Final Fantasy IV map corruption problem

[FlamePurge]

Hello, everyone! I asked this on Forum Slickness, but I figure asking around couldn't hurt.

I'm creating a Final Fantasy II US v1.1 mod--it's very near completion, but I have one last issue: some map corruption! These three maps, exclusive to the ending as far as I know, became horribly altered into some mess of garbage.

1. Palom, the little girl, and the toads at the lake in Mysidia.


2. Rydia, Leviathan, Asura, and the summons in the Throne Room in the Land of Summon Monsters. (It wasn't like this when I went there in the game itself!)


3. The topmost part of the Castle of Dwarf.


Does anyone here know how I might go about remedying this? I already compared the map data between both versions, and they appear to be identical.

Thank you so much in advance; have a lovely day!

[Bregalad]

Are you sure this bug comes from map corruption ? I doubt they'd store the same map two times (once for the game and once for the ending).

[FlamePurge]

Well, I'm not sure what it is, to be honest. I'm calling it "corruption" because I don't know what else to call it. And yes, the throne room and top of the castle are copies of their respective maps, apparently; they have two separate listings in the ROM, plus they weren't corrupted when I actually went there in-game.

[Lenophis]

Are you sure this bug comes from map corruption ? I doubt they'd store the same map two times (once for the game and once for the ending).

This is Square we're talking about, and with them nothing surprises me. Fun fact: FF6 uses the same map 3 times for the Beginner's School in Narshe, 16 times for the various Weapon/Armor/Item shops for some towns, and some towns have 3 maps themselves for various alterations during storyline.

Vivify, after looking at these screenshots again, it almost appears as though the tilesets were changed, and not necessarily the maps themselves. Have you changed any tilesets for any maps that may have appeared unused?

[STARWIN]

While I'm not an experienced hacker, I can think of two approaches.

1. Basic approach: set a graphics-related breakpoint that triggers when the corrupted tiles are chosen/drawn (not sure of graphics details). Have tracing on and backtrack the causality. If the amount of asm is too much to handle, try to get a similar trace from the unmodified game, and see where the logic path differs.

2. Inferior approach: If you can save conveniently enough before the ending and if the savegames between your hack and the original version are compatible enough, you could try undoing hack components and re-finish the game to narrow down the cause. (I assume that savestates won't work for this, but not sure)

I agree with the tileset point, worth checking. You could compare screenshots of corrupted/proper to see if the pattern itself changes.

[FlamePurge]

Vivify, after looking at these screenshots again, it almost appears as though the tilesets were changed, and not necessarily the maps themselves. Have you changed any tilesets for any maps that may have appeared unused?

The only tile set change I made was altering the circled S save points to the five-pointed star found in the Sylph Cave. Do you think I should try reverting them to the S ones and see if that makes a change?

I compared the original screenshots with the corrupted ones, and it all just looks like garbage to me. The dwarfs in the last screenshot have even had their positions moved around.

While I'm not an experienced hacker, I can think of two approaches.

1. Basic approach: set a graphics-related breakpoint that triggers when the corrupted tiles are chosen/drawn (not sure of graphics details). Have tracing on and backtrack the causality. If the amount of asm is too much to handle, try to get a similar trace from the unmodified game, and see where the logic path differs.

2. Inferior approach: If you can save conveniently enough before the ending and if the savegames between your hack and the original version are compatible enough, you could try undoing hack components and re-finish the game to narrow down the cause. (I assume that savestates won't work for this, but not sure)

I agree with the tileset point, worth checking. You could compare screenshots of corrupted/proper to see if the pattern itself changes.

I'm not smart enough for 1, and 2 sounds extremely tedious. So, I guess that's it for this project. Thank you for your suggestions, though.

I guess I really shouldn't try hacking if I don't have the knowledge of doing complicated things.

[STARWIN]

I think it is natural that knowledge follows from ambition. That is, people figure out how something works if they really want to know. Or in reverse, no one figures anything out if the condition is that they must know everything about it first. No need to respect knowledge to that extent.

But you are right about tediousness. Being lazy is a reasonable choice, especially if this bug is cosmetic and outside of gameplay.