epigonone identifies the pointers that choose which tiles are drawn in the Prologue screen; they are visible here in TM; experimenting with different values @ $0051EF will likely place different nearby tiles into those 4 spots on the Prologue screen. Initial value is FD 86.

At FE 86 the 4 tiles chosen are quite close, on the next row. They are NOT immediately adjacent to the 4 selected by the initial value…

The pointer was only advanced by 1 bit (FD → FE) yet 4 tiles are skipped. The routine that processes this pointer’s value must “jump” those 4. Expectation is that when testing FF 86, it will skip the next 4, and choose the 4 below the initial selection.

This is indeed what happens with the pointer set at FF 86.

It’s unclear if these tiles are used in this romhack for a character sprite. 🤔 If they are not, then the tiles for the Moons can live here instead, and the issue is solved!

OK, wait… these 6 tiles look like they assemble into Anna.

Confirmed, by narrowing the rom view down to 2 columns from its default of 16.

Since the very next 4 tiles, which truly appear to be unused, are not targetable by this pointer… a space earlier (higher up, visually) will need to be found.

Scrolling up and looking for space that’s either the first or last color in the palette probably means those tiles’ data corresponds to all 00 or all FF in the rom, which is freespace. 4 tiles at the proper alignement first appear on the row at 0B6000.

Hitting the → key to shift the tiles until these are in the upper left corner will make their offset shown in the bottom of the TM window.

So getting the pointer for the Prologue moon tiles to aim at 0B6100 instead of where it’s aiming at when its value is FD 86 is the goal. So what are the initial tiles’ offset?