
Sun Mar 29 15:18:57 1998 Uwe Bonnes <elektron.ikp.physik.tu-darmstadt.de> * [msdos/int21.c] More verbose Trace messages Sun Mar 29 15:03:30 1998 Alexandre Julliard <julliard@lrc.epfl.ch> * [loader/ne_image.c] Terminate relocation chains on 0 offset. Sun Mar 29 02:35:45 1998 James Juran <jrj120@psu.edu> * [windows/caret.c] [windows/class.c] [windows/clipboard.c] Fixed more USER32 ordinal numbers in function documentation. Sat Mar 28 22:40:23 1997 Andreas Mohr <100.30936@germany.net> * [controls/desktop.c] Return 1 for WM_NCCREATE in DesktopWndProc. * [controls/menu.c] Fixed crash when destroying a top popup win that was only created but not shown. * [ole/typelib.c] [if1632/typelib.spec] Implemented OABuildVersion. * [windows/win.c] Finally fixed ugly bug in GetParent32 that affected window placement: Didn't return 0 for windows that have neither WS_POPUP nor WS_CHILD set. Now it acts exactly like Windows (at least I hope so ;) Sat Mar 28 02:26:43 1998 Dimitrie O. Paun <dimi@mail.cs.toronto.edu> * [DEVELOPERS-HINTS] [documentation/debug-msgs] Updated documentation on debug messages * [include/button.h] [controls/button.c] [include/windows.h] [include/combo.h] [controls/combo.c] [controls/commctrl.c] [controls/icontitle.c] [controls/listbox.c] [controls/menu.c] [controls/scroll.c] [controls/widgets.c] Changed some 16-bit code to 32-bit code. I am positive this will not break anything. * [controls/uitools.c] Renamed the prefix of some functions from UITOOLS_ to UITOOLS95_ to reflect the fact that they implement Win95 look. Maybe we will have a UITOOLS31_ in the future. Anyhow, the change is completely internal to this file. * [*/*.c] Changed a lot of [f]printf into appropriate debug messages. Fri Mar 27 19:56:12 1998 Marcus Meissner <msmeissn@cip.informatik.uni-erlangen.de> * [documentation/status/] Added directdraw and directsound status. * [if1632/thunk.c][ole/*.c][win32/kernel32.c] Added some documentation. Moved FUNC004 to thunk.c since it seems to be a 16->32 thunk (calling conventions and the like still unknown). Fri Mar 27 09:59:32 1998 Morten Eriksen <mortene@pvv.org> * [loader/resource.c][windows/keyboard.c] Implemented CopyAcceleratorTable() and CreateAcceleratorTable(). * [include/compobj.h][ole/compobj.c][relay32/ole32.spec] Added implementation of StringFromGUID2(). Tue Mar 26 23:12:05 1998 Eric Kohl <ekohl@abo.rhein-zeitung.de> * [misc/imagelist.c][relay32/comctl32.spec] Implemented new debugging interface. Added missing functions (some are empty stubs). Changed ordinals in comctl32.spec (now dll version 4.72 comatible). * [objects/oembitmaps.c][include/windows.h] [include/bitmaps/*] Added missing restore bitmaps for Win95 look. Added missing Win95 cursors. Thu Mar 26 10:18:20 1998 Douglas Ridgway <ridgway@winehq.com> * [programs/view/Makefile.in] [programs/view/view.c] Fix compilation, Aldus placeable metafile loading. * [graphics/metafiledrv/init.c] [relay32/gdi32.spec] [objects/metafile.c] Implemented various 32 bit versions of regular metafile API. * [objects/enhmetafile.c] Implemented rendering of a couple of dozen records. Tue Mar 24 20:06:39 1998 Matthew Becker <mbecker@glasscity.net> * [memory/*.c] Updated documentation for API manpages. Mon Mar 23 09:09:00 1998 Alex Korobka <alex@pharm.sunysb.edu> * [misc/winsock.c] Fixed 32-bit DNS lookup. Mon Mar 23 23:54:47 1998 Luiz Otavio L. Zorzella <zorzella@nr.conexware.com> * [multimedia/init.c] New code for the initialization and gathering of information about the MIDI devices, for future use. Thu Mar 19 00:59:29 1998 Jim Peterson <jspeter@nospam.birch.ee.vt.edu> * [windows/win.c] Eliminated inaccurate setting of cs->hMenu in WIN_CreateWindowEx, along with superfluous SetMenu32 call. Mon Mar 16 17:40:53 1998 Petter Reinholdtsen <pere@td.org.uit.no> * [debugger/dbg.y] [debugger/Makefile.in] [Makefile.in] Install wine.sym in $(exec_prefix)/lib and make sure the debugger finds it. Sun Mar 15 22:36:35 1998 Michael Veksler <mveksler@tx.technion.ac.il> * [dummy.c] [if1632/gdi.spec] Fixed stub declaration of GDI_532, so Hebrew-Word2 kind of works.
75 lines
3.4 KiB
Plaintext
75 lines
3.4 KiB
Plaintext
This file contains information on the current implementation of the DirectDraw
|
|
API.
|
|
|
|
The DirectDraw API is being used in a lot of current computergames. Its API
|
|
layer is represented by the functions in the Win32 DLL ddraw.dll and the
|
|
IDirectDraw* interfaces (that can be created using the API functions).
|
|
|
|
WINE currently implements a very basic set of the DirectDraw functionality
|
|
in graphics/ddraw.c. This implementation uses the XFree86-DGA extension
|
|
to get very fast access to the graphics card framebuffer and doublebuffering
|
|
features. It is as fast as the MS equivalent for the stuff that is implemented.
|
|
Several applications already work, see below.
|
|
Problems of the implementation using XFree86-DGA:
|
|
|
|
- XFree86 cannot switch resolution nor depth on the fly.
|
|
This is a problem with X and unavoidable.
|
|
Current solution is to pop up a MessageBox with an error for
|
|
mismatched parameters and advice the user to restart the X server
|
|
with the specified depth and/or screensize.
|
|
- The rest of the functionality that has to be implemented will have
|
|
to be done in software and will be very slow.
|
|
- This requires WINE to be run as root user so XF86DGA can mmap the
|
|
framebuffer into the addressspace of the process.
|
|
- Blocks all other X windowed applications.
|
|
|
|
TODO:
|
|
- Add more implementations of the DirectDraw API.
|
|
Primary candidate is a raw Xlib implementation (one with/one without
|
|
XShm extension).
|
|
An additional layer (basically a HAL) will be required to decide on
|
|
case-by-case basis which implementation to use.
|
|
|
|
Status:
|
|
- Diablo [640x480x8]:
|
|
The movies play with speed comparable to the MS DDRAW one.
|
|
The game doesn't work, it somehow tries to write into 2 lines _BEFORE_
|
|
the start of the surface. I have tried to investigate this, but it is
|
|
in code that is automatically generated.
|
|
|
|
- WingCommander 4 / Win95 Patch [640x480x8]:
|
|
The intromovie plays, in 8 bit mode (to reconfigure wc4, run wine
|
|
"wc4w.exe -I"). The 16bit mode looks broken on Mystique, but I think
|
|
this is due to the unusual colorweight (565 rgb) of the Mystique in 16bit
|
|
mode. [Specifying it in DDPIXELFORMAT didn't help.]
|
|
|
|
Requires to be run in 640x480xdepth mode (doesn't seem to heed
|
|
DDSURFACEDESC.lPitch).
|
|
You can even fly the first mission with Maniac (ignore the weird sounds),
|
|
but it crashes as soon as you arrive at Blue Point Station...
|
|
|
|
- Monkey Island 3 [640x480x8]:
|
|
Goes to the easy/hard selection screen, then hangs due to multithreaded
|
|
windows-messagehandling problems.
|
|
|
|
- DiscWorld 2 [640x480x8]:
|
|
[Crashes with 'cli' in WINE released version. Yes. Privileged instructions
|
|
in 32bit code. Will they ever learn...]
|
|
Plays through nearly all intro movies. Sound and animation skip a lot of
|
|
stuff (possible DirectSound asynchronization problem).
|
|
|
|
- XvT [640x480x16]:
|
|
Shows the splash screen, then fails with missing Joystick. I have
|
|
investigated but failed to find a reason for this.
|
|
|
|
- Tomb Raider 2 Demo (using 8 bit renderer) [640x480x8]:
|
|
I have played nearly all of the DEMO level using Keyboard Controls.
|
|
Sound is a bit weird. Keyboard/Mouse handling too due to incomplete
|
|
DirectInput.
|
|
|
|
- WingCommander Prophecy Demo (using software renderer) [640x480x16]:
|
|
[Crashes with an invalid opcode (outb, probably to test for 3dFX) in the
|
|
WINE release version.]
|
|
Plays trailer, hangs in selection screen (no keyboard input, DirectInput
|
|
problem).
|