Rooba
/
ReC98
mirror of https://github.com/nmlgc/ReC98.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity
ReC98/th02/main/spark.hpp

48 lines
1.2 KiB
C++
Raw Normal View History

[Reverse-engineering] [th02] Sparks: Structure First taste of how sprites are managed in this game. Expressing the current and previous position in terms of the VRAM page is kind of quirky, but not as bad as expected. What *is* bad, however, is that the most natural next function is undecompilable thanks to an inconsistent `LEAVE` instruction in its epilog 🙄 And `codegen.hpp` has almost none of the instructions we would need to hide the SI and DI registers from the compiler, and I'm *really* not in the mood for another such mess anyway. I'd really like to try patching these mistakes in a separate build step instead, but unfortunately I'm out of research money as well. So let's go for something else next. Part of P0237, funded by Yanga.
2023-03-18 17:05:05 +00:00
/// Spark sprites
/// -------------
[Maintenance] #include each header's dependencies within the header itself OK, this is the big one. We still keep using `#include` guards only where we absolutely need to, but with each header now being valid in isolation, this can now actually help *minimize* the length of each translation unit's `#include` list. Turns out that after removing all the duplicates, we only *actually* need to guard 29 headers across all 5 games. Part of P0285, funded by [Anonymous] and iruleatgames.
2024-05-25 18:31:09 +00:00
#include "th02/main/entity.hpp"
#include "th01/math/subpixel.hpp"
[Reverse-engineering] [th02] Sparks: Structure First taste of how sprites are managed in this game. Expressing the current and previous position in terms of the VRAM page is kind of quirky, but not as bad as expected. What *is* bad, however, is that the most natural next function is undecompilable thanks to an inconsistent `LEAVE` instruction in its epilog 🙄 And `codegen.hpp` has almost none of the instructions we would need to hide the SI and DI registers from the compiler, and I'm *really* not in the mood for another such mess anyway. I'd really like to try patching these mistakes in a separate build step instead, but unfortunately I'm out of research money as well. So let's go for something else next. Part of P0237, funded by Yanga.
2023-03-18 17:05:05 +00:00
enum spark_render_as_t {
SRA_DOT = 0, // 1×1-pixel dot
SRA_SPRITE = 14, // Hardcoded 8×8-pixel spark sprite
};
struct spark_t {
entity_flag_t flag;
uint8_t age;
// Stores the current and previous position, indexed with the currently
// rendered VRAM page.
SPPoint screen_topleft[2];
SPPoint velocity;
spark_render_as_t render_as;
int8_t unused_1;
// Defaults, initialized at the beginning of each stage
// ----------------------------------------------------
unsigned char angle;
SubpixelLength8 speed_base;
int8_t unused_2;
spark_render_as_t default_render_as;
// ----------------------------------------------------
};
static const int SPARK_COUNT = 64;
extern spark_t sparks[SPARK_COUNT];
// Index of the next spark_t instance within [sparks] to be set when adding a
// new spark, regardless of its [flag].
extern unsigned int spark_ring_i;
// Number of spark_t instances within [sparks] between two successive
// SRA_SPRITE instances. All others get initialized to SRA_DOT.
extern uint8_t spark_sprite_interval;
// Sparks are transitioned to F_REMOVE after their [age] reaches this value.
extern uint8_t spark_age_max;