The listing is complete: 32,768 bytes, none unexplained. That does not mean everything about the cartridge is known. What is not, is here.
CLOSED on 2026-09-09, at theNestruo's request. It is a drink that refills the player's energy bar when both bars are nearly empty, and the sign at 0x7525 is its announcement. Measured in openMSX with (0xE450) = 1: the bars go from (0x08, 0x0C) to (0x24, 0x0C). It is in the findings.
CLOSED on 2026-09-24, at theNestruo's request (issue #2). From the fight screen's own bands: the twelve wave screens dump the same patterns and colours as the fights, byte for byte, in openMSX (tools/omsx_oleadas.tcl). What was wrong was the reader here — the nibbles backwards, and the curtain at 0x41D1 that wipes the title before a game starts was missing — not the cartridge. The twelve are in the game.
The slot scan at 0xBF6C is a game looking for another game, which is not the Game Master mechanism. The obvious question is whether it appears anywhere else in the RC-7xx range. The trail to follow is a very early call out of INIT that touches 0xFCC1 and ENASLT.
Not checked across the rest of the series yet.
The scan tells the two builds of RC-725 apart, 5A 47 for the easy one and 23 70 for the hard one, and stores the same 1 in (0xE450) either way. So the cartridge goes to the trouble of recognising two builds separately and then does not distinguish them. Whether that is a leftover, a second table that was never used, or something read elsewhere, is not known.
0x44E4 fills the strip at 0xE2C0 with ten sceneries, 0 to 9, but the round wraps at eight (0x4311). Entries 8 and 9 are never read. Whether that is a leftover from a longer game or just a round number is not known.
Scene 6 (0x425E) handles the endings and 0x58F9 holds their captions, which read as a congratulation. They have not been reached in play here, so what the sequence looks like end to end is described from the code and not from having seen it.