https://bugs.winehq.org/show_bug.cgi?id=40180
Bug ID: 40180 Summary: Heroes Of Might & Magic VI: flying squares instead of smoke Product: Wine Version: 1.9.4 Hardware: x86 OS: Mac OS X Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs@winehq.org Reporter: isakov-sl@bk.ru
Created attachment 53688 --> https://bugs.winehq.org/attachment.cgi?id=53688 Dark green squares flying over swamps
I notice flying squares instead of smoke over acid swamps. wine-1.9.3-371 last testing version (~ 1.9.4) wine-staging-1.9.2 as well.
There is no essencial in the plain log. Tell me what can I trace to localize the issue?
https://bugs.winehq.org/show_bug.cgi?id=40180
--- Comment #1 from Matteo Bruni matteo.mystral@gmail.com --- Just a guess, does the message "Blending into a sRGB render target with no GL_ARB_framebuffer_sRGB support, expect rendering artifacts." appear in a +d3d_shader trace?
https://bugs.winehq.org/show_bug.cgi?id=40180
--- Comment #2 from Sergey Isakov isakov-sl@bk.ru --- (In reply to Matteo Bruni from comment #1)
Just a guess, does the message "Blending into a sRGB render target with no GL_ARB_framebuffer_sRGB support, expect rendering artifacts." appear in a +d3d_shader trace?
No such message. And this extension seems to se supported ~~~ trace:d3d:parse_extension_string - "GL_ARB_framebuffer_sRGB". trace:d3d:parse_extension_string FOUND: GL_ARB_framebuffer_sRGB support. ~~~
About blending I see shaders using them ~~~ trace:d3d_shader:shader_glsl_compile void main() trace:d3d_shader:shader_glsl_compile { trace:d3d_shader:shader_glsl_compile float m; trace:d3d_shader:shader_glsl_compile vec3 r; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_position = vs_in0; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_blendweight = vs_in1; trace:d3d_shader:shader_glsl_compile vec3 ffp_attrib_normal = vs_in3; trace:d3d_shader:shader_glsl_compile float ffp_attrib_psize = vs_in4; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_diffuse = vs_in5; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_specular = vs_in6; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_texcoord0 = vs_in7; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_texcoord1 = vs_in8; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_texcoord2 = vs_in9; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_texcoord3 = vs_in10; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_texcoord4 = vs_in11; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_texcoord5 = vs_in12; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_texcoord6 = vs_in13; trace:d3d_shader:shader_glsl_compile vec4 ffp_attrib_texcoord7 = vs_in14; trace:d3d_shader:shader_glsl_compile ffp_attrib_blendweight[0] = 1.0; trace:d3d_shader:shader_glsl_compile vec4 ec_pos = vec4(ffp_attrib_position.xyz, 1.0); trace:d3d_shader:shader_glsl_compile gl_Position = ffp_projection_matrix * ec_pos; --
And such messages ~~~ trace:d3d_shader:shader_record_register_usage Not recording register of type 0xa and [0x1][0xffffffff]. trace:d3d_shader:shader_record_register_usage Not recording register of type 0xa and [0][0xffffffff]. trace:d3d_shader:shader_record_register_usage Not recording register of type 0xa and [0x2][0xffffffff]. trace:d3d_shader:shader_record_register_usage Not recording register of type 0xa and [0x3][0xffffffff]. ~~~
I have a version where the bug absent but I don't know which patch is a key. May be 39057? It influences on log.
MacOSX 10.9.5 Radeon HD6770M 0x6740 Turks family Used native libraries d3dx9_36, d3dx9_40, d3dcompiler_43.
https://bugs.winehq.org/show_bug.cgi?id=40180
--- Comment #3 from Sergey Isakov isakov-sl@bk.ru --- That's all. This is the dupe of bug 38216. The key for good picture is ~~~~ $ git diff diff --git a/dlls/wined3d/directx.c b/dlls/wined3d/directx.c index 5f995ce..346c590 100644 --- a/dlls/wined3d/directx.c +++ b/dlls/wined3d/directx.c @@ -1846,6 +1846,7 @@ cards_amd_binary[] = {"HD 7800", CARD_AMD_RADEON_HD7800}, {"HD 7700", CARD_AMD_RADEON_HD7700}, /* Northern Islands */ + {"Turks", CARD_AMD_RADEON_HD6600}, {"HD 6970", CARD_AMD_RADEON_HD6900}, {"HD 6900", CARD_AMD_RADEON_HD6900}, {"HD 6800", CARD_AMD_RADEON_HD6800}, ~~~~~~
Log before (+d3d) ~~~~ trace:d3d:wined3d_guess_card Unrecognized renderer "AMD Radeon Turks XT Prototype OpenGL Engine", falling back to default. trace:d3d:wined3d_adapter_init_gl_caps Found (fake) card: 0x1002 (vendor id), 0x71c2 (device id). trace:d3d:fixup_extensions Applying driver quirk "AMD normalized texrect quirk". trace:d3d:quirk_amd_dx9 GL_ARB_texture_non_power_of_two advertised on R500 or earlier card, removing. ~~~~~
Log after ~~~~ trace:d3d:fixup_extensions Applying driver quirk "Init texcoord .w for Apple ATI >= r600 GPU driver". trace:d3d:quirk_texcoord_w Enabling vertex texture coord fixes in vertex shaders. ~~~~ I think these fixes is a key for the good picture.
https://bugs.winehq.org/show_bug.cgi?id=40180
--- Comment #4 from Sergey Isakov isakov-sl@bk.ru --- First quirk is wrong for me. Second not affects.
https://bugs.winehq.org/show_bug.cgi?id=40180
--- Comment #5 from Sergey Isakov isakov-sl@bk.ru --- Re-confirm with 1.9.10
https://bugs.winehq.org/show_bug.cgi?id=40180
winetest@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |00cpxxx@gmail.com, | |dimesio@earthlink.net, | |winetest@luukku.com
--- Comment #6 from winetest@luukku.com --- (In reply to Sergey Isakov from comment #3)
That's all. This is the dupe of bug 38216. The key for good picture is
$ git diff diff --git a/dlls/wined3d/directx.c b/dlls/wined3d/directx.c index 5f995ce..346c590 100644 --- a/dlls/wined3d/directx.c +++ b/dlls/wined3d/directx.c @@ -1846,6 +1846,7 @@ cards_amd_binary[] = {"HD 7800", CARD_AMD_RADEON_HD7800}, {"HD 7700", CARD_AMD_RADEON_HD7700}, /* Northern Islands */ + {"Turks", CARD_AMD_RADEON_HD6600}, {"HD 6970", CARD_AMD_RADEON_HD6900}, {"HD 6900", CARD_AMD_RADEON_HD6900}, {"HD 6800", CARD_AMD_RADEON_HD6800},
https://bugs.winehq.org/show_bug.cgi?id=40180
--- Comment #7 from winetest@luukku.com --- (In reply to Sergey Isakov from comment #3)
That's all. This is the dupe of bug 38216. The key for good picture is
https://bugs.winehq.org/show_bug.cgi?id=40180
--- Comment #8 from Sergey Isakov isakov-sl@bk.ru --- Wine-2.0-rc4-38-ga528847 The bug is still here
https://bugs.winehq.org/show_bug.cgi?id=40180
Sergey Isakov isakov-sl@bk.ru changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |DUPLICATE Status|UNCONFIRMED |RESOLVED
--- Comment #9 from Sergey Isakov isakov-sl@bk.ru --- The bug seems to be a duplicate of 38216 and cured by the same orkaround
*** This bug has been marked as a duplicate of bug 38216 ***
https://bugs.winehq.org/show_bug.cgi?id=40180
--- Comment #10 from Sergey Isakov isakov-sl@bk.ru --- Test again. Strange, looks like the bug is resolved!
https://bugs.winehq.org/show_bug.cgi?id=40180
Bruno Jesus 00cpxxx@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #11 from Bruno Jesus 00cpxxx@gmail.com --- Closing duplicate bugs.