Engines with "simple" savenames would support "Run in background" in
save/load dialog and gradual save slots unlocking. Other engines
save/load feature would be locked until save sync is over.
Phant1, PQ:SWAT, GK2, and Phant2 all have different audio maps
and audio volumes on each CD. In order to make this work within
ScummVM, where CDs are never swapped, each RESOURCE.AUD for these
games must be renamed to RESAUD.00x and each RESOURCE.SFX renamed
to RESSFX.00x.
Added games are:
Hoyle Classic Games
Hoyle Bridge
Hoyle Children's Collection
Hoyle Solitaire (CD and Hard Drive versions)
Additionaly, kGetConfig was modified to support two settings used by
these games: "laptop" and "jumpto".
The demo versions of these games were using a very different engine -
SCI1.1 vs SCI2/SCI2.1. Thus, we split them into different game IDs,
to avoid mixing specific game checks for them, as well as specific
game workarounds, which are different for the demos than the full
versions. Also, the demos should be working when SCI32 is disabled.
For these games, we don't use ADGF_DEMO, to avoid game IDs like
foodemo-demo
Removed the superfluous initForDetection() function, which was not
updated in commit 2f17ba2b0a and
caused the fallback detector to crash because of uninitialized
variables
Auto-saving is not used by SCI, but slot 0 is ScummVM "standard"
for auto-saving, that's why it's not available for saving anymore.
Jones still uses slot 0 for saving/restoring (because it's
hardcoded and changing it would break it somewhat)
Deleting + restoring is still possible of couse.
Instead of choosing Windows as platform, users can now also simply
click this option for Gabriel Knight 1 + King's Quest 6
Defaults to high resolution graphics
This means tick-based times are saved to save games, as in SCI32
engine, instead of seconds, which are not accurate enough. It
also means places in SCI engine that need to access game ticks
should do so through g_sci instead of g_system or g_engine.
Original name + description were inaccurate. We don't actually
do undithering (besides the view undithering), but we instead
skip the EGA dithering code.
This bug occurs because in the cases specified in the bug report, the main
loop hasn't run fully yet, and there is a mini loop running instead (e.g.
inside Print())
Hopefully, this catches most cases where the crash occurs, but it needs more
testing to find if there is any other such case.
