This was done after the players had already been created. To ensure that everything gets set properly it is necessary to pass the desired sample rate to the device's constructor and let it make sure that a proper sample rate gets set.
- implemented a fallback both for sound font lookup and for MIDI device selection so that if some non-working combination is set up, the player will fall back to something that works.
In file included from /usr/include/c++/6/math.h:36:0,
from src/vectors.h:43,
from src/sound/i_soundinternal.h:7,
from src/sound/i_sound.h:39,
from src/sound/i_musicinterns.h:5,
from src/sound/timidity/timidity.cpp:36:
/usr/include/c++/6/cmath:407:11: error: ‘::pow’ has not been declared
VC\include\type_traits(1468): warning C4800: 'int': forcing value to bool 'true' or 'false' (performance warning)
src\sound\musicformats\music_midistream.cpp(832): note: see reference to function template instantiation 'std::function<bool (int)>::function<MIDIStreamer::SetMIDISource::<lambda_...>,int,void>(_Fx)' being compiled
src/sound/timiditypp/reverb.h:467:6: error: declaration of ‘TimidityPlus::lfo TimidityPlus::<anonymous struct>::lfo’ [-fpermissive]
src/sound/timiditypp/reverb.h:61:8: error: changes meaning of ‘lfo’ from ‘struct TimidityPlus::lfo’ [-fpermissive]
This reverts commit 8f7a503561.
# Conflicts:
# src/sound/timiditypp/playmidi.cpp
Something in here wasn't working as intended, and since it needs better synchronization let's redo it entirely.
This may cause problems because the player runs in a different thread than the input code.
Instead the play thread will now copy their values to local variables when it starts generating output.
The text input field for the configs have been removed. Instead it will now present a list of soundfonts that are found in specific locations.
For that it will look in the 'soundfont' directories of the FileSearch.Directories entry of the config.
Acceptable file formats here are SF2 and zipped GUS patch sets. These zipped patch sets need to have a timidity.cfg at the root and refer to all containing data by relative path. References to outside files are not allowed here.
It is still possible to refer to sound fonts elsewhere on the hard drive by manually entering a path at the console - but these won't show in the menu and will get lost if one cycles through the list of available options.
Of the available softsynths, FluidSynth will only list SF", Wildmidi will only list GUS patch sets and Timidity++ and GUS will list both.
Please note that although the GUS synth can read SF2, the output appears to be broken die to some old bug.
This is necessary to write a universal, device independent wave dumper for MIDIs.
With each format inheriting from the main player class it is not possible to create a generic dumper player.
PathExpander has been integrated into the sound font manager and PseudoMidiDevice was only used as the base class for the old Timidity++ device with the external executable.
- allow the GUS to change sound fonts at run time.
- implemented sound font manager support in the GUS synth. This works but also made me realize that the SF2 support of this synth has been rather broken, apparently forever.
- added the missing handler for reading Timidity configs from the lump directory.
- fixed incomplete init of the GUS synth when no soundfont can be located.
- fixed bad initialization of WildMidi sampling frequency. It would default to 11.025 kHz if no valid rate was set. Now it will use the sound device's native frequency.
- final tweak to volume adjustment: Do this upon extraction for better results.
If this is done via the synth's parameters it will affect channel balance.
The given factor of 5 is necessary to roughly match the output of FluidSynth.
This source has been heavily edited to remove the deep integration with the provided UI modules and to eliminate use of global variables and puts everything into a small number of C++ classes.
src\sound\musicformats\music_libsndfile.cpp(119): warning C4267: 'argument': conversion from 'size_t' to 'long', possible loss of data
src\sound\musicformats\music_libsndfile.cpp(139): warning C4267: 'argument': conversion from 'size_t' to 'long', possible loss of data
src\sound\musicformats\music_libsndfile.cpp(145): warning C4267: 'argument': conversion from 'size_t' to 'long', possible loss of data
src\sound\musicformats\music_libsndfile.cpp(167): warning C4800: 'int': forcing value to bool 'true' or 'false' (performance warning)
src\sound\musicformats\music_libsndfile.cpp(178): warning C4267: 'argument': conversion from 'size_t' to 'long', possible loss of data
src\sound\musicformats\music_libsndfile.cpp(240): warning C4267: 'argument': conversion from 'size_t' to 'long', possible loss of data
* do not use the global temp directory. Instead create one in the AppData folder.
* removed lots of unneeded code from tmpfileplus.
* use C++ strings in there.
This prevents Timidity++ from playing externally, so it now always goes through
the sound system. Aside from the issues relating to controlling playback, that
functionality has been busted for who knows how long anyway since Open() always
tries to create a pipe and sound stream regardless of the calculated size.
Rather than building a command line that's going to be manually split into
individual arguments passed to execvp, build the individual arguments directly.
For non-Windows systems, read() may be non-blocking and can return less than
the requested amount if the timidity process hasn't written enough audio yet.
src/gl/scene/gl_sprite.cpp:685:34: warning: '&&' within '||' [-Wlogical-op-parentheses]
src/polyrenderer/scene/poly_sprite.cpp:297:34: warning: '&&' within '||' [-Wlogical-op-parentheses]
src/swrenderer/scene/r_opaque_pass.cpp:975:35: warning: '&&' within '||' [-Wlogical-op-parentheses]
src/sound/mididevices/music_timiditypp_mididevice.cpp:548:30: warning: comparison of integers of different signs: 'int' and 'size_t' (aka 'unsigned long') [-Wsign-compare]
Add code to find the timidity executable, and split the command line into separate arguments by spaces.
For some reason, this doesn't work, although reverting 64e96c5f makes timidity work again.
This helps to avoid flooding of output with the following messages on every MIDI song change:
fluidsynth: error: Unable to open file "/usr/share/sounds/sf2/FluidR3_GS.sf2"
fluidsynth: error: Couldn't load soundfont file
fluidsynth: error: Failed to load SoundFont "/usr/share/sounds/sf2/FluidR3_GS.sf2"
fluidsynth: error: Unable to open file "/usr/share/sounds/sf2/FluidR3_GM.sf2"
fluidsynth: error: Couldn't load soundfont file
fluidsynth: error: Failed to load SoundFont "/usr/share/sounds/sf2/FluidR3_GM.sf2"
Use snd_streambuffersize CVAR to set buffer size in kilobytes, from 16 to 1024
Name of existed CVAR from removed FMOD sound backend is used
Increased default size of stream buffer to 64 kB like it was in FMOD backend
This fixes the following compilation warnings:
src\sound\oalsound.cpp(1588): warning C4244: 'argument': conversion from 'unsigned int' to 'ALfloat', possible loss of data
src\sound\oalsound.cpp(1796): warning C4244: 'argument': conversion from 'unsigned int' to 'ALfloat', possible loss of data
Stopping a source in the middle of playback can causing undesirable "popping"
from a sudden change in amplitude. Setting the source gain to 0 gives OpenAL
the opportunity to ramp the source's output to 0, where it can be safely
stopped without causing a sudden change in amplitude.
This isn't supported with the OpenAL backend, and instead of trying to make a
workaround for it, a better approach would probably be to make a FileReader
implementation that handles URLs.
src/sound/oalsound.cpp:1288: error: non-constant-expression cannot be narrowed from type 'uint32_t' (aka 'unsigned int') to 'ALint' (aka 'int') in initializer list [-Wc++11-narrowing]
- added arrays to the config to hold entries for the softsynths' config files. This is not active yet, but will later be used to give the user a list of config options instead of having to type it by hand.
The needed headers are now included in the repo, which for these libraries is possible thanks to a stable ABI (at least on Windows, the other platforms still need to be checked but the headers only add, never remove or change existing content.)
The big advantage of this setup is that it allows building the project on Windows without any necessary setup - all that needs to be provided is the DLLs from the binary package.
This still requires some fixes for macOS and Linux. On MacOS the proper library names are missing and the ones for Linux are not verified. Both platforms should work, though, if the dynamic loading is disabled.
src/sound/oplsynth/musicblock.cpp:3:10: fatal error: 'muslib.h' file not found
src/sound/oplsynth/oplio.cpp:410:12: error: use of undeclared identifier 'cos'
src/sound/oplsynth/oplio.cpp:410:41: error: use of undeclared identifier 'sin'
This is somewhat brute-force thanks to the surprising lack of good documentation for the Ogg headers. The only other option would have been some rather bloated library for a function that should be 25-30 lines at most.
The idea is to have more control on the game side instead of dealing with these formats in the backend, which was done for FMod because it already had the decoders implemented.
However, with OpenAL this setup makes no sense and only complicates future extensions that can be better handled at a higher level.
(Is there anyway to tone down GCC's warning level? It outputs too many false positives for potentially uninitialized variables in which the genuine errors get drowned.)
- enable precompiled headers for all non-system-specific MIDI devices.
- moved the native Windows and Mac MIDI devices into their respective sections in the project file so that they won't get compiled on the other ones.
This one was particularly nasty because Windows also defines a DWORD, but in Windows it is an unsigned long, not an unsigned int so changing types caused type conflicts and not all could be removed.
Those referring to the Windows type have to be kept, fortunately they are mostly in the Win32 directory, with a handful of exceptions elsewhere.
A major part of this device's implementation details about how to handle the callback were not encapsulated by the device class at all, they were #ifdef'd into the streamer class.
This puts everything into the device class which now exposes a clean interface to the rest of the game with no special handling aside from calling two additional virtual functions that are empty for the other devices
* make the critical section local to the respective platform instead of polluting everything with system specific symbols.
* moved system specific class declarations into the source file instead of having them in the global header.
This commit temporarily disables the Windows system device because it cannot be done without polluting the global header and still needs a bit of refactoring.
The approach being used here caused the entire sound system to be infested by windows.h, just to avoid copying around a handful of variables in one place, effectively preventing any compiling optimization.
Windows will now use the same internally defined structure for all MIDI processing which only for actual submission to the system player will be converted to the internal format.
According to MinGW headers (which is used for compiling libmpg123) ssize_t type must be 64-bit in size on 64-bit Windows
Moreover it was impossible to build GZDoom with the recent versions of libmpg123 because of wrong type redifinition
Tested MP3 music on Deus Vult II with Release x64 and OpenAL backend
libmpg123 spews quite a lot of debug stuff in stdout when encountering files like WAV or Ogg Vorbis, while libsndfile is silent when encountering an MP3 file.
- Replaced GTK/OS X (note different from Cocoa) clipboard code with SDL clipboard API.
- Removed requirement to link to GTK in order to compile with GTK support.
- GTK is no longer init'd if the GTK IWAD picker is not used.
- Our usage of GTK is such that the dynamic loader can work with both GTK2 and GTK3 depending on what's installed.
- Since we're accumulating a lot of library loaders I've built a generic interface as FModule which replaces TOptWin32Proc and the loaders in the OpenAL and Fluidsynth code.
* everything related to scripting is now placed in a subdirectory 'scripting', which itself is separated into DECORATE, ZSCRIPT, the VM and code generation.
* a few items have been moved to different headers so that the DECORATE parser definitions can mostly be kept local. The only exception at the moment is the flags interface on which 3 source files depend.