Age | Commit message (Collapse) | Author | |
---|---|---|---|
2014-06-17 | End current draw surface before drawing cinematic | Zack Middleton | |
Fixes drawing StretchPic *before* CIN_DrawCinematic resulting in cinematic being drawn before the StretchPic. | |||
2014-06-17 | OpenGL2: Use RGBA16F format for HDR. | SmileTheory | |
RGB16F is not 4-byte aligned and not supported by certain hardware. | |||
2014-06-17 | #6059: OpenGL2: Read depths from resolve fbo when msaa is on in RB_TestFlare(). | SmileTheory | |
2014-06-17 | OpenGL2: Match glsl data type names: matrix_t -> mat4_t, vec*i_t -> ivec*_t | SmileTheory | |
2014-06-17 | OpenGL2: Match glsl data type names: matrix_t -> mat4_t, vec*i_t -> ivec*_t | SmileTheory | |
2014-06-17 | OpenGL2: Calculate bitangent in shader, and store normal/tangent as ↵ | SmileTheory | |
normalized byte arrays. | |||
2014-06-17 | Fix comparing unsigned values to < 0 in gl2 | Zack Middleton | |
2014-06-17 | Fix flares being invisible after vid_restart | Zack Middleton | |
flareCoeff was only set if r_flareCoeff cvar was set as modified. Cvars are set as modified when created or changed. | |||
2014-06-17 | OpenGL2: Remove srfTriangle_t, and use glIndex_t instead. | SmileTheory | |
2014-06-17 | OpenGL2: Small optimizations and comments in world VBO creation and surface ↵ | SmileTheory | |
merging. | |||
2014-06-17 | OpenGL2: Prefer VBO/IBO sizes of around 4MB. | SmileTheory | |
2014-06-17 | OpenGL2: Reimplement soft overbright to avoid a framebuffer blit. | SmileTheory | |
2014-06-17 | OpenGL2: Some small shader optimizations. | SmileTheory | |
2014-06-17 | OpenGL2: Revisit fragment tangent space calculation, and remove tangent ↵ | SmileTheory | |
space lighting. | |||
2014-06-17 | OpenGL2: Reduce glsl shader count by using a uniform to disable textures. | SmileTheory | |
2014-06-17 | Remove vertexLightmap from textureBundle_t, it's never qtrue | Zack Middleton | |
2014-06-17 | Remove unused shader state code from tr_local.h | Zack Middleton | |
2014-06-17 | Merge last two commits into renderergl2 | Zack Middleton | |
2014-06-17 | IQM int/float vertex arrays were already swapped | Zack Middleton | |
2014-06-17 | Add support for IQM int blend indices and float blend weights | Zack Middleton | |
Integer blend indices are converted to bytes at load, ioq3 doesn't allow more than 128 joints. Heavily based on code by @zippers. | |||
2014-06-17 | Support IQMs with joints and no poses | Zack Middleton | |
2014-06-17 | OpenGL2: Merge bsp surface structs into a single struct, and more cleanup. | SmileTheory | |
2014-06-17 | OpenGL2: Use an idea from ETXreal and separate world VBO into separate VBOs ↵ | SmileTheory | |
by shader. | |||
2014-06-17 | OpenGL2: Clean up texmod calculations, and improve vertex animation handling. | SmileTheory | |
2014-06-17 | OpenGL2: Fix TCGEN_ENVIRONMENT stages not rendering correctly. | SmileTheory | |
2014-06-17 | Revert "5097 - Menu corrupted on start (on IRIX)" | Zack Middleton | |
Calling glClear( any bits ) in GL_SetDefaultState makes connect screen black using the proprietary NVidia driver on Windows and GNU/Linux. This reverts commit 92573270de98a20cd78842bd2043cfe07bc5b2e1. | |||
2014-06-17 | OpenGL2: default to RGBM lightmap, and add r_floatLightmap for old behaviour. | SmileTheory | |
2014-06-17 | OpenGL2: Remove AGEN_FRESNEL(superceded by cubemap patch), and some small ↵ | SmileTheory | |
fixes and optimizations. | |||
2014-06-17 | OpenGL2: Disable GL_ARB_seamless_cube_map by default, caused huge frame ↵ | SmileTheory | |
drops on some hardware | |||
2014-06-17 | OpenGL2: reduce varying usage in lightall shader. | SmileTheory | |
2014-06-17 | OpenGL2: Fix for "unsupported framebuffer format" bug | SmileTheory | |
2014-06-17 | OpenGL2: Some shader cleanup | SmileTheory | |
2014-06-17 | OpenGL2: Speed up fog shader slightly. | SmileTheory | |
2014-06-17 | OpenGL2: Fix inaccurate RGBM calculation. | SmileTheory | |
2014-06-17 | OpenGL2: Use RGBM instead of RGBE encoding for lightmaps. | SmileTheory | |
2014-06-17 | #5979: Cubemap support for opengl2. | SmileTheory | |
2014-06-17 | Fix crash caused by too many surfaces in skin | Zack Middleton | |
2014-06-17 | Fix upside-down SSAO. | SmileTheory | |
2014-06-17 | #5999: OpenGL2: Don't use lightall shader with deforms or adjustColorsForFog. | SmileTheory | |
2014-06-17 | #6004: OpenGL2: Make any CGEN_LIGHTING_DIFFUSE stages use lightall shader. | SmileTheory | |
2014-06-17 | Add assignment below declarations otherwise compiling fails with msvc | Henry Stratmann III | |
Signed-off-by: Zack Middleton <zturtleman@gmail.com> | |||
2014-06-17 | Misc fixes to IBO code to correctly use sizeof(glIndex_t) | Anthony Pesch | |
2014-06-17 | Fix r_mergeLightmaps 0 crashing OpenGL2 renderer | Zack Middleton | |
tr.fatLightmapStep was 0 and caused modulus division by 0. | |||
2014-06-17 | Fix r_nocurves in OpenGL2 renderer | Zack Middleton | |
Check r_nocurves in surface cull. While r_nocurves is enabled, don't use merged surfaces because merged curves would be renderer. (note: surface merging is enabled by r_mergeLeafSurfaces) | |||
2014-06-17 | Fix r_speeds 4 in OpenGL2 renderer | Zack Middleton | |
2014-06-17 | Fix IQM tess buffer overflow | Zack Middleton | |
Set the variables that use tess.numVertexes after calling RB_CHECKOVERFLOW() as it may set tess.numVertexes to 0! Could cause visual issues and error "RB_EndSurface() - SHADER_MAX_VERTEXES hit". | |||
2014-06-17 | Only allow model meshes to have SHADER_MAX_VERTEXES - 1 vertexes | Zack Middleton | |
The last index is used to check for buffer overflows. See RB_CheckOverflow and RB_EndSurface. | |||
2014-06-17 | Don't try to fog flares with fogNum 0 | Zack Middleton | |
fogNum 0 means no fog. Shaders don't try to apply fog color if fogNum is 0. This was done to make things more consistent and fix a crash in iortcw MP using a user made map on GNU/Linux x86_64. | |||
2014-06-17 | Improve warning for shader missing opening brace | Zack Middleton | |
2014-06-17 | Check for shaders without closing brace | Zack Middleton | |
Shaders without closing brace can eat shaders in other files. Pass depth to SkipBracedSection instead of reparsing text as it messed up parse line numbers. |