SVN handles and Git names are merged into one person.
Remove GitLab CI/CD configuration
Synthesis builds are now kicked off with a post-commit hook.
remove LIBXMP_CORE_DISABLE_IT preprocessor definition
This seems to fix memory corruption issues with .xm playback... ???
engine, audiolib: clean up most SDL_MAJOR_VERSION checks so they're relative to version 2
comment the reason SetWindowGrab exists
Fix trains not moving spawn points with higher indexes than player count.
use timerGetPerformanceCounter() instead of timerGetHiTicks() for CON performance profiling
move gameHandleEvents() (and therefore system event pumping) back to directly in the main loop
I'm hoping this is OK now that the issues with SDL and GetClipCursor() have been worked around.
bypass SDL_SetWindowGrab on Windows
make CON_GETFLORZOFSLOPE and CON_GETCEILZOFSLOPE TROR aware
add yax_getflorzofslope() and yax_getceilzofslope()
Guard new !(o & RS_TOPLEFT) behavior behind TEXT_VARHEIGHT
Use MF_Minifont struct for shadowminitext and creditsminitext
revert Vorbis handling to standard memory allocation instead of cache1d
clean up classic mode sloped sprite drawing a little
flip CON_USE_COMPUTED_GOTO dispatch macro logic around so the branch is taken less often
change random_angle hack in CON_MOVE to match CON_AI
zero player return_to_center when setting horizRecenter to false
interpolate player view angle when dead
ALSA code cleanup/standardization
Merge branch 'AudioLib_ALSAMIDI' into 'master'
audiolib: Import ALSA MIDI support from JFAudioLib and patch through the menu system as appropriate.
source/audiolib: Initial commit of ALSA MIDI code from jfaudiolib, with EDuke32 audiolib-specific changes as required.
fix OOB memory access in CON_FOR
set .horizRecenter when handling SK_CENTER_VIEW
This seems so obvious that I almost think I intentionally left it out before, but I can't remember why I would have.
zero .horizAngleAdjust and .horizSkew after checking playerquitflag instead of before
I don't think this actually makes a difference, but it's more consistent.
allow SE_14_SUBWAY_CAR to move STAT_PLAYER sprites
Fixes #18.
hoist checks for ST_2_UNDERWATER outside of loops in SE_14_SUBWAY_CAR processing
remove interpolation fixups from subway
These aren't needed anymore after Striker's patch to improve overall interpolation.
fix APLAYER radius damage distance calculation regression
Fixes #20.
slightly clean up fix16 usage for volume
disable error message printed when m32help.hlp is missing
Clarify clipupdatesector() log message a bit
Derp...
Probably actually fixes #4. Regardless, I want to note that this is really sub-optimal.
Properly display non power of 2 floor aligned sprites in 2d mode
Fix G_PlaySoundWhileNoInput()
Fixes #2.
Fix crash with CONs that only define one skill
Fixes #3.
Fix really stupid error with CON_NULLOP branch detection
Fixes #4.
Fix a crash with tabledivide64()
Fixes #1.
fix stupid input scaling bug
slash resets sprite slope
Use pragma pack in sw/src/network.h instead of the locally defined
PACKED macro. This is more consistent with the current Duke3D codebase, and further fixes build with older versions of MinGW GCC, in which attribute packed is broken without specifying -mno-ms-bitfields.
Remove unused wfe_Clock variable
sw/src/game.cpp: Add missing wfe_Esc callback from
SW 1.2 after minor adjustments. Originally recreated for sw-src-and-build-addendum-20190129.7z.
Fix usage of possibly wrong address in sw/src/player.cpp:DoPlayerHorizon
mmulti.cpp: Change PAKRATE to 250 (4ms) as previously done in SVN r365,
and also as currently in use in EDuke32-OldMP.
Merge branch 'Engine_StutterFixes' into 'master'
build/src/sdlayer.cpp: Add new CVARs r_glfinish & r_finishbeforeswap.
Update source/build/src/sdlayer.cpp, source/build/src/polymost.cpp, source/build/src/baselayer.cpp, source/build/include/baselayer.h files
build/src/sdlayer.cpp: Add new CVAR r_glfinish to make 'glFinish()' optional.
- Issues requiring glFinish() aren't really apparent on Linux, therefore not needed. - Running glFinish() can have an detrimental impact of Polymer users depending on whether they use the HRP or not.
build/src/sdlayer.cpp: Add new CVAR r_finishbeforeswap.
- glFinish() after swapping stabilises stuttering issues and has minimum performance impact. - glFinish() before swapping stabilises should be better but sometimes isn't, and is needed for KMT not to tear. Offer as an option.
