip B) that needed to be displayed on-screen. Thing is, for some reason, the right 1K of VROM (chip A) is loaded into Pattern Table #1 where it's supposed to be while the vram-write routine is writing the numbered tile's data to vram. But, here's the problem, once the screen refreshes it isn't the tile that was on chip A that gets displayed, but the one chip B. It *is* the right tile number and all; it's just the wrong chip that it's getting it from.
I'm not sure why this is. I'm thinking it have something to do with how the PPU refreshes, but I'm not sure as I've never really studied the PPU beyond basic function. If someone could maybe shed some light on this, I'd be most appreciative. The game uses Mapper #16, by the way.
Thanks.
Now, the 1K of VROM that contains the number tiles (0-9) (I'll call this chip A) is swapped in and out with another 1K of VROM that contains other graphics (chip B) that needed to be displayed on-screen. Thing is, for some reason, the right 1K of VROM (chip A) is loaded into Pattern Table #1 where it's supposed to be while the vram-write routine is writing the numbered tile's data to vram. But, here's the problem, once the screen refreshes it isn't the tile that was on chip A that gets displayed, but the one chip B. It *is* the right tile number and all; it's just the wrong chip that it's getting it from.
I'm not sure why this is. I'm thinking it have something to do with how the PPU refreshes, but I'm not sure as I've never really studied the PPU beyond basic function. If someone could maybe shed some light on this, I'd be most appreciative. The game uses Mapper #16, by the way.
Thanks.