SVN handles and Git names are merged into one person.
Add r_texcompr 2, which compresses ART in addition to hightile.
Add HICR_ARTIMMUNITY and DAMETH_ARTIMMUNITY, which currently function identically to NOTEXCOMPRESS and NODOWNSIZE combined, and are only used when:
1. Using an ART tile 2. Using a hightile defined through the tilefromtexture "texture" subtoken 3. Using a hightile that, if downsized, will be smaller than the ART tile for that tilenum (tenuous)
Refactor all GL internal texture format decision making into uploadtexture().
This directly achieves the following: * Eliminated codedup. * Recently introduced GL ES performance downsampling now applies to model skins. * Replaces the alpha/art format distinction with {RGB/no alpha, RGBA/one-bit alpha (ART transparency), RGBA/full alpha}. * Cleanly allows multiple fallback formats for each type. * Ready for further commits.
Make voxels take on the current basepalette, like they should.
Fix voxels so palette colors are treated as 8-bit, not 6-bit. Overlooked in r5349.
Enumify the lower bits of the method/dameth argument to various functions.
Bookend the recursive call to polymost_drawpoly as part of the fullbright pass with bglDepthFunc(GL_EQUAL) and bglDepthFunc(GL_LEQUAL) to ensure that the fullbright overlay is always visible and never fights with the sprite itself.
Remove "indrawroomsandmasks" variable and its only use in polymost_drawpoly as a condition for the fullbright pass.
Connect spriteext[].x/ypanning to model UVs (range 0-255).
Clean up formatting of parallax sky code and comments, in part so that the floor and ceiling sections are line-for-line consistent. DONT_BUILD.
don't redundantly adjust globalcursectnum in polymost_drawrooms().
See r5024 and the added comment.
fix wall and floor sprites clipping into or z-fighting with surfaces (replaces glDepthRange() hacks for sprites), add r_nofog cvar for debugging, reduce the total number of OpenGL calls by only calling glFogi() to set the fog mode when initializing the renderer or changing r_usenewshading.
r_npotwallmode, emulating 8-bit for walls w/ nonpow2 height textures.
When that mode is enabled (see below for caveats), wall textures that have a non-power-of-two height (call it 'h') will be modified to look like in classic: Let 'H' be the next power of two greater than 'h'. The texture will be uploaded with height 'H', made up from 'h' hlines of the original texture, followed by 'H'-'h' first hlines of the same. No panning "corrections" will take place. The mode is disabled by default. Caveats/notes: * the mode requires that r_hightile is disabled * it is not implemented in Polymer * in the Lunatic build, it is ineffective when a VX map is loaded, as those display walls with NPOT height textures correctly
in NPOT wall ypanning "correction", behave like Polymer in one corner case.
Specifically, the handling of one-way walls ([1] in Mapster32) was slightly different.
make r_usetileshades 2 the default.
This mode does not take a sector's visibility values into account for uniform darkening of a texture. It is more faithful to classic with respect to fog/visibility *distances*, but may be less so as far as the "tint" of the texture due to visibility attenuation is concerned.
introduce r_usenewshading 3 and make it the default.
With this mode, the near distance of the linear fog is always 0, unlike being negative with r_usenewshading 2.
handle HUD model 'fov' DEF token.
This "simply" requires an appropriate factor in the projection matrix. Also, get rid of a special-case factor for >=1.6 aspects (making HUD models wider then, I think).
make spriteext[].{x,y,z}off independent of xdimen/viewingrange/aspect.
The offsets (that will translate the axes for .pitch/.roll rotation) were calculated improperly with screen sizes other than 1280 x 1024 and were dependent on other related variables. The fix proceeded empirically, i.e. mostly trial, error, and some educated guesswork. (It's telling that the magic constant 1280 is needed.) A test for Lunatic is added to test.lua that will rotate SEENINES. Bug pointed out by Mblackwell.
fix my r_usetileshades==2 sub-method.
Don't touch TerminX's r_usetileshades==1 or it being the default.
fix 'zadd' DEF token handling. (Polymer has it right.)
Fix ART offset and face sprite flipping interaction.
factor out y panning coefficient calculation code.
don't attempt to draw sprite with void tile, preventing div-by-0s later
fix mirror-related oob access analogous to r1872's fix for classic.
When drawing from inside the mirror, there may be no 'bunches' (BUILD lingo, not TROR ones) collected, but following code used the first one unconditionally.
