Commit Graph

295 Commits

Author SHA1 Message Date
Rémi Verschelde
c1f687c681
Merge pull request #21387 from hpvb/fix-15324
Fall back to GLES2 if GLES3 is not working
2018-08-27 16:54:22 +02:00
Juan Linietsky
bca706b361
Revert "Fix some 3D texture issues" 2018-08-27 10:12:07 -03:00
Rémi Verschelde
3dc63a710f
Merge pull request #21469 from akien-mga/glsl-clang-format
Style: Enable clang-format on GLSL shaders
2018-08-27 10:12:24 +02:00
Rémi Verschelde
4226d56ca9 Style: Enable clang-format on GLSL shaders
As of clang-format 6.0.1, putting the `/* clang-format off */` hint
around our "invalid" `[vertex]` and `[shader]` statements isn't enough
to prevent a bogus indent of the next comments and first valid statement,
so we need to enclose that first valid statement in the unformatted chunk.
2018-08-27 07:34:14 +02:00
elasota
30e20b6278 Partial fixes for 3D texture issues 2018-08-26 20:22:56 -04:00
Hein-Pieter van Braam
08f452d1a9 Fall back to GLES2 if GLES3 is not working
This adds a static is_viable() method to all rasterizers which has to be
called before initializing the rasterizer. This allows us to check what
rasterizer to use in OS::initialize together with the GL context
initialization.

This commit also adds a new project setting
"rendering/quality/driver/driver_fallback" which allows the creator of a
project to specify whether or not fallback to GLES2 is allowed. This
setting is ignored for the editor so the editor will always open even if
the project itself cannot run. This will hopefully reduce confusion for
users downloading projects from the internet.

We also no longer crash when GLES3 is not functioning on a platform.

This fixes #15324
2018-08-26 16:40:46 +02:00
Juan Linietsky
62233423c7 Fix generation of env map, closes #18880 2018-08-24 13:31:31 -03:00
Rémi Verschelde
1ae7a78d8b
Merge pull request #21364 from akien-mga/shaders-style
Style: Fix code formatting in GLSL shaders
2018-08-24 15:17:38 +02:00
Rémi Verschelde
52466d57e9 Make some debug prints verbose-only, remove others 2018-08-24 14:59:01 +02:00
Rémi Verschelde
d442f3d0aa
Merge pull request #21351 from akien-mga/print_verbose
Add print_verbose to print to stdout only in verbose mode
2018-08-24 14:58:36 +02:00
Rémi Verschelde
e68b96928b Style: Fix code formatting in GLES2 shaders 2018-08-24 14:50:59 +02:00
Thomas Herzog
f123694b4e
Merge pull request #21357 from karroffel/gles2-shader-language-f-to-pay-respects
[GLES2] fix wrong shader compiler output
2018-08-24 12:14:40 +02:00
Thomas Herzog
56bab6509c
Merge pull request #21318 from karroffel/gles2-shader-fixes
[GLES2] attempt to fix some android problems
2018-08-24 12:14:20 +02:00
Thomas Herzog
2fa4b50027 [GLES2] fix wrong shader compiler output
The `f` postfix was working fine on dekstop GL but not on
some mobile drivers.
2018-08-24 12:10:54 +02:00
Thomas Herzog
211e0fa4aa [GLES2] fix multimesh byte colors
When using float colors, everything works fine, but when using
the byte format, the colors were read incorrectly.
2018-08-24 11:23:42 +02:00
Rémi Verschelde
de59fe04e7 Add print_verbose to print to stdout only in verbose mode
Equivalent of the cumbersome:
if (OS::get_singleton()->is_stdout_verbose())
	print_line(msg);
2018-08-24 09:23:20 +02:00
Rémi Verschelde
b2494d3898
Merge pull request #21325 from karroffel/gles2-texture-flags
[GLES2] fix texture flags not having an effect
2018-08-23 22:22:33 +02:00
Thomas Herzog
8884b3daad [GLES2] fix texture flags not having an effect 2018-08-23 18:25:55 +02:00
Thomas Herzog
0edb50f629
Merge pull request #21321 from karroffel/gles2-canvas-modulate
[GLES2] fix CanvasModulate not working
2018-08-23 18:15:42 +02:00
Thomas Herzog
0f59cc179c [GLES2] fix CanvasModulate not working
That's what happens when you forget about your debug
shortcuts.

Fixes #21025
2018-08-23 17:38:20 +02:00
Thomas Herzog
5ebc70f288 [GLES2] fix depth for alpha pass 2018-08-23 17:25:00 +02:00
Thomas Herzog
d61fc6a97b
Merge pull request #21317 from karroffel/gles2-canvas-texture-sampler-long-branch-name
[GLES2] fix canvas_item sampler allocation
2018-08-23 17:11:00 +02:00
Thomas Herzog
a7bedaf935 [GLES2] attempt to fix some android problems 2018-08-23 17:08:16 +02:00
Thomas Herzog
da876907fd [GLES2] fix canvas_item sampler allocation 2018-08-23 16:46:43 +02:00
Thomas Herzog
8e7454e5d7 enable hardware skeletons 2018-08-23 16:13:45 +02:00
elasota
35f6ba5c5d BPTC support 2018-08-21 22:56:04 -04:00
Marcin Zawiejski
bd5ff205b2 Revert "Batch GLES2 draw calls"
This reverts commit f55039b194.

The GLES2 batching seems to require more testing and tweaking in order
to actually make the performance better on Android devices. It's been
proved with #21184 that the current implementation has it's drawbacks
therefore I suggest reverting the commit for now.
2018-08-20 02:10:14 +02:00
Thomas Herzog
b4ce1e3005
Merge pull request #20965 from dragmz/gles2_batching
Batch GLES2 draw calls
2018-08-18 18:37:52 +02:00
dumitru.stama
5a74b15b3d Removed the lines forcing black screen on android GLES2 2018-08-17 22:26:30 -07:00
Thomas Herzog
0ef85b55e0 [gles2] fix black bar artifacts 2018-08-15 17:12:21 +02:00
Rémi Verschelde
4c6cf97257
Merge pull request #20941 from codingHahn/fix-normals-on-shader
Fix handling of normals that approach 1
2018-08-15 08:40:16 +02:00
Marcin Zawiejski
ab1e460f53
Reduce verbosity for unsupported GLES2 bg modes 2018-08-14 15:11:19 +02:00
Marcin Zawiejski
f55039b194 Batch GLES2 draw calls
Adds GLES2 draw calls batching for the same render list item that uses
multiple rasterizer commands (e.g. Label node; a node with multiple
GDScript draw_* calls).
2018-08-14 10:47:16 +02:00
Thomas Herzog
3d2984c962 [gles2] added ImmediateGeometry rendering 2018-08-13 16:16:23 +02:00
Nick Hahn
66eba18969 Fix handling of normals that approach 1 2018-08-12 12:45:43 +02:00
Marcin Zawiejski
a0afaaeeb9 Fix mat4 uniform writes 2018-08-08 21:39:43 +02:00
Thomas Herzog
1800664e80
Merge pull request #20805 from karroffel/gles2-misc-fixes
unhacking some GLES2 code
2018-08-08 17:28:35 +02:00
Thomas Herzog
73fe08be2e GLES2 refactors
This commit unhacks some parts of the 3D rendering.

Most notably:

 - possibility to use negative texture units
   (no longer weird manual index allocation for user samplers)

 - refactoring of light code, now sorts in a different way,
   should yield better performance

 - fixes a crash while saving (because of "Illegal instruction" execution)
   when using a decent compiler (clang, it's clang. Thanks GCC for not telling me about UB).
2018-08-08 17:09:44 +02:00
Juan Linietsky
b67f982723
Merge pull request #20681 from DavidSichma/master
Keeping track of discard
2018-08-08 09:54:59 -03:00
Hugo Locurcio
31c12f05c4
Tweak the default SpatialMaterial properties
Roughness is now set to 1 by default and albedo is now white,
even on meshes that do not have any materials defined.

This means there is no longer a visual difference between a
mesh with no materials defined and a mesh with a default
SpatialMaterial defined.
2018-08-07 17:04:30 +02:00
Juan Linietsky
edde52c8de Added proper import support for 3D and Array textures 2018-08-06 14:56:06 -03:00
Juan Linietsky
07a982e614
Merge pull request #20597 from karroffel/tex3d
add 3D textures
2018-08-06 07:20:00 -03:00
Marcin Zawiejski
477f496162
Fix uniform set for mat3 and mat4 2018-08-04 07:30:25 +02:00
David Sichma
c101dd5fa6 Keeping track of discard
Shader compilation now keeps track of the discard key word.

Previously only variables were monitored. But discard, which needs
special treatment in some cases, went unnoticed by the compiler as
discard is not a variable but a flow control.

This commit adds monitoring for discard.
2018-08-02 21:55:02 +02:00
Thomas Herzog
b4d3f541e7 add 3D textures 2018-07-30 12:20:27 +02:00
karroffel
b64171e79c add initial GLES2 3D renderer 2018-07-27 14:15:46 +02:00
Hein-Pieter van Braam
0e29f7974b Reduce unnecessary COW on Vector by make writing explicit
This commit makes operator[] on Vector const and adds a write proxy to it.  From
now on writes to Vectors need to happen through the .write proxy. So for
instance:

Vector<int> vec;
vec.push_back(10);
std::cout << vec[0] << std::endl;
vec.write[0] = 20;

Failing to use the .write proxy will cause a compilation error.

In addition COWable datatypes can now embed a CowData pointer to their data.
This means that String, CharString, and VMap no longer use or derive from
Vector.

_ALWAYS_INLINE_ and _FORCE_INLINE_ are now equivalent for debug and non-debug
builds. This is a lot faster for Vector in the editor and while running tests.
The reason why this difference used to exist is because force-inlined methods
used to give a bad debugging experience. After extensive testing with modern
compilers this is no longer the case.
2018-07-26 00:54:16 +02:00
Rémi Verschelde
e38051970c
Merge pull request #20448 from volzhs/duplicated-code
Remove duplicated codes
2018-07-25 22:20:47 +02:00
Rémi Verschelde
237148e80f
Merge pull request #18368 from Gamblify/RasterizerEngineSync
sync rasterizers with engine
2018-07-25 20:33:37 +02:00
volzhs
946be8be50 Remove duplicated codes 2018-07-25 23:34:25 +09:00