----Message d'origine----
Date: Fri, 2 Dec 2005 14:07:24 -0800 De: Toan T Nguyen nguyenthetoan@gmail.com A: wine-devel@winehq.org Sujet: wglGetPixelFormatAttribivARB unexpected RenderType(100)
Hi,
I compile and run Quake2Evolved in linux using winelib. Although every seems fine, there's this error printed out when Q2E scans for the best pixel format:
err:opengl:wglGetPixelFormatAttribivARB unexpected RenderType(100)
I look up my /usr/include/GL/glxATI.h and found these lines:
#ifndef GLX_ATI_pixel_format_float #define GLX_ATI_pixel_format_float 1
#define GLX_RGBA_FLOAT_ATI_BIT 0x00000100
#endif // GLX_ATI_pixel_format_float
Bests,
Toan
Hi
Interesting and fun extension :) Try attached patch if it works (and it should) forward it to wine-patches
Regards, Raphael
Hi,
Unfortunately, now Q2E cannot detects any valid pixel format. Here is the Q2E output from wine-0.9.2
============================== Initializing OpenGL driver ...getting DC: succeeded ...getting gamma ramp: failed fixme:opengl:wglMakeCurrent (0x398,0x47c91210) fixme:opengl:wglMakeCurrent created a delayed OpenGL context (0x7c3108d8) for 0x7c310740 fixme:opengl:wglMakeCurrent make current for dis 0x7c021e68, drawable 0x3400002, ctx 0x7c3 108d8 fixme:opengl:wglMakeCurrent returning True ...19 pixel formats found ...PIXELFORMAT 2 rejected, improper flags ...PIXELFORMAT 3 rejected, insufficient stencil bits (0 < 8) ...PIXELFORMAT 4 rejected, improper flags ...PIXELFORMAT 6 rejected, improper flags ...PIXELFORMAT 7 rejected, insufficient stencil bits (0 < 8) ...PIXELFORMAT 8 rejected, improper flags ...PIXELFORMAT 10 rejected, improper flags ...PIXELFORMAT 11 rejected, insufficient stencil bits (0 < 8) ...PIXELFORMAT 12 rejected, improper flags ...PIXELFORMAT 14 rejected, improper flags ...PIXELFORMAT 15 rejected, insufficient stencil bits (0 < 8) ...PIXELFORMAT 16 rejected, improper flags ...PIXELFORMAT 17 rejected, software emulation err:opengl:wglGetPixelFormatAttribivARB unexpected RenderType(100) ...PIXELFORMAT 18 rejected, software emulation err:opengl:wglGetPixelFormatAttribivARB unexpected RenderType(100) ...PIXELFORMAT 19 rejected, software emulation fixme:opengl:wglMakeCurrent ((nil),(nil)) fixme:opengl:wglMakeCurrent returning True ...hardware acceleration found ...PIXELFORMAT 1 selected ...creating GL context: succeeded ====================================
And here is the output from winecvs + your patch:
==================================== Initializing OpenGL driver ...getting DC: succeeded ...getting gamma ramp: failed ...19 pixel formats found ...PIXELFORMAT 1 rejected, insufficient color bits (1 < 32) ...PIXELFORMAT 2 rejected, improper flags ...PIXELFORMAT 3 rejected, insufficient color bits (1 < 32) ...PIXELFORMAT 4 rejected, improper flags ...PIXELFORMAT 5 rejected, insufficient color bits (1 < 32) ...PIXELFORMAT 6 rejected, improper flags ...PIXELFORMAT 7 rejected, insufficient color bits (1 < 32) ...PIXELFORMAT 8 rejected, improper flags ...PIXELFORMAT 9 rejected, insufficient color bits (1 < 32) ...PIXELFORMAT 10 rejected, improper flags ...PIXELFORMAT 11 rejected, insufficient color bits (1 < 32) ...PIXELFORMAT 12 rejected, improper flags ...PIXELFORMAT 13 rejected, insufficient color bits (1 < 32) ...PIXELFORMAT 14 rejected, improper flags ...PIXELFORMAT 15 rejected, insufficient color bits (1 < 32) ...PIXELFORMAT 16 rejected, improper flags ...PIXELFORMAT 17 rejected, software emulation ...PIXELFORMAT 18 rejected, software emulation ...PIXELFORMAT 19 rejected, software emulation ...no hardware acceleration found ...failed to find an appropriate PIXELFORMAT ...destroying window VID_AppActivate( 0 ) ...restoring display settings ...shutting down QGL ==========================
I can send you the source file of Q2E where OpenGL is initialized if you want.
Bests, Toan
On 12/2/05, fenix@club-internet.fr fenix@club-internet.fr wrote:
----Message d'origine----
Date: Fri, 2 Dec 2005 14:07:24 -0800 De: Toan T Nguyen nguyenthetoan@gmail.com A: wine-devel@winehq.org Sujet: wglGetPixelFormatAttribivARB unexpected RenderType(100)
Hi,
I compile and run Quake2Evolved in linux using winelib. Although every seems fine, there's this error printed out when Q2E scans for the best pixel format:
err:opengl:wglGetPixelFormatAttribivARB unexpected RenderType(100)
I look up my /usr/include/GL/glxATI.h and found these lines:
#ifndef GLX_ATI_pixel_format_float #define GLX_ATI_pixel_format_float 1
#define GLX_RGBA_FLOAT_ATI_BIT 0x00000100
#endif // GLX_ATI_pixel_format_float
Bests,
Toan
Hi
Interesting and fun extension :) Try attached patch if it works (and it should) forward it to wine-patches
Regards, Raphael
FYI, here is the code in question:
http://cvs.sourceforge.net/viewcvs.py/q2e/q2e/source/win32/glw_win.c?rev=1.1...
The function where pixelFormat is initialized is GLW_ChoosePixelFormat.
Toan
On 12/2/05, fenix@club-internet.fr fenix@club-internet.fr wrote:
----Message d'origine----
Date: Fri, 2 Dec 2005 14:07:24 -0800 De: Toan T Nguyen nguyenthetoan@gmail.com A: wine-devel@winehq.org Sujet: wglGetPixelFormatAttribivARB unexpected RenderType(100)
Hi,
I compile and run Quake2Evolved in linux using winelib. Although every seems fine, there's this error printed out when Q2E scans for the best pixel format:
err:opengl:wglGetPixelFormatAttribivARB unexpected RenderType(100)
I look up my /usr/include/GL/glxATI.h and found these lines:
#ifndef GLX_ATI_pixel_format_float #define GLX_ATI_pixel_format_float 1
#define GLX_RGBA_FLOAT_ATI_BIT 0x00000100
#endif // GLX_ATI_pixel_format_float
Bests,
Toan
Hi
Interesting and fun extension :) Try attached patch if it works (and it should) forward it to wine-patches
Regards, Raphael
On Saturday 03 December 2005 09:20, Toan T Nguyen wrote:
FYI, here is the code in question:
http://cvs.sourceforge.net/viewcvs.py/q2e/q2e/source/win32/glw_win.c?rev=1. 13&view=auto
The function where pixelFormat is initialized is GLW_ChoosePixelFormat.
Toan
Hi,
behavior seems strange to me
can you provide me output of gxlinfo and output of (using patch) WINEDEBUG="+opengl" q2....
Thx
Regards, Raphael
Here they are.
On 12/4/05, Raphael fenix@club-internet.fr wrote:
On Saturday 03 December 2005 09:20, Toan T Nguyen wrote:
FYI, here is the code in question:
http://cvs.sourceforge.net/viewcvs.py/q2e/q2e/source/win32/glw_win.c?rev=1. 13&view=auto
The function where pixelFormat is initialized is GLW_ChoosePixelFormat.
Toan
Hi,
behavior seems strange to me
can you provide me output of gxlinfo and output of (using patch) WINEDEBUG="+opengl" q2....
Thx
Regards, Raphael
On Monday 05 December 2005 04:41, Toan T Nguyen wrote:
Here they are.
On 12/4/05, Raphael fenix@club-internet.fr wrote:
On Saturday 03 December 2005 09:20, Toan T Nguyen wrote:
FYI, here is the code in question:
http://cvs.sourceforge.net/viewcvs.py/q2e/q2e/source/win32/glw_win.c?re v=1. 13&view=auto
The function where pixelFormat is initialized is GLW_ChoosePixelFormat.
Toan
Hi,
behavior seems strange to me
can you provide me output of gxlinfo and output of (using patch) WINEDEBUG="+opengl" q2....
Thx
Regards, Raphael
Thx
Can you retry with this patch ?
Thx
Regards, Raphael
Thanks very much. It works. There's a small probelm worth mentioning however: Q2E requires 32 color bits, 24bits depth, 8bits stencils. The ATI binary driver supports 24bits only. In wine-0.9.2, somehow wglGetPixelFormat returns 32 color bits but the current patch returns 24 only. I easily modified q2e to accept both 24bits and 32bits in Linux. Other programs might not be happy.
Toan
On 12/5/05, Raphael fenix@club-internet.fr wrote:
On Monday 05 December 2005 04:41, Toan T Nguyen wrote:
Here they are.
On 12/4/05, Raphael fenix@club-internet.fr wrote:
On Saturday 03 December 2005 09:20, Toan T Nguyen wrote:
FYI, here is the code in question:
http://cvs.sourceforge.net/viewcvs.py/q2e/q2e/source/win32/glw_win.c?re v=1. 13&view=auto
The function where pixelFormat is initialized is GLW_ChoosePixelFormat.
Toan
Hi,
behavior seems strange to me
can you provide me output of gxlinfo and output of (using patch) WINEDEBUG="+opengl" q2....
Thx
Regards, Raphael
Thx
Can you retry with this patch ?
Thx
Regards, Raphael
On Monday 05 December 2005 19:36, Toan T Nguyen wrote:
Thanks very much. It works. There's a small probelm worth mentioning however: Q2E requires 32 color bits, 24bits depth, 8bits stencils. The ATI binary driver supports 24bits only. In wine-0.9.2, somehow wglGetPixelFormat returns 32 color bits but the current patch returns 24 only. I easily modified q2e to accept both 24bits and 32bits in Linux. Other programs might not be happy.
WGL_COLOR_BITS_ARB cannot be 32 (was a bug i have corrected recently) see http://www.nvidia.com/dev_content/nvopenglspecs/WGL_ARB_pixel_format.txt
WGL_COLOR_BITS_ARB * The number of color bitplanes in each color buffer. For RGBA * pixel types, it is the size of the color buffer, excluding the * alpha bitplanes. For color-index pixels, it is the size of the * color index buffer.
So for 32 bits FBConfig, it should be 24 (8 bits per RGB channels) + 8 for Alpha channel
If Q2E expect 32 bits for WGL_COLOR_BITS_ARB its a bug :)
And 24 bits limitation is a X limitation (X display cannot support alpha-channel) not an ATI limitation. Anyway, 24 bits displays usually support 32 bits using GLX :)
Toan
Regards, Raphael
Hi, thanks for the clarification. Everything is fine then. Thank you for the patch.
BTW, do you know how to improve OpenGL performance in wine? I get about 10-30% drop in FPS compared to the Windows version (the card is overclocked about 10% both in CPU and memory frequencies in windows, so that might explains some of the performance drop). I set UseDGA to Y in "wine regedit" but that doesn't seem to help at all.
Thanks again, Toan
WGL_COLOR_BITS_ARB cannot be 32 (was a bug i have corrected recently) see http://www.nvidia.com/dev_content/nvopenglspecs/WGL_ARB_pixel_format.txt
WGL_COLOR_BITS_ARB
The number of color bitplanes in each color buffer. For RGBA
pixel types, it is the size of the color buffer, excluding the
alpha bitplanes. For color-index pixels, it is the size of the
color index buffer.
So for 32 bits FBConfig, it should be 24 (8 bits per RGB channels) + 8 for Alpha channel
If Q2E expect 32 bits for WGL_COLOR_BITS_ARB its a bug :)
And 24 bits limitation is a X limitation (X display cannot support alpha-channel) not an ATI limitation. Anyway, 24 bits displays usually support 32 bits using GLX :)
Toan
Regards, Raphael
BTW, do you know how to improve OpenGL performance in wine? I get about 10-30% drop in FPS compared to the Windows version (the card is overclocked about 10% both in CPU and memory frequencies in windows, so that might explains some of the performance drop). I set UseDGA to Y in "wine regedit" but that doesn't seem to help at all.
There are many factors * Overclocking, as you mentioned * Speed of the 3D driver. * Resource usage of background processen * Overhead of Wine of course, but that should be minimal for OpenGL. * and many more
The UseDGA has no influence on this, this affects only 2D acceleration of DirectDraw games.
On Monday 05 December 2005 23:53, Toan T Nguyen wrote:
Hi, thanks for the clarification. Everything is fine then. Thank you for the patch.
was Nothing
BTW, do you know how to improve OpenGL performance in wine? I get about 10-30% drop in FPS compared to the Windows version (the card is overclocked about 10% both in CPU and memory frequencies in windows, so that might explains some of the performance drop). I set UseDGA to Y in "wine regedit" but that doesn't seem to help at all.
10-30% is not wine overhead (as wine overhead for opengl is very small) i think it's a driver problem: no linux graphic drivers have the maturity andd performances of windows drivers :(
Thanks again, Toan
Regards, Raphael