terminx
c1aff49e4c
Improve usage of initprintf mutex
...
git-svn-id: https://svn.eduke32.com/eduke32@7908 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:46:35 +02:00
terminx
064516c050
Re-implement our mutex_lock() family of functions using SDL's atomic spinlocks
...
git-svn-id: https://svn.eduke32.com/eduke32@7907 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:46:32 +02:00
terminx
9021b885d6
Eliminate pos_as_vec2 from anything that had it
...
git-svn-id: https://svn.eduke32.com/eduke32@7905 1a8010ca-5511-0410-912e-c29ae57300e0
# Conflicts:
# source/build/src/build.cpp
# source/build/src/polymost.cpp
2019-09-20 15:46:28 +02:00
terminx
81882aa1a3
Fix issue clipping against sloped ceilings
...
git-svn-id: https://svn.eduke32.com/eduke32@7903 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:44:21 +02:00
pogokeen
4d22a44a5d
tiles.cpp: backup & restore picsiz, walock, & waloff alongside other per tile arrays in artClearMapArt() and artSetupMapArt()
...
This fixes an issue with viewscreens as tiles that are setup game-side after a map is loaded could have their sizes reset to 0 while not being properly cleared from walock/waloff
git-svn-id: https://svn.eduke32.com/eduke32@7892 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:43:58 +02:00
pogokeen
5fad944ed1
engine.cpp: make renderSetTarget() more robust so that passing an xsiz or ysiz <= 0 is not accepted and can't cause a segmentation fault
...
git-svn-id: https://svn.eduke32.com/eduke32@7890 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:43:55 +02:00
pogokeen
a844fd4b13
polymost.cpp: Fix issue where tiles whose dimensions had grown would not have their sizes updated in GL and could fail to update their texture data
...
git-svn-id: https://svn.eduke32.com/eduke32@7888 1a8010ca-5511-0410-912e-c29ae57300e0
# Conflicts:
# source/build/src/polymost.cpp
2019-09-20 15:42:29 +02:00
terminx
6931d448ff
Oops
...
git-svn-id: https://svn.eduke32.com/eduke32@7884 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:34:30 +02:00
terminx
7553bb043f
Skip drawing invalid polygons in polymost_drawpoly()
...
git-svn-id: https://svn.eduke32.com/eduke32@7880 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:34:18 +02:00
terminx
95598e08f9
getwalldist() tweak
...
git-svn-id: https://svn.eduke32.com/eduke32@7878 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:34:14 +02:00
terminx
9f9a158946
Replace runtime pow2 calculations with pow2char LUT usage
...
I don't know that this is any faster, but there's something to be said for standardization and consistency. I will be making most of this stuff use bitmap_set/test/clear() soon.
git-svn-id: https://svn.eduke32.com/eduke32@7876 1a8010ca-5511-0410-912e-c29ae57300e0
# Conflicts:
# source/build/src/build.cpp
# source/build/src/polymost.cpp
2019-09-20 15:34:08 +02:00
terminx
2a3c80f95d
Fix issue with sounds being expired from cache
...
git-svn-id: https://svn.eduke32.com/eduke32@7874 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:33:05 +02:00
terminx
1a0c56f0f1
This adds back in the engine's old automapping feature
...
This was removed a long time ago because Duke3D and Shadow Warrior don't use it.
git-svn-id: https://svn.eduke32.com/eduke32@7873 1a8010ca-5511-0410-912e-c29ae57300e0
# Conflicts:
# source/build/src/polymer.cpp
# source/build/src/polymost.cpp
2019-09-20 15:32:13 +02:00
terminx
1a83ec15e4
Fix silly branching logic in engineLoadBoard()
...
git-svn-id: https://svn.eduke32.com/eduke32@7871 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:29:38 +02:00
terminx
c7e403619b
Don't blow away wall cstat bit 14 for use as a temp flag in the editor
...
Patch from Nuke.YKT.
git-svn-id: https://svn.eduke32.com/eduke32@7870 1a8010ca-5511-0410-912e-c29ae57300e0
# Conflicts:
# source/build/src/build.cpp
2019-09-20 15:29:35 +02:00
terminx
a949665f26
Work around an issue with far plane clipping with voxels without mipmaps in software mode
...
Patch from Nuke.YKT.
git-svn-id: https://svn.eduke32.com/eduke32@7869 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:29:20 +02:00
terminx
9335b339b0
Emulate slope overflow in Polymost in order to match the rest of the engine's slope algorithms
...
Patch from Nuke.YKT.
git-svn-id: https://svn.eduke32.com/eduke32@7868 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:29:19 +02:00
terminx
0f4ccaf970
Hopefully work around more Polymost polygon leakages
...
Patch from Nuke.YKT.
git-svn-id: https://svn.eduke32.com/eduke32@7867 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:29:17 +02:00
terminx
acc0b527a9
Cache the distrecip[] table
...
Patch from Nuke.YKT.
git-svn-id: https://svn.eduke32.com/eduke32@7866 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:29:15 +02:00
terminx
0091acc858
"Another dumb polymost optimization. Probably worth few fps in heavy scenes"
...
Patch from Nuke.YKT
git-svn-id: https://svn.eduke32.com/eduke32@7865 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:29:12 +02:00
terminx
0652b5b6d6
Make ksgn() return int instead of int32_t
...
git-svn-id: https://svn.eduke32.com/eduke32@7864 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:29:09 +02:00
terminx
26272f6755
if 0 out our klabs() function and just define it as abs()
...
The compiler generates much better code for this now.
git-svn-id: https://svn.eduke32.com/eduke32@7863 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:29:06 +02:00
terminx
752cf77887
Untabify a couple of headers
...
git-svn-id: https://svn.eduke32.com/eduke32@7862 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:29:05 +02:00
terminx
f84e161e52
Wahhh, I'm MSVC and I cry about applying unary negation to a boolean
...
git-svn-id: https://svn.eduke32.com/eduke32@7859 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:28:59 +02:00
terminx
751b734795
Fix issue rendering TROR when the camera's z coordinate is right on the portal
...
git-svn-id: https://svn.eduke32.com/eduke32@7858 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:28:57 +02:00
terminx
6e4951bdb6
Another clipping fix
...
git-svn-id: https://svn.eduke32.com/eduke32@7857 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:28:53 +02:00
pogokeen
a0231cf90c
polymost_scansector(): avoid using a culling method not designed for voxels when handling 1-sided wall aligned sprites that have voxel replacements
...
git-svn-id: https://svn.eduke32.com/eduke32@7854 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:28:45 +02:00
pogokeen
83e863422a
Fix mixing offscreenrendering (render targets) and mirrors. Don't allow recursing beyond 1 level of mirroring through multiple levels of render targets.
...
git-svn-id: https://svn.eduke32.com/eduke32@7853 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:28:42 +02:00
pogokeen
5377b128a9
Fix rendering skyboxes to a tile by altering renderSetTarget() & renderRestoreTarget() to set & reset xdim & ydim so that G_ShowView() can render while a render target is set,
...
and additionally modifying G_SetupCamTile() to call EVENT_DISPLAYROOMSCAMERATILE after calling renderSetTarget()
git-svn-id: https://svn.eduke32.com/eduke32@7852 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:28:41 +02:00
terminx
e0e517aaed
Revise r7838 again
...
git-svn-id: https://svn.eduke32.com/eduke32@7847 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:26:22 +02:00
terminx
f3e4dd4463
Patch from Nuke.YKT to work around precision-based inaccuracy problems in Polymost TROR
...
git-svn-id: https://svn.eduke32.com/eduke32@7846 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:25:31 +02:00
terminx
1d14991acc
Add MERGE_NODES macro for Polymost... this will be used in the following commit
...
git-svn-id: https://svn.eduke32.com/eduke32@7845 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:25:28 +02:00
terminx
ee92304087
Patch from Nuke.YKT to clip bunches to screen boundaries in Polymost
...
git-svn-id: https://svn.eduke32.com/eduke32@7844 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:25:24 +02:00
terminx
47fa1eccd4
Patch from Nuke.YKT to fix sloped floor and ceiling textures wiggling around in Polymost
...
git-svn-id: https://svn.eduke32.com/eduke32@7843 1a8010ca-5511-0410-912e-c29ae57300e0
# Conflicts:
# source/build/src/polymost.cpp
2019-09-20 15:24:54 +02:00
terminx
a3b6d0e70b
Revise r7838: one less dumbass typo
...
git-svn-id: https://svn.eduke32.com/eduke32@7842 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:57 +02:00
terminx
28c778c8a3
Automatic source file dependency generation for GNUmakefile
...
This automatically saves the dependencies used for a given object file during the first compilation, then uses the generated data for each subsequent compile to determine if the object needs to be rebuilt. Why? Because if I see one more warning about the C++ One Definition Rule, I'm going to fucking snap.
git-svn-id: https://svn.eduke32.com/eduke32@7839 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:50 +02:00
terminx
565383e1d2
Revise r7835: instead of not testing wall sprites, we'll just ignore the side blockers
...
git-svn-id: https://svn.eduke32.com/eduke32@7838 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:47 +02:00
terminx
4beaeb5426
Don't check clip objects for sprites in the keepaway() loop in clipmove()
...
This prevents some hilarious issues with deadly wall decorations. The "side blocker" clip lines that already exist for these sprites should already prevent the kinds of clipping issues this loop is intended to address, anyway.
git-svn-id: https://svn.eduke32.com/eduke32@7835 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:40 +02:00
terminx
dad28df269
OK, this is just a bit too restrictive
...
git-svn-id: https://svn.eduke32.com/eduke32@7834 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:38 +02:00
terminx
dc7ec3b2fe
Fix additional clipmove() issue
...
git-svn-id: https://svn.eduke32.com/eduke32@7832 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:33 +02:00
terminx
1da1331299
Clean up raytrace() and rename it to cliptrace()
...
git-svn-id: https://svn.eduke32.com/eduke32@7831 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:31 +02:00
terminx
490b7249a2
Allow reading vec3_whatever_t as vec2 directly
...
It's a union in a struct in a union in a struct. Oh god, what have I done.
git-svn-id: https://svn.eduke32.com/eduke32@7830 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:28 +02:00
terminx
52db0e37f7
Additional clipmove() reworks
...
This fixes a LOT of clipping errors but needs thorough testing.
git-svn-id: https://svn.eduke32.com/eduke32@7829 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:25 +02:00
terminx
17d3c302a3
Switch a few iterators in clipmove() from int to native_t
...
git-svn-id: https://svn.eduke32.com/eduke32@7828 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:23 +02:00
terminx
417318c271
clipmove() patch from Nuke.YKT:
...
"In some edge cases posx/y might be 'inside' of neighbor sector. I've added double check for this case"
git-svn-id: https://svn.eduke32.com/eduke32@7827 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:20 +02:00
terminx
df28860bad
Add divscale64()
...
git-svn-id: https://svn.eduke32.com/eduke32@7826 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:18 +02:00
terminx
dc0130625a
Minor pushmove() cleanup
...
git-svn-id: https://svn.eduke32.com/eduke32@7825 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:15 +02:00
terminx
ed457895f1
Replace check_floor_curb() with cliptestsector()
...
git-svn-id: https://svn.eduke32.com/eduke32@7824 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:12 +02:00
terminx
0fb77f8af3
Use corrected slope heights in getzrange() and pushmove()
...
git-svn-id: https://svn.eduke32.com/eduke32@7823 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:10 +02:00
terminx
2482cddbb5
Add getcorrectzsofslope()
...
git-svn-id: https://svn.eduke32.com/eduke32@7822 1a8010ca-5511-0410-912e-c29ae57300e0
2019-09-20 15:23:08 +02:00