82521353f5
d3d9: Make the shader test more precision tolerant.
2008-04-07 12:05:06 +02:00
c65020646c
d3d9: Run the visual test with hardware vertex processing.
2008-04-07 11:49:39 +02:00
5b7ad3ad61
d3d9: Fix a few ok conditions in the visual tests.
2008-04-07 11:49:27 +02:00
b245c37ce6
d3d9: Accept nvidia hardware vp garbage.
...
Apparently nvidia cards do not like 3.0 vshader + fixed function fragment
processing, so accept their result. This has no implementation implications
for Wine because we need the code to support the 3.0+ffp combination to
handle 3.0+1.x and 3.0+2.0 shader combinations too
2008-04-07 11:49:20 +02:00
3ec49a32f2
d3d9: Read from a slightly different spot in a test.
2008-04-07 11:48:46 +02:00
b094c95d85
d3d9: Remove some test trace lines.
2008-04-07 11:48:33 +02:00
de5a8db58b
d3d9: Improve an ok line.
2008-04-03 12:01:39 +02:00
b4f52938e4
Spelling fixes.
2008-04-02 23:18:21 +02:00
685d14818c
d3d9: Make the shading test even more precision tolerant.
...
This is needed on both windows and wine, seems to be a hardware thing.
2008-04-02 20:28:47 +02:00
20a2afe6aa
d3d9: Add a test for temporary results.
2008-04-02 20:06:50 +02:00
ab86a4e963
wined3d: Catch D3DUSAGE_AUTOGENMIPMAP and D3DUSAGE_RENDERTARGET combos.
2008-04-01 20:31:59 +02:00
4f5cb4fb76
d3d9: Add some visual tests for Direct3DDevice9_StretchRect().
2008-04-01 13:41:49 +02:00
9625d76b5e
d3d9: Add a visual test for the DP2ADD shader instruction.
2008-04-01 13:41:45 +02:00
76014205de
wined3d: Fix a postpixelshader_blending bug.
2008-03-31 12:14:33 +02:00
61f7dc006a
d3d9: Test post pixelshader blending.
2008-03-28 15:49:52 +01:00
82d25338aa
d3d9: Add a basic test for multiple rendertargets.
2008-03-27 11:18:59 +01:00
4777f26d61
Spelling fixes.
2008-03-27 11:10:07 +01:00
7bb58929b7
d3d9/tests: Test that GOURAUD and PHONG modes are equal.
2008-03-27 11:09:11 +01:00
5c93f092c8
d3d9: Make the visual test pass on GeForce Go 7600.
2008-03-26 11:18:33 +01:00
aaced1f5ba
d3d9: Add a projected texture test.
2008-03-25 19:39:34 +01:00
740b56e87b
d3d9: Allow for driver precision differences in some visual tests.
2008-03-22 13:10:12 +01:00
1a933d6b44
d3d9: Add visual test for D3DRS_SHADEMODE states.
2008-03-22 13:10:08 +01:00
27364e8454
d3d9: Add a visual test case for the D3DSAMP_SRGBTEXTURE sampler state.
2008-03-20 11:43:12 +01:00
8ffad44397
d3d9/tests: Fix compilation on systems that don't support nameless unions.
2008-03-10 19:12:54 +01:00
d0b8fd0a13
d3d9: Fix a few ok() comparisons to not always return true in the visual tests.
2008-03-07 11:13:47 +01:00
5f14acc0bf
d3d9: Add a point size test.
2008-03-07 11:01:07 +01:00
8a3ef776f4
wined3d: gl_FragCoord isn't exact.
2008-03-07 11:01:07 +01:00
c00c89eb51
d3d9: Add a test for the vPos register.
...
The vPos register returns the unnormalised position of the current
fragment to the pixel shader. This test tests for its functionality in
both onscreen and offscreen render targets. On- and offscreen targets
are used because of the inverted y coordinate.
2008-03-07 11:01:07 +01:00
73ba896e8e
d3d9: Add a test testing two texbem instructions in one shader.
2008-03-05 19:22:53 +01:00
d7098076e1
d3d9: Implement Direct3DCreate9Ex + tests.
2008-03-03 20:50:29 +01:00
aedb8873c0
d3d9: Fix an ok line.
2008-03-03 11:58:19 +01:00
438fa44d43
d3d9: Add tests showing the D3DUSAGE_RENDERTARGET and D3DUSAGE_AUTOGENMIPMAP are mutually exlusive.
2008-02-29 10:54:53 +01:00
4ac1b94d96
wined3d: Improved CheckDeviceType.
2008-02-27 14:52:25 +01:00
3ed963dc5d
d3d9: Fix a ok line.
2008-02-25 12:37:04 +01:00
708e217506
d3d9: ATI drivers do not handle D3DUSAGE_QUERY_LEGACYBUMPMAP properly.
2008-02-25 12:36:30 +01:00
9850e4b529
d3d9: Add a non-shader reversed fog test.
2008-02-25 12:36:22 +01:00
22f379ccac
d3d9: Some Windows drivers do not like reversed fog.
...
The fog test used start = 0.9 and end = 0.1. This is perfectly ok in
theory, but some Windows drivers do not like it. Since this test tests
how fog is calculated with Vertex and Pixel shaders put the fog range in
the normal order(start = 0.1, end = 0.9), this fixes a number of fog
failures on ATI cards on Windows.
2008-02-19 14:48:34 +01:00
1a906aace7
d3d9: Do not expect consistent failures.
...
The drivers for some radeon cards are not prepared for the things the
fixed function type test does and fail inconsistently. Thus skip the
color comparison if one of the two draws fails, and do not write an
extra complaint about the inconsistency
2008-02-19 14:48:20 +01:00
27dd3428d7
d3d9: Add a stencil+culling test.
2008-02-19 14:48:08 +01:00
c2d97b2329
wined3d: Add a test for fixed function bump mapping.
2008-02-18 13:34:25 +01:00
4c8e218252
Assorted spelling fixes.
2008-02-15 12:09:24 +01:00
9950cb4bf5
wined3d: Switch the visual tests back to D3DDEVTYPE_HAL.
2008-02-15 09:40:12 +01:00
81de2fa10a
wined3d: Add a test for the vFace register.
2008-02-15 09:40:03 +01:00
ef4119029b
d3d9: Implement the device lost state.
2008-02-07 13:05:37 +01:00
b322f81b75
wined3d: Reject unsupported modes when restting the device.
2008-02-07 13:05:27 +01:00
b6ecb56eef
d3d9/tests: Accept the rejection of non-standard fixed function attrib types.
2008-02-05 12:41:48 +01:00
334e1600f2
d3d9/tests: Be less picky about X8L8V8U8 precision.
2008-02-05 12:41:39 +01:00
8daf8322ba
wined3d: Some texture transform test corrections.
2008-02-05 12:38:06 +01:00
635daf8b5d
wined3d: D3DTADDRESS_WRAP on conditional NP2 textures clamps to edge.
...
Conditional NP2 textures in Direct3D do not support D3DTADDRESS_WRAP
texture addressing. Similarly, GL_ARB_texture_rectangle textures do
not support GL_REPEAT. This patch adds a test which shows that _WRAP
clamps to the texture edge and implements that.
2008-02-05 12:37:49 +01:00
451d85a71c
d3d9: Some drivers allow non-aligned offsets.
2008-02-05 12:33:28 +01:00
450c634ab8
d3d9: Fix the type of three loop variables.
2008-02-04 11:11:20 +01:00
fc49a338bb
d3d9/tests: Initialize decl.
2008-01-23 12:05:32 +01:00
e11d7c63b4
d3d9 tests: Correct some typos.
2008-01-22 11:27:24 +01:00
44b52b128c
Assorted spelling fixes.
2008-01-16 12:48:29 +01:00
80953d62d7
d3d9: Reject IDirect3DDevice9::Reset with active default pool resources.
2008-01-14 15:12:00 +01:00
a77249ad14
d3d9/tests: Simplify condition in test_vshader_input().
2008-01-02 12:38:42 +01:00
2cb3f982a6
d3d9: Initialize pVertexBuffer.
2007-12-27 14:13:30 +01:00
998426e6b7
d3d9: Accept D3DDEVTYPE_REF devices in the vertex declaration test.
2007-12-21 13:44:21 +01:00
47a8a6bd44
wined3d: Implement and test vertex fixups with per stream offsets.
2007-12-21 13:44:16 +01:00
e27cf6ff8f
wined3d: Per stream offsets are 4 byte aligned.
2007-12-21 13:44:06 +01:00
8ff517eeb9
wined3d: Vertex attributes are 4 byte aligned.
2007-12-21 13:43:58 +01:00
a0b81e1c44
wined3d: Add a test for POSITIONT vertex fixups.
2007-12-21 13:15:09 +01:00
d8209b5de3
d3d9: Extend the fixed function vertex declaration test.
2007-12-21 13:14:49 +01:00
aaa0e5090a
wined3d: Add a test for 16 bit floats.
2007-12-20 12:10:16 +01:00
2d90449500
wined3d: Fixed function vertex attribute types are flexible.
2007-12-20 10:49:42 +01:00
f1c27eda6b
Assorted spelling fixes.
2007-12-18 14:16:11 +01:00
df77e28854
wined3d: Implement D3DFMT_G16R16.
2007-12-18 14:08:18 +01:00
de6aba11d4
d3d9: Add a note about a breakage in the refrast.
2007-12-18 13:22:27 +01:00
110facbac4
d3d9: Remove a few more incorrect D3DLOCK_DISCARDs.
2007-12-18 13:22:22 +01:00
b9860a9b27
d3d9: Do not set NONE MIN and MAG filters.
2007-12-17 12:30:04 +01:00
297b88486e
d3d9: D3DLOCK_DISCARD is not valid on managed textures.
2007-12-17 12:26:09 +01:00
b060390c73
d3d9: The refrast returns an error on invalid clear rects.
2007-12-17 12:26:01 +01:00
5f7a9cea38
Assorted spelling fixes.
2007-12-07 16:35:32 +01:00
f700cbe74b
wined3d: Add an alpha blending test.
2007-12-07 16:26:43 +01:00
0e8c13e630
wined3d: A more detailed occlusion query test and fixes.
2007-12-07 15:27:11 +01:00
f4d2449244
wined3d: Test clearing regarding the scissor test and colorwrite.
2007-12-06 13:48:43 +01:00
fce9abd095
wined3d: Clear honors the viewport.
2007-12-04 14:32:39 +01:00
3cc253c557
wined3d: Enabling too many lights is silently ignored.
...
I'm resending this patch because my reply to Henri's concern came too late.
Henri noted that I am enabling lights that do not exist. Existing tests show
that if no light is assigned to the index, LightEnable creates a light with a
set of default parameters, so the tests should be fine.
From 9ee4c61805b50886f79e87d744b52f27b7b00b4e Mon Sep 17 00:00:00 2001
From: Stefan Doesinger <stefan@codeweavers.com>
Date: Thu, 29 Nov 2007 13:22:47 +0100
Subject: [PATCH] WineD3D: Enabling too many lights is silently ignored
This patch adds tests for all d3d versions that show that Windows
pretends that enabling more lights than supported succeeds. D3D_OK is
returned, and the light is reported as enabled.
What is not tested in this patch is the rendering output of this
situation, thus the FIXME is still written.
2007-12-03 13:44:07 +01:00
0f39b29da0
wined3d: mov to a0.x does a floor(), not a round to nearest.
2007-12-03 13:43:54 +01:00
ba5c61ba20
wined3d: Fog is applied after sRGB correction.
2007-11-27 16:01:44 +01:00
02ebbd1fe5
d3d9: Add a test for vertex shader input matching.
2007-11-27 15:18:28 +01:00
5c3151b2e7
d3d9/tests: Fix the trailing '\n' in an ok() call.
2007-11-16 12:38:27 +01:00
e4f8a2da2b
wined3d: Depth stencil fixes.
2007-11-14 11:58:36 +01:00
9f41a359b6
wined3d: Partially revert "Get rid of the conditionals in shader_glsl".
2007-11-14 11:58:22 +01:00
b4b10d8bc2
d3d9: Release the window after the visual tests.
2007-11-09 12:35:50 +01:00
4454096db2
d3d9: Be less picky in the D3DFMT_X8L8V8U8 test.
2007-11-08 12:18:30 +01:00
6209b36acb
wined3d: mov to a0.x does a floor(), not a round to nearest.
2007-11-08 11:55:31 +01:00
d2c166bbb7
d3d9: Initialize the test rectangle correctly.
2007-11-07 12:08:40 +01:00
c390b6044b
d3d9: Remove some spamy debug traces from the texture test.
2007-11-07 12:08:15 +01:00
50ea23efd6
d3d9: Use HAL devices in some tests.
2007-11-07 12:07:52 +01:00
5fc54b5459
wined3d: Shader Model 3.0 varying tests.
2007-11-07 12:03:04 +01:00
393ed4a070
wined3d: Make resource memory 32 byte aligned.
2007-11-01 12:38:45 +01:00
a022e8b402
wined3d: Add a (nested) loop test.
2007-10-12 10:59:17 +02:00
3587dfeb48
d3d9: Release pixel shaders after we're done with them in the visual test.
2007-10-10 19:52:22 +02:00
6486a79e2a
d3d9: Do not rely on CreateVertexShader failing.
2007-10-09 12:31:36 +02:00
700c602eb6
d3d9: Fix crash in texture test.
2007-10-09 12:31:32 +02:00
4a31362d53
d3d9/tests: Initialize shader to NULL.
2007-10-03 11:01:47 +02:00
4c51db896c
wined3d: Check for NULL streams.
2007-09-20 15:53:19 +02:00
02114ac21f
wined3d: The D3DSI_COISSUE flag changes the behavior of cnd.
2007-09-14 14:26:26 +02:00
c96236ff3e
wined3d: Add a test for the cnd instruction.
2007-09-14 14:26:08 +02:00
7248354067
wined3d: Pixel shader 1.x constants are clamped to [-1;1].
2007-09-14 14:23:45 +02:00
160258b637
wined3d: Implement mipmap auto generation.
2007-09-12 12:10:25 +02:00
41bf0d48db
wined3d: Fix WINED3DFMT_X8L8V8U8.
2007-09-11 13:48:15 +02:00
422e0ba1cc
wined3d: texkill ignores the .w only in ps 1.x.
2007-09-11 13:48:08 +02:00
a006d2568b
wined3d: Fix texdepth instruction.
2007-09-11 13:47:49 +02:00
a305af0b30
wined3d: Fix the z range.
2007-09-11 13:47:36 +02:00
d0fdb1ea7e
d3d: Add a test for double surface locking.
2007-08-31 11:25:19 +02:00
a116879507
Assorted spelling fixes.
2007-08-30 12:26:02 +02:00
4438b3be06
wined3d: Texture transform flag test.
2007-08-27 18:28:48 +02:00
04d62b4322
wined3d: fogstart == fogend means full fog.
2007-08-21 13:35:48 +02:00
86b991c851
wined3d: Fix R32F and R16F.
2007-08-17 12:01:38 +02:00
b220905e4a
wined3d: Reinstall the internal reference on the current index buffer.
2007-08-16 11:35:06 +02:00
427723cc37
d3d9: Comment out some failing fog tests.
2007-08-08 15:32:03 +02:00
2d15cb8b27
d3d: Add a simple offscreen rendering test.
2007-08-08 15:32:00 +02:00
dbd338c018
wined3d: WINED3DSAMP_MAXMIPLEVEL fixes.
2007-08-06 12:05:39 +02:00
9ae22f1a76
d3d9: Do not fail if d3d9 is not available.
2007-08-06 12:05:31 +02:00
39a8f1b2a4
d3d9: Deal with failing surface creation.
2007-08-06 12:04:58 +02:00
17fe57e212
d3d9: Check for volume texture support before running the volume test.
2007-08-03 12:30:58 +02:00
878414a5e6
d3d9: Some visual test fixes.
2007-08-03 12:30:52 +02:00
7a8455094d
d3d9/tests: Fix use of uninitialized variable (Coverity).
2007-06-25 13:23:50 +02:00
078523f73e
wined3d: Present does not clear the depth stencil.
2007-06-14 13:22:19 +02:00
b70bc73196
d3d9: Add a test for calling DrawIndexedPrimitive with a NULL index buffer.
2007-06-04 12:58:42 +02:00
e6196f9ae7
d3d9/tests: Fix the Visual C++ double to float conversion warnings.
2007-05-29 13:47:01 +02:00
fd578240f2
d3d9/tests: Fix compilation with Visual C++ 2005.
2007-05-29 12:12:41 +02:00
d476313a49
direct3d: Remove misspelled constants as they exist correctly spelled.
2007-05-21 12:38:54 +02:00
5c48d504fc
wined3d: Destroying the resource frees all private data.
...
Also adds a test for ddraw, d3d8 and d3d9.
2007-05-11 11:22:31 +02:00
1316b7f5e4
d3d9: Add a surface locking test.
2007-05-04 13:41:21 +02:00
71fe9ae7da
d3d9/tests: Fix fog with shader test.
2007-04-23 13:55:35 +02:00
725f4bfeb1
d3d9: Rework the converted vertex declaration management.
...
Instead of creating a converted declaration each time SetFVF is
called, exactly one declaration is created for each FVF(on demand) and
stored for the lifetime of the device. This avoids memory leaks and
makes keeping track of converted declarations easier. Wether a
declaration is converted from a fvf or not is now a static information
inside the declaration. Those declarations are not destroyed in
VertexDeclaration::Release, they stay for the lifetime of the
device. This keeps us free from tracking the declaration through
stateblocks
2007-04-18 13:13:45 +02:00
c4eb3de8d8
d3d9: Fix vertex decl test.
2007-04-13 11:42:24 +02:00
9580c3f6c3
d3d9: Add a test for the converted vertex decl.
2007-04-12 16:11:39 +02:00
46c5d223ff
d3d9: Break out two helper functions from test_fvf_to_decl.
2007-04-12 16:11:13 +02:00
35dd693a79
d3d9: Fix the circular converted vertex declaration reference.
2007-04-12 16:10:48 +02:00
c343fb1cdf
wined3d: Make CreateCubeTexture fail when not supported.
2007-04-02 11:28:39 +02:00
76d153144f
d3d9/tests: Add test for texbem instruction.
2007-03-23 12:32:30 +01:00
7b31e91291
d3d9/tests: Add test for fog with shader.
2007-03-23 12:31:13 +01:00
01199072dd
wined3d: Fix LockRect memory location calculation for WINED3DFMT_DXT*.
2007-03-17 19:51:34 +01:00
1283772a26
d3d9: Add a LockRect test for DXTn formats.
2007-03-16 11:20:42 +01:00
fd64779fd0
d3d9: Add a cubemap wrapping test.
2007-03-16 11:20:08 +01:00
4b445124d3
direct3d: Add a fixed function fog test.
2007-03-15 12:14:50 +01:00
e343652338
d3d9/tests: Use skip as is done in the other tests.
2007-03-12 12:21:53 +01:00
26344ce3f2
d3d9: Add a mova rounding test.
2007-03-07 11:52:26 +01:00
5f159452ef
wined3d: Clearing the depth stencil buffer fails without a depth stencil buffer.
2007-03-06 22:17:03 +01:00
4a16dbbbec
wined3d: More Clear fixes.
...
Test for rectangles with x1 > x2 and y1 > y2. Empire earth passes such rects.
2007-02-27 12:20:10 +01:00
dd93e1fb7f
wined3d: Floating point shader constant setting.
...
Test how shader constant limits are checked in Set*ShaderConstantF.
The vertex shader tests it based on the limits reported in the caps,
the pixel shader tests tries to find the limit manually because I
could not find a cap structure member specifying the pixel shader
constant limit.
Set*ShaderConstantF returns an error as soon as start or start + count
are bigger than the constant limit.
2007-02-21 10:56:01 +01:00
93494f2ac0
Assorted spelling fixes.
2007-02-20 20:30:51 +01:00
bbcf98209c
wined3d: No bounds checking is done on sampler / texture numbers.
2007-02-20 11:49:42 +01:00
266fc77cf0
wined3d: Lighting is calculated for vertices without normals.
...
It still should be disabled for transformed vertices though.
2007-02-19 13:08:35 +01:00
03bc11fa16
wined3d: Visual test framework.
2007-02-19 13:07:44 +01:00
a41ba013b6
wined3d: Test for supported queries.
...
This test contains a table of known supported / known unsupported
queries. In theory wined3d could just report the queries it supports,
but applications may make some assumtions regarding supported /
unsupported queries and stop working if they find a query unexpectadely
supported or unsupported. It also tests what is supposed to happen if an
unsupported query is created.
2007-02-15 17:21:58 +01:00
Rico Schüller