Every byte of the cartridge is accounted for and every routine carries a comment, but that does not mean everything is understood. What is still open, without dressing it up:
For these we know who loads them and where they end -from the loader and from where the next block starts- but not what each byte is for:
| block | bytes | who loads it |
|---|---|---|
| 0x4B7A | 13 | 0x4B6E, which copies them into the sprite buffer |
| 0x4B87 | 8 | 0x419B, and they are the small "KONAMI" on the title |
| 0x5886 | 40 | 0x5849, eight bytes from 0x5886 plus three times the level |
| 0x5C58..0x5C6D | 21 | four different loaders, all in the machine's play |
| 0x6A9C / 0x6AA8 | 12 and 12 | 0x6A85 and 0x6A53, the force and reach of a shot |
The 0x5886 one is the oddest: the stretches overlap, because the index advances by three and eight bytes are copied. That is measured, but whether the overlap is deliberate or there are simply spare bytes has not been settled.
0x4CF6 copies 21 bytes to 0xE07C from three starts -0x4D0A, 0x4D1B and 0x4D2C- that are 17 bytes apart. So each copy takes four bytes of the next one. They are declared that way, with the overlap stated, but whether anyone reads those four extra bytes is not known.
The level is known to come in at three places: the trajectory table at 0x5D37 -five levels by six zones-, the points threshold at 0x5C5E and the easiest level's latch. What has not been measured is how much each of the three shows in an actual match, or which weighs most.
The table at 0x6225 has twelve entries and matches to the byte by the first-entry fit, just like the scene one. It dispatches the phases of a rally. But the twelve have not been walked one by one in the emulator to see which comes up when: what there is, is the reading of the code.
0x6D15 points at four scripts that draw two vertical columns at the sides of the table, on a diagonal. What they paint and who picks them is known -bit 2 of the frame clock, so they flicker- but not what they represent: the crowd, the net seen edge-on, or the edge of the room.
If you know the answer to any of these -or find an error in what is published- it would be welcome: the point of this series is that the listing can be believed, and for that it has to be correctable.