Entry 93

Story starts to the ending: events 15-31, where each opens, and what they run into

2026-09-25scripttest

Filescrates/piney-game/src/start.rscrates/piney-game/src/main.rsdocs/engine/event-vm.md

82's story starts stopped at event 14. Infection's main story goes on as one chain to its ending. --mode story:N now reaches all of it.

The chain

Each event from 14 to 31 has one open condition on the story, event_done of the event before. 31 is ENDING, which also wants game_status 2, the desktop. So the story is events 1-4 and 10-31, and a start at event N replays every earlier one with ccEventFlagSet, as before.

Where each opens

Each event's first block that names a place gives it:

• The town starts. They enter Mac Anu from Log in, as 11 and 13 do.
• The party. Logging out empties it, so these starts need none.
• The existing places are enough. Every event from 15 on opens on the desktop, the board or Mac Anu, which the starts already had.

The replay's virus_core

virus_core at level 1 calls WORLD_MAN::SimGenerateCode (main 0x0019e5c0) on the area's words. That sets the field generator's seed and randcnt and WORLD_MAN's generated area, not the save, and a start's own scene set-up makes them again. The replay host implements it as nothing to keep. The protect bit and the protect items the instruction also handles are the interpreter's.

What the new starts run into

Each ran 4,000 frames with X every 25 frames.

• Starts 15-21 and 23-30. They open their event and run without a host default. Only a talk-through was tried, so a start that needs a walk somewhere has only begun.
• 31, the ending. Its stream 15 plays on the desktop's set-up.
• 22 (PIRO02). piros_colour and its wait (begin and end) are missing. More in the list below.

Checked

• The story-start test. story_starts_open_their_event now covers all 24 starts: each opens its event's first block, and no earlier event plays again.
• The rest. The workspace's tests, clippy, fmt and the docs check pass.