libretro/scummvm
1
0
Fork 0
mirror of https://github.com/libretro/scummvm.git synced 2024-12-12 03:56:20 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

SCI: Use a simpler atan implementation for kGetAngle in SCI1 and newer games

Browse Source 
SCI1 games (QFG2 and newer) use a simpler and more accurate atan implementation
for kGetAngle. This properly fixes bug #3540976.
This commit is contained in:
Filippos Karapetis 2012-08-21 03:28:34 +03:00
parent 5df8c99768
commit 9568b78bab
1 changed files with 4 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
engines/sci/engine/kmath.cpp
View File

@ -84,27 +84,10 @@ reg_t kSqrt(EngineState *s, int argc, reg_t *argv) {
* accurate.
*/
uint16 kGetAngleWorker(int16 x1, int16 y1, int16 x2, int16 y2) {
// TODO: This has been implemented based on behavior observed with a test
// program created with SCI Studio. However, the return values have subtle
// differences from the original, which uses custom implementation of atan().
// The differences in the return values are the cause of bug #3540976
// and perhaps bug #3037267 as well.
// The results of this function match the expected results of SCI0, but not
// SCI1 (hence the bug in Longbow). We need to find the point in history
// when this function was changed.
// HACK: Return the expected value for Longbow, scene 150 (bug #3540976).
// This is a temporary solution, till the function returns the expected
// results.
if (g_sci->getGameId() == GID_LONGBOW && g_sci->getEngineState()->currentRoomNumber() == 150) {
if (x1 == 207 && y1 == 88 && x2 == 107 && y2 == 184)
return 226;
}
#if 0
// A simpler atan2-based implementation
return (int16)(360 - atan2((double)(x1 - x2), (double)(y1 - y2)) * 57.2958) % 360;
#endif
// SCI1 games (QFG2 and newer) use a simple atan implementation. SCI0 games
// use a somewhat less accurate calculation (below).
if (getSciVersion() >= SCI_VERSION_1_EGA_ONLY)
return (int16)(360 - atan2((double)(x1 - x2), (double)(y1 - y2)) * 57.2958) % 360;
int16 xRel = x2 - x1;
int16 yRel = y1 - y2; // y-axis is mirrored.