diff options
Diffstat (limited to 'engines/agi/op_cmd.cpp')
| -rw-r--r-- | engines/agi/op_cmd.cpp | 22 |
1 files changed, 22 insertions, 0 deletions
diff --git a/engines/agi/op_cmd.cpp b/engines/agi/op_cmd.cpp index 7418d043be..873c0c5a11 100644 --- a/engines/agi/op_cmd.cpp +++ b/engines/agi/op_cmd.cpp @@ -757,11 +757,33 @@ cmd(erase) { cmd(position) { vt.xPos = vt.xPos2 = p1; vt.yPos = vt.yPos2 = p2; + + // WORKAROUND: Part of the fix for bug #1659209 "AGI: Space Trek sprite duplication" + // with an accompanying identical workaround in position.v-command (i.e. command 0x26). + // These two workarounds together make up the whole fix. The bug was caused by + // wrongly written script data in Space Trek v1.0's scripts (At least logics 4 and 11). + // Position-command was called with horizontal values over 200 (Outside the screen!). + // Clipping the coordinates so the views stay wholly on-screen seems to fix the problems. + // It is probable (Would have to check better with disassembly to be completely sure) + // that AGI 2.440 clipped its coordinates in its position and position.v-commands + // although AGI 2.917 certainly doesn't (Checked that with disassembly) and that's why + // Space Trek may have worked better with AGI 2.440 than with some other AGI versions. + // I haven't checked but if Space Trek solely abuses the position-command we wouldn't + // strictly need the identical workaround in the position.v-command but it does make + // for a nice symmetry. + if (g_agi->getFeatures() & GF_CLIPCOORDS) + g_agi->clipViewCoordinates(&vt); } cmd(position_f) { vt.xPos = vt.xPos2 = _v[p1]; vt.yPos = vt.yPos2 = _v[p2]; + + // WORKAROUND: Part of the fix for bug #1659209 "AGI: Space Trek sprite duplication" + // with an accompanying identical workaround in position-command (i.e. command 0x25). + // See that workaround's comment for more in-depth information. + if (g_agi->getFeatures() & GF_CLIPCOORDS) + g_agi->clipViewCoordinates(&vt); } cmd(get_posn) { |