The object ROM sheet | Ghosts'n Goblins

Draft. This sheet was read from the original 1985 drawing and cross-checked against MAME, but it has not yet been walked pin by pin against a physical board. The logic described here is what the drawing shows.

▸This sheet is part ofHow sprites work

The subsystem followed end to end, across every sheet that builds it.

What this sheet is for

Six sprite ROM positions, the shift registers that serialise what they return, and the logic that picks which ROM answers.

The drawing shows six 27128s at 1L, 2L, 3L, 1N, 2N and 3N. Ninety-six kilobytes of sprite artwork.

The jumper block

J2 to J6 are pads for zero-ohm links, and the drawing spells out the rule: short J2, J3 and J5 for 27256 parts, or J4 and J6 for 27128s. The difference is where A14 goes and which pin carries the chip select, since the two devices disagree about pin 27.

The photographed board takes the 27256 option in part. Column L holds a 27128 at the top and a 27256 at the bottom with the middle socket empty; column N is the same. Two small parts and two large ones still add up to the same 96K the drawing intends.

Why the data lines look scrambled

The ROM data pins are not wired straight through to the shift registers. Following the drawing, the order looks wrong.

It is deliberate. Reordering bits in copper costs nothing, and doing it here saves whatever logic would otherwise be needed to rearrange them later. The wiring is the operation. This catches every reader at least once.

The tile ROM's data lines are not wired straight to the shifters. Reordering them in copper costs nothing and saves the logic that would otherwise be needed to rearrange the bits. It makes the drawing look wrong until you realise the wiring IS the operation.

Every chip on this sheet

Click any of them to open the schematic with that chip selected in both panels.

← open the schematic for this sheet