Commit Graph

33 Commits

Author SHA1 Message Date
Gregory Hainaut c5d35d434c glsl: reduce geometry shader complexity
output 1 strip of 2 triangles instead of 2 strips of 1 triangle.

Potentially it would reduce the geometry shader overhead. And it
might avoid a middle line in sprite in some AMD GPU/driver/OS
bad combination
2017-07-12 22:04:03 +02:00
Gregory Hainaut 38010432ee gsdx ogl: handle float integral coordinate for depth sampling
Fix depth effect regression on Nemo (issue #1979)
2017-07-12 22:03:47 +02:00
Gregory Hainaut 61a7c747e1 gsdx-ogl: alternate implementation of half pixel offset
The previous implementation of HPO adds an offset on vertex position. It
doesn't always work beside it moves the rendering window.

The new implementation will add a texture offset so that instead to sample
the middle of the GS texel, we will sample the middle of the real texture texel.

It must be manually enabled with
* UserHacks_HalfPixelOffset_New = 1 (keep a small offset as intended by GS effect)
* UserHacks_HalfPixelOffset_New = 2 (no offset)

v2: always apply a 0.5 offset in case of float coordinates (Tales of Abyss)
Might break other games but few of them uses float coordinates to read
back the target
2016-11-29 17:22:02 +01:00
Gregory Hainaut c2229e3c0b gsdx-ogl: add a texture offset uniform parameter to vertex shader
It would be used for a new implementation of the half pixel offset hack

Hopefully it doesn't badly impact the perf on low end iGPU
2016-11-29 17:22:02 +01:00
Gregory Hainaut e80ca0fa92 gsdx ogl: of course some driver don't work with GL_ARB_enhanced_layouts
Windows and (AMD or Intel)
2016-10-23 12:48:23 +02:00
Gregory Hainaut 8fdf973e66 gsdx glsl: add basic support of GL_ARB_enhanced_layouts (GL4.4)
Hardcode location of interface to the location 0. If I understand the
spec correctly (unlikely), variable in interface will get successive
location.

Goal is to reduce driver work. Instead to compute some location based on
name matching approach (and silly validation), the driver can now use
static allocation.

Tests on future Mesa 13 are welcome
2016-10-19 23:03:39 +02:00
Gregory Hainaut 1fb2c66a21 gsdx ogl: Unscale line
Line thickness will be increased to N pixels (N is the upscaling factor).

Code will also be enabled by UserHacks_unscale_point_line = 1
2016-08-22 18:18:12 +02:00
Gregory Hainaut 018895067b gsdx ogl: restore code to unscale point
enabled by UserHacks_unscale_point_line = 1

Point will be transformed into a NxN square sprite. (N is the upscaling factor)
2016-08-22 18:18:12 +02:00
Gregory Hainaut 5d49a6b685 gsdx ogl: replace 4 VS shader variation by an AND mask
Perf will be roughly the same. However there is a single VS for all
the HW emulation.
2016-06-01 09:29:56 +02:00
Gregory Hainaut 959abe64f8 gsdx ogl: implement wildhack on the CPU
Speed impact is likely small and the plan is only to keep a single Vertex Shader
2016-06-01 09:29:56 +02:00
Gregory Hainaut 1558220f74 gsdx glsl: create a common header definition
* avoid duplication between all the files for UBO
* remove various interface define
2016-04-24 10:54:39 +02:00
Gregory Hainaut f751f70b1e gsdx ogl: GL_ARB_clip_control is now mandatory 2016-04-07 21:57:54 +02:00
Gregory Hainaut bc73195193 gsdx-ogl: pack more tightly the FS UBO
Merge TA vec2 + Af vec1 into a single vec4
2016-03-10 18:29:05 +01:00
Gregory Hainaut 9ba470fc63 gsdx-ogl: properly handle t_float in GS
f8c442cf76 misses the swap of t_float in GS

Fix regression in Okami

Bonus: factorize VS_WILDHACK in the VS
2016-02-21 11:48:30 +01:00
Gregory Hainaut f8c442cf76 gsdx-ogl: make VS more generic
Texture coordinate could be dummy/float/int integral/int normalized.

Old behavior:
* VS was in charge to select the texture coordinate
* int integral format wasn't supported

New behavior:
* Always compute all formats
* FS will be in charge to select the good format

Impact:
* VS will be slightly slower but it reduces shaders permutation from
   little to 0 (won't be bad for CPU)
* FS speed isn't impacted as 2 separate code paths were already required
  to support both format
* Rasterizer will be 33% slower but unlikely to be the limited factor of
  the GPU
* In future we could directly use the integral format in the FS.

V2: remove useless PSin_t
2016-02-18 19:02:05 +01:00
Gregory Hainaut bdc8038e7e glsl: set a size to gl_ClipDistance array
The purpose is to avoid issue on MS-Intel driver without
a dedicated hack in the compilation.

Code doesn't use it so I suspect others implement to discard the
statement.
2015-08-25 18:58:39 +02:00
Gregory Hainaut bfadd884c9 glsl: expand tab into space
The mix of the 2 was awful
2015-08-22 13:34:02 +02:00
Gregory Hainaut a46c11b088 gsdx-ogl: delete various leftovers 2015-08-22 13:33:32 +02:00
Gregory Hainaut 7002ff3ec3 gsdx-ogl: move texture scale from vs_cb to fs_cb
It avoid useless update of the vs_cb.
2015-08-22 13:33:04 +02:00
Gregory Hainaut ea9e608288 gsdx-ogl: fix texture palette mask 2015-07-17 21:08:49 +02:00
Gregory Hainaut b4c04ed00a gsdx-ogl: wipeout subroutine code
Code was completey bitrotten
Code was a partial test (and yet 500 lines already)
Shader is more and more complex and multithreading support greatly
reduce the cost of shader switch
2015-07-17 21:08:49 +02:00
Gregory Hainaut 91fbe6f108 gsdx-ogl: add some code to fix black netting on some renderings
Code is not yet enabled because it requires extensive test

The idea is to replace point by a 1 pixels sprite with the help of
a geometry shader. In 4x, point will be replaced by a 4x4 sprite.
2015-07-11 14:35:35 +02:00
Gregory Hainaut 5e7ce63ed1 glsl: add a geometry shader to upscale GS_POINT
The purpose is to avoid upscaling issue on POINT rendering (black netting on some FMVs)
2015-07-11 14:35:34 +02:00
Gregory Hainaut 9023f5f0db glsl: drop useless ifndef
I often mix them with the real value...
.
2015-05-08 19:28:17 +02:00
Gregory Hainaut 530e4ce776 gsdx-ogl: drop hack that rescale primitive (to avoid upscale glitch)
Rendering is bad. It renders sprites at native resolution.
2015-05-06 19:09:13 +02:00
Gregory Hainaut eaa0dae305 glsl: drop invariant qualifier
Fix issue #510 and potentially it allows the compiler to  better optimize the code
2015-04-27 18:52:55 +02:00
Gregory Hainaut 46ff4dc3d3 gsdx-ogl: hardware unit only support normalization of 4 bytes...
(At least on recent AMD GPU)
2015-04-27 18:51:59 +02:00
Gregory Hainaut 301fe869c7 glsl: improve the logz handling
This way only 0 and 1 will conflict
2015-04-26 17:51:33 +02:00
Gregory Hainaut 6a3250cad2 gsdx-ogl: fix logz management (FFXII)
The issue is that z+1 overflow if z is UINT max
log2(z) overflow if z is 0

So let's use an or 1 to avoid both overflow
2015-04-25 20:23:26 +02:00
Gregory Hainaut 36514bd95f glsl: fog is a single byte
Give a chance to the driver to optimize if possible
2015-04-24 21:37:37 +02:00
Gregory Hainaut 62489f42f1 gsdx-ogl: add an optimization note for later
Only 1 byte of fog is useful
2015-04-20 07:18:09 +02:00
Gregory Hainaut 31f8c065db gsdx-ogl: implement a new hack UserHacks_UnscaleSprite for opengl
UserHacks_UnscaleSprite = 1 will unscale flat sprites
UserHacks_UnscaleSprite = 2 will unscale all  sprites (don't work well so far)

The idea of the hack is to redo the interpolation of texture coordinate
based on the non-upscaled pixel position.

It avoids various glitches but sprites aren't upscaled anymore (so no
more anti-aliasing, potentially a coefficient can be added).
2015-04-20 07:18:08 +02:00
Gregory Hainaut 16e6d0d305 glsl: move shader into a separate directory
Only keep glsl_source.h for clarity
2015-04-19 18:49:02 +02:00