EDuke32 Changelog Unofficial archive

What changed in EDuke32, build by build

2,117 builds · 8,866 commits · 25 authors · October 2009 – August 2026

Commits per year

Showing 2016 · select it again to see every year
SVN era · r1519 – r8798 Git era · voidpoint.io, from april 2020
Filters · 3 active

Showing 19 builds · 33 commits

Export CSV
Display options
Commit details:

December 2016 1 build · 1 commit

November 2016 1 build · 1 commit

October 2016 4 builds · 9 commits

Build 5888

9 Oct 2016 · 13 commits · 2 authors SVN
Open build
  • General

    GL .blend support.

    Hendricks266 r5888 9 Oct, 00:55 Engine
    Show details

    Consult me if you are interested in this feature. TODO: Polymer.

  • General

    In 8-bit, change the precedence of .alpha and .blend.

    Hendricks266 r5887 9 Oct, 00:55 Engine
    Show details

    Previously: Non-zero alpha stacked with cstat translucency in an imitation of Polymost, potentially clobbering .blend if alphatabs are used. Now: If cstat translucency is active, .alpha is ignored.

  • Remove the Polymer aspect ratio option, as it was always broken, and I don't know what purpose it has besides configuring non-square pixel stretching, which we have a long way to go in terms of configuring anyway.

    Hendricks266 r5879 written 2 Oct 2016 FixEngine
    Show details

    Use r_pr_customaspect from the console instead.

  • General

    Some float optimizations in the engine.

    Hendricks266 r5877 written 2 Oct 2016 Engine
Open the build to see all 13 commits

September 2016 2 builds · 2 commits

June 2016 2 builds · 3 commits

May 2016 3 builds · 7 commits

Build 5716

3 May 2016 · 9 commits · 1 author SVN
Open build
  • Factor out the ETC block preparation by moving the ETC function decision from Polymost_SendTexToDriver into Polymost_PickETCFunction.

    Hendricks266 r5714 3 May, 17:25 Engine
  • Allow compressed ART tiles (r_texcompr 2) to be cached.

    Hendricks266 r5713 3 May, 17:25 Engine
  • Tweak the "don't upload textures with dimension > xdim" GL ES rule to limit to the power of two greater than xdim instead of less than it.

    Hendricks266 r5711 3 May, 17:25 Engine
  • Clean up and rearrange some texcache code. No functional changes.

    Hendricks266 r5710 3 May, 17:25 Engine
  • Call texcache_calcid once per texture instead of twice. Clean it up a little while we're at it.

    Hendricks266 r5709 3 May, 17:25 Engine
Open the build to see all 9 commits

Build 5707

2 May 2016 · 7 commits · 1 author SVN
Open build
  • Add "artquality" subtoken to textures, model skins, and skyboxes, which grants the texture HICR_ARTIMMUNITY. In effect, this implies "nodownsize" and signals that texture compression would be undesirable, but okay using r_texcompr mode 2 (aggressive).

    Hendricks266 r5705 2 May, 11:29 Engine
Open the build to see all 7 commits

April 2016 1 build · 1 commit

March 2016 3 builds · 5 commits

Build 5650

7 Mar 2016 · 6 commits · 1 author SVN
Open build
  • Add one-bit alpha detection for hightile and model skins.

    Hendricks266 r5649 7 Mar, 03:22 Engine
  • Add r_texcompr 2, which compresses ART in addition to hightile.

    Hendricks266 r5648 7 Mar, 03:21 Engine
  • Add HICR_ARTIMMUNITY and DAMETH_ARTIMMUNITY, which currently function identically to NOTEXCOMPRESS and NODOWNSIZE combined, and are only used when:

    Hendricks266 r5647 7 Mar, 03:21 Engine
    Show details

    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)

Open the build to see all 6 commits

February 2016 1 build · 2 commits

Build 5643

28 Feb 2016 · 15 commits · 2 authors SVN
Open build
  • Refactor all GL internal texture format decision making into uploadtexture().

    Hendricks266 r5642 28 Feb, 22:34 Engine
    Show details

    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.

  • Rename HICR_NOCOMPRESS to HICR_NODOWNSIZE, CACHEAD_NOCOMPRESS to CACHEAD_NODOWNSIZE, DAMETH_NOCOMPRESS to DAMETH_NODOWNSIZE, and HICR_NOSAVE to HICR_NOTEXCOMPRESS.

    Hendricks266 r5640 28 Feb, 22:34 Engine
Open the build to see all 15 commits

January 2016 1 build · 2 commits

Build 5532

7 Jan 2016 · 19 commits · 1 author SVN
Open build
  • General

    Introduce USE_GLEXT macro and use it to conditionally compile out all GL ARB function calls when compiling for GL ES platforms.

    Hendricks266 r5526 7 Jan, 17:33 Engine
    Show details

    This replaces the previous hack where we kept the dynamic pointers null and simply never tried calling any of them by a conspiracy of default runtime settings and disabled menu options. Code affected: VBO support for models, VPX GLSL decoding, texture compression (texture cache), and multitexturing (glow/detail maps). This commit also replaces EDUKE32_GLES conditionals with USE_GLEXT where appropriate. I didn't touch polymer.c because it depends too heavily on extensions for conditionally compiling them to make sense.

  • General

    Fix POLYMER=0 warnings introduced in r5496.

    Hendricks266 r5516 7 Jan, 17:32 FixEngine
Open the build to see all 19 commits