Christoph Oelckers
fda4293013
- draw the movie franes outside of the tile system to reduce the palette maintenance.
...
The movie code has always been the part causing the most issues because it alters the palettes - the rest treats them as static data.
2020-04-13 23:58:45 +02:00
Christoph Oelckers
fac004b4d2
- palette code cleanup
2020-04-13 23:58:44 +02:00
Christoph Oelckers
d46ea7481f
- handle RRRA E1L2's fog in the shader instead of hacking the shade tables.
2020-04-12 08:30:43 +02:00
Christoph Oelckers
5584c726ba
- file system update and adjustment.
2020-04-12 08:30:40 +02:00
Christoph Oelckers
66ecb05eb8
- use Printf as the only command to print to the console.
...
All uses of OSD_Printf, initprintf etc. have been renamed.
2020-04-12 08:30:38 +02:00
NY00123
3610609659
Add a workaround for possible rounding errors in calculations of
...
sintable and radarang. sintable[512] was different in a 32-bit build.
From-SVN: r8796
2020-04-08 00:38:48 +02:00
Christoph Oelckers
5af0217db7
- got rid of the software rendering blend tables.
2020-03-29 17:23:57 +02:00
Christoph Oelckers
41545b644a
- cleaned out most of the software renderer's static global variables.
2020-03-29 16:59:49 +02:00
Christoph Oelckers
7392b699e4
- cleaning out a few more parts of the software renderer.
2020-03-29 15:54:02 +02:00
Christoph Oelckers
3f69044770
- removed all references to sdlayer.cpp as all that was left in there was only needed for software rendering.
...
- removed now unused a-c.cpp.
2020-03-29 15:22:07 +02:00
Christoph Oelckers
0de4b134da
- stripped out most of the software renderer.
...
Now that the camera textures can be handled through the hardware renderer this isn't needed anymore.
2020-03-29 14:55:09 +02:00
Christoph Oelckers
621a9338e5
- render camera textures with the hardware renderer.
...
This removes the last active use case for the software renderer and allows rendering the camera views at a higher resolution.
For Shadow Warrior this necessitated a split of JS_DrawMirrors, because it processed cameras and mirrors in the same loop which cannot be done with the hardware renderer.
2020-03-29 14:01:46 +02:00
Christoph Oelckers
73a4e0d1fe
- made Blood's FullMap flag global and removed the hacks to show the full automap.
...
Also refactor show2dsector into a bit array to get rid of the bit shifting code at play here.
2020-03-07 19:30:18 +01:00
nukeykt
2bf65c9108
Exhumed: Use old ksqrt implementation
2020-03-04 17:57:15 +01:00
Christoph Oelckers
348325c879
- removed some editor-only code from engine.cpp
2020-03-01 19:12:18 +01:00
Christoph Oelckers
39c96d58c6
- fixed bug with wallsprite calculation trashing angles of voxel sprites.
...
Fix was missed from upstream due to a merge conflict.
2020-02-22 16:56:20 +01:00
Christoph Oelckers
a5fc374b59
- always recalculate the projection matrix.
...
Trying to cache this is not useful, it barely saves time and was causing display errors.
# Conflicts:
# source/build/src/polymost.cpp
2020-02-09 15:53:07 +01:00
Christoph Oelckers
215cb14401
- use a non-0 depth bias for sprites.
...
Hopefully this helps fixing the z-fighting issue with wall sprites in some Blood maps.
2020-02-06 18:44:06 +01:00
nukeykt
03b1f65a52
Polymost: fix sprite rendering issues
...
# Conflicts:
# source/build/src/engine.cpp
# source/build/src/polymost.cpp
2020-02-03 19:08:51 +01:00
Magnus Norddahl
305e6fa617
Move a few more vars to engine.cpp
2020-01-28 22:54:57 +01:00
Magnus Norddahl
113c0d8b34
Move engine variables not managed by polymost.cpp to engine.cpp
2020-01-28 22:41:07 +01:00
Christoph Oelckers
809b687969
- removed the Build color matching code
...
For the handful of lookups it got used for the needed data is too large and the lookup not precise enough. ZDoom's BestColor yields better results, does not need any tables and for the small amount of lookups being performed is more than adequate.
2020-01-28 10:31:59 +01:00
Christoph Oelckers
4c6265e246
- fixed movie palette in Blood.
2020-01-26 12:10:23 +01:00
Christoph Oelckers
e8231f904f
- I have no idea why a fov of exactly 90° does not work.
...
Changing this by one in either direction makes the problem go away.
2020-01-25 11:56:13 +01:00
Christoph Oelckers
f8a807e3bb
- cleaned out a tiny bit of the garbage that has accumulated in compat.h.
...
This file really needs to go away, we do not need this kind of "compatibility".
2020-01-25 10:56:30 +01:00
Christoph Oelckers
dc5b8d27f8
- infrastructure for savegame pics.
...
Savepic generation implemented for Duke 3D, but results in a black image.
2020-01-12 23:16:21 +01:00
CommonLoon102
bc4a4bea23
Add possibility to move voxels/models via maphacks
...
# Conflicts:
# source/build/src/engine.cpp
# source/build/src/polymer.cpp
# source/build/src/polymost.cpp
# source/duke3d/src/lunatic/con_lang.lua
# source/rr/src/lunatic/con_lang.lua
2020-01-11 22:23:53 +01:00
Christoph Oelckers
df7c5a7067
- use proper math instead of crash-prone libdivide here.
2020-01-08 00:06:24 +01:00
Christoph Oelckers
fb985d2503
- more macOS compilation fixes
...
Still not complete…
2020-01-07 01:11:19 +01:00
Christoph Oelckers
3380420de9
- converted the clear screen commands.
2020-01-02 23:56:35 +01:00
Christoph Oelckers
e01c161258
- removed some dead variables.
2020-01-01 13:11:44 +01:00
Christoph Oelckers
18099e9179
- removed the pure software render surface and deleted a few unused variables.
...
With all the 2D refactorings thhe softsurface won't work anymore.
This also revealed a bug with the fullscreen variable, a few places were still using the old one from the SDL backend.
2020-01-01 12:36:48 +01:00
Christoph Oelckers
d09b83d4a5
- moved the last remaining function out of baselayer.cpp and removed that file.
...
g_logFlushWindow was deleted entirely because with the current console this is not needed anymore.
2020-01-01 12:01:26 +01:00
Christoph Oelckers
05e381ff6d
- fixed screen clearing for the automap.
...
This should only affect the active window, not the entire screen.
2019-12-31 19:50:27 +01:00
Christoph Oelckers
ef5292b4ae
- don't use Build's clipping code to clip automap parts to the screen.
...
Not really surprisingly this is not compatible with a real triangulator. Fortunately it's also not really needed.
2019-12-31 19:38:50 +01:00
Christoph Oelckers
1890df98f9
- process lines through the 2D drawer.
2019-12-31 19:02:55 +01:00
Christoph Oelckers
2f8d472d7d
- cleaned up the map drawer by using a real triangulator.
...
Immediate benefit: almost 200 lines of bona-fide Kencode go to the dumpster.
2019-12-31 17:23:29 +01:00
Christoph Oelckers
d825282726
- removed a few software rendering parts from the automap drawer.
...
These won't be needed anymore.
2019-12-31 15:41:12 +01:00
Christoph Oelckers
964e303dd7
- draw the textures on the automap with the 2D drawer.
2019-12-31 15:05:08 +01:00
Christoph Oelckers
ad24a1ce31
- place weapon sprites in a separate render list.
...
They need to be drawn in a different pass than the 2D overlay HUD so the backend must have them separately.
2019-12-30 20:16:51 +01:00
Christoph Oelckers
7ea053bd90
- reconnected the HUD model drawer.
...
This will have to be done differently later when the postprocessor is fully working, but for now it's sufficient to just render the model right away.
2019-12-30 20:16:51 +01:00
Christoph Oelckers
6459f4e532
- refactored rotatesprite to really use the 2D drawer.
...
Mostly working, except clipping and weapon sprites.
2019-12-30 19:37:23 +01:00
hendricks266
6465feb438
Refactor tsprite creation into renderAddTSpriteFromSprite
...
git-svn-id: https://svn.eduke32.com/eduke32@8521 1a8010ca-5511-0410-912e-c29ae57300e0
# Conflicts:
# source/build/src/polymer.cpp
# source/duke3d/src/m32exec.cpp
2019-12-29 22:22:19 +01:00
Christoph Oelckers
cc332486b4
- cleaned out the 2D drawing code.
...
With the new backend there will always be just one page, never more, so the RS_PERMS case will never be entered.
In addition, since the software renderer has already been nonfunctional due to lacking support from the 2D drawer, its 2D components have also been removed. Its main remaining purpose, drawing camera textures, remains unaffected by this.
2019-12-29 15:46:48 +01:00
Christoph Oelckers
737bf15ad8
- added GZDoom's postprocessing/presentation code.
...
Compiles but only draws a black screen. Something must be missing but no idea yet what that might be.
2019-12-28 22:36:47 +01:00
Christoph Oelckers
f6dee38d28
- route all 2D drawing through the 2D drawer unconditionally.
...
This is needed so that the postprocessor receives a clean 3D view to process without messing up the 2D parts.
2019-12-28 19:10:23 +01:00
Christoph Oelckers
6373b75d22
Merge branch 'master' into powerslave
...
# Conflicts:
# source/common/gamecontrol.cpp
# source/common/version.h
# wadsrc/static/engine/grpinfo.txt
2019-12-26 14:46:14 +01:00
Christoph Oelckers
62e9112133
- renamed the internal resource directory to "engine" and routed most literal mentions of the engine name through version.h
...
All this comes from a time when I didn't use version.h so it's better to do it the same way as GZDoom to allow easy renaming of the engine.
2019-12-26 14:04:53 +01:00
hendricks266
8e6a54a1e4
Mostly clean up the codebase in preparation for tspritetype != uspritetype
...
Remaining exceptions:
SW - ConnectCopySprite
CON and M32Script - pSprite/pUSprite
git-svn-id: https://svn.eduke32.com/eduke32@8519 1a8010ca-5511-0410-912e-c29ae57300e0
# Conflicts:
# source/build/include/polymer.h
# source/build/src/engine.cpp
# source/build/src/engine_priv.h
# source/build/src/polymer.cpp
# source/build/src/polymost.cpp
# source/duke3d/src/astub.cpp
# source/duke3d/src/game.h
# source/duke3d/src/m32common.cpp
# source/duke3d/src/m32exec.cpp
# source/duke3d/src/m32structures.cpp
# source/kenbuild/src/bstub.cpp
# source/kenbuild/src/game.cpp
# source/sw/src/jnstub.cpp
# source/sw/src/jsector.cpp
2019-12-26 08:58:25 +01:00
Christoph Oelckers
1a8f11e01d
Merge branch 'master' into powerslave
...
# Conflicts:
# source/build/src/sdlayer.cpp
# source/common/menu/menu.cpp
# source/sw/src/config.cpp
# source/sw/src/game.cpp
2019-12-26 00:21:04 +01:00