8e5c79e439
d3d9: Remove superflous NULL checks (Coverity).
2011-06-27 13:28:13 -05:00
11bf170ce0
d3d9/tests: Add some tests for IDirect3DDevice9::UpdateSurface().
2011-06-16 11:16:22 +02:00
708d94212b
wined3d: Don't clamp vertex depth values.
...
Since we report D3DPMISCCAPS_CLIPTLVERTS.
2011-06-06 16:18:33 +02:00
301c6ef4e5
d3d9/tests: Added unbound sampler test for pixel shaders.
2011-06-03 13:41:48 +02:00
480259ca2c
d3d9/tests: Fix some ok() messages in the visual tests.
2011-05-25 17:54:38 +02:00
cf757a6361
d3d9: Use a pixel shader for the vs 3.0 input test.
2011-05-25 12:23:32 +02:00
d8224118ab
d3d: Make the INTZ test more reliable.
2011-05-25 12:23:24 +02:00
ea34d7aaba
d3d9/tests: Add a test for unwritten pixel shader outputs.
2011-05-25 11:49:27 +02:00
6d31824a6e
d3d9/tests: Fix the depth clamping test.
2011-05-25 11:48:29 +02:00
e7c55939c4
d3d9: Drop the fixed function varying test.
...
Windows drivers return different values, some don't even draw. I don't know
any app that depends on this.
2011-05-24 12:47:28 +02:00
f6b30a17b3
d3d9: AMD cards don't clip transformed vertices.
2011-05-24 12:47:28 +02:00
77c1454cf6
d3d9: Skip the YUV tests when the driver confuses the U and V channels.
2011-05-20 10:34:37 +02:00
803bad0959
d3d9: Account for another different behavior in the viewport test.
2011-05-20 10:34:30 +02:00
5107796fc7
d3d9: Disable the zero instance test.
...
It fails on modern Nvidia cards (Win7, Geforce 9600).
2011-05-19 10:09:08 +02:00
5920de5ffb
Assorted spelling and case fixes.
2011-05-09 12:09:27 +02:00
ed3f5081a3
d3d9/tests: Check if formats are supported for render targets in pixelshader_blending_test().
2011-05-06 14:23:46 +02:00
86d090ee8e
d3d9/tests: Add a depth blit test.
2011-04-06 21:17:53 +02:00
f212a56819
d3d9/tests: Add a test for partial depth buffer copies.
2011-04-05 11:13:56 +02:00
5e9884a488
d3d9: Test depth stencil size mismatches.
2011-03-16 16:31:54 +01:00
66f5234f3e
d3d9/tests: Make sure to use return values (LLVM/Clang).
2011-02-17 18:33:42 +01:00
c04f8ce055
d3d9/tests: Add a test for clip planes.
2011-02-15 17:24:26 +01:00
82a4373cd0
wined3d: Check if formats are supported as render target in srgbwrite_format_test().
...
Floating point formats in particular may not be supported everywhere.
2011-02-02 12:23:34 +01:00
0fb6005fa8
d3d9/tests: Check another return value in maxmip_test() (LLVM/Clang).
2011-02-02 12:21:20 +01:00
af641eb4fd
d3d9: Add an sRGB format test.
2011-02-01 12:04:35 +01:00
73c4fbcc51
d3d9: Work around an odd nvidia windows bug.
2011-02-01 12:03:24 +01:00
4125fa62b9
d3d9/tests: Check ColorFill() return codes in color_fill_test() (LLVM/Clang).
2011-02-01 12:01:06 +01:00
91ac0c37b1
wined3d: Disable vertex shader output clamping.
...
It looks like NV50+ hardware gets you infinities and NaN's in D3D as well for
most things, so we should only need special handling for pow and nrm.
2011-01-25 12:32:38 +01:00
ac2afd08ae
d3d9/tests: Don't draw undefined data in stream_test().
2010-12-07 11:33:20 +01:00
27b72fa847
d3d9/tests: Use GetRenderTargetData() for reading back pixels.
...
This is the recommended way to read back render targets. It's also much faster
on current wined3d, but that's mostly a side effect of imperfections in our
resource location management.
2010-11-23 13:15:29 +01:00
4f651808fc
wined3d: Disallow empty surfaces.
2010-11-18 12:37:36 +01:00
10ccba52e7
d3d9/tests: Fix some failure messages.
2010-11-17 14:01:14 +01:00
e279fa2afd
d3d9/tests: Add a test for INTZ.
2010-11-05 17:15:37 +01:00
c1b1b6f184
d3d9/tests: Make some variables static.
2010-10-01 13:36:20 +02:00
d911622722
d3d9/tests: Cleanup maxmip_test() a bit.
...
Having the failure message match the actual draw call in particular makes
debugging failures a lot nicer.
2010-09-21 14:16:54 +02:00
b099eeb8c4
d3d9/tests: Test NVDB (depth bounds test) behavior.
2010-09-16 10:12:39 -05:00
bb201bb510
d3d9/tests: Read from the appropriate pixels in test_constant_clamp_vs().
2010-07-30 11:36:03 +02:00
7b3d200ab3
d3d9/tests: Fix an ok() message in pointsize_test().
2010-07-29 17:33:35 +02:00
da0582a624
d3d9/tests: Add a test for handling floating point specials in shaders.
2010-07-29 17:33:35 +02:00
a554d1e296
d3d9: Remove the shader version varying tests.
...
My Windows 7 installation rejects mixing ps_3_0 with anything other than
vs_3_0 and vice versa, making testing how those versions pass varyings
moot. Since I have never seen an app mixing 3.0 and non-3.0 shaders I
see no reason to keep the tests.
2010-07-26 18:38:32 +02:00
0b630e1430
d3d9/tests: Add a test for clearing render targets in an MRT.
2010-07-23 11:46:51 +02:00
74015090ae
d3d9/tests: Introduce a vertex shader to the MRT test.
2010-07-23 11:46:41 +02:00
8284f70de8
d3d9: Use a vertex shader in the vpos test.
2010-07-21 13:23:59 +02:00
bb3fb05c38
d3d9: Use a vertex shader in the vFace test.
2010-07-21 13:23:59 +02:00
ba0a5ed104
d3d9: Use a vertex shader for the nested loop test.
...
Win7 (or newer Nvidia drivers) don't allow ps_3_0 fragment processing without
vertex shaders. This is interesting on its own, but not part of this test.
2010-07-21 13:23:59 +02:00
e534e2e2ac
d3d9: Win7+Nvidia returns c2 in COISSUE'd CND.
2010-07-21 13:23:59 +02:00
24adf40ee1
d3d9: Remove the fog vs sRGB test.
2010-07-21 13:23:59 +02:00
0190b6cb90
d3d9: Use color_match in the coissue tests.
2010-07-06 19:28:27 +02:00
b50a008549
d3d9: Windows7 doesn't draw with Viewports bigger than the framebuffer.
...
This is my Geforce 9600 setup here, the failure doesn't occur on all
windows 7 systems. I don't think we can change wined3d's behavior as a
number of games are broken on my Windows installation because they do
not like the extra parameter validation.
2010-07-06 19:28:27 +02:00
cb4ba6a862
d3d9: Ignore a Windows 7 failure in the d3d9 depth clamp test.
2010-07-06 19:28:27 +02:00
dd6cabcc1e
d3d9: Pass count=0 when clearing the whole surface.
2010-07-05 12:59:46 +02:00
218b9cec6f
d3d9: Nvidia drivers don't like unused output formats.
...
Newer nvidia drivers return garbage when using a COUNTx where x is bigger
than the amount of coordinates the texture can consume. So stick to COUNT3 for
volume textures. The tests don't need COUNT4 anyway.
2010-07-05 12:59:46 +02:00
95aaaf2f12
d3d9/tests: Add a test for offscreen point sprites.
2010-06-11 16:22:30 +02:00
c3cda92cf4
d3d9/tests: Remove an invalid point size test.
...
This test tests point size clamping behaviour when POINTSIZE_MAX <
POINTSIZE_MIN, but that behaviour isn't consistent between D3D drivers.
2010-05-28 16:22:26 +02:00
1c12ba44a7
d3d9/test: Introduce point_match() to simplify testing point sizes.
2010-05-28 16:22:26 +02:00
8c92f805f6
d3d9/tests: Don't depend so much on point rasterization rules in pointsize_test().
...
We can create a separate test for point rasterization if we really care about
that, but I doubt we do.
2010-05-28 16:22:26 +02:00
2210b9a6ac
d3d9/tests: Fix a typo.
2010-05-27 14:46:43 +02:00
123bfa20f5
d3d9/tests: Add a shadow sampler test.
2010-05-17 13:20:03 +02:00
2661e9abf7
d3d9/tests: Fix test_cube_wrap() for different texture filtering schemes.
2010-05-03 13:50:37 +02:00
67f43418d0
d3d9: Add an initial ColorFill regression test.
2010-04-14 18:00:47 +02:00
6ecdad06a3
d3d9/tests: Add some depth buffer tests.
2010-04-12 11:33:20 +02:00
cc20a0da59
d3d9/tests: Initialize a variable (Coverity).
2010-03-31 16:12:32 +02:00
342c6c6de5
d3d9: Add source/destination and backbuffer tests for StretchRect.
2010-03-30 13:55:28 +02:00
48d6167c29
d3d9: Remove old debug messages from the tests.
2010-03-17 16:10:15 +01:00
69861cf873
wined3d: Get rid of the depth range hack.
2010-01-27 11:16:07 +01:00
983f96b82f
d3d9: Add depth clamp tests.
2010-01-27 11:15:49 +01:00
14b343fcd0
d3d9/tests: Destroy the window after we release the device.
...
Although either order is supposed to work, the former crashes with Wine.
2009-12-03 15:26:52 +01:00
bdad5ace2a
d3d9/tests: Trace the viewport in clear_test().
...
There are some failures in this test on Win7 that appear to be related to the
initial viewport. Perhaps this trace will help narrow it down.
2009-12-01 13:16:22 +01:00
7f961c0796
d3d9: Abort the x8l8v8u8 test if the format is not supported.
2009-11-12 15:18:09 +01:00
b49a9cbfd2
d3d9: Run the visual tests in a window.
2009-11-09 19:59:33 +01:00
69612bb241
d3d9: Make the visual test read the backbuffer.
2009-11-09 19:59:29 +01:00
d17cd37d3b
d3d9/tests: Remove an unused color result.
2009-10-19 11:45:30 +02:00
023446f109
d3d9/tests: Check some more return values (LLVM/Clang).
2009-10-16 13:37:12 +02:00
2b5ff09160
d3d9/tests: Fix position of EndScene in visual test.
2009-10-15 12:06:17 +02:00
fcd0bc8f54
d3d9: Reduce the YUV test precision.
2009-10-08 12:03:20 +02:00
eda00d97ad
d3d9: Test viewports that are bigger than the surface.
2009-10-02 15:06:38 +02:00
dc45657f2c
wined3d: Fix vs_2_0 sgn.
...
vs_2_0's sgn has two extra parameters - temporary registers for the
instruction. This patch fixes the crash of the d3d9:visual test on windows and
properly handles the vs_2_0 version of sgn.
2009-10-02 12:36:59 +02:00
6fa09d1d22
d3d9: Fix some typos in visual test.
2009-09-22 20:19:58 +02:00
edf1c50b54
wined3d: Handle per-texture max LOD level.
...
GL_TEXTURE_BASE_LEVEL matches the basetexture::SetLOD functionality.
D3DSAMP_MAXMIPLEVEL essentially does the same as SetLOD. The test included in
this patch shows that the smallest mipmap level is used.
2009-08-31 12:18:41 +02:00
ffb38da024
wined3d: Try to force a top-left filling convention.
2009-08-19 14:45:09 +02:00
ba87299402
d3d9/tests: Detect empty graphics driver name.
2009-08-14 11:26:44 +02:00
3b06fc92f4
wined3d: Use EXT_provoking_vertex to match Direct3D's provoking vertex convention.
2009-08-10 13:52:50 +02:00
3115f3357f
d3d9/tests: Use color_match() in a few more places.
2009-07-10 12:13:32 +02:00
7d3beb6b79
d3d: Add a test for releasing the device after destroying the window.
2009-06-29 12:30:32 +02:00
d8669f9e41
d3d9/tests: Use color_match ((logical || with non-zero constant).
2009-06-23 11:47:38 +02:00
2364ce08e5
d3d9: Add a SGN test.
2009-06-22 11:26:14 +02:00
b93136e0c3
d3d9: Add an aL indexing test.
2009-06-22 11:26:07 +02:00
1e0b1e5487
d3d9: Use color_match() in the fog test.
2009-06-19 13:52:56 +02:00
f75dadfeb7
d3d9: Accept an alternative oFog->ps_3_0 handling.
2009-06-17 12:19:57 +02:00
6492622350
wined3d: sincos for vertex shaders.
...
SCS is unfortunately a fragment program only instruction. If we have the NV
extensions we can use SIN and COS. Otherwise we have to approximate sine and
cosine with a taylor series. Luckily we're provided with the necessary
constants by the application.
2009-05-27 11:56:53 +02:00
31f5b01f9a
wined3d: Add a point size test.
...
It shows that our current behavior (enabling coord_replace on all
units) is correct.
2009-05-08 12:17:04 +02:00
758777d1e2
d3d9: Fix a copy/paste error in test.
2009-04-27 13:23:19 +02:00
2df012a01a
d3d9/tests: Use LONG instead of long.
2009-03-23 12:41:14 +01:00
8e64efeeec
d3d9/tests: Fix some ok() strings in the fog test.
2009-03-12 13:09:54 +01:00
1b9a6fb4e9
d3d9: Fix some sign compare warnings.
2009-03-09 14:44:09 +01:00
7a4ed15884
d3d9/tests: Fix typos in text.
2009-03-04 15:45:43 +01:00
4aa0591c96
d3d9: Add an alpha blending test.
2009-02-24 16:57:30 +01:00
e51f2a0e6e
d3d9: Test table fog and exp fog in the shader fog test.
2009-02-11 12:22:00 +01:00
693d9ded47
wined3d: Properly set the fog frag coord according to the FOGTABLEMODE.
...
This also fixes the test that an earlier patch broke, so the todo_wine can be
removed again.
2009-02-11 12:21:44 +01:00
ace65a527f
d3d9: Initialize hr (Coverity).
2009-01-28 12:10:28 +01:00
50109aa969
wined3d: Get rid of last_was_foggy_shader & friends.
...
The fog settings do not depend on wether the shader writes to oFog or not,
instead they depend on the FOGVERTEXMODE and FOGTABLEMODE settings, and if a
vertex shader is bound at all.
It works the same way as with the fixed function, and having a vertex shader
is the same as using pretransformed vertices, just that the fog coord comes
from the shader instead of the specular color:
FOGTABLEMODE != NONE: The Z coord is used, oFog is ignored
FOGTABLEMODE == NONE, with VS: oFog is used
FOGTABLEMODE == NONE, no VS, XYZ: Z is used
FOGTABLEMODE == NONE, no VS, XYZRHW: diffuse color is used
2009-01-20 12:45:40 +01:00
Marcus Meissner