thirdparty/scummvm
2
0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

SCI: adding patch for kq5 harpy volume

Browse source 
fixes actual game bug #3042277 - sierra switched in that case to highest volume possible

svn-id: r51999
This commit is contained in:
Martin Kiewitz 2010-08-11 20:39:13 +00:00
parent ff8b296c29
commit 60b2750209
2 changed files with 63 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

61
engines/sci/engine/script_patches.cpp
View file

@ -51,6 +51,7 @@ struct SciScriptSignature {
// - if not EOS, an adjust offset and the actual bytes
// - rinse and repeat
// ===========================================================================
// stayAndHelp::changeState (0) is called when ego swims to the left or right
// boundaries of room 660. Normally a textbox is supposed to get on screen
// but the call is wrong, so not only do we get an error message the script
@ -189,6 +190,7 @@ const SciScriptSignature gk1Signatures[] = {
{ 0, NULL, 0, 0, NULL, NULL }
};
// ===========================================================================
// this here gets called on entry and when going out of game windows
// uEvt::port will not get changed after kDisposeWindow but a bit later, so
// we would get an invalid port handle to a kSetPort call. We just patch in
@ -252,7 +254,63 @@ const SciScriptSignature hoyle4Signatures[] = {
{ 0, NULL, 0, 0, NULL, NULL }
};
// ===========================================================================
// at least during harpy scene export 29 of script 0 is called in kq5cd and
// has an issue for those calls, where temp 3 won't get inititialized, but
// is later used to set master volume. This issue makes sierra sci set
// the volume to max. We fix the export, so volume won't get modified in
// those cases.
const byte kq5SignatureCdHarpyVolume[] = {
34,
0x80, 0x91, 0x01, // lag global[191h]
0x18, // not
0x30, 0x2c, 0x00, // bnt [jump further] (jumping, if global 191h is 1)
0x35, 0x01, // ldi 01
0xa0, 0x91, 0x01, // sag global[191h] (setting global 191h to 1)
0x38, 0x7b, 0x01, // pushi 017b
0x76, // push0
0x81, 0x01, // lag global[1]
0x4a, 0x04, // send 04 (getting KQ5::masterVolume)
0xa5, 0x03, // sat temp[3] (store volume in temp 3)
0x38, 0x7b, 0x01, // pushi 017b
0x76, // push0
0x81, 0x01, // lag global[1]
0x4a, 0x04, // send 04 (getting KQ5::masterVolume)
0x36, // push
0x35, 0x04, // ldi 04
0x20, // ge? (followed by bnt)
0
};
const uint16 kq5PatchCdHarpyVolume[] = {
0x38, 0x2f, 0x02, // pushi 022f (selector theVol) (3 new bytes)
0x76, // push0 (1 new byte)
0x51, 0x88, // class SpeakTimer (2 new bytes)
0x4a, 0x04, // send 04 (2 new bytes) -> read SpeakTimer::theVol
0xa5, 0x03, // sat temp[3] (2 new bytes) -> write to temp 3
0x80, 0x91, 0x01, // lag global[191h]
// saving 1 byte due optimization
0x2e, 0x23, 0x00, // bt [jump further] (jumping, if global 191h is 1)
0x35, 0x01, // ldi 01
0xa0, 0x91, 0x01, // sag global[191h] (setting global 191h to 1)
0x38, 0x7b, 0x01, // pushi 017b
0x76, // push0
0x81, 0x01, // lag global[1]
0x4a, 0x04, // send 04 (getting KQ5::masterVolume)
0xa5, 0x03, // sat temp[3] (store volume in temp 3)
// saving 8 bytes due removing of duplicate code
0x39, 0x04, // pushi 04 (saving 1 byte due swapping)
0x22, // lt? (because we switched values)
PATCH_END
};
// script, description, magic DWORD, adjust
const SciScriptSignature kq5Signatures[] = {
{ 0, "CD: harpy volume change", PATCH_MAGICDWORD(0x80, 0x91, 0x01, 0x18), 0, kq5SignatureCdHarpyVolume, kq5PatchCdHarpyVolume },
{ 0, NULL, 0, 0, NULL, NULL }
};
// ===========================================================================
// this is called on every death dialog. Problem is at least the german
// version of lsl6 gets title text that is far too long for the
// available temp space resulting in temp space corruption
@ -303,6 +361,7 @@ const SciScriptSignature larry6Signatures[] = {
{ 0, NULL, 0, 0, NULL, NULL }
};
// ===========================================================================
// It seems to scripts warp ego outside the screen somehow (or maybe kDoBresen?)
// ego::mover is set to 0 and rm119::doit will crash in that case. This here
// fixes part of the problem and actually checks ego::mover to be 0 and skips
@ -417,6 +476,8 @@ void Script::matchSignatureAndPatch(uint16 scriptNr, byte *scriptData, const uin
// hoyle4 now works due workaround inside GfxPorts
// if (g_sci->getGameId() == GID_HOYLE4)
// signatureTable = hoyle4Signatures;
if (g_sci->getGameId() == GID_KQ5)
signatureTable = kq5Signatures;
if (g_sci->getGameId() == GID_LSL6)
signatureTable = larry6Signatures;
if (g_sci->getGameId() == GID_SQ5)