https://bugs.winehq.org/show_bug.cgi?id=40465
Bug ID: 40465 Summary: The sting demo is mainly black Product: Wine Version: 1.9.7 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs@winehq.org Reporter: jeremielapuree@yahoo.fr Distribution: ---
Created attachment 54223 --> https://bugs.winehq.org/attachment.cgi?id=54223 colsole output with ddraw and d3d7 channels enabled
The Sting demo does not crash anymore. But the screen is a lot black. Not completely black although. It does not look like the well known bug : black screen for ddraw game. It is surely related to the d3d3 version of ComputeSPhereVisibility that does not behave as the d3d7 version
https://bugs.winehq.org/show_bug.cgi?id=40465
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- URL| |http://www.fileplanet.com/6 | |2038/60000/fileinfo/The-Sti | |ng!-Demo-(Large)
https://bugs.winehq.org/show_bug.cgi?id=40465
--- Comment #1 from joaopa jeremielapuree@yahoo.fr --- Indeed, this is a problem of ComputeSphereVisibility in the d3d3 version. I returned D3DVIS_INSIDE_FRUSTRUM and the whole scene was visible (but with some too much enlighted places)!!
https://bugs.winehq.org/show_bug.cgi?id=40465
--- Comment #2 from joaopa jeremielapuree@yahoo.fr --- Does anyone know where I could download the SDK of direct 7 to compare ComputeSphereVisibility in Directx7 and Directx 3
https://bugs.winehq.org/show_bug.cgi?id=40465
--- Comment #3 from Matteo Bruni matteo.mystral@gmail.com --- I don't think there are legitimate places to download those old SDKs anymore.
They might not be necessary though. d3d7 returns the D3DSTATUS_* flags while d3d3 uses the D3DVIS_* flags, like you noticed. It should be possible to modify d3d_device3_ComputeSphereVisibility to convert from the d3d7-style return flags to the d3d3-style ones. Also (unrelated bug) notice that the d3d7 version should also check visibility with the user clipplanes, which apparently were a new feature in that DirectX version.
Anyway, the first step is to cleanup and port the ComputeSphereVisibility test from d3d.c to ddraw7.c and ddraw4.c.
https://bugs.winehq.org/show_bug.cgi?id=40465
--- Comment #4 from joaopa jeremielapuree@yahoo.fr --- Bug still there in current wine
https://bugs.winehq.org/show_bug.cgi?id=40465
winetest@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winetest@luukku.com
--- Comment #5 from winetest@luukku.com --- (In reply to joaopa from comment #1)
Indeed, this is a problem of ComputeSphereVisibility in the d3d3 version. I returned D3DVIS_INSIDE_FRUSTRUM and the whole scene was visible (but with some too much enlighted places)!!
Can you share how you did it? I could test it against other bugs.
https://bugs.winehq.org/show_bug.cgi?id=40465
--- Comment #6 from joaopa jeremielapuree@yahoo.fr --- In line 4606 of the file https://github.com/wine-mirror/wine/blob/master/dlls/ddraw/device.c
; change the line return IDirect3DDevice7_ComputeSphereVisibility(&device->IDirect3DDevice7_iface, centers, radii, sphere_count, flags, return_values);
with
return 0;
https://bugs.winehq.org/show_bug.cgi?id=40465
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution|--- |FIXED
--- Comment #7 from joaopa jeremielapuree@yahoo.fr --- The bug is fixed. There are graphical glitches, but they deserve their own bug reports.
Marking this one as FIXED.
https://bugs.winehq.org/show_bug.cgi?id=40465
Matteo Bruni matteo.mystral@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |8bfc9e66143e8cb079432ba9418 | |7f4e9baeb08d9
https://bugs.winehq.org/show_bug.cgi?id=40465
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #8 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 2.5.