dhewg
8003056c18
Attempt at fixing annoying markdown crap
2012-06-20 23:27:02 +02:00
dhewg
b9cc8548dc
Remove leftover X11 header
2012-06-15 18:25:48 +02:00
Daniel Gibson
94cd0ee559
Fix several bugs from iodoom3 bugtracker
...
rhyskidd@gmail.com found them (with PVS studio IIRC), reported them and posted
patches.
Some of the patches were incorrect so I rewrote them.
2012-06-02 10:12:22 +02:00
Daniel Gibson
1e087d9bf6
Fix idMatX::IsOrthonormal()
...
The original implementation was pretty broken (but not used anyway),
it is now fixed and improved a bit (got rid of one inner loop).
This (at least part of the problem) was detected by PVS-Studio,
see http://www.viva64.com/en/b/0120/ Fragment 3
2012-05-13 03:04:34 +02:00
Daniel Gibson
5b2c78b9ff
Fix wrong length used in memset() in idMegaTexture::GenerateMegaMipMaps()
...
it should be tileSize, not sizeof( tileSize ), as it was also used for
allocating that block of memory
2012-05-13 02:52:12 +02:00
Daniel Gibson
56873e1fec
Fix memleak in snd_cache.cpp (from Harry Jeffery)
...
This fix is stolen from https://bugzilla.icculus.org/show_bug.cgi?id=5345
2012-05-13 02:31:37 +02:00
Daniel Gibson
6a23b17509
Work around broken libs that call setlocale(LC_ALL, "");
...
Because that can fuck up parsing floats if your locale uses ',' as a radix.
LADSPA plugins that may be used by ALSA do this. A PITA to debug.
2012-05-13 00:04:17 +02:00
dhewg
45a73e2dcd
Add the build target "Profile" suitable for profiling
2012-01-19 00:53:16 +01:00
dhewg
a18650b27e
Remove initial R_SetColorMappings() call
...
This happens before the SDL GL window is set up, which just
results in a warning. Another call will properly set the
gamma table after the window is initialized.
Make gammaTable a stack variable. This is sufficient, since SDL
already restores the gamma values for us on exit.
2012-01-19 00:53:13 +01:00
dhewg
0e3500b919
Accept both alt keys for alt-enter
...
Reported by Ensiform.
2012-01-19 00:22:47 +01:00
dhewg
3e066baf91
Map the right alt key
...
SDLK_RALT was missing, it is independent of SDLK_MODE.
Reported by Ensiform.
2012-01-19 00:19:51 +01:00
dhewg
6d6c5c8ce7
Fix "unreachable code" warnings
2012-01-15 14:13:44 +01:00
dhewg
33bb970596
README: Move the OSX note
...
It is less confusing.
2012-01-14 22:36:11 +01:00
dhewg
08c758984f
README: Use markdown and UTF8 encoding
2012-01-14 21:39:05 +01:00
dhewg
794a01603b
README: Rename to README.md
2012-01-14 21:22:16 +01:00
dhewg
2d4c18173d
README: Use MinGW as an example for Win64 builds
...
The tree contains MSVC specific asm which won't compile with
a MSVC 64bit compiler.
2012-01-14 20:50:12 +01:00
dhewg
b23e73dea7
README: Clarify OSX OpenAL situation
2012-01-14 20:48:09 +01:00
dhewg
668802f09a
Cleanup OpenAL includes
...
We use OpenAL Soft on all platforms.
2012-01-14 20:43:50 +01:00
dhewg
c5cc860409
README: Add a CHANGES section
...
Describe what has been done on this tree.
2012-01-14 20:43:48 +01:00
dhewg
7aad5a21d3
README: Describe dhewm 3
2012-01-14 20:43:35 +01:00
dhewg
898f690fc4
This is dhewm 3
...
Rename project, now at https://github.com/dhewm
2012-01-14 20:06:05 +01:00
dhewg
d472149426
Get rid of unused defines in Licensee.h
2012-01-14 17:59:37 +01:00
dhewg
db4aca78d6
Get rid of CPU_EASYARGS
...
Never use the event callbacks with mismatching prototypes.
2012-01-14 15:19:38 +01:00
dhewg
b7ce0ef149
CMake: Use config.h for BUILD_OS and BUILD_CPU
...
This fixes, among others, FreeBSD 64bit builds where the build
system used a "amd64" suffix and runtime "x86_64".
2012-01-14 15:19:12 +01:00
dhewg
827124431e
CMake: Detect libcurl and use it when found
...
Move ID_ENABLE_CURL to config.h
2012-01-14 15:09:29 +01:00
dhewg
545d98fc08
CMake: Create config.h
...
Use config.h for configure time settings.
2012-01-14 15:09:27 +01:00
dhewg
9108c4ec9c
CMake: Use CMAKE_SOURCE_DIR
...
Instead of CMAKE_CURRENT_SOURCE_DIR.
2012-01-14 15:02:26 +01:00
dhewg
b5c66c088c
Get rid of ID_OPENAL
...
This is always required for clients.
2012-01-14 15:02:26 +01:00
dhewg
dca4721c8f
Get rid of the scons build system
...
This only works for linux and cmake works on all platforms.
Hopefully this prevents invalid reports.
2012-01-14 15:01:19 +01:00
dhewg
1a5499e6f5
Fix compilation of the dedicated server
...
Add missing OpenAL stubs.
2012-01-14 13:26:01 +01:00
dhewg
2b9e2e5193
More fixes regarding event saving and restoring
...
Fix silly issue from 3d692b58
(sizeof(E_EVENT_SIZEOF_VEC)).
Align entity values of events to the native pointer size, which
is noticeably faster on x86_64 and fixes another assert() in
debug builds.
Past savegames with entities in events are not compatible.
2012-01-13 01:45:19 +01:00
dhewg
76ecb1c863
README: Remove iodoom3 note
...
This reverts commit ee4420557d
.
2012-01-12 22:07:02 +01:00
dhewg
3d692b58de
Prevent assert() while saving and restoring events
...
The argument size of each event is checked in debug builds
and this change was missing from the x86_64 commits.
Surprisingly this didn't trigger yet with the original game,
only with the mod 'Classic Doom 3'.
2012-01-12 19:05:52 +01:00
dhewg
91bca094c0
Fix out of bounds access for buggy UIs
...
Seen with the main menu on the mod 'Classic Doom 3'.
2012-01-12 19:05:51 +01:00
dhewg
3e07208cae
New CVar s_device to choose the audio device to use
2012-01-11 22:52:05 +01:00
dhewg
35bd70ea46
Improve environment suit sfx
...
Set AL_LOWPASS_GAIN like EAX did under the hood.
Set the same filter via AL_DIRECT_FILTER on sources.
This should bring the overall sfx closer to EAX:
<kaan_> it sounds awesome
All infos provided by KittyCat from #openal.
2012-01-11 19:29:35 +01:00
dhewg
ad48afdbd7
Fix compilation via scons
...
SConscript.core was missing jpeg_memory_src.cpp.
Reported by ameyp.
Fixes #2 .
2012-01-11 18:56:38 +01:00
dhewg
16958d070b
Fix [record|play|time]Demo for x86_64
...
These are still fragile due to restored stale pointers (!).
But at least the vanilla demos can now be used to check for
performance regressions.
2012-01-10 22:54:01 +01:00
dhewg
d8acc43aca
Port the D3XP environment suit effect to EFX
...
Use a AL_LOWPASS_GAINHF filter to muffle sfx.
2012-01-10 17:45:37 +01:00
dhewg
8ea0f8f0d5
Clamp EFX reverb effect values
...
The values might be slightly out of range due to the scale
conversion. Clamp them so OpenAL doesn't refuse them.
2012-01-10 17:45:37 +01:00
dhewg
bdbee3543c
Cleanup EFX code
...
Add error checking.
Add debug printf()s, enable via EFX_VERBOSE.
2012-01-10 17:45:37 +01:00
dhewg
b8f9829255
Use the correct unit when passing values for EFX
...
The gain values are expressed as millibels in the .efx files
and need to be converted to normalized floating points.
Get the EAX4 EAXREVERBFLAGS_DECAYHFLIMIT from the "flags" token
and set AL_EAXREVERB_DECAY_HFLIMIT accordingly.
All infos provided by KittyCat from #openal, many thanks!
2012-01-09 00:52:30 +01:00
dhewg
ece2adbdcd
Port EAX to EFX
...
First attempt at porting the EAX reverb code to EFX.
This only works when the ALC_EXT_EFX extension is supported by
the OpenAL vendor (which is not the case for the OSX supplied
framework, use OpenAL soft instead).
The current stable version of OpenAL Soft (v1.13 as of this
writing) can barely handle this additional workload, current
master is highly recommended when using this feature.
2012-01-08 22:04:56 +01:00
dhewg
1ce658e835
Get rid of memory status functions
...
Unused.
2012-01-08 12:24:58 +01:00
dhewg
fe7b32d47e
Get rid of dead prototype
2012-01-08 12:24:58 +01:00
dhewg
6a1d6cb842
Always allow multiple instances
2012-01-08 12:24:58 +01:00
dhewg
438685e6e6
Get rid of Sys_FlushCacheMemory()
...
Stub on all platforms.
2012-01-08 12:24:58 +01:00
dhewg
9542663097
Get rid of Sys_DefaultCDPath()
...
Stub on all platforms.
2012-01-08 12:24:58 +01:00
dhewg
79fa35d0ec
Get rid of ID_MCHECK
...
mcheck() is not thread safe, use MALLOC_CHECK_ or TCMalloc
instead.
2012-01-08 12:24:57 +01:00
dhewg
5f8712902c
Return at least 64MB VRAM in case the detection fails
2012-01-08 12:24:57 +01:00