Commit graph

85 commits

Author SHA1 Message Date
Christoph Oelckers
d2fa4d0ff9 - most stat stuff done. 2018-03-12 19:33:48 +01:00
Rachael Alexanderson
6aa724a9b4 - add 'gl_riskymodernpath' for computers that worked before the GL >= 3.3 modern path enforcement, since some did, but they seemed to be the exception rather than the rule. 2018-01-03 21:46:04 -05:00
Rachael Alexanderson
8454e09694 - fixed: I was using the wrong check to determine which renderer was in use. Ooops. It now checks for the actual (active) renderer rather than the state of vid_glswfb 2017-09-12 14:32:29 -04:00
Rachael Alexanderson
79fc219afd - allow OpenGL 3.0 MESA drivers to use vid_glswfb 2017-09-12 08:18:22 -04:00
Christoph Oelckers
c74356a1c0 - made OpenGL 3.3 a requirement for running the modern render path. Also force '-glversion 3' to use the modern render path, even though actual OpenGL 3.0 won't support it.
This will exclude Intel HD3000 and many obsolete drivers for NVidia and ATI which should have been updated long ago.
2017-06-28 17:57:04 +02:00
svdijk
c698f10257 Fix setting of the gl_legacy_mode cvar. 2017-06-05 14:45:24 +02:00
Christoph Oelckers
a4231ffd00 - more handling of some very broken old ATI drivers. 2017-05-19 21:33:48 +02:00
Christoph Oelckers
762ba13cd9 - store the last found OpenGL version in the INI so that modern GL related options can be removed when running old hardware with software rendering. 2017-05-08 19:30:51 +02:00
Magnus Norddahl
85a9984807 - Add OpenGL ES 3 support to GL renderer 2017-04-06 23:34:42 -04:00
Christoph Oelckers
9e70771da3 - Added a check to allow shader storage buffers on GL 4.3 Intel drivers. It doesn't work if GLSL version is set to 4.0 and the feature activated via extension. 2017-03-11 20:14:18 +01:00
Magnus Norddahl
9c4b11b671 Add OpenGL ES support to Linux target and enable it for ARM devices 2017-01-25 07:18:26 +01:00
Christoph Oelckers
fcb1c1edc4 - only print GL extensions to the log, but not to the console.
The list has become so long by now that it's more of a distraction than help. Besides, searching on-screen for specific extensions is futile anyway.
2017-01-05 12:01:00 +01:00
Christoph Oelckers
c87836c3f3 - fixed check for Mesa driver. 2017-01-05 11:51:15 +01:00
Christoph Oelckers
96623b3052 - disabled OpenGL 3.0 on the Open Source Mesa driver for Linux because it appears to be broken. 2017-01-03 12:01:03 +01:00
Christoph Oelckers
8b6e09ca09 - changed the license of the OpenGL renderer to LGPL v3.
This was done to clean up the license and to ensure that any commercial fork of the engine has to obey the far stricter requirements concerning source distribution. The old license was compatible with GPLv2 whereas combining GPLv2 and LGPLv3 force a license upgrade to GPLv3. The license of code that originates from ZDoomGL has not been changed.
2016-09-14 20:01:13 +02:00
Christoph Oelckers
3389a5a74e - removed most of the specific options for legacy hardware and consolidated them in one variable (does not work yet.) 2016-09-01 11:52:52 +02:00
Christoph Oelckers
8f535997f8 - enable core profile by default on GL 3.x.
After doing some profiling it was very obvious that this has better performance than client arrays. Persistent buffers are still better, though, especially for handling dynamic lights.
2016-08-29 11:33:20 +02:00
Christoph Oelckers
0f0dc2c852 - implemented buffers for GL 3.x. These only get mapped during the data collection pass so the order of some things is different here. 2016-08-29 10:43:03 +02:00
Christoph Oelckers
597e10116f - added missing framebuffer function to fudging block. 2016-08-26 08:30:47 +02:00
Magnus Norddahl
1594cc9570 Only use KHR_debug if available 2016-08-22 19:25:13 +02:00
Christoph Oelckers
250be72939 - added handling for drawing with uniform vertices. It draws something but in the wrong place. Right now I have no idea what's happening... 2016-08-22 15:31:23 +02:00
Magnus Norddahl
5eeac830eb Clear and InvalidateFramebuffer optimization 2016-08-17 17:37:13 +02:00
Christoph Oelckers
9c9edbffdb - on older ATI drivers where the clip planes are broken, only disable the clip planes themselves, but not OpenGL 3.x so that dynamic lights still work as intended.
- only disable clip planes on Windows, but not on Linux or macOS.
- If a driver reports full OpenGL 4.5 support, assume that all features are working properly.
2016-08-08 12:55:09 +02:00
Christoph Oelckers
6a66d0255d - use a dedicated vertex buffer for rendering the wipes.
- fixed: The postprocessing shaders clobbered the render state's vertex buffer info by bypassing and not notifying it of the change.
2016-08-06 14:12:40 +02:00
Christoph Oelckers
7576e85cef - force use of shaders on all hardware, unless -noshader is specified.
Right it's only for easy testing, so only the code that does the feature checking has been disabled.
2016-08-06 13:06:55 +02:00
Christoph Oelckers
2e555e6dab - use client arrays on compatibility profiles instead of calling glBegin/glEnd.
This eliminates most behavioral differences for FFlatVertexBuffer between both operating modes, now the only difference is where the buffer is located.
2016-08-06 12:03:16 +02:00
Magnus Norddahl
5849c83028 Added bloom and tonemap to menus
Added gl_renderbuffers CVAR that disables render buffers
Added patch shader support to FShaderProgram
Added OpenGL 2 fallback support to render buffers
2016-07-29 21:31:20 +02:00
Christoph Oelckers
0e14f00b51 - fixed caching of texture sampler state for old hardware.
- fixed handling of CLAMP_XY_NOMIP sampler mode. This cannot be lumped together with CLAMP_XY because it has different mipmap requirements.
2016-05-03 13:39:41 +02:00
Christoph Oelckers
50ba1ecde8 - add menu patching for GL 2.x mode.
- use stdint types in model code, because we have to start somewhere with the transition.
2016-05-01 13:09:13 +02:00
Christoph Oelckers
c29e96d369 - print buffer info only if the buffer type is used.
- removed some GL specs that are not informative in GZDoom's context.
2016-04-30 16:31:09 +02:00
Christoph Oelckers
50ab301bd8 - fixed: If we want to support pre-GL3 hardware, we may not require the presence of glGetStringi. 2016-04-30 16:23:32 +02:00
Christoph Oelckers
913e3df7e3 - fixed display of alpha textures without shaders.
As it turned out, the translation's alpha channel was clobbered by the whole setup.
2016-04-26 19:11:32 +02:00
Christoph Oelckers
7e6e2bc0a8 - stated adding GL 2.0 compatibility handling 2016-04-26 18:24:02 +02:00
Christoph Oelckers
eb27057853 - enable shader-less rendering. Still very incomplete, but at least the basics are working. 2016-04-26 16:44:03 +02:00
Christoph Oelckers
3b1500438d - got rid of VAO's. they have no impact on performance and are a general nuisance if some render paths cannot use them.
- cleaned up gl_interface and removed gl.compatibility.
2016-04-26 16:26:34 +02:00
Christoph Oelckers
f066457a48 - add shader patching to allow running the engine with GLSL 1.2.
- made some initial preparations for the shader-less fallback path.
2016-04-26 15:26:52 +02:00
Christoph Oelckers
090d13b915 - updated gl_interface to deal with older GL versions.
- added the framework for allowing multipass textured dynamic lights again.
2016-04-26 15:20:26 +02:00
Christoph Oelckers
0d17459c36 - do not disable sampler objects with -gl3. 2016-04-23 15:47:51 +02:00
Christoph Oelckers
c786b65727 - allow the engine to run without sampler objects. This will create some overhead in the texture code, but that's still better than having to error out on those few systems where it may be an issue. 2016-04-22 19:54:51 +02:00
Christoph Oelckers
fc57180d7e - added necessary state to implement 3D light splitting using hardware clipping planes. 2016-01-30 23:01:11 +01:00
Christoph Oelckers
abf11f1a12 - print an error message if GL_ARB_sampler_objects cannot be found.
GZDoom requires this extension and all supported hardware has drivers implementing it but there still seem to be people out there who stuck to some older, obsolete drivers that don't.
2015-07-13 09:49:02 +02:00
Christoph Oelckers
f273b6c46d - fix uninitialized variable. 2015-04-27 21:25:16 +02:00
Christoph Oelckers
717c82ee4a - On Intel, we have to disable shader storage buffers because their complier contains a bug disallowing unsized arrays. 2014-12-01 09:58:23 +01:00
Christoph Oelckers
c773a500f2 - only activate shader storage buffers on modern drivers also supporting GL_ARB_buffer storage. Early adopters of this extension (i.e. older AMD drivers mainly) tend to implement this badly. 2014-09-17 09:01:16 +02:00
Christoph Oelckers
32f08adaf3 - moved some code to better places.
- allow GL version 3.0 in Windows, too.
2014-09-14 23:01:57 +02:00
Ralgor
5cc43137a1 Only require OpenGL 3.0 compatibility profile. 2014-09-14 14:43:42 -05:00
Ralgor
cfc8f3dbbf Global GL render context shouldn't be initialized inside of gl_PrintStartupLog, since it's not compiled in non MSC builds. 2014-09-14 14:42:14 -05:00
Christoph Oelckers
bf03d02228 - print OpenGL profile type in startup log. 2014-08-24 01:09:44 +02:00
Christoph Oelckers
86d37e06f9 - lowered requirements of GL 2.x to OpenGL 3.3.
There was one issue preventing the previous 2.0 betas from running under GL 3.x: The lack of persistently mapped buffers.
For the dynamic light buffer today's changes take care of that problem.
For the vertex buffer there is no good workaround but we can use immediate mode render calls instead which have been reinstated.

To handle the current setup, the engine first tries to get a core profile context and checks for presence of GL 4.4 or the GL_ARB_buffer_storage extension.
If this fails the context is deleted again and a compatibility context retrieved which is then used for 'old style' rendering which does work on older GL versions.

This new version does not support GL 3.2 or lower, meaning that Intel GMA 3000 or lower is not supported. The reason for this is that the engine uses a few GL 3.3 features which are not present in the latest Intel driver.
In general the Intel GMA 3000 is far too weak, though, to run the demanding shader of GZDoom 2.x, so this is no real loss. Performance would be far from satisfying.

A command line option '-gl3' exists to force the fallback render path. On my Geforce 550Ti there's approx. 10% performance loss on this path.
2014-08-19 15:56:33 +02:00
Christoph Oelckers
b2860a1d63 - it looks like glProgramUniform is not working correctly with Intel drivers, so better forget about it for setting the view and projection matrices. Even on NVidia the time difference can only be measured in microseconds per frame so it's not a big loss. 2014-08-02 11:57:42 +02:00