Fall back to the DOS soundtracks in Windows CD versions if the user picks
a non-General MIDI music device, as the Windows tracks only contain MIDI
music
Bug #3493317 ("SCUMM: Detecting Loom PCE without 16bpp support crashes")
is caused by an assert() in detection algorithm. In case an MD5 is found
the md5table, but the variant from the md5table is not found in
detection_tables.h this assert triggers. However since certain variants
can be left out compile-time this situation can occur. By ignoring
instead of assert()-ing the entry ScummVM will no longer abort but
continue the detection process.
This is a regression from 86a817beb5.
Formerly Kyra always queried the render_mode and initializes its Screen
functionality accordingly, even when the game actually only supported VGA.
While in the "Edit game..." options only VGA was selectable, the global options
still allowed for example EGA to be selected and thus messing up the internal
state. I only check the render_mode setting for EoB 1 and 2 now, which are
the only games which support the setting inside Screen.
The BMP decoder getPalette function definition is now identical to
the other image format decoders subclassed from ImageDecoder.
This also fixes a overloaded virtual warning reported by salty-horse.
This is the third and final part of the patch supplied by Ignaz Forster
on bug #3423955 ("CRUISE: Slow / unresponsive game behaviour") for
avoiding blocking graphical updates during user wait loops.
The removal of the continue statement fixes the update of background
animations while waiting for user to respond i.e. try opening a locked
door on the upper deck of the boat and check the sea background
animation, but it also has the side effect of allowing hotspots to
respond during this period if the user clicks fast enough.
However, this bug was also present in the original interpreter, and
a workaround has been added to inhibit this.
This is the second part of the patch supplied by Ignaz Forster on
bug #3423955 ("CRUISE: Slow / unresponsive game behaviour") for
avoiding blocking graphical updates during user wait loops.
This removes the check for fastMode from the cursor update code and
other code in this "idle" loop, and moves it down to qualifying
only a few of the function calls.
This is the first part of the patch supplied by Ignaz Forster on
bug #3423955 ("CRUISE: Slow / unresponsive game behaviour") for
avoiding blocking graphical updates during user wait loops.
Hotspots passed during the idle loop do _not_ give any user feedback
without this patch.
The original patch has been split to allow for better understanding of
the changes and to allow fine grained bisection if this introduces any
regressions.
This is a situation that would occur when a user copies drscula save
files from one device to another, without copying the corresponding
index, or copies the index without all corresponding save files.
This removes the large static tables from the binary (which saves 500K
to 1Mb of binary size) and replaced them with a class which generates
the required tables as needed in RAM. This has been tested with QDM2
and shows no obvious performance degredation and Memprof shows no
significant rise in RAM usage.
