Christoph Oelckers
4a7b1aada7
Merge remote-tracking branch 'remotes/origin/modern'
...
# Conflicts:
# wadsrc/static/shaders/glsl/main.fp
2018-07-20 10:19:07 +02:00
Magnus Norddahl
8a500a25f5
- rewrite the user shader support for materials - new syntax is to create a 'Material ProcessMaterial()' function
2018-07-15 23:01:40 +02:00
Christoph Oelckers
124fe63d00
Merge branch 'master' into modern
2018-07-14 10:24:41 +02:00
alexey.lysiuk
c29651da95
- restored initial grayscale conversion in fragment shader
...
Accidental change of red color multiplier was discovered during investigation of https://forum.zdoom.org/viewtopic.php?t=61126
2018-07-02 17:31:48 +03:00
Christoph Oelckers
4937848123
- refactoring of fixed colormap stuff to have it better organized and to reduce the number of uniforms in the main shader.
...
This removes 3 uniforms, consisting of 9 floats. Those were merged into other values that never get used at the same time.
It also moves the costly setup of the fixed colormap out of the render state into the 2D processing code.
Since 3D forces use of render buffers now, it is no longer necessary to draw the entire scene with the colormap active, meaning it can be handled more efficiently.
2018-06-16 22:40:44 +02:00
Christoph Oelckers
612fb40f3a
- don't use uniforms related to fixed colormaps for other things.
...
The fixed colormap is a per-scene global setting that normally does not need to change ever during rendering of a scene so it's easily shoved aside into a static uniform buffer.
Having to change this buffer for inconsequential stuff should be avoided, especially when there's other uniforms that are just as good to hold these values.
2018-06-14 21:28:03 +02:00
Christoph Oelckers
ad7aaa8f2a
- specify fragment output locations in the shader source.
...
OpenGL has been supporting this since version 3.3 and Vulkan requires it so it's the way to go.
2018-06-12 23:52:33 +02:00
Christoph Oelckers
286846a8ec
2D must consider brightmaps.
2018-04-27 08:59:59 +02:00
Christoph Oelckers
832df6d43a
- implemented desaturation for the legacy GL renderer's 2D.
...
A lot of work that's only needed to be able to use the hardware renderer's 2D code with the software renderer.
2018-03-30 18:14:42 +02:00
Christoph Oelckers
811d96e07e
D3D 2D code refactored but not tested yet.
2018-03-29 23:21:25 +02:00
Christoph Oelckers
a40e085a46
- 2D drawer implementation complete for hardware renderer.
...
This also replaces DTA_ColormapStyle with proper implementations of its components. As implemented it was a very awkward mixture of various effects that already existed in a separate form. As a result of its implementation it required additional but completely redundant shader support which could be removed now. As a side effect of this change a new DTA_Desaturate option was added.
2018-03-29 16:21:21 +02:00
Christoph Oelckers
25999c1c6a
- looks like the 2D drawer sources did not get committed...
...
- This also adds the missing draw modes to the GLSL shader.
2018-03-28 19:07:43 +02:00
Christoph Oelckers
d586d774cf
- adjusted GL renderer to the new rules for alpha textures. This means that the translation fallback is only needed for the legacy renderer now. Modern GL can use the regular texture and will apply a grayscale mapping instead of merely using the red channel.
2018-03-22 20:42:17 +01:00
Magnus Norddahl
07fa310f23
- move material light modes to their own subshader lumps
2018-02-20 00:13:05 +01:00
Magnus Norddahl
eb39e88682
- clean up the main.fp light handling code so that a single ApplyDynLights function applies all dynamic light
2018-02-19 02:01:33 +01:00
Magnus Norddahl
3207d8aef7
- generate uniform declarations from c++
2018-02-09 23:29:31 +01:00
Magnus Norddahl
769867475c
- Replaced max(dot(a,b), 0.0) with clamp as some rounding errors caused pow to receive negative values when then value was subtracted from 1.0 (undefined glsl behavior)
...
- Fixed that surface angle attenuation was getting applied twice
2018-02-03 23:56:55 +01:00
Magnus Norddahl
7b9a334f77
- Change PBR materials to use quadratic light falloff
2018-02-03 13:17:09 +01:00
Magnus Norddahl
747ef8324e
- Improve sector light slightly for PBR
2018-01-29 19:09:24 +01:00
Magnus Norddahl
6deb51f23c
- disable the tonemapping to let it bloom instead
2018-01-27 00:42:17 +01:00
Magnus Norddahl
a6da1d356a
- add PBR materials to main.fp
2018-01-27 00:22:15 +01:00
Magnus Norddahl
7a59bcde4c
- move glossiness and specular level to GLDEFS
2018-01-25 19:53:55 +01:00
Magnus Norddahl
1875902414
- Do not flip the normal based on face direction
2018-01-23 23:59:58 +01:00
Magnus Norddahl
81c6808d2a
- Add specular and normal map handling to main.fp
2018-01-23 23:10:28 +01:00
Magnus Norddahl
e045fb57c9
- Declare new textures for specular and pbr modes
2018-01-23 20:51:48 +01:00
Magnus Norddahl
5f36b86013
- Add dynamic spot lights
2018-01-04 17:58:11 +01:00
Magnus Norddahl
a6d4bfc748
- Added linear and nearest shadowmap filters (enabled with defines)
2017-08-13 13:49:02 +02:00
Magnus Norddahl
5c7b1ee8e1
- Significantly improve shadowmap light performance for faces with a large number of lights but where only a few applies to each individual fragment
2017-08-08 23:28:42 +02:00
Rachael Alexanderson
8d3d271484
- tweaked the banded software lightmode shader a bit more, looks a little closer (but still not quite...) to real ZDoom in paletted mode
2017-07-21 18:45:08 -04:00
Magnus Norddahl
ddd1b629c3
- Calculate globvis/r_visibility in r_utility and use it in GL and poly renderers
2017-07-09 19:01:34 +02:00
Rachael Alexanderson
1562d64580
- fixed: in the availability of gl_fogmode == 0, default to having square fog instead of radial fog in gl_lightmode == 8
2017-06-25 09:37:14 -04:00
Magnus Norddahl
b6e035e796
- Implemented radial fog support in software light mode
2017-06-25 15:02:34 +02:00
Rachael Alexanderson
4229389680
- fixed: banded sw light was broken in a recent commit. Added back the 32 light levels and also made it prefer darker shades.
2017-06-21 19:10:39 -04:00
Magnus Norddahl
28acf2ad06
- Fix OpenGL software diminishing light to be identical to truecolor swrenderer and softpoly
...
All there is left is to make globVis an uniform and move the LightVisibility::SetVisibility calculations out of the software renderer.
That will make it 100% correct for all r_visiblity values and aspect ratios.
2017-06-22 00:01:57 +02:00
Magnus Norddahl
85a9984807
- Add OpenGL ES 3 support to GL renderer
2017-04-06 23:34:42 -04:00
Magnus Norddahl
59ec97d2d5
Fix shadow map acne and the attenuate flag
2017-03-10 22:08:55 +01:00
Magnus Norddahl
b660493051
Add menu option for disabling shadow maps and detecting if storage buffers are available or not
2017-03-10 19:10:40 +01:00
Magnus Norddahl
cb40c369cd
Added PCF shadows
2017-03-08 01:35:07 +01:00
Magnus Norddahl
818b72fbf5
Reduce margin a little
2017-03-08 00:33:42 +01:00
Magnus Norddahl
6df3b3fbca
Changed the light collision structure uploaded to the GPU to be a binary tree using AABBs for the nodes instead of a BSP plane
2017-03-07 15:58:22 +01:00
Magnus Norddahl
8515f9720a
1D shadow maps are now working
2017-03-04 09:14:01 +01:00
Magnus Norddahl
0d1deddae5
Bind shadow map texture for main.fp and sample from the shadowmap texture
2017-03-02 19:10:57 +01:00
Magnus Norddahl
538d516c9a
Upload shadow map index for each light to main.fp
...
Move storage buffer binding location
2017-03-02 18:07:47 +01:00
Christoph Oelckers
12d073a27d
- objectcolor stuff.
...
# Conflicts:
# src/gl/renderer/gl_renderstate.cpp
2017-01-28 19:06:21 +01:00
Rachael Alexanderson
3154652885
- fixed possible AMD compilation error.
2017-01-24 22:24:53 +01:00
Rachael Alexanderson
9a777f719b
- Added "gl_bandedswlight" to reduce the software light emulation gradient to 32 levels.
2017-01-24 22:24:53 +01:00
Magnus Norddahl
057060022a
Merge remote-tracking branch 'gzdoom/master' into ssao
2016-11-27 09:59:57 +01:00
Christoph Oelckers
bea625a42c
- added an ATTENUATE flag to dynamic lights, this is set by default for attached lights. For placed lights this is off, because it'd interfere with many existing maps that depend on unattenuated lights.
2016-10-23 20:42:48 +02:00
Magnus Norddahl
14c1a77f8a
Fix AmbientOcclusionColor bug where pixelpos was assumed to be in eye space coordinates
2016-10-13 18:08:04 +02:00
Magnus Norddahl
00e72028ef
Add another gbuffer with normal data and make ssao pass use it
2016-10-05 07:57:27 +02:00